cells; sometimes they are twice the number; occasionally they resemble little pores or holes below the summit, as in the Antirrhinum.

Examples. Digitalis, Primula, Rhododendron.

XX. AMPHISARCA. (Amphisarca, Desv.)

Many-celled, many-seeded, superior, indehiscent; indurated or woody externally, pulpy internally.

Examples. Omphalocarpus, Adansonia, Crescentia.

B. Pericarpium fleshy.

XXI. TRYMA. (Tryma, Watson.)

Superior, by abortion one-celled, one-seeded, with a two-valved indehiscent endocarpium, and a coriaceous or fleshy valveless sarcocarpium.

Example. Juglans.

XXII. NUCULANIUM. (Nuculanium, Rich.; Bacca, Desvaur.)

Two or more celled, few- or many-seeded, superior, indehiscent, fleshy, of the same texture throughout, containing several seeds, improperly called nucules by the younger Richard. This differs scarcely at all from the berry, except in being superior.

Examples. Grape, Achras.

XXIII. HESPERIDIUM. (Hesperidium, Desv. Rich.)

Many-celled, few-seeded, superior, indehiscent, covered by a spongy separable rind; the cells easily separable from each other, and containing a mass of pulp, in which the seeds are embedded. The pulp is formed by the cellular tissue, which forms the lining of the cavity of the cells: this cellular tissue is excessively enlarged and succulent, is filled with fluid, and easily coheres into a single mass. The external rind is by M. De Candolle supposed to be an elevated discus of a peculiar kind, analogous to that within which the fruit of Nelumbium is seated; and perhaps its separate texture and slight connexion with the cells of the fruit seem to favour this supposition. But it is difficult to reconcile with such a hypothesis the continuity of the rind with the style and stigma, which is a sure indication of the identity of their origin; and it is certain that the shell of the ovarium and the pericarpium are the same. The most correct explanation of this structure is to consider the rind a union of the epicarp and sarcocarp, analogous to that of the drupa.

Example. Orange.

Sect. 2. Fruit inferior.
A. Pericarpium dry.

XXIV. GLANS. (Glans, Linn., Desv.; Calybio, Mirb.; Nucula, Desvaux.) fig. 166.

One-celled, one- or few-seeded, inferior, indehiscent, hard, dry; proceeding from an ovarium containing several cells and several seeds, all of which are abortive but one or two; seated in that kind of persistent involucre called a cupule. The pericarpium is always crowned with the remains of the teeth of the calyx; but they are exceedingly minute, and are easily overlooked. Sometimes the gland is solitary, and quite naked above, as in the common oak; sometimes there is more than one completely enclosed in the cupule, as the beech and sweet chestnut.

Examples. Quercus, Corylus, Castanea.

XXV. CYPSELA. (Akena, Necker; Akenium, Rich.; Cypsela, Mirb.; Stephanoum, Desv.) fig. 149. 150.

One-seeded, one-celled, indehiscent, with the integuments of the seed not cohering with the endocarpium; in the ovarium state evincing its compound nature by the presence of two or more stigmata; but nevertheless unilocular and having but one ovulum. Such is the true structure of the Achenium; but as that term is often applied to the simple superior fruits, called Nux by Linnæus, I have thought it better, in order to avoid confusion, to adopt the name Cypsela.

Examples. All Compositæ.

XXVI. CREMOCARPIUM. (Cremocarpium, Mirb.; Polakenium, or Pentakenium, Rich.; Carpadelium, Desv.) fig. 155. 160, 161.

Two to five-celled, inferior; cells one-seeded, indehiscent, dry, perfectly close at all times; when ripe separating from a common axis. M. Mirbel confines the application of Cremocarpium to Umbelliferæ: but it is better to let it apply to all fruits which will come within the above definition. It will then be the same as Richard's Polakenium, excluding those forms in which the fruit is superior. The latter botanist qualifies his term Polakenium according to the number of cells of the fruit: thus when there are two cells it is diakenium, three triakenium, and so on. M. De Candolle calls the half of the fruit of Umbelliferæ mericarp.

Examples. Umbelliferæ, Aralia, Galium.

XXVII. DIPLOTEGIA. (Diplotegia, Desv.)

One- or many-celled, many-seeded, inferior, dry, usually bursting either by pores or valves. This differs from the capsule only in being adherent to the

XXVIII. POMUм. Apple, or Pome. (Melonidium, Rich.; Pyridium, Mirb. ; Pyrenarium, Desvaur; Antrum, Mænch.) fig. 167.

Two or more celled, few-seeded, inferior, indehiscent, fleshy; the seeds distinctly enclosed in dry cells, with a bony or cartilaginous lining, formed by the cohesion of several ovaria with the sides of the fleshy tube of a calyx, and sometimes with each other. These ovaria are called parietal by M. Richard. Some forms of Nuculanium and this differ only in the former being distinct from the calyx.

Examples. Apple, Cotoneaster, Cratægus.

XXIX. PEPO. (Peponida, Rich.)

One-celled, many-seeded, inferior, indehiscent, fleshy; the seeds attached to parietal pulpy placentæ. This fruit has its cavity frequently filled at maturity with pulp, in which the seeds are embedded; their point of attachment is, however, never lost. The cavity is also occasionally divided by folds of the placenta into spurious cells, which has given rise to the belief that in Pepo macrocarpus there is a central cell, which is not only untrue but impossible. Examples. Cucumber, Melon, Gourd.

XXX. BACCA. Berry. (Bacca, L.; Acrosarcum, Desvaux.) fig. 162.

Many-celled, many-seeded, inferior, indehiscent, pulpy; the attachment of

the seeds lost at maturity, when they become scattered in the substance of the pulp. This is the true meaning of the term berry; which is, however, often otherwise applied, either from mistaking nucules for seeds, or from a misapprehension of the strict limits of the term.

Example. Ribes.

XXXI. BALAUSTA. (Balausta, Officin. Rich.)

Many-celled, many-seeded, inferior, indehiscent; the seeds with a pulpy coat, and attached distinctly to their placenta. The rind was called Malicorium by Ruellius.

Example. Pomegranate.

CLASS IV. Collective Fruits.

ANTHOCARPI.

Fruit of which the principal characters are derived from the thickened floral

envelopes.

XXXII. DICLESIUM. (Dyclesium, Desvaux; Scleranthum, Manch; Cataclesium, Desvaux; Sacellus, Mirb.)

Pericarpium indehiscent, one-seeded, enclosed within an indurated perian

thium.

Examples. Mirabilis, Spinacia, Salsola.

XXXIII. SPHALEROCARPUM.

authors.)

(Sphalerocarpum, Desv.; Nux baccata of

Pericarpium indehiscent, one-seeded, enclosed within a fleshy perianthium.
Examples. Hippophäe, Taxus, Blitum, Basella.

XXXIV. SYCONUS. (Syconus, Mirb.)

A fleshy rachis, having the form of a flattened disk, or of a hollow receptacle, with distinct flowers and dry pericarpia.

Examples. Ficus, Dorstenia, Ambora.

XXXV. STROBILUS. Cone. (Conus, or Strobilus, Rich., Mirb.; Galbulus, Gærtn.; Arcesthide, Desvaur; Cachrys, Fuchs; Pilula, Pliny.) fig. 168.

An amentum, the carpella of which are scale-like, spread open, and bear naked seeds; sometimes the scales are thin, with little cohesion; but they often are woody, and cohere into a single tuberculated mass.

The Galbulus differs from the Strobilus only in being round, and having the heads of the carpella much enlarged. The fruit of the Juniper is a Galbulus, with fleshy coalescent carpella. Desvaux calls it Arcesthide. Example. Pinus.

A spike or raceme converted into a fleshy fruit by the cohesion, in a single mass, of the ovaria and floral envelopes.

Examples. Ananassa, Morus, Artocarpus.

R

172. Seed of a Garden Bean.

173. The same, after germination has just begun, and the testa is thrown off. 174. Fruit of Mirabilis Jalapa, with the embryo commencing the act of germination by protruding the radicle. 175. The same, disentangled from the pericarp, and become a young plant. 176. A section of the seed of Sterculia, with the embryo inverted in the midst of albumen. 177. The embryo of Pinus, taken out of the seed, to show its numerous cotyledons. 178. The same, after germination has advanced a little. 179. Seed of Oxalis, with the revolute elastic epidermis of the testa. 180. Seed of Salsoia radiata divided vertically, and showing the annular dicotyledonous embryo, rolled round the albumen. 181. Embryo of the same, taken out of the seed. 182. Section of the seed of Cyclamen, showing the transverse embryo lying in the midst of albumen. 183. Section of the fruit of a Grass, with the lateral embryo at the base. 184. The same, with germination just beginning. 185. The same, after germination is completed, and the monocotyledonous embryo become a young plant. 186. Section of seed of Scirpus, with germination begun; the solitary cotyledon is retained within the testa, the plumule and radicle are growing beyond it. 187. Section of a Grass seed germinating; the plumula is directed upwards like a slender horn; the cotyledon is at its base, adhering to the albumen. 188. Seed of Commelina germinating; the cauliculus is protruded, is emitting radicles from its end, and has pushed aside the lid called embryotega.

As the fruit is the ovary arrived at maturity, and is therefore subject to the same laws of structure as the latter; so is the seed the ovule in its most perfect and finally organised state, and constructed upon exactly the same plan as the ovule. But as the fruit, nevertheless, often differs from the ovary in the suppression, or addition, or modification of certain portions, so is the seed occasionally altered from the precise structure of the ovule, in consequence of changes of like nature.

The seed is a body enclosed in a pericarp, is clothed with its own integuments, and contains the rudiment of a future plant. It is the point of developement at which vegetation stops, and beyond which no increase, in the same direction with itself, can take place. In a young state it has already been spoken of under the name of ovule; to which I also refer for all that relates to the insertion of seeds.

That side of a seed which is most nearly parallel with the

axis of a compound fruit, or the ventral suture or sutural line of a simple fruit, is called the face, and the opposite side the back. In a compound fruit with parietal placenta, the placenta is to be considered as the axis with respect to the seed; and that part of the seed which is most nearly parallel with the placenta, as the face. Where the raphe is visible, the face is indicated by that.

When a seed is flattened lengthwise it is said to be compressed, when vertically it is depressed; a difference which it is of importance to bear in mind, although it is not always easy to ascertain it: for this purpose it is indispensable that the true base and apex of the seed should be clearly understood. The base of a seed is always that point by which it is attached to the placenta, and which receives the name of hilum the base being found, it would seem easy to determine the apex, as a line raised perpendicularly upon the hilum, cutting the axis of the seed, ought to indicate the apex at the point where the line passes through the seed-coat; but the apex so indicated would be the geometrical, not the natural apex: for discovering which with precision in seeds, the natural and geometrical apex of which do not correspond, another plan must be followed. If the skin of a seed be carefully examined, it will usually be found that it is composed in great part of lines representing rows of cellular tissue, radiating from some one point towards the base, or, in other words, of lines running upwards from the hilum and meeting in some common point. This point of union or radiation is the true apex, which is not only often far removed from the geometrical apex, but is sometimes even in juxtaposition with the hilum, as in mignionette: in proportion, therefore, to the obliquity of the apex of the seed will be the curve of its axis, which is represented by a line passing through the whole mass of the seed from the base to the apex, accurately following its curve. If the lines above referred to are not easily distinguished, another indication of the apex resides in a little brown spot or areola, hereafter to be mentioned under the name of chalaza. Where there is no indication either externally or internally of the apex, it may then be determined geometrically.