§ 6. If now you take this oblique form of sail, and cut it into any number of required pieces down to its mast, as in Fig. 21, A, and then suppose each of the pieces to contract into studding-sails at the side, you will have whatever type of divided leaf you choose to shape it for. In Fig. 21, A, B, I have taken the rose as the simplest type. The leaf is given in separate contour at C; but that of the mountain ash, A, Fig. 22, suggests the original oval form which encloses all the subdivisions much more beautifully. Each of the studding-sails in this ash-leaf looks much at first as if he were himself a mainsail. But you may know him always to be a subordinate, by observing that the inequality of the two sides which is brought about by accidental influences in the mainsail, is an organic law in the studding-sail. The real leaf tries to set itself evenly on its mast; and the inequality is only a graceful concession to circumstances. But the subordinate or studding-sail is always _by law_ larger at one side than the other; and if he is himself again divided into smaller sails, he will have larger sails on the lowest side, or one more sail on the lowest side, than he has on the other. He always wears, therefore, a servant's, or, at least, subordinate's dress. You may know him anywhere as not the master. Even in the ash leaflet, of which I have outlined one separately, B, Fig. 22, this is clearly seen; but it is much more distinct in more finely divided leaves.[2] [Illustration: FIG. 22] § 7. Observe, then, that leaves are broadly divisible into mainsails and studding-sails; but that the word _leaf_ is properly to be used only of the mainsail; leaflet is the best word for minor divisions; and whether these minor members are only separated by deep cuts, or become complete stalked leaflets, still they are always to be thought of merely as parts of a true leaf. It follows from the mode of their construction that leaflets must always lie more or less _flat_, or edge to edge, in a continuous plane. This position distinguishes them from true leaves as much as their oblique form, and distinguishes them with the same delicate likeness of system; for as the true leaf takes, accidentally and partially, the oblique outline which is legally required in the subordinate, so the true leaf takes accidentally and partially the flat disposition which is legally required in the subordinate. And this point of position we must now study. Henceforward, throughout this chapter, the reader will please note that I speak only of true _leaves_, not of _leaflets_. [Illustration: FIG. 23.]