130

ORIGIN OF THE PLACENTA.

pels of which the compound ovary was formed originally met in the center and developed placentas from their margins in the same manner as in ordinary axile placentation, but that subsequently the walls of the ovary grew more rapidly than the dissepiments, so that the connection between them was soon destroyed; and that from this cause, and also from the great subsequent development of the placenta, the septa ultimately became almost or quite broken up, so that the placenta was left free in the cavity of the ovary. This theory is strengthened by the fact that in several of the Caryophyllacea we often find dissepiments in the young ovary, and even traces of these at the lower part of the mature ovary; hence it may be concluded that these are the remains of dissepiments which have become ruptured on account of the unequal development of the parts of the ovary. In the Primrose, however, and many other plants which have a free central placenta, no traces of dissepiments can be found at any period of the growth of the ovary. The formation of such a free central placenta can not therefore be well explained upon the marginal theory, as the carpels have never had any connection with it except at their bases. Hence this kind of placentation has been supposed by many botanists not to be formed from the carpels at all, but to be a prolongation of the axis, which bears ovules, instead of buds as is the case with branches. It seems most probable therefore that the origin of the free central placenta is sometimes from the axis, as in the Primrose; and that at others, as in the Caryophyllaceæ, it is originally axile, and becomes ultimately free by the obliteration of the dissepiments.

(2) The Style.-The style usually arises from the geometrical summit of the ovary of which it is a continuation in an upward direction: it is then termed apicilar or apical. In other cases the apex of the ovary becomes inflected toward the side or base, from the carpel or carpels

of which it is formed being folded like ordinary leaves in reclinate vernation; the style then becomes lateral as in the Strawberry (Fig. 161), or basilar as

in Alchemilla (Fig. 162).

The style is generally directly continuous with the ovary, in which case it is more or less persistent, and then forms a more or less evident part of the fruit (page 135); at other times, however, the style always falls off after the process of fertilization is completed, in which case it is said to be deciduous, and FIG. 161.-One of the has no connection with the fruit.

FIG. 161.

FIG. 162.

carpels of the Strawberry with a lateral style. FIG. 162.-Carpel of Alchemilla with a basilar style. The stigma is capitate.

When the style is basilar or lateral, and the ovary to which it is attached more or less imbedded in the thalamus, it frequently appears to spring from the latter part; such an arrangement is called a gynobase, and the ovary is said to be gynobasic. Thus, in the Labiata and Boraginaceæ, the four ovaries are free, but the styles become connected and form a central column, which appears therefore to be a prolongation of the thalamus.

When two or more styles are united into one body, this is termed a compound style. This adhesion may take place either entirely, when the style is termed simple, or, more properly, entire; or the union is more or less incomplete as we proceed toward its apex, and corresponding terms are used accordingly. These terms are similar to those previously mentioned in describing the degrees of division in the other parts of the plant: thus the style is said to be cleft, partite, etc., according to the depth to which it is divided, and further characterized as bifid, trifid, bipartite, tripartite, etc., according to the number of its divisions.

The style is also subject to variations of form, as cylindrical, filiform, etc.; or when flattened and colored like

a petal, as in the Iris (Fig. 163, sty), it is said to be petaloid. Again, the surface of the style may be either smooth, or covered in various ways with glands or hairs. These hairs when situated on the style frequently serve the purpose of collecting the pollen as it is discharged from the anther, and are hence termed collecting hairs (Fig. 165, pc). In certain natural orders, as the Goodeniaceæ and the Lobeliaceæ, the hairs form a little ring below the stigma (Fig. 164, i), to which the term of indusium has been given.

FIG. 163-Pistil of a species of Iris 0. Ovary. sty. Petaloid styles. stig. Stigmas.

(3) The Stigma.-The stigma has been already described (page 122) as being connected with the placenta by means of the conducting tissue of the style; hence it may be considered as a portion of the placenta prolonged up

ward, but differing from it in not bearing ovules.

The stigmas of a syncarpous pistil are generally oppo

site to the cells of the ovary, and alternate with the dissepiments, but it sometimes happens that half the stigma of one carpel unites with a similar half of that of the adjoining carpel, and thus it becomes alternate with the cells, and opposite to the dissepiments, which are here, however, imperfect (Fig. 158).

The term stigma is only properly applied to that portion of the style which is destitute of epidermis, and which secretes

рс

FIG. 165.

FIG. 164. FIG. 164-Upper part of the style and stigma of Leschenaultia formosa. t. Style. s. Stigma. i. Indusium. FIG. 165.-Upper_part of the style, t, of a Composite Plant dividing into two branches, which are covered above by collecting hairs, pc. s. True stigma.

the stigmatic fluid; but it is often improperly given to

DIVISIONS OF THE STIGMA.

133

mere divisions of the style. Thus, in the species of Iris (Fig. 163), the three petaloid portions of the style are in descriptive botany commonly termed petaloid stigmas; whereas the stigma is properly confined to a little transverse space, stig, near the apex of each division.

In a syncarpous pistil the stigmas may be either united or distinct; in the latter case, instead of looking upon these separate parts as so many distinct stigmas, it is usual to describe them as if they were portions of but one; thus we speak of the stigma as bifid, trifid, etc., or as bilobate, trilobate, etc., according to the number and character of its divisions. Thus the term lobe is usually applied when the divisions are thick; or when these are flattened and strap-shaped, as in the Compositæ, the stigma is fissured or cleft; or when flattened into plates or bands they are termed lamella. The number of these divisions in the majority of instances corresponds to the number of carpels of which the pistil is composed; and, if the compound ovary of the latter is two or more celled, the number of cells will generally correspond also to the divisions of the stigma.

IV. The Thalamus.-The extremity of the peduncle or pedicel, or any part of the axis upon which the parts of a solitary flower are arranged, has been variously distinguished by botanists as the thalamus, receptacle, and torus. The use of these names indifferently has led to much confusion; and the uncertainty is still further increased in consequence of the terms receptacle and torus being also sometimes applied in a different sense. Thus, that of receptacle is employed in a special manner, as already mentioned (page 82); while the term torus is used by some botanists as synonymous with disk. To prevent confusion, therefore, it would be far better to limit the terms receptacle and torus to their special applications, and use the term thalamus only as defined above, and as it is employed in this work.

134

THE THALAMUS.

In the majority of plants the thalamus is a little flattened surface or point, but in other plants it becomes. much enlarged, and then assumes a variety of appearances, and thus modifies to a considerable extent the form of the flower. Thus in the plants of the order Magnoliaceæ generally, the thalamus is cylindrical (Fig. 166, a);

FIG. 166.-Central part of the flower of the Tulip-tree (Liriodendron tulipifera). The thalamus, a, is more or less cylindrical. c, c. Carpels. e, e. Stamens. FIG. 167.-Section of the flower of the Strawberry. The thalamus is nearly hemispherical, and bears a number of separate carpels on its upper portion. FIG. 168.-Section of the fruit of the Raspberry, showing the conical thalamus, 7.

in the Raspberry (Fig. 168, 7) and species of Ranunculus it is conical; in the Strawberry (Fig. 167), hemispherical; in Nelumbium (Fig. 169, thal), it is a large tabular expansion, in which there are a number of cavities containing

thal

the separate carpels. In the Rose it forms a concavity upon which the carpels are placed.

In the Primulaceæ, Santalaceæ, and in all cases where the placenta is free from the wall of the ovary from its earliest appearance, the thalamus becomes prolonged into FIG. 169.-thal. Thalamus of the cavity of the ovary and forms the placenta (Fig. 160). At other times the thalamus becomes prolonged beyond the ovary, as in the Geraniacea and Umbelliferæ; this prolongation is termed a carpophore. In the species of Geranium (Fig.