cannot be derived directly or indirectly from Gaimar or the Lay, about the only possible arrangement that remains is to derive it from this lost French version. This explains all the points which the Interpolation has in common with both Gaimar and the Lay. It accounts for those points in which the Interpolation agrees with one of the French versions in opposition to the other, in which case an agreement with either Gaimar or the Lay would establish the form of the romance taken in any incident by the lost French original. This arrangement further makes possible the preservation in the Interpolation of elements lacking in both Gaimar and the Lay, but which may have existed in the lost French version, or even in still earlier forms of the romance. Inasmuch as this arrangement clears up old difficulties and presents no with his bride. Gaimar constantly gives the impression of having been condensed from an original, and in one instance at least this seems to have resulted in confusion. Sigar, in reassuring Havelok the morning after the attack by the six youths, says:

Kore vus aim plus ke ne fis hier
Quant vus asis a mon manger.

Gaimar, 669-670.

"I love you now more than I did yesterday when I placed you at my table." But Gaimar makes no mention whatever of Sigar's placing Havelok at his table the preceding day and the allusion cannot well be explained unless it is assumed that Gaimar had an original in which there was some such mention. The Lay (lines 675-694) does tell about the entertainment of Havelok at dinner on the preceding day, an incident also found in the English romance (lines 1660-1745). This is additional evidence for the lost original of Gaimar and the Lay, and for a relationship between this lost version and the English romance.

Dr. W. H. Schofield suggests that the probable date of this lost version seems to be established by the references to Arthur in Gaimar, the Lay and the Interpolation. In each case the reference stands in connection with an invasion of Denmark to demand or collect tribute. This must have been in the lost version, which therefore could not have been written before Geoffrey of Monmouth's history, and which must have been written before Gaimar. This leaves 1136 and 1150 as the outside dates, with the probabilities in favor of a middle point, somewhere between 1140 and 1145. This mention of Arthur furnishes new evidence for the immediate popularity of Geoffrey.

new ones, it may be regarded as settied that the Interpolation goes back to the lost French version which was also the source of Gaimar and the Lay.

So far the Lambeth Interpolation has been examined in its relation with the French versions of Havelok. It has, however, one or more incidents in common with the English romance, while in other details it differs from all other extant forms of the story. The most striking point in common with the English romance in opposition to the French is the reason assigned for Havelok's leaving Grimsby and going to Lincoln. It will be remembered that Gaimar passes over this portion of the story and merely makes Havelok say that he left home when Grim was dead (line 371). In the Lay, Grim, believing that the boy would still regain his heritage, tells Havelok to go to the court in order to hear instruction and learn sense (lines 157-187). This sounds very much like the custom common in the romances of chivalry of sending a youth to court to learn knightly accomplishments. The Lambeth Interpolation, however, gives a different reason. It says that Havelok was brought up by Grim and his wife. as their own child

Til he was mykel & mighti, and man of mykel cost,
That for his grete sustinaunce, nedly serue he most.

Lambeth, 25-26.

The next line says he took his leave and went to the court. The passage is not in itself very clear, but it certainly seems to mean that he became large and strong, that it required so much to sustain him that he must work for his living and that on this account he left Grim. The interpolator did not find any suggestion of this in Gaimar or the Lay. The incident, however, corresponds remarkably with the English version. The English writer makes constant reference to Havelok's great appetite. The boy thinks he eats too much and determines to go to work (lines 788-810). When the famine arises so that Grim does not have enough to eat for

himself and his family, he advises Havelok to go to Lincoln and find work (824-852). As it is extremely improbable that the interpolator had more than one source, and as it is likewise improbable that this agreement with the English romance is accidental, it seems to be clear that this must have been an element of the story in an early form, and its preservation in the Interpolation shows that it was also found in the lost French version. Gaimar omitted it in his condensation, the Lay changed it in giving the romance its courtly tone, while the English version and the Interpolation have preserved the original.' It is also to be noted that in several details in which the Interpolation agrees with one of the French versions in opposition to the other, it agrees also with the English romance. Such are the humble position of Grim and the defeat of the Danish usurper in general battle, common to Gaimar, the English romance and Lambeth; and the marriage of Havelok, the finding of Grim dead, and the calling out of the host, common to the Lay, the English romance and Lambeth. This agreement with the English makes all the more positive the derivation of the Lambeth Interpolation from the common source of Gaimar and the Lay which was evident from a comparison of the French versions. Incidentally it shows the difficulties in the way of any attempt to derive the English romance from either of the extant French versions.

There are several details in which the Lambeth Interpolation is unique. (1) Gunter, Havelok's father, is identified with the Guthrun or Gormo who fought against Alfred in

'There are also two minor points in which the Interpolation agrees with the English romance. (1) It is said in the Interpolation that while Havelok is at the court all the folk love him (line 44). The English writer says that knights, children, young and old, all love him (lines 955-958). (2) According to the Lambeth version Edelsi hears that Havelok has come to the coast (line 75). In the English, Godrich hears that Havelok has come into England (lines 2531-2547). In the French versions nothing is said about the usurper's hearing of the return of Havelok before he sends his defiance.

the ninth century, the only allusion to Arthur being the statement that the invasion of Denmark was to collect tribute which he had formerly taken. It is not safe to make much of this for the story is interpolated at this point in the chronicle merely because Langtoft, by confusion of names perhaps, called Gunter the father of Havelok. An interesting question is involved as to the historical basis of the Havelok legend, which, however, need not be discussed here. (2) When the Danish king is killed, his queen escapes to the sea with Havelok and meets Grim on the shore (lines 12-16). There is reason for believing this to be a feature going back to the original form of the story. There is nothing in Gaimar's condensed account of the early Danish events to contradict the assumption. The Lay makes an unquestionable modification here in that it calls Grim a baron, and has the queen and child entrusted to him in a castle (lines 53-68). The English version, too, shows an entire modification of the early Danish events in order to carry out an extended duplication of the English part of the story. Thus in both England and Denmark the king knows he is going to die, he summons his barons, and he entrusts his kingdom and infant heir to an earl who takes an oath and afterwards usurps the kingdom. Of the two series of events that in England must have been the original because it is also found in the French Lay. There is, therefore, little or nothing in the English romance to show what its original had to say about the early Danish part of the story. (3) The Lambeth Interpolation states definitely that Havelok was brought up by Grim and his wife as their own child and regarded himself as such (lines 23-24). This is implied in both Gaimar and the Lay, in both of which Havelok does not know who he is until told by Grim's relatives. In the English romance Havelok is apparently at all times conscious of his

This duplication of events was suggested by G. Wittenbrinck, in a dissertation, Zur Kritik und Rhythmik des altenglischen Lais von Havelok dem Dänem, Burgsteinfurt, 1891, p. 5.

position, though it may be hard to reconcile this with his inaction and indifference. In this the Lambeth version seems to represent best the original form of the story. (4) The king is influenced in marrying Argentille by a "chere" which he has seen her make to Coraunt (line 46). This touch is probably an addition on the part of the scribe. (5) Edelsi, after his defeat, voluntarily makes Argentille heir to Lyndeseye (lines 79-80). These details, some of them significant, add to the importance of the Lambeth version in the discussion of Havelok, for it must be borne in mind that it is possible for this brief analysis of the story, interpolated in a late manuscript of a chronicle, to preserve elements belonging to the original legend.

This agreement with the English romance in certain details and the preservation in others of traces of a lost original make all the more conclusive the observation that the Lambeth Interpolation is derived from a form of the story earlier than Gaimar or the Lay, for in no other way could these incidents have come down to the interpolating scribe. That this early form of the story was identical with the lost source of Gaimar and the Lay has already been shown. The present investigation, therefore, may be regarded as giving the Lambeth Interpolation, for the first time, its proper place in the development of the romance.

Of the more general results obtained by the investigation the most noteworthy is the additional light thrown on the lost French version in octosyllabic rimed couplets,-the common source of Gaimar and the Lay. With merely Gaimar and the Lay to work with, it is not always possible to determine accurately what form of any particular incident was taken by this lost version. It is frequently evident, where the two differ, that one of the extant French versions represents the

1

1As Kupferschmidt has suggested, additional light may be thrown on the lost version, by the Havelok episode in the Brute, but in view of the possible contamination with the English version, indicated by the name Birkabeyn, it is not here considered.