authority, to illustrate, in some of the most striking instances, the connection existing betwixt the places of accent in Horatian metre, and the legitimate structure of the verse. For a complete and satisfactory view of the subject in all its bearings, the reader must be referred to the pages of the original article.

Before proceeding to the task itself de Metris Horatianis, it may be prudent if not necessary to define the end proposed in that labour. In all the metres I shall hope to make the structure of the verse as well as the scansion understood; while in the most important of them, especially those much in use for exercises of imitative composition, I shall indulge greater latitude of practical detail. Agreeably to that view and purpose, the nineteen denominations in Horace will be treated of thus. Those in the Epodes, (numbered I. to VII.,) which in the just order of publication now stand first, may in general be dispatched with less of ceremonious regard; and those in the Odes, (VIII. to XII.,) hardly lyrical measures, either from difficult construction abandoned after one or two experiments, or from whatever cause not much favoured by Horace, will occupy a briefer notice, but still with sufficient correctness. Thus, a larger consideration may be fairly granted to verses of that species (XIII. to XVII.) which has its character from the Choriambus interposed: and a still more extensive space will thus be allowed to the two principal metres (XVIII. and XIX.) in the Sapphic and the Alcaic stanza.

METRES 1. TO VII.

And first of the Epodes; of which the name has been, in P. D. p. 10, already explained.

METRE I. EPODE I. Iambic trimeter (1) and dimeter (2).

The scansion of the trimeter (1) runs thus, in all its admissible feet, some of them very rarely admitted.

To the exclusion of the anapest in 5to from the table here given (a foot so frequent in the Senarii of Seneca) there are but three instances of apparent objection, which when examined disappear.

Ep. 11. 35.

Pavidumque leporem et advenam lăqueō gruem,

(which is a very rare instance of a line otherwise irregular.) Read laqueo in two syllables, as alveō,

3 C. VII. 28. Tusco denatat alveō.

These two peculiarities,

Ep. v. 79.

- XI. 23.

Priusque cœlum sidet infĕrĭūs mari.

Nunc gloriantis quamlibet muliĕrculam.

turning similarly on the vowels, may safely be placed on the same footing with Virgil's Geo. II. 482.

and as that is pronounced Fluvyorum in three syllables, we need not scruple surely so to pronounce inferyus and mulyerculam likewise.

The structure of the trimeter requires a cæsura, or division of words, after the penthemimer usually.

Ep. 11. 1. Beatus ille qui procul negotiis...

or after the hepthemimer sometimes,

Ibid. 19. Ut gaudet insitiva || decerpens pyra.

And here comes in the curious question of accentual cadence.

The words of the Romans then were accented by an invariable rule (and do we not now follow a similar rule in reading Latin?) namely, that if the penult be long, it shall be accented, as amábam; if it be short, the antepenult shall be accented, whether long or short, as amáveram, potúeram.

All monosyllables were accented; and in these pages, if not so marked, may be always so reckoned. Only, enclitics like que, ve, ne, must be taken as forming part of the preceding word. Thus,

Ep. 11. 45. Claudénsque téxtis....

Now for the application. The Iambic trimeter in its legitimate construction, whatever other accent it has, takes a leading one on the sixth place: or if not so, then it usually requires to be accented on the fourth and the eighth. Thus,

Ep. 1. 3.

Paratus omne Ca'saris periculum.. 7. Utrumne jússi persequémur otium.

The leading accent on the sixth, though generally with penthemimeral cæsura, and at the beginning of a word as in Cæsaris above, may yet be otherwise circumstanced, as, Ep. 1. 15. Roges, tuum labóre quid juvem meo.

and

Ep. 11. 19. Ut gaudet insitíva decerpens pyra.

Nay, more:

Ep. 1. 19. Ut assidens implúmibus pullis avis,

is an Iambic verse, because the sixth is accented: whereas Ut assidens púllis ávis deplumibus,

though exactly the same as to quantities, would not be a verse, on account of the wrong position of the accents. A similar remark may be made on

Ep. XI. 15. Quod si meis inæ ́stuat præcordiis,

as having neither cæsura; and yet it is an Iambic verse from the correctness of its leading accent.

In long words constituted like inverecundus, if not in

others, the Romans may seem to have admitted a secondary accent on the first syllable: a supposition which would restore propriety to the following verse,

Ep. xi. 13. Simul caléntis inverecúndus deus.

It remains to be noticed as a singular fact, that of verses constructed like

Ep. 1. 7. Utrumne jússi | persequémur | otium,

many examples occur in Terence as his Prologues alone may testify, in the Epodes of Horace, and in the Fables of Phædrus; yet in the Tragedies which bear the name of Seneca, not a single instance can be found of a verse constituted like that above.

Iambic dimeter (2).

The scansion of this verse as to its predominant and admissible feet:

excluded, seems inadmissible. Casimir Sarbievius uses it,

Epod. 1. 18. Metuenda jaceret fulmina.

As to the structure, it is worthy of observation, that a verse from its composition accented like the following,

C. Sarb. Ep. v. 2. Non núdus énsium tímor,

though a just dimeter as to quantity, is never found in Horace, nor in any author of the early ages.

Now as the iambus itself is used freely in the third place, this combination might have occurred very often, if it had not been purposely avoided. In what then does the faultiness of that line consist? Clearly in this, that it bears the accentual cadence peculiar to a very different kind of verse, the Glyconic of which verse,

4 C. III. 11. Et spíssæ némorum cómæ.

is a very common form; and it may be better contrasted perhaps with Dr. Herbert's fictitious example,

Et spíssa móntium cóma,

than with the real faulty dimeter given above from Sarbievius.

METRE II. EPODE XI.

Iambic dimeter (1) as in Metre

1. (1), and with Elegiambus (2) so called.

(1) Petti, nihil me sicut antea juvat

(2) Scribere versiculos | amore perculsum gravi.

(2) This verse consists of two parts; the first like the latter half of the Dactylic pentameter

but unlimited in the mode of its construction; with the last syllable indifferently long or short, and even as in vv. 14, 24, with the hiatus of a long vowel before a vowel initial in the next portion of the verse.

Fervidiora mero | arcana promorat loco.

Vincere mollitiâ | amor Lycisci me tenet.

The second part is the Iambic dimeter (but without initial) as in Metre 1. (2).

For the meaning of the term asynartete (arvvágτntos) applied to the metre (2) of this distich, Dr. Bentley's learned note affords full satisfaction: the following extract contains the essential part of it." Sub primis Poetica artis initiis simplici pede versus decurrebant, Heroicus dactylo, Trochaicus et Iambicus uterque suo; nisi ubi pes omnibus illis cognatus, Spondeus, interponebatur, quo versus, ut Noster ait, tardior paullo graviorque ad aures veniret. Postea, ut varietatis gratiam aucuparentur, cola quædam sive partes Heroici versus cum colis Trochaici generis vel Iambici, et vicissim, in unum versum miscebant; unde magnus novorum versuum numerus illico nascebatur: quos Græci magistri 'AovvaρTýτous, hoc est, inconnexos vocabant; quia alterum colon altero