in Book 9 was written before the mention of the two friends in Book 5, on account of the great fulness of the description in the later book, though it is of course possible. Drances is more fully described on his last appearance in Book 11 than on his first introduction earlier in the same book; and in fact such varieties are indispensable to the freedom of poetical narrative. As usual, M. Ribbeck points out various lines in Book 5 which he regards as stop-gaps or interpolations; and as usual, I am unable to agree with him. In v. 120 Virgil is only following his ordinary custom of reiterating a thing in a new form: v. 403 contains one of his most characteristic verbal manipulations: v. 467 would never have been suspected by any one who was not, like Peerlkamp, suspicious by profession: the repetition of v. 538 in v. 572 may be a carelessness, but it may be a deliberate imitation of Homeric commonplace: v. 802 is an amplification, but a sonorous and effective one: in v. 440 the picture of the preceding line is given more Vergiliano, with a circumstantial difference: if v. 603 savours of the commentator rather than of the poet, a large number of similar lines in the Iliad and Odyssey must be due not to Homer but to the Scholiasts. V. 506 has nothing to do with the dove's clapping her wings, but is a translation, as I have shown in my commentary, of Iliad 23. 869. V. 290 is not free from difficulty: but there is no reason why "consessu" should not mean "in consessum," the place of assembling being spoken of as the assembly, and though there may be no authority for making "exstructo a substantive, it can be amply supported by analogies like "suggestum." I cannot think that vv. 82, 83 make an inappropriate close to Aeneas' brief address. "Non licuit" is to be understood as in Book 4, v. 550, where it is similarly introduced at the end of a speech, not as a mere statement of fact, but as a passionate ejaculation. Lastly, I do not see that we need seriously arraign either the poet or a copyist for "ab Ida," v. 254, closely following "frondosa Ida" in v. 252. If anything, it is Virgil's own carelessness, though a slight one: but perhaps, as I have already suggested, the repetition may be intended to show that the two actions, Ganymede hunting and Ganymede carried off, are represented as taking place on the same spot.

د"

In the Ninth Book M. Ribbeck finds fault with vv. 367 foll. for their obscurity. They are not quite plain, certainly: but the difficulty does not seem to arise from the poet's carelessness, but from his habit of indirect narrative. Ladewig seems substantially right in supposing that a legion was on its way from Laurentum, in answer to a request from Turnus, and that the horsemen were sent on before to announce its approach. They were reaching the Rutulian camp just as Nisus and Euryalus were leaving it and so the two parties naturally came into collision. M. Ribbeck's own explanation" immo speculatum ultra Rutulorum castra missi, dum ex urbe reliqui profecti castra Troiana oppugnant, equites illi iam redibant," I do not clearly understand: at any rate it appears to ignore the obvious opposition between "praemissi" and "moratur." With the passage generally we may compare Book 11, vv. 511 foll. Whether the mention of Numa v. 454 and that of Asilas v. 571 are carelessnesses or pieces of indirect information is not easy to say, and matters but little. There is perhaps some awkwardness in the omission of Turnus' name v. 749 but Virgil probably did it intentionally, meaning to be rapid and vivid. V. 403, as read in the oldest MSS., can scarcely be right: so there, as in a few other places, we must suppose that the later copies have contrived to preserve the true reading. Vv. 85 foll. can hardly be explained as a dittographia," as neither the first line nor the two last could well stand alone: it is better to suppose that Virgil means to tell us indirectly that besides pines, there were pitch-trees and maples in the grove, while we admit that there is a clumsiness in "lucus " following “silva" and a strange ambiguity in "arce summa." The gifts promised to Nisus and Euryalus vv. 263 foll. certainly do seem extravagant in some respects, and the mention of a single bowl where other things are double is awkward: there accordingly we may say that Virgil's later

thoughts would perhaps have corrected his earlier. I am glad to see that M. Ribbeck does not quarrel, as some tasteless critics have done, with the striking and pathetic passage vv. 314 foll.: he is far, however, from seeing its full beauty. The explanation "tamen" does not lie in the word "inimica," though that word was no doubt chosen intentionally, to strike a note of melancholy, but in the suppressed thought, “perituri quidem ipsi." I know nothing more touching in Virgil than the manner in which he has chosen to indicate what he will not mention, nothing more disheartening in criticism than the blindness of the commentators to this wonderful stroke of art. The "loci Albani" (v. 387) are a puzzle: but a puzzle does not necessarily imply a corruption or a carelessness. Vv. 151 and 363 can be explained, though the last is awkward, so that it is rash to pronounce them interpolations: v. 777 is necessary for the full close of the paragraph, and to obliterate it shows a want of feeling for rhythm. Vv. 146, 147 are somewhat inconsistent with the end of Turnus' speech: but he may well be supposed to change his mind, and after proposing an attack, decide on deferring it to the morrow. They would come in awkwardly where M. Ribbeck places them, after v. 72: indeed, if they had appeared in the MSS. anywhere in that neighbourhood, the critic would probably have condemned them as a "dittographia" of v. 51.

The Seventh Book, according to M. Ribbeck, is especially faulty in the latter part, the account of the origin of the war and the catalogue, though the earlier part also shows signs of imperfection. V. 444 he thinks may have been filled up by an interpolator, not improbably, but Heumann had been before him in the supposition: in suspecting the latter part of v. 571 he follows Heyne, but the notion is less plausible. He next points out some lacunae, mostly on insufficient grounds: one after v. 242, on account of the change of nominative in v. 243; a second after v. 543, on account of the difficulty in "convexa ;" a third after v. 663, where it must be confessed there is some abruptness; a fourth after v. 695, where the verb can easily be supplied by a zeugma; while in v. 535 the same craving for a verb after "seniorque Galacsus" leads him to the same suspicion of incompleteness. Then come "dittographiae,” vv. 75—77 for v. 74, a needless supposition, and vv. 624-627 for vv. 638-640, an injurious one, the lines in question being first dislocated and then condemned. Other dislocations are the transposition of vv. 395, 396, where a natural variety is sacrificed to an ill-advised endeavour after regularity, and that of vv. 698-702 and vv. 703-705, of which the same may be said. The two "vanissimi versus "vv. 146, 147 are attributed to an interpolator, because after telling us that food had run short, Virgil would not have talked of "instaurant epulas," while there is no evidence that the wine had ever been removed, as is implied in v. 134. But Virgil plainly means that having come to the end of their meal, they renew it in honour of the discovery of their new home, not necessarily by eating more, but by drinking, and especially by libation. To M. Ribbeck's "Otiosi praeterea videntur vv. 386 et 575," I can only reply, "Ribbeckio fortasse et Peerlkampio, sed non mihi." It is satisfactory, however, that M. Ribbeck does not follow Peerlkamp in condemning the whole exordium of the book, vv. 1—35, in which he thinks him excessive, "nimius." I should like to see the Peerlkamp who I could have written the lines.

The Eighth Book M. Ribbeck thinks unusually finished and accurate. He approves of Heyne's notion of getting rid of the hemistich in v. 41, by combining it with the latter part of v. 49 and striking out the intervening lines; an alteration which might appear tempting to one who, like Heyne, would be troubled by the inconsistency with Book 3, but need not attract persons who, like M. Ribbeck and myself, accept Conrads' theory expounded above. Peerlkamp, M. Ribbeck considers, has "demonstrated" that the latter part of v. 13 and the whole of v. 14 were not written by Virgil; the fact being that the language used is that of intentional exaggeration, such as would be consciously or unconsciously employed by the Italian princes and their emissaries. M.

Ribbeck puts v. 654 after v. 641, following the example of the Parma edition; but if he had sufficiently realized the fact that Virgil is describing not the historical scenes as we may conceive them to have taken place, but as they would have been represented on the shield, he would scarcely have quarrelled with the old order. V. 3, according to him, is not only superfluous but perverse, as the stirring up of steeds and armour ought to follow the moving of men's minds, not to precede it. But Virgil meant to represent Turnus' fiery spirit as kindling the spirits of others, and so he represents him like a roused war-god, shaking his bridle-rein and smiting on his shield, and thus exciting the Italian tribes. We may compare the description in the simile Book 12, vv. 331 foll. V. 149 is condemned with Peerlkamp, to the injury of the rhythmical effect of the passage. Vv. 283, 284 are rejected after a suspicion of Heyne's; a second instance of M. Ribbeck's antipathy to the notion of a renewed banquet. V. 601 is called in question, because a Latin poet would hardly have thought it necessary to specify that Silvanus was god of agriculture and cattle. The specification would be pardonable as a mere piece of poetical surplusage like "Mars armipotens :" it is laudable when we consider that Virgil, though a Latin poet, is identifying himself with the Trojan new-comers, and, as it were, explaining Italian customs for their benefit. Lastly, while rejecting Peerlkamp's view that vv. 666-670 are a grammarian's addition, M. Ribbeck censures the passage as inappropriate. I might myself be inclined to question it if I were certain that I thoroughly comprehended Virgil's conception of the shield as it is, I remember the warning that one ought to be sure that one understands a writer's ignorance before one professes oneself ignorant of his understanding. If my readers are as tired as I am of discussing similar criticisms in similar words, they will be glad to hear that in this Book at any rate there are no supposed cases of "dittographia."

:

In the Tenth Book M. Ribbeck suspects the latter half of v. 27, "nec non exercitus alter," of being an interpolation. A stop-gap it may be, as it does not seem particularly forcible but there is no reason to doubt that Virgil wrote it. Servius indeed does not explain it but why should he? V. 20 is objected to because we have not previously heard of Turnus as riding in a chariot: but he may have appeared in a chariot nevertheless, as M. Ribbeck admits he does later in this book, v. 440. The mention of Capua, v. 145, is thought frigid: why so, more than the mention of the competitors in the ship-race in Book 5, as founders of Roman families? Virgil has named Mnestheus and other Trojan worthies with some honourable addition, and he naturally does the same in the case of Capys. Vv. 109, 110 are complained of, perhaps because they are not understood. Jupiter declines to entertain the question whether the advantage gained by the Italians in investing the Trojan camp is due to their own favouring destiny on the one hand, or on the other to the mistake of the besieged in allowing their leader to leave them and to the malignant warning conveyed to Turnus by Juno. Perhaps there is something inconsistent in condemning the Trojans implicitly because Aeneas, following a divine intimation, left them to go to Evander: but that is no ground for doubting the integrity of the text. In v. 475 Pallas, like a Homeric warrior, having thrown his spear, prepares to come to close quarters with his sword: but he has no opportunity of doing so. We are not obliged to suppose that everything is related in the precise order in which it occurs in v. 474 the spear is thrown in vv. 476 foll. we follow its course: the drawing of the sword doubtless took place while Turnus was levelling his own spear. The simile in vv. 804 foll. may be a little overloaded; but this is a Homeric fault, and makes us realize the picture more completely. From some of M. Ribbeck's other remarks I do not greatly differ. The Arcadian cavalry, whom we left in the Eighth Book, meet us rather unexpectedly in vv. 238 foll. of the present. Turnus in v. 285 is said to encounter the Trojans as they land with his whole force: we hear nothing of those who were to continue (v. 285) the

:

blockade of the camp. The enlargement of Ascanius and his followers is noticed v. 604 with strange and inconvenient brevity. Altogether, the conclusion, "Ergo non satis diligenter ac plene haec relata sunt" does not seem an unwarrantable one. But I cannot agree that vv. 270-275 are out of place where they stand. We are meant to see Aeneas as the Rutulians saw him, as the fleet came nearer and nearer, a glorious and terrible presence, like that of a comet or of Sirius. For the time we think of Aeneas and of him only, so that we do not need to be told in v. 270, whose head is a blaze of light. He is described, as he appears to the enemy, just as in the parallel passage in the Twenty-second Iliad, Achilles is described as he appears to Priam. Then, when we have looked at him sufficiently with their eyes, we are told that what appals them does not appal their leader, and our sympathy reverts to Turnus in consequence.

Like the rest, the Eleventh Book is declared to contain instances of incompleteness, interpolation, and dislocation. V.87 is said to be incomplete, the poet having probably intended to insert the complaints of Acoetes; a supposition required neither by literary propriety, which would rather reject the notion of a third lament, in addition to those of Aeneas and Evander, nor by grammar, which is amply satisfied by making “sternitur” the verb of the sentence, "keeps throwing himself on the ground." The only reason for suspecting v. 822 is the use of the infinitive of habit in a relative sentence, which is an arbitrary objection enough, when we consider that the historical infinitive is found after "cum" (see Madvig's Lat. Gr. § 392); while the integrity of the passage is strongly supported by the parallel Book 4, vv. 421 foll. Vv. 537-584 would certainly be a monstrous parenthesis, if they were a parenthesis at all; but something more conclusive than the dogmatic "errat magnopere Servius cum ceteris" must be urged before a judicious editor will abandon the milder alternative of making them part of Diana's speech. The proposed transposition of vv. 264, 265 after v. 268 destroys the effect of "Ipse Mycenaeus," &c., coming at the end of a series of enumeration; and there is more force in making "invidisse deos" a sorrowful exclamation, like "non licuit" Book 4, v. 550 spoken of above (I am adducing of course a rhetorical, not a grammatical parallel), than in constructing it with "quid referam," as if Diomede were likely to dismiss his own misfortune among the et ceteras. I need hardly discuss the question of transposing vv. 469-472, as M. Ribbeck is not quite certain whether they ought to go after v. 476 or v. 485, and finally has recourse to his usual panacea "nimirum ne hunc quidem locum satis absolverat poeta." I may say, however, that the lines appear to me perfectly in place; there is a general rush to the walls; even Latinus sees that he must break up the council; and the preparations for defence immediately begin. The lament of Evander is pronounced too garrulous, and various things in it are excepted against. "The Trojans ought not to be called Phryges' (v. 170), except in contempt;" a sweeping statement, to which Book 1, v. 468, is a sufficient answer. "It is vain to say that if Pallas had been as old and as strong as Turnus he would have killed him (vv. 173 foll.);" why is it no praise to say that of two well-matched warriors one would have prevailed? "Vv. 179-181 (Meritis-imos ') are redundant and feeble;" they are rather difficult, but I should call them forcible and appropriate. M. Ribbeck once thought v. 80 a “ dittographia;" he now thinks it spurious. I do not see why it should be either, though it is not particularly striking. Vv. 523 foll. he suspects to be a repetition of Book 7, vv. 565 foll., as if, because there is a gorge (or rather, as appears to be the fact, a pond under a hill) in one spot, there could not be a wooded defile in another. Lastly, v. 607 is condemned as harsh in itself, "ardescit" being not even suited to "fremitus," much less to "adventus," and as absolutely needless after vv. 597 foll. I should myself have said that "ardescit " was a word which none but a poet like Virgil could have used, suggesting the comparison of an approaching flame with its heat and glare, while at the same time we are made to think of the actual glow of the rapid advance and the warm breath of the horses. But tastes differ.

In the Twelfth Book three instances of incompleteness are noted, v. 218, vv. 732, 733, and vv. 889-893. In the first the language is confessedly harsh and obscure, and until some parallel shall be produced, we need not hesitate to admit that Virgil has expressed himself carelessly. The connexion between vv. 732, 733, is like that of which I spoke in the Ninth Book: the poet passes from the regular narrative to the unexpressed thoughts of Turnus, who feels that he is undone "ni fuga subsidio subeat." So far from being incomplete in a poetical sense, the passage is highly finished. The third passage would hardly have been excepted against by any one not possessed by an unseasonable spirit of logical precision. Aeneas first tells Turnus plainly that he can fly no longer but must stand and fight, and then tauntingly bids him to transform himself as he pleases, soar into the air or dive into the depth. Vv. 879-881 and 882-884 M. Ribbeck apparently thinks a "dittographia:" but Juturna may be allowed a little amplification in her parting lament. The speech of Latinus on ratifying the treaty is complained of; vv. 203, 205 are thought too exaggerated for Virgil, and interfere, it is urged, with the construction of the following lines: the poet however has followed Homer, who introduces the appeal to the sceptre with still less attention to regularity, though in each case the reader receives the impression intended, that of physical impossibility that the thing spoken of or hinted at should take place. 'Haud nescia rerum (v. 227) is not an idle supplement, but an epic mannerism. Why we are to suspect vv. 439 foll. "et te animo-Hector" because we know that interpolators were in the habit of filling up imperfect lines does not appear. I need hardly defend vv. 563, 564, as M. Ribbeck admits that Wagner has excused them sufficiently. The transposition of vv. 515, 516, adopted from Peerlkamp, is ingenious but unnecessary. To remove vv. 801, 802 from their present place and insert them after v. 831 (which is M. Ribbeck's last proposal) is to remove from Jupiter's first speech the one touch of playfulness which shows that he means to conciliate even where his commands are peremptory. If it is replied, as M. Ribbeck replies in his "emendationes Vergilianae," that Jupiter in his first speech ought to be simply peremptory, and that the time for conciliation is afterwards, when Juno has declared herself willing to submit, we may rejoin by urging the inconsistency of the words "precibusque inflectere nostris " (v. 800) with the tone of unqualified command which M. Ribbeck bids us expect. Here, as in so many other places, the critic is misled by failing to appreciate the free play of feeling, which, both in poetry and in impassioned rhetoric, refuses to be bound by the strict rules of logical sequence.

دو

What more I have to say about M. Ribbeck's Prolegomena relates chiefly to the last section of his thirteenth chapter, that in which he defends the various conjectures which he has introduced into the text. I will notice, however, one or two points in transitu, from the earlier sections of the same chapter.

In speaking of the Verona fragment (p. 275), M. Ribbeck proposes to read "velis" for "ventis" in Aen. 3. 705. "Dare vela" is of course common enough: and perhaps for that very reason Virgil did not use it here. M. Ribbeck is quite right in saying that "datis ventis" is not to be supported from 3. 61, where "dare classibus austros" has a different meaning: but he gives no reason why "datis ventis," in the sense of winds vouchsafed by the gods, is improper or unnatural. Does he forget the common expression "ventis vocatis," which may be said to be almost correlative to "ventis datis," or the many passages in the classics where the gods are said to send favouring winds? In the section on the Vatican fragment he defends his conjecture "num" for "cum” Aen. 9. 513, by attacking the received reading as expressing a state of things which is contradicted by what happens immediately afterwards. But that is precisely Virgil's object he throws himself by turns into the feelings of the two contending parties: the Trojans hurl down stones: the Rutulians (so to say) laugh at them, and declare that under their penthouse they can endure everything with content and even pleasure; in