that the principal part, the Lancelot portion, runs parallel with the hitherto unpublished third part of the French Prose Romance of Lancelot in the Mort d'Arthus. It agrees in the main much more closely with that version than with Malory, and, according to Dr. Schüler, represents a more archaic phase of the legends than those hitherto known. The immediate original, however, from which the translation has been made, was an Italian version, and probably one hailing from Tuscany.
If we were quite sure of the exact reading of the Romance words found in the Hebrew text, and also that they had not been modified by the copyist, who may have lived in Italy, there would still be great difficulties, on linguistic as well as on literary grounds, in assuming the existence of so old an Italian translation of any of the Arthurian legends as to permit the present version to be based on an Italian original. The text is a combination of at least two or three legends, and reference is moreover made to others, such as a detailed history of the Quest, a history of Merlin, and another history of Lancelot. Out of these legends this "Morte d'Arthur," if I may call it so, has been compiled, and it agrees with the latter in its composition and partly in the contents. It presupposes an entire cycle of Arthurian prose romances, and that some at least had been joined together to form a corpus of such legends. The legends used may have been old, and therefore short, although the compiler has reduced them to a smaller compass. I am not aware, however, of any version of these legends so brief and succinct as the Hebrew, but, in spite of this, I do not believe that the translator had of his own accord abbreviated his original beyond the modest proportions mentioned by him and not exceeding more than a few pages in the whole book.
A problem is hereby raised for the students of these legends,—to find out whether such a short compilation already existed in the first half of the thirteenth century,