SCINDIT SESE its due force, and set aside for the moment the deceptive adjunct REDUCTOS, in order to perceive that in the words QUIBUS OMNIS AB ALTO FRANGITUR INQUE SINUS SCINDIT SESE UNDA, Virgil must speak, not of the reflux of the wave or sea, or of the form in which the wave or sea recedes from the shore, but of the advance of the sea forwards between the prominences of the island; for how, except by its flowing up between those prominences, is it possible that it should divide itself, or be divided by them: FRANGITUR INQUE

SINUS SCINDIT SESE. Compare OviD, Metamorph. XV. 739, where, speaking of the insula Tiberina, he says:

"Scinditur in geminas partes circumfluus amnis,

Insula nomen habet, laterumque a parte duorum,
Porrigit æquales media tellure lacertos;"

the sole difference between which view and that given by Virgil is, that here the water is described as divided by the whole island, and into two parts only, while in Virgil's view it is described as divided, not by the whole island, but by its several projections or promontories, and therefore into several parts or sinuses. Compare also OVID, Metam. XIV. 51:

"Parvus erat gurges curvos sinuatus in arcus;" where the idea is the same as that in the text, except that Virgil's sinuses are sharply re-entrant, while Ovid's are gently curved. This interpretation, long ago proposed by Turnebus, and adopted by Burmann, but forgotten, it would seem, by modern commentators, is so far from being contradicted or invalidated as to be even confirmed by REDUCTOS, which, (first), is not a participle, but an adjective, corresponding exactly to odoratam (En. VII. 13), inaccessos (En. VII. 11), and numerous other adjectives with participial terminations; nay, is so much an adjective, as to be capable of comparison: “ut qui singulis pinxerunt coloribus, alia tamen eminentiora, alia reductiora fecerunt" (QUINCTIL. Instit. XI. III. 46); and (secondly) means, as clearly shown by

the passage just quoted from Quinetilian, standing backward, or in the back-ground, in comparison of something which is more prominent; precisely the ideal which the mathematicians express by the term re-entrant. So reducta valle (En. VI. 703), is not a deep or long valley, but a valley standing back or re-entrant from the plain; i. e. extending backward from the plain toward the interior between two ranges of hills; not a sunk valley, or one upon which you look down, but one on a level with, and an offset from, the plain, and into which you look from one end. And so also, in the passage before us, the sinuses into which the edge of the sea is divided by the prominences of the island are reducti, re-entrant between those prominences, offsets of the sea; or, as expressed by Livy (Lib. XXVI.) in his description of the port of Carthago Hispanica: introrsum retracti. Compare Mela, HII. 1. "Frons illa aliquamdiu rectam ripam habet; dein modico flexu accepto, mox paullulum eminet; tum reducta iterum, iterumque rectâ margine jacens, ad promontorium quod Celticum vocamus, extenditur." Having differed so widely from the above-quoted commentators (and I am not ashamed to add even from my own earlier opinion, expressed in the Classical Museum No. 19, and quoted by Forbiger in his 3rd Edition) in my interpretation of each of the three words, SINUS, SCINDIT, and REDUCTOs, I am inclined to differ from them, besides, in the interpretation of the word UNDA, which I understand to mean here, not fluctus, or a great wave or billow rolling in from the deep, and breaking violently on the island, but the sea, or, if I may so say, the undulant itself; a sense in which the term is so frequently used, not only by Virgil (ex. gr. Georg. I. 360, III. 340. &c.), but by all other Latin writers. So understood, UNDA seems to me to harmonize better (a) with the present quietude of the sea after the miraculous stilling of the storm, and (b) with the words, SCINDIT SESE IN SINUS REDUCTOS, the

re-entrant sinuses being less properly constituent parts of individual waves than of the sea itself. Nor let it be said that FRANGITUR contradicts this idea, and points to billows breaking with great force, for we find the selfsame term used to express the common breaking of the sea upon the shore in calm weather, in the words: "Qua vada non spirant, nec fracta remurmurat unda."

En. X. 291.

SINUS therefore, in the passage before us, is applied to the sea in the identical sense in which it is applied to it, not only by Virgil himself elsewhere, and other Latin writers, but in the familiar proper names, Sinus Adriaticus, Sinus Tarentinus, Sinus Saronicus, &c.; a sense, it will be observed, directly opposite to that in which it is applied to the female breast, the sails of a ship, or the dress; the term in these latter applications preserving its original meaning of a concavity, hollow, or depression, while in its application to the sea it means a projection corresponding to, and accurately filling up, an opposite concavity or hollow. This remarkable deviation from, or exception to, the original and still general meaning of the word as applied to other objects, has, no doubt, arisen (as in the case of our own bay) from its having been found convenient in practice to extend the application of a term, which originally and in strictness signified only a hollow or sinuosity of the shore, to the arm of the sea filling it up. Compare Vossius's definition of the word in his Etymol. "Sinus de mari dicitur metaphorice, quia ut in homine sinus est quod brachiis comprehenditur, ita et in mari sinus est maris pars quasi brachiis terræ interjecta. Græcis est xohлоç; unde Itali 'golvo' dicunt pro κολπω.” A similar interpretation will, I think, be found to answer for Georg. IV. 420, where the same words occur again, and where the meaning is: a mountainous promontory runs into the sea, presenting on the exposed side a number of inlets, into which

the sea beats, and on the sheltered side the cave of Proteus, and a safe roadstead for ships. Voss and La Cerda understand REDUCTOS SINUS of the two inlets or arms by which the sea communicates round the island with the port behind: an interpretation to which there seems to me to be these two great objections: first, that it is wholly inapplicable to the words where they occur again in the fourth Georgic; and secondly, that we cannot doubt that, if such had been his meaning, Virgil would (like Ovid in his description of the Insula Tiberina, above quoted) have added either geminos or duos, to indicate that he spoke of two particular inlets, and not of an indefinite number. The mystification under which Caro and Dryden endeavor to conceal their ignorance of their author's meaning amounts almost to

nonsense:

"Questa si sporge co' suoi fianchi in guisa,
Ch' ogni vento, ogni flutto, d'ogni lato
Che vi percuota, ritrovando intoppo
O si frange, o si sparte, o si riversa."

"Broke by the jutting land on either side,
In double streams the briny waters glide."

CARO.

DRYDEN.

GEMINIQUE MINANTUR IN CELUM SCOPULI. "Tam alti sunt ut videantur tendere in cœlum: minas murorum, infra IV. 88, muros præaltos dixit." WAGNER, Virg. Br. En. "Minantur (ire or ascensum) in cœlum: the expression is most poetically beautiful."-TRAPP.

"Rise on each side huge rocks, two o'er the rest
Menace the skies."

BERESFORD.

"Velut respiciat ad gigantum conatus cœlum oppugnantium." GESNER. This is not the meaning: first, because it is always directly, and not through the medium of a preposition, that minari governs the object threatened: compare the numerous examples of the use of

this word adduced by the lexicographers; and (especially in point, though not adduced by them) Silius Italicus's "Saxa minantia cœlo" (IV. 2); and Propertius's "Coloque minantem Cœum" (III. IX. 47); and secondly, because to have described the SCOPULI as threatening the sky had been to introduce an idea foreign from the subject, and distractive of the reader's attention from the main object, the security and privacy of the harbour, to the danger of the sky. I therefore understand MINANTUR in our text to be taken absolutely, i. e. irrespectively of an object, and to mean, rise with a bold, towering, or, if the reader prefer it, threatening aspect. Compare, first, En. VIII. 668, where we have precisely the same predication applied to the identical word scopulus: "Et te, Catalina, minaci

Pendentem scopulo;"

where the meaning can be no other than a threatening-looking, or, as we say, bold, towering cliff. Compare, secondly, En. II. 628:

Et tremefacta comam concusso vertice nutat;"

where the meaning is not (with Gesner and Dryden) minatur casum, but the very opposite: stands boldly; resisting, not yielding to, the attack; as proved by the words, usque and donec; still preserves its bold, towering, fearless attitude, until &c. in confirmation of which interpretation observe that the word nutat, added here by way of explanation, means where it is again similarly employed by Virgil, viz. En. IX. 682, nod in a menacing manner. Compare, thirdly, En. IV. 88: "Pendent opera interrupta, minæque

Murorum ingentes, æquataque machina cœlo;"

not (with Servius) eminentiæ murorum, quas pinnas dicunt, but, the threats of the walls, i. e. the high, towering, threatening-looking walls themselves. And here observe the complementary clause: æquataque machina