644 SCALE-LEAVES, FOLIAGE-LEAVES, FLORAL-LEAVES. 



is to say, there are flowers in which the outer lower stamens are flat, leafy ex- 

 pansions, whilst the upper are i-educed to anthers and filamentous supports. Of 

 course this is only a view which it would be unwise to insist upon after the 

 foregoing strictures. 



This supposition does not exclude the fact that in many instances a single 

 whorl of carpels is developed, and that the carpels of this whorl not only form 

 the capsule, but that at the same time ovules may arise from them. At one 

 time the teeth of the margins of these carpels become ovules; at another, whole 

 segments of a leaf are metamorphosed into ovules; again, in another instance, 

 groups of cells have given rise to ovules over the midribs of the carpels; and 

 lastly, innumerable ovules may have developed from the whole inner surface of 

 the carpels. 



The internal structure of the cavity of the ovary is still further complicated 

 by the fact that the end of the axis in one case rises up like a hemisphere or 

 truncated column in the centre of the capsule, while in other instances the end 

 of the axis is hollowed into a pit, and sometimes even deeply excavated. In 

 consequence of these manifold arrangements, very different relations between carpels 

 and axis naturally follow, and the most various constructive plans result, which, 

 however, will be more suitably discussed in the second volume when considering 

 the individual families, especially the Primulacese and Onagracese. 



In whatever way the ovules may be explained, they exhibit a gi-eat agreement 

 in structure. In them may be distinguished the nucellus {nueleus), surrounded 

 by two, or less frequently by only one coat (integumentum), and also the portion 

 by which the ovule is connected with its substratum, the placenta. Usualty this 

 has the form of a stalk or filament (funiculus), and then the ovules appear, as it 

 were, suspended in the interior of the ovary. When the ovule is straight, and is 

 a direct continuation of the funicle, it is called orthotropous; if the straight ovvde 

 is hung on a thread-like support, but reversed, and more or less fused with the 

 support, it is said to be inverted or anatropou8\ when it is curved, the designation 

 campylotropous is used. The coats do not completely inclose the ovules, but at 

 one pole a spot which bears the name of micropyle is left uncovered. 



As already remarked, the style corresponds to a leaf-stalk as regards its 

 position and relation to the other portions of the pistil. In the one-leaved pistil 

 its foi-m frequently resembles a leaf-stalk, especially in papilionaceous plants. 

 If the ovary of a one-leaved pistil be regarded as arising from the sheathing 

 portion, and the style from the stalk of a leaf, it will be easily conceived that 

 the style appears to be aflixed to one side of the ovary. The lateral position 

 of the style can be clearly understood if we imagine that the sheathing portion 

 of the ovary is swollen up like a vesicle, as it is on the foliage-leaves of Umbel- 

 liferas, or that it bears large stipules as in the Cinquefoil (Potentilla). In the 

 one-leaved ovary of the cinc^uefoils the style in fact is not seen to spring from 

 the apex of the ovary, but looks as if it had grown out laterall}- from its capsule. 

 In pistils which are built up of many carpels arranged in a whorl, and having 



