^{Note 1 page 432} The paper is intended to supplement and in part to supersede the earlier paper, which appeared in the Publications of the Association for 1889. It restates and reinforces the theory of the earlier paper. Certain errors in detail which do not affect the truth of the main thesis, I need not specify; one sentence, however, that beginning “Personification addresses itself” (p. 189; p. 49 of the reprint) I wish to cancel as entirely inadequate, and in part incorrect. At the time of writing the sentence I must hare had in mind merely alphabetic personification.

^{Note 1 page 435} In his life of Milton, Mark Pattison, whose classical scholarship is unquestioned, has the following sentence (p. 192):—“The power of metaphor, i. e., of indirect expression, is, according to Aristotle, the characteristic of genius.” The reference is undoubtedly to the passage in the Poetics quoted above. Whately, in his Rhetoric translates the same passage by the words “a mark of genius.” I question whether the foregoing translations do not attribute to Aristotle's words,—,— more meaning than they will bear. On the other hand, Wharton's translation, “a proof of cleverness,” seems to understate the force of the original. Several eminent classical scholars have been so kind as to give me more exact translations of the passage. Two suggested independently “natural ability;” this rendering, which I have adopted, is also employed better in preserving the significance of the first part of the compound in . Perhaps, however, the word “ability” preserves the force of; if so, I should prefer not to employ three words in order to translate one. George Eliot (Mill on the Floss, Bk. II, ch. 1) translates the phrase by “a sign of high intelligence.” The natural temptation is to give to the words all the meaning that they will bear.

^{Note 1 page 438} See the first paragraph of Charles Lamb's essay on “Poor Relations” for an amusing list of descriptive epithets that are not used as Kenningar, though many of them are capable of conversion in to Kenningar.

I should like to plead for the introduction into our text-books of the name Keening. If we can adopt and use with ease Greek words such as Synecdoche, Metonymy, Metaphor, which even to most of those who use them are mere names, surely we can adopt a word which is much more nearly English, and which is already known to students of Old English. Epithet (a Greek word) is not so good name as Kenning; and it is possible to give to the latter word a definite meaning. The word, if anglicized, would naturally receive an English plural.

^{Note 1 page 441} Quoted also by McElroy, The Structure of English Prose, p. 240.

^{Note 1 page 445} In order to assure myself that the foregoing paragraph was not superfluous or overstated, before sending it to press I examined with reference to the point under discussion twelve modern rhetorics, from Blair's (1788) to a book published in 1892. Ten of these twelve books give definitions of allegory that are inaccurate; one (intentionally) gives no definition; the definition in the twelfth book is correct.

^{Note 1 page 448} Blair (Lecture XIV) says,—“This distinction . . . is of no great une; as nothing can be built upon it in practice; neither is it always very clear.”

President D. J. Hill, in his Science of Rhetoric (p. 203), saya,—“ Quintilian's distinction between tropea and figura is of no practical value.”

Professor Bain, in his English Composition and Rhetoric (Vol. I, p. 135), says,—“ The distinction is artificial, and turns on a point that has little relevance to the leading uses of the Figura in Style.”

^{Note 1 page 449} Quintilian, Inst. Orator. ix, 1, 4:—Figura, sicut nomine ipso patet, conformatio quaedam a communi et primum se offerente ratione.