First, as to the form of the narrative. This begins with verse, then turns to prose, and throughout drops again at intervals into poetry in a friendly way like Mr. Wegg. Now this is a form of writing not unknown in other branches of literature, the _cante-fable_, of which “Aucassin et Nicolette” is the most distinguished example. Nor is the _cante-fable_ confined to France. Many of the heroic verses of the Arabs contained in the _Hamâsa_ would be unintelligible without accompanying narrative, which is nowadays preserved in the commentary. The verses imbedded in the _Arabian Nights_ give them something of the character of a _cante-fable_, and the same may be said of the Indian and Persian story-books, though the verse is usually of a sententious and moral kind, as in the _gâthas_ of the Buddhist Jatakas. Even as remote as Zanzibar, Mr. Lang notes, the folk-tales are told as _cante-fables_. There are even traces in the Old Testament of such screeds of verse amid the prose narrative, as in the story of Lamech or that of Balaam. All this suggests that this is a very early and common form of narrative. Among folk-tales there are still many traces of the _cante-fable_. Thus, in Grimm's collection, verses occur in Nos. 1, 5, 11, 12, 13, 15, 19, 21, 24, 28, 30, 36, 38_a_, _b_, 39_a_, 40, 45, 46, 47, out of the first fifty tales, 36 per cent. Of Chambers' twenty-one folk-tales, in the _Popular Rhymes of Scotland_ only five are without interspersed verses. Of the forty-three tales contained in this volume, three (ix., xxix., xxxiii.) are derived from ballads and do not therefore count in the present connection. Of the remaining forty, i., iii., vii., xvi., xix., xxi., xxiii., xxv., xxxi., xxxv., xxxviii., xli. (made up from verses), xliii., contain rhymed lines, while xiv., xxii., xxvi., and xxxvii., contain “survivals” of rhymes (“let me come in--chinny chin-chin”; “once again ... come to Spain;” “it is not so--should be so”; “and his lady, him behind”); and x. and xxxii. are rhythmical if not rhyming. As most of the remainder are drolls, which have probably a different origin, there seems to be great probability that originally all folk-tales of a serious character were interspersed with rhyme, and took therefore the form of the _cante-fable_. It is indeed unlikely that the ballad itself began as continuous verse, and the _cante-fable_ is probably the protoplasm out of which both ballad and folk-tale have been differentiated, the ballad by omitting the narrative prose, the folk-tale by expanding it. In “Childe Rowland” we have the nearest example to such protoplasm, and it is not difficult to see how it could have been shortened into a ballad or reduced to a prose folk-tale pure and simple.