History, it has been said, is philosophy teaching by examples. Unhappily, what the philosophy gains in soundness and depth the examples generally lose in vividness. A perfect historian must possess an imagination sufficiently powerful to make his narrative affecting and picturesque. Yet he must control it so absolutely as to content himself with the materials which he finds, and to refrain from supplying deficiencies by additions of his own. He must be a profound and ingenious reasoner. Yet he must possess sufficient self-command to abstain from casting his facts in the mould of his hypothesis. Those who can justly estimate these almost insuperable difficulties will not think it strange that every writer should have failed, either in the narrative or in the speculative department of history. It may be laid down as a general rule, though subject to considerable qualifications and exceptions, that history begins in novel and ends in essay. Of the romantic historians Herodotus is the earliest and the best. His animation, his simple-hearted tenderness, his wonderful talent for description and dialogue, and the pure sweet flow of his language, place him at the head of narrators. He reminds us of a delightful child. There is a grace beyond the reach of affectation in his awkwardness, a malice in his innocence, an intelligence in his nonsense, an insinuating eloquence in his lisp. We know of no writer who makes such interest for himself and his book in the heart of the reader. At the distance of three-and-twenty centuries, we feel for him the same sort of pitying fondness which Fontaine and Gay are said to have inspired in society. He has written an incomparable book. He has written something better perhaps than the best history; but he has not written a good history; he is, from the first to the last chapter, an inventor. We do not here refer merely to those gross fictions with which he has been reproached by the critics of later times. We speak of that colouring which is equally diffused over his whole narrative, and which perpetually leaves the most sagacious reader in doubt what to reject and what to receive. The most authentic parts of his work bear the same relation to his wildest legends which Henry the Fifth bears to the Tempest. There was an expedition undertaken by Xerxes against Greece; and there was an invasion of France. There was a battle at Plataea; and there was a battle at Agincourt. Cambridge and Exeter, the Constable and the Dauphin, were persons as real as Demaratus and Pausanias. The harangue of the Archbishop on the Salic Law and the Book of Numbers differs much less from the orations which have in all ages proceeded from the right reverend bench than the speeches of Mardonius and Artabanus from those which were delivered at the council-board of Susa. Shakspeare gives us enumerations of armies, and returns of killed and wounded, which are not, we suspect, much less accurate than those of Herodotus. There are