all this is done in order that now, when Antony has been checked, we might supplicate another to allow himself to be placed in the position which Antony held, or was it not rather that the State should be its own master, and at its own disposal? unless, perchance, our opposition was directed, not against bondage in general, but against a particular kind of bondage.' The clause an ... res publica, expressing, as it does, the real reason why opposition was directed against Antony, comes in awkwardly in the midst of the other ironical sentences. Perhaps we should read reponit, an pateretur ut esset, 'that we should ask the other, who is putting himself into the place of Antony, whether he will be graciously pleased to allow the State to be its own master.' i. 18. 4. Videtur enim esse indoles, sed flexibilis aetas. With indoles Wesenberg proposes to supply bona; but there is no necessity to do so: indoles by itself can mean 'natural excellence,' 'capacity': cp. Att. x. 12. 7, est enim indoles; Plaut. Rud. 424, tum quae indoles in saviost; Liv. i. 3. 1, tanta indoles in Lavinia erat. ii. 3. 5. Duabus rebus egemus, Cicero, pecunia et supplemento, quarum altera potest abs te expediri, ut aliqua pars militum istinc mittatur nobis vel secreto consilio adversus Pansam vel actione in senatu, altera, quae magis est necessaria, neque meo exercitui magis quam reliquorum. This sentence has no proper ending. Ernesti omits quae after altera, and Wesenberg reads alteraque. But altera . . . alteraque is, I think, a solecism. Possibly ab ipso senatu, or words to that effect, have dropped out between senatu and altera. As regards the transference of forces, the authority, though formally resting with the Senate, was practically in the hands of the general, and the Senate would not interfere further than to make a recommendation, except on important and critical occasions. But the case was different with grants of funds; such grants were both formally and practically made by the Senate. For the diminished control exercised by the Senate over the armies of the provincial governors during the last century of the Republic see Willems Le Sénat, ii. 646 ff. L. C. PURSER. CATULLUS 53. Risi nescioquem modo e corona, 'Di magni salaputtium disertum !' Haec in the fourth verse seems somewhat awkward. I venture to suggest that Catullus wrote Admirans ait ec manusque tollens, i.e. extollensque manus. For ec- see vi. 13. A. P. IN NOTES ON VALERIUS FLACCUS. Na paper in the last number of this Journal I made a number of tentative suggestions which had occurred to me during a first study of the Argonautica. A large number of these suggestions I should now, after more careful consideration, decline to defend; but my observations have not been quite fruitless, as they have called forth Mr. Postgate's Annotations in the JOURNAL OF PHILOLOGY (xxii., p. 307 sqq.), and elicited some emendations from the editor of HERMATHENA. In the present paper I propose to deal with some other difficulties in the poem, but will first make brief remarks on a few of the passages dealt with in my first paper. i. 147. Mr. Postgate's explanation is right. 213. I find that in reading regem I have been anticipated by Kiessling. 529, 530. No change is necessary; Bumann's explanation is correct. 749. artus is right. Compare Virgil, Aen., iv. 336, dum spiritus hos regit artus. ii. 235. Read obduntque (see Mr. Postgate's note, p. 308). 414. raptus is right. 454. Kurtz (Zeitschrift für die österreichischen Gymnasien, xxviii. 610, has well defended flebile succedens, cum fracta remurmurat unda. 455. uacuum is right. 626. Kurtz (ib.) anticipated me in the emendation caelamina. iii. 120. Mr. Postgate's sinistrum is, I think, certain. 594. Ph. Wagner's nunc motas (so Mr. Palmer) is probably right. iv. 674. uel fallis is the true reading, as Mr. Herbert Greene pointed out to me (so Mr. Postgate, p. 311). One or two other passages will be rehandled below. In treating the text of Valerius, it is important to realize that the poet had not finally revised even the early part of his work. There are not only clear indications of this fact, but it may be shown with probability that V. was ultimately derived from a text in which Valerius had jotted down additions which he intended to make before publication, but was prevented (by death) from incorporating in his poem. This question has been well discussed by J. Peters, De C. Valerii Flacci vita et carmine, p. 14 sqq., who has added much to the remarks of Thilo. There is a clear instance of the need of revision in ii. 332, where an abrupt transition leaves the story incomplete (see Peters, p. 21). In vii. 423 we are told casually that Iphis is dead, and in i. 441 that Iphis is to die; in revising Valerius would assuredly have introduced a narrative of his death. There is an inconsistency between vi. 507 and 750 (Peters, 18). In v. 477 Cretheus is (by inadvertence) stated to have been an ancestor of both Phrixus and Iason, whereas he was brother of Athamas, who was father of Phrixus. This lapse would have been corrected on revision (Peters, 20). The true explanation of v. 565, 566— qualis ab Oceano nitidum chorus aethera uestit is, no doubt, that suggested by Bulaeus, that the poet wrote both lines provisionally, intending to make a final choice of one. The same is to be said of vii. 201, 201 a— hoc satis; ipsa etiam casus spectare supremos ei mihi ne casus etiam spectare supremos. i. 778-84 are to be regarded as having been written in the margin by the poet, who intended subsequently to work them in, making the necessary alterations in his original text. And so iii. 273, cur etiam flammas miseros que moramur honores, was a marginal addition of Valerius, perhaps intended to come after 1. 310 (as Thilo suggests), but inserted by the copyist in an inappropriate place. Some cases of tautology would perhaps have been altered by Valerius in revising-e.g. iii. 139, elatae; 140, delataque (which, however probable, is not absolutely certain, as V has delicataque); 254, uersique; 256, conuersa. The repetition of a word or phrase is not a sufficient cause for suspecting our text. For the same reason I would not, with Bährens and Schenkl, question such a verse as i. 851— et loca et infernos almae uirtutis honores. Valerius wrote ii. 642 longaque iam populis inperuia lucis eoae, as it stands, but he would probably have altered it. I. 398. insequeris casusque tuos expressa, Phalere, uacua V, naeua M. Heinsius proposes patula, Schenkl uasta. uacua, which can only mean that the snake had left the tree, is superfluous with lapsus ab arbore. Valerius fully appreciated the possibilities of uacuus (cp. iii. 589, frangit et absentem uacuis sub dentibus hostem, and vii. 528, uacuo furit ore per auras), and if he wrote uacua here we may assume that he would have replaced it subsequently by some other epithet. It seems to me, however, |