As a general rule, in plants with stipulate leaves, each leaf is provided with two stipules. But when the leaves are opposite, the two on the same side of the node often coalesce, forming a single interpetiolar stipule, as in the case of Cephalanthus ( fig. 78). That this coalescence is secondary is shown by the fact that the distal portions only of the veins of the two stipules have united. Now in the Stellate also, this must have been the original condition, but the interpetiolar stipules have been greatly developed to serve assimilative purposes, the veins having mean while united completely to form a midrib. The increase in size has advanced until in Callum the stipules are of the same size and form as the leaves and morphologically indistinguishable from them, except in G. Wolium where the stipples are smaller. In this condition they remain in the broader-leaved species, as G. pilosum lit., G. latifolium Michx. and G. lanceolatum Torr. But in the narrower-leaved species, a still greater foliar expan sion being desirable, separation has been re-accomplished, proceed ing probably from the tip downward, as is illustrated in Bubia peregrina L. with whorls of four. In this species stipules are occasionally found with two midribs (fig. 79), most widely sepa rated at the apex or even coalescing toward the base. In Galium Aparine L. and other species in which the number of stipules is abnormal, we may suppose this condition to have arisen from a repetition of the process of division which has produced the six leaved whorls. This is not improbable, since even in the four leaved forms the stipules have already entirely lost their original morphological character and have taken on a more generalized nature, making them fit material for development along new lines of evolution. Embryological evidence is not wholly wanting, al though the family stands so near the head of the plant series. In Galium Aparine L., in common with the six-leaved species, the earlier whorls are of four leaves only, representing the ancestral condition. In Rubia tinctorium L., the opposite leaves of the subterranean portion of the stem are exstipulate. At the first aerial node there is a whorl of four, interpetiolar stipules being present, and in the higher whorls there are six leaves.* This is a series of long range, though lacking in intermediate steps.
Another case in which there is present a nodal girdle from which the stipular bundles arise is that of Humulus Lupulus (fig. 80), but there are three bundles in each leaf-trace. They are placed at about equal distances around the circumference of the stem, and the girdle-bundles proper occupy only about one-third of the periphery on each side. From them a part of the stipular bundles arise, the remainder originating directly from the lateral bundles of the leaf-traces.
It would be to small purpose that examples should be fnrther multiplied. From those already cited we may confidently deduce the following conclusions : 1. The sheathing petiole has its origin independently of the true petiole and is formed by a concomitant development of the lateral and central-basal portions of the primitive leaf.
2. The ligule is a special development of the apical parts of the lateral portions of the primitive leaf along the ridge between the sheathing petiole and the distal parts of the leaf. It may be sup plied with veins either by the marginal bundles of the sheath or by tangential branches from those entering the blade. The sheathing petiole may disappear by degeneration, rendering the axillary as in many species of Potamogeton.
3. The ochrea is related to the ligule and is generally associated with the sheathing petiole. It consists of the apical tissues de veloped in those cases where the sheathing petiole completely sur rounds the stem or did so in the ancestral condition. The part of the ochrea posterior to the lamina or petiole may be called its ligular portion and is usually supplied by bundles arising tan gentially from the main ones.
4. The lateral portions of the primitive leaf, when separated in greater or less degree, constitute stipules in the usual acceptation of the term. They are variously modified by subsequent evolu tionary changes, by increased development, by basal or total degeneration, by secondary adnations and various textural modi fications. They receive their vascular bundles typically as branches of the lateral ones of the leaf-trace.
5. The lateral portions of the primitive leaf therefore represent in potential the ligule, the ochrea, the margins of sheathing peti oles and stipules, but they are often incorporated with the other portions as the wings of petioles and as lateral basal portions of leaf-blades.