[24] The reader will find this question discussed in Wagner's Aulularia; where references are given to the original German authorities.
[25] Dactylic poetry is not here included, as its progress is somewhat different. In this metre we observe: (1) That when a dactyl or spondee ends a word, the natural and metrical accents coincide; e.g.—ómnia, súnt mihi, prorúmpunt. Hence the fondness for such easy and natural endings as claudúntur lúmina nócte, common in all writers down to Manilius. (2) That the caesura is opposed to the accent, e.g.—árma virúmque cáno | Troiae | qui. These anti-accentual rhythms are continually found in Virgil, Ovid, &c. from a fondness for caesura, where the older writers have qui Troiae, and the like. (3) That it would be possible to avoid any collision between ictus and accent, e.g.—scílícet ómnibus ést labor ímpendéndus et ómnes: inveteráscit et aégro in corde senescit, &c. But the rarity of such lines after Lucretius shows that they do not conform to the genius of the language. The correspondence thus lost by improved caesura is partially re-established by more careful elision. Elision is used by Virgil to make the verse run smoothly without violating the natural pronunciation of the words; e.g.—mónstrum horréndum infórme; but this is only in the Aeneid. Such simple means of gaining this end as the Lucretian sive volúptas est, immortáli súnt, are altogether avoided by him. On the whole, however, among the Dactylic poets, from Ennius to Juvenal, the balance between natural and metrical accent remained unchanged.
[26] Most of the verses extant in this metre will be found in Wordsworth's Fragments and Specimens of Early Latin.
[27] A good essay on this subject is to be found in Wordsworth's Fragments p. 580, sqq.
CHAPTER III.
[1] Scipio quoted Homer when he saw the flames of Carthage rising. He is described as having been profoundly moved. And according to one report Caesar's last words, when he saw Brutus among his assassins, were kahi se teknon.
[2] The reader will find them all in Wordsworth.
[3] Brut. xviii. 71, non digna sunt quae iterum legantur.
[4] Ep. ii. 1, 69.
[5] Liv. vii. 2.