characteristic in the organization of verse in the mediæval Latin, that 'carmen rhythmicum' came to signify a rimed poem, and the later Latinists used the word 'rhythmus' precisely in the sense of 'rime.'" Schipper goes on to inquire whether this mediæval Latin poetry was the means of introducing rime to the Germanic peoples. Wackernagel held it as certain that Otfried learned his rimes from Latin poems; but as Otfried declares that his poetry is intended to take the place of useless popular songs, it seems probable that he was not using an innovation, and that rime was already popular in Old High German. Indeed, it appears sporadically in the _Hildebrandslied_ and the _Heliand_. Grimm observes that it is theoretically unlikely that alliteration should have vanished suddenly and rime as suddenly have taken its place. The gradual development of rime from assonance among the Romance peoples suggests the same thing. The early appearance of rime by the side of alliteration is illustrated by C. F. Meyer (_Historische Studien_, Leipzig, 1851) from the Norse _Edda_, from _Beowulf_, Cædmon, etc. In the later Anglo-Saxon period rime was evidently securing a considerable though uncertain hold. The Riming Poem suggests what development rime might have had in English without the incoming of Romance influences. Grimm observes that, while to his mind alliteration was a finer and nobler quality than rime, there was need of a stronger sound-likeness which should hold the attention more firmly by its unchanging place at the end of the line. Schipper remarks, finally, that the fact that the use of rime in connection with alliteration was especially popular in the southern half of England, where the Romance influence was strongest, may suggest why in the more purely Germanic northern half alliteration remained the single support of the poetic form well into the fifteenth century.