But choice of words was of comparatively small importance in determining the style of Isokrates, between whom and the elder ‘austere’ school the essential point of contrast was just this, that they relied much on words, while he relied almost wholly on composition. It was Isokrates who developed, though he did not originate, the idea of a literary prose-rhythm. The Greek theory distinguished a ¹ music proper to the continuous ( συνεχής) exertion of the voice in prose-declamation from the music of its exertion at intervals ( διαστήματα) in singing ² . As singing can scarcely charm the ear or make claim to beauty until it has brought itself under definite laws, partly of rhythm, partly of harmony, so oratorical prose cannot give artistic pleasure until it has become, in its proper measure, rhythmical. This implies the bringing out of that musical element which is inherent in all language; and the technical Rhetoric early began to take account of the proserhythm into which this element must be wrought. Thus Aristotle ³ discusses the relative merits for rhetorical prose of the dactyl—which is too epic for ordinary use—the iambus, which is too common to give any distinctive effect—the trochee, which is too light—and the paeon, which he thinks on the whole the most serviceable,—the ‘first’ paeon (¯˘˘˘) for the beginning of the period, the ‘fourth’ (˘˘˘¯) for the end. Poetry has its strict correspondence of rhythms and its precision of metres. Prose has its irregular rhythms and its wandering melody in the fall of syllables—rhythms and metres not bound by any rigid framework, yet reducible to certain general laws which the attentive ear can discover, and which the skilful speaker can apply in ever-varying combinations ⁴ . Now the mistake of Gorgias had consisted in trying to bring the essentially free rhythms and metres of prose too near to the strict rhythms and metres of verse. Thrasymachos of Chalkedon was probably more judicious ⁵ . But Isokrates was the earliest great artist in the rhythm proper to prose ⁶ ,—so distinctly so, that Cicero more than once calls him its discoverer ⁷ . Great artist as he was, however, he was only a developer, not a perfecter; and the chief reason why he fell short of the highest excellence seems to have been this, that he sought too constantly to base his rhythms on a certain type of composite period.