In the previous chapter, on recitative, we discussed the way in which musical punctuation seeks to express the most predominant structures of language, those of prose. In this chapter, we now turn the tables to view musical punctuation, still as it seeks to imitate language, but from the perspective of music's most predominant structures. The general consensus among eighteenth-century music theorists, as expressed above by Koch, is that the basics of musical punctuation are to be derived from the most recognizable and distinctly defined forms of the day, what we are describing as the verse-like structures of music. These verse forms come in both vocal and instrumental varieties, but it is the latter which best depict the pleasing regularity of meter, accent, and cadence. True, the melodies to songs and odes (the chorale and figured melodies) make good punctuation models: John Gunn [45] adapts forty favorite Scotch airs for the violoncello (or the violin or German flute) in 1793 as familiar models for the study of proper phrasing; and Türk recommends in 1789 “short songs set for the clavichord by good composers.” But the character of these melodies still depends on the contents of the poetry, and their overall form on the structure of the strophes. These compositions, according to Koch, presuppose a certain knowledge of vocal music and the rules of language, a two-sided and therefore more difficult branch of composition. And as we understand from Schuback, even for those who have acquired the necessary knowledge of declamation, the manipulation of poetry to the rhythm and meter of melody is a matter of constant compromise and little pleasure.