dispelling drunkenness; and the Lord Bacchus taught him the toil of the winedresser which loves the flowers, how to dig around the vines a watering ditch, to cut off the tips of the old dead branches, and foster the young shoots of the vine that they may bear the wine producing clusters."

The palace of King Staphylus is inhabited by mere abstractions, like the house of Alma in the Fairie Queene. The king himself, Queen Methe, Prince Botrys and the lusty old man Pithos are no more nor less than what their names would indicate. But in spite of their shadowy unreality the scenes in which they take part have a fantastic charm, heightened by the splendid setting of the fairy palace agleam with silver and gold and mosaic work of precious stones. Here at a great feast at which the king welcomed the god, for the first time they tasted wine. And first Maron (Od. IX, 197), Bacchus' charioteer, danced supported by a Satyr on each side, holding a wine skin slung around his neck, and a wine cup. Then Queen Methe became drunk and Bacchus had to hold her to keep her from falling, and the cheeks of Prince Botrys, who was now full of wine, were flushed with a purple glow, and King Staphylus too was drunk. Both father and son bound their heads with ivy, and Botrys performed a most elegant pas seul, spinning around with steps that rivalled each other as they flewδεξιὸν ἐκ λαίοιο μετήλυδα ταρσὸν ἑλίσσων— evidently a kind of double shuffle. Then the king danced with his arm around his son's neck, and the queen joined them and made a third with her husband and son on either side, an arm around the neck of each-a lovely trio. "And lusty old Pithos shook his grey hair to the breeze, full of the sweet liquor up to his very teeth, and danced heavy with wine, twirling his staggering foot, and the sweet drops that flew from his hiccuping lips made his tawny beard white with foam. All day they drank, and the cups were still being drawn when shadowy darkness covered all the eventide earth-and the still night clad the dark in her own colors, patterning the heavens with her robe of stars."

The king sent Bacchus on his way with prophecies of victory, but while the god was traversing Syria, King Staphylus died. Dionysus returned at the news to comfort the sorrowing queen, by interpreting to her the meaning of their names. A funeral agon is celebrated for the dead king, consisting of a contest in song between Erechtheus against Oeagros, won by the latter, and a con

test in pantomime-Maron dancing the rivalry of Aristaeus with honey against Bacchus with wine for the favor of the gods—and Silenus performing a variety of difficult dances, and finally turning into a river. After the funeral Bacchus clad the queen in a new crimson peplos, washed Pithos and bade him put on a shining white chiton, and throw away his old one, befouled with the ashes of weeping. Then he made Prince Botrys open the royal wardrobes and put on the purple-dyed robes worn by his father, and all feasted again till evening. The last we hear of this charming royal household only Nonnus would have told as he does: "The long ranks of feasters took the gift of sleep by turns upon the deep-strewn couches within the hall. And Pithos and Maron went up upon one bed, belching up the fragrant draught of the nectar of the winepress, and made each other drunk on the like breath each breathed forth all night long! But Eupetale, the nurse of Lyaeus, lighted a torch and spread for Botrys and for Dionysus a double purple covered bed for the two of them, but in the thalamus nearby, apart from the Satyrs and removed from Dionysus, the attendants spread a golden couch for the queen."

The University of South Carolina.

L. P. CHAMBERLAYNE.

[The note that follows pleads as a justification of its inclusion in this group of classical studies its relation to the broadly interested syntactician in whose memory this volume is published. In a larger and still unpublished investigation, of which this note is a by-product, my friend and former colleague, Professor Bain, showed his interest by helping me over many a rough place when I wandered into the field of classical syntax. And he was always so ready to lend me the aid of his fine syntactical feeling in weighing judgments about English constructions that I cannot feel any lack of appropriateness in the subject-matter of my slight contribution to the memory of one who de constructione verborum nil a se alienum putavit. J. F. R.] In an article published in Modern Philology1 a year ago I showed that John Lydgate has left in his poetry the first recorded frequent use of the unambiguous do auxiliary in English. Although in that article I called attention to the fact that, contrary to his practice in his poetry, Lydgate did not use do periphrastic tenses in his prose piece, The Serpent of Division, and pointed out the same sort of inconsistency between the poetry and the prose of Lydgate's contemporary, John Capgrave, I had at the time of putting the paper into print no satisfactory reason to account for the general use of the do auxiliary in poetry before it appeared in prose. Dietze's theory of metrical convenience to explain the more frequent occurrence of do periphrastic tenses in modern English poetry than in modern English prose would not satisfy my curiosity in regard to the cause of Lydgate's frequent use of a construction that had appeared very sparingly in the written record before his day. Here I return to my knitting long enough to offer a simple and reasonable theory to account for Lydgate's carrying over into poetry what I suppose was, before his elevation of it, a construction employed only in the spoken language.

To facilitate his rime, not to fill his rhythm, Lydgate used do periphrastic tenses. The use of do periphrastic tenses gave him the opportunity of substituting the infinitive for inflected forms of small rime value, and of throwing the infinitive to the end of the verse-line. The end-line infinitive after the do auxiliary furnished him the chance of a rime with other infinitives, especially with infinitives after the established auxiliaries, and with any other part of speech. The inflected third person singular indicative and the inflected preterites, largely -ed and -t preterites, gave him, on the contrary, slight opportunity for rimes except with similarly inflected forms.

1 "The Do Auxiliary-1400 to 1450," Modern Philology, XII, 7, January 1915, pp. 449-456. 2 Das Umschreibende Do in der neuenglischen Prosa, Jena, 1895, pp. 18-19.

From the following list of all the rime pairs in the Temple of Glas of which an infinitive after an auxiliary do furnishes one member, it may be seen with what sort of rime elements Lydgate joined the end-line infinitives of do periphrastic tenses:

19-20, roche:did approche; 79-80, liue:dide rife; 115-116, tre:did fle; 119-120, did sue:to transmue; 133-143, did obeie:to conuei; 229-230, grace:dop pace; 311-312, benigne:doþ resigne; 370-372, did enclyne:fyne; 511-513, do specifie:fantasie; 671-672, to take:dop awake; 846-847, dop enbrace:grace; 944-945, disease: dide sease; 1026-1028, dop suffise:deuyse (infinitive); 1054-1056, did abraide:

seide; 1055-1057, did fele:wele; 1232-1233, founde (past part.):dide wounde; 1265-1267, do assure:tendure; 1279-1281, wele (adjective):did knele;

brace:space; 1356-1358, done appere:chere; 1364-1365,

That Lydgate took full advantage of the rime convenience offered by the possible end-line position of the infinitive after auxiliary do is made clear by the statement that of the one hundred and twentyone instances of his use of this construction which I cited in my former article, one hundred and nineteen cast the infinitive into the riming position.1

In making rime use of a colloquialism that Gower, Chaucer, and Hoccleve avoided, Lydgate merely added one more means of throwing the easily riming infinitive to the end of the line. For the same rime purpose, earlier and better poets than Lydgate had generously employed the infinitive after the established auxiliaries. While Chaucer, for instance, used in the Prologue to the Canterbury Tales only fortyfive per cent of the infinitives after the established auxiliaries to carry the rime, he had employed in Book I of Troilus and Creseide sixty-two per cent, and in The Clerkes Tale seventy-five per cent, of the infinitives after the established auxiliaries to bear the rime burden. Approximately six per cent of the rimes in Book I of Troilus and Creseide and in The Clerkes Tale depend upon this infinitive rime device. And in the use of the established auxiliary plus infinitive a mere manipulation of the word-order would bring the infinitive into the riming position; Lydgate's practice, on the contrary, involved a choice between forms of identical

3 Twenty examples in the Esop; twenty one in the Temple of Glas; thirteen in the Troy Book (II. 1-4000); thirty in Resoun and Sensuallyte (l. 1-4100); twenty in the Pilgrimage of the Life of Man (ll. 1-4000); seventeen in Secres of Olde Philosoffres (ll. 1-1491).

4 Fourteen of sixteen consecutive examples of do periphrastic tenses in Capgrave's verse Life of St. Katharine send the infinitive to the end of the line.

5 The stanzaic-form of Troilus and Creseide and of The Clerkes Tale makes more demand upon rime than does that of the Prologue.

6 Compare Tennyson's practice in In Memoriam: only one-fifth of one per cent of the lines obtain their rime in this manner.

meaning: the periphrastic form provided the means of an easy rime; the inflected form offered a more difficult rime possibility.

As far as the record shows, Lydgate was the first maker of verse to employ a rime device that weak versifiers have found of great help;7 that many good poets have not always been above using; and that Doctor Johnson, in these well-known words, excused the poet Cowley for having overworked: "The words 'do' and 'did,' which so much degrade in present estimation the line that admits them, were in the time of Cowley little censured or avoided." And it is a rime device in perfect harmony with the monkish mechanism of Lydgate's verse. JAMES FINCH ROYSTER.

The University of Texas.

7 A good specimen of recent Coogleresque verse (Domnold, A Legend of Ireland in the Thirteenth Century, by F. W. Grattan, 1911) on some of its pages runs the percentage of do plus infinitive rimes as high as fifteen.

8 For example, in an early poem, A Dream of Fair Women (288 II.), Tennyson uses periphrastic do tenses four times apparently only for the sake of an easy rime. In a blank-verse poem, more than four times as long as A Dream of Fair Women, Lancelot and Elaine (1418 II.), but one do periphrastic tense is found, and the infinitive is not at the end of the line. Furthermore, in Lancelot and Elaine, past tenses in -'d and -t end approximately three per cent of the blank-verse lines!

The Works of the English Poets, VII, 27.