[25] Heinze, 94. Miss Crump analyzes the probable changes of revision. Her theory that Book III survives from an earlier plan in which it stood first, and that Vergil probably intended to revise it entirely, has grave difficulties. [26] Paul LeJay, _L’Énéide_, lxviii; Heinze, 263. For the detail of the composition of single books and groups see also Heinze, 180, 448, 453. For instances of peripety, see Heinze, 223, 323. [27] Heinze, 120, is hardly extravagant in maintaining that this is beyond any other ancient achievement of the kind. [28] Heinze, 133. [29] Woodberry, 132. See also J. R. Green, “Æneas, a Virgilian Study” in _Stray Studies from England and Italy_, 227. [30] Heinze, 271 seq.; W. Warde Fowler, _The Religious Experience of the Roman People_, lecture xviii. [31] Ovid (_Tristia_, II. 535) says that the fourth book was the most popular. [32] Sellar, 395. For careful discussion of this whole aspect of Vergil’s diction, see Heinze, 410-427. [33] That even Dido’s desperate plea, as well as the calm reply of Æneas, proceeds from point to point, not all readers will agree with Heinze (425-6, on _Æneid_ IV. 305). The variations of rhythm in this passage would surely be used by a sympathetic reciter to suggest agitation. But Vergil’s general neglect of the familiar means of asyndeton and hyperbaton (see, for example, _De sublimitate_, xxi-xxii) to suggest emotional disorder shows a characteristic distrust of incoherence. [34] Heinze, 279. [35] How freely Latin authors transferred it to poetic may be seen in Horace’s _Ars Poetica_ (see below, Chapter viii). Compare Plutarch, _Quomodo adolesc._, x, below, page 244. [36] Sellar, 101, quotes Comparetti: “an elaboration of language which disdains or is unable to say a plain thing in a plain way.” [37] See above, page 210. [38] “During all the years in which Virgil brooded over it and wrought upon it, he kept his material ... in fusion, not crystallized and hardened into final shape” (Mackail, 78); i.e., he continued to adjust. [39] _Eximios tauros_, _farre pio_, etc., noted, among other critics, by E. Nageotte, _Histoire de la littérature latine_, 334. Apropos of Vergil’s incomparable command of the resources of his language, Nageotte adds happily that a “tache de rouille antique a son effet prévu dans la gamme des couleurs environnantes” (324). [40] Woodberry, 125. [41] Woodberry, 111. [42] See above, page 206. [43] See above, page 205. [44] See above, page 203.