The Letters of Sidonius. 2 vols. Translated with Introduction and Notes by O. M. Dalton. 7 × 4¾, clxxxiii + 86 + 268 pp. Oxford: Clarendon Press, 1915. 3s. 6d. n. each vol.

inserted here by the editors, who would not wish to excise it altogether, has great probability. But I should have agreed with them in deprecating its excision. In this connexion, the remarks on the Virgilian hemistichs in Mr. Warde Fowler's appendix are worth study. It is to be hoped that he will go into the subject more fully. In more cases than he is inclined to allow, they are indications of unfinished work. The note of Suetonius which he cites is not that only one of them comes under this head, but that only one of them leaves the syntax incomplete ; which is a different thing. I find it impossible to believe that Virgil deliberately left them unfinished in order to obtain a metrical variation which, however fascinating, would be illegitimate as art. One instance, in xi, 390, may be cited. It seems to me certain that semper erit and fulsus ego, as they stand, cannot be Virgil's final mind; and that in his revision, he would either have completed the line beginning semper erit, or, as I think more probable, have struck these words out, making fulsus ego take their place. In themselves they are needless, and actually weaken the force of what precedes them,