The commentator who should seek far afield for the source of a poem dealing with the ceremonies of the latter night would, I think, be treated to an amused smile. Even supposing that Keats needed any source other than the conversation with "Mrs. Jones," there was popular rumour and the day itself. I cannot leave this aspect of the poem, however, without just glancing at two of the most recent suggestions as to source. They are both ingenious, but each is, I believe, beside the mark, although for very different reasons.

The first of these suggestions was made some seventeen years ago by Dr. Henry Noble MacCracken.1 It is to the effect that certain analogies to the Eve of St. Agnes are to be found in Boccaccio's early prose romance, Il Filocolo. To my mind, these analogies are by no means so cogent as they seem to Dr. MacCracken, and for a very simple reason: every one of them belongs to the stock in trade of romantic narrative and is almost as much a commonplace of the type as is the existence of a love episode itself. In Il Filocolo, the two lovers, Florio and Biancofiori, he a Moorish prince of Spain, she a Christian damsel, are brought up together, but later separated. Their childish affection for each other has ripened into passion with the years, and their enforced separation is a great grief to both. On one occasion, Florio learns that Biancofiori is shut up with her ladies in an impregnable tower. On the eve of a festival, he manages to get himself conveyed into the tower hidden in a basket of roses. Emerging therefrom, he throws himself on the mercy of Biancofiori's aged attendant, Glorizia, begging her to get him speech with her mistress. Glorizia promises to hide him in Biancofiori's bed-curtains. She then seeks out Biancofiori and invents a fictitious dream in which, she tells the lady, she beheld Florio entering Biancofiori's chamber while she slept. Biancofiori, much comforted by the dream, mingles in the festivities, al

The Source of Keats's 'Eve of St. Agnes,' in Modern Philology. Vol. V. 1907-1908.

though she cannot conceal her melancholy mood. Evening arrived, and Florio duly conducted to his hiding place, Biancofiori comes into the chamber, where, assisted by Glorizia, she undresses, the old crone taking care to keep her impatience for Florio at stretch by suggesting that he may perchance come, that he may not, in short that either event is possible, but that neither can be predicted. Biancofiori at last in bed, Glorizia leaves her, and after a time the lady falls into a troubled sleep from which she is soon aroused by Florio making impassioned love to her. As he does so, two magic carbuncles suddenly glow with a soft light, causing the chamber to become as bright as day. Biancofiori, who has been dreaming the actual scene, cannot at first reconcile herself to an awakening which she hardly comprehends to be a realization. Florio finally convinces her that the fact and the dream are identical. The lovers vow themselves to each other, a ring is exchanged, and the night is passed in complete understanding and mutual satisfaction.

Boiled down to absolute resemblance, what do these parallels amount to? Next to nothing, really. A young man and a young woman are deeply in love with each other, but separated by untoward circumstances. Keats's untoward circumstances take the form of a family feud dividing their two houses. Keats need only have thought for a moment of Romeo and Juliet to have conceived that much of his plot, and the old attendant as go-between is but a duplicate of the nurse in Shakespeare's play and in any one of a dozen old romances. A lover concealed in his lady's chamber is part of the machinery of romantic narration the world over, and Keats's method of handling the scene is as unlike Boccaccio's in Il Filocolo as can well be imagined. Keats's lovers flee away into the night together. Nothing of the sort happens in Il Filocolo. It has been suggested that the magic carbuncles duplicate Keats's moonlight through the stained glass window. But why? Keats needed

no carbuncles to remind him of the moon. We know that he was reading the old Spanish romance, Palmerin of England, in Southey's translation just about this time, and I have already shown how many were the purely colour passages he underscored in that book.1

It will be seen, therefore, that this whole episode from Il Filocolo bears only the very vaguest resemblance to the structure of the Eve of St. Agnes. As to the question of whether Keats could ever have read, or been told of, Il Filocolo, I think we must accord it a decided negative. At the time Keats wrote the Eve of St. Agnes, he could not read Italian. He read the Decameron in an English translation,2 and Dante in the translation of the Rev. H. F. Cary. In the following September, Keats had started to read a little Italian, beginning with Ariosto, "not managing more than six or seven stanzas at a time," but that he had reached any such point by the beginning of the year, that he had even looked at an Italian book with a view to reading it at that date, there seems no possible reason for supposing. And why on earth should he have wanted to read this extremely dull early work of Boccaccio when there were so many better books which he might have read? For in the original he must have read it, if he read it at all, since there was no English translation, and only a French seventeenth century one which there is no reason to believe he had ever come across. More cogent than mere speculation on the subject is the fact that Hunt suggests no such source for the poem, and he himself was the most likely person to have turned Keats's attention to Il Filocolo if there had been any turning at all, which it seems perfectly evident there was not. Boccaccio took parts of his tale from the old French romance, Florice et Blanchefleur. It would take more study than I propose to give to the subject to deter

1 See Vol. I, p. 103; also Appendix C.

2 Sir Sidney Colvin quotes Woodhouse to the effect that the translation read by Keats was that published by Allen Awnmarch. Fifth Edition. 1624.

mine where Boccaccio departed from, and where he kept to, his original. Suffice it to say that a fragmentary English medieval version of the romance, quoted by George Ellis in his Specimens of Early English Metrical Romance published in 1806, a book which Keats may have read, contains none of the particular incidents upon which Dr. MacCracken bases his theory. In the light of all this, I think we may dismiss Dr. MacCracken's speculation as interesting, but without a leg to stand on.

The second suggestion of source which I have mentioned was made three or four years ago by Miss Martha Hale Shackford.1 Miss Shackford considers that she has found certain close scenic and verbal parallels between Keats's poem and Mrs. Radcliffe's Mysteries of Udolpho. And the truth is she has found them; but, owing to the perfectly unrelated manner in which she has been obliged to cull them from Mrs. Radcliffe's novel, they prove nothing more than a vein of reminiscence, if I may so express it. Miss Shackford presents her theory with perfect temperateness and common sense, and without demanding more for it than it will bear. The only question is whether it can be made to bear anything at all of real importance. It is quite true, as Miss Shackford says that:

"The setting of Mrs. Radcliffe's story possessed many elements that seemed revived by Keats. There was the solid grandeur of an ancient Gothic castle, with shadowy galleries, mysterious staircases, moonlit casements, and gorgeous apartments hung with arras glowing with medieval pageantry. The feudal life with old retainers serving an arrogant master and his carousing friends is pictured in both works."

Yet there is an essential difference here, for Mrs. Radcliffe's story is laid frankly in the eighteenth century, its mediæ

1 The Eve of St. Agnes and the Mysteries of Udolpho, by Martha Hale Shackford. Reprinted from the Publication of the Modern Language Association of America. Vol. XXXVI. No. 1. 1921.

valism is pure pastiche, while Keats's tale is chivalric legend throughout.

For the midnight scene in the chapel in the Eve of St. Agnes, Miss Shackford finds striking resemblances in two widely separated scenes in the Mysteries of Udolpho, one in Chapter VIII, the other in Chapter XXXI. For various references to the old woman in the poem, there are others to an old woman in Chapter XLIII and XLIV of the novel. The young woman guiding the old woman down the staircase in St. Agnes is very closely tallied in the last of these chapters, but the errands which bring about this parallel of events are entirely different.

Miss Shackford cites as important the journey through winding passages to a room, and indeed the analogies here are rather striking. In Stanza XIII of the poem, Porphyro .. follow'd through a lowly arched way,"

and eventually

"... found him in a little moonlight room

Pale, lattic'd, chill, and silent as a tomb."

And in Stanza XXI are these lines:

" '... Safe at last

Through many a dusky gallery, they gain

The maiden's chamber, silken, hush'd and chaste.'

In Chapter XXXII of Udolpho we read: "I have only to go... along the vaulted passage and across the great hall and up the marble staircase, and along the north gallery and through the west wing of the castle, and I am in the corridor in a minute," and another passage of much the same purport appears two chapters farther on. While toward the beginning of Udolpho there are several resemblances: "As she passed along the wide and empty galleries, dusky and silent, she felt forlorn and appre