An introduction to physiological and systematical botany/Chapter 24

From Wikisource
Jump to: navigation, search



I proceed to a compendious view of the Linnæan Classes and Orders, which will serve to illustrate many things in the preceding pages.

Class 1. Monandria. Stamen 1.

This contains only two Orders.

1. Monogynia. Style 1. Here we find the beautiful exotic natural order called Scitamineæ, consisting of Cardamoms, Ginger, Turmerick, &c., hitherto a chaos, till Mr. Roscoe, in a paper printed in the 8th vol. of the Linnæan Society's Transactions, reduced them to very natural and distinct genera by the form of the filament. See Exot. Bot. t. 102, 103, 1068.

Salicornia, Engl. Bot. t. 415 and 1691, and Hippuris, t. 763, are British examples of Monandria Monogynia.

Valeriana (Class 3) has some species with one stamen.

2. Digynia. Styles 2. Contains Corispermum, Fl. Græc. t. 1, Blitum, Curt. Mag. t. 276, and a few plants besides.

CLASS 2. Diandria. Stamens 2.—Orders 3.

1. Monogynia. This, the most natural and numerous Order, comprehends the elegant and fragrant Jasmineæ, the Jasmine, Lilac, Olive, &c, also Veronica, Engl. Bot. t. 2, 1027, 623, 783, &c.—and a few labiate flowers with naked seeds, as Salvia, Engl. Bot. t. 153, 154, Rosemary, &c., natural allies of the 14th class; but having only two stamens, they are necessarily ranged here in the artificial system.

2. Digynia consists only of Anthoxanthum, a grass, Engl. Bot. t. 647, which for the reason just given is separated from its natural family in the third class.

3. Trigynia has only Piper, the Pepper, a large tropical genus.

CLASS 3. Triandria. Stamens 3.—Orders 3.

1. Monogynia. Valeriana, Engl. Bot. t. 698, 1591 and 1531, is placed here because most of its species have three stamens. See Class 1. Here also we find the sword-leaved plants, so amply illustrated in Curtis's Magazine, Iris, Gladiolus, Ixia, &c., also Crocus, Engl. Bot. t. 343, 344, 491, and numerous grass-like plants, Schœnus, Cyperus, Scirpus, see Fl. Græc. v. 1, and Engl. Bot. t. 950, 1309, 542, 873, &c.

2. Digynia. This important Order consists of the true Grasses; see p. 127. Their habit is more easily perceived than defined; their value, as furnishing herbage for cattle, and grain for man, is sufficiently obvious. No poisonous plant is found among them, except the Lolium temulentum, Engl. Bot. t. 1124, said to be intoxicating and pernicious in bread. Their genera are not easily defined. Linnæus, Jussieu, and most botanists pay regard to the number of florets in each spikelet, but in Arundo this is of no moment. Magnificent and valuable works on this family have been published in Germany by the celebrated Schreber and by Dr. Host. The Fl. Græca also is rich in this department, to which the late Dr. Sibthorp paid great attention. Much is to be expected from scientific agriculturists; but Nature so absolutely, in general, accommodates each grass to its own soil and station, that nothing is more difficult than to overcome their habits, insomuch that few grasses can be generally cultivated at pleasure.

3. Trigynia is chiefly composed of little pink-like plants, or Caryophylleæ, as Holosteum, Engl. Bot. t. 27.

Tillæa muscosa, t. 116, has the number proper to this order, but the rest of the genus bears every part of the fructification in fours. This in Linnæan language is expressed by saying the flower of Tillæa is quadrifidus[1], four-cleft, and T. muscosa excludes, or lays aside, one fourth of the fructification.

Class 4. Tetrandria. Stamens 4.—Orders 3.

1. Monogynia. A very numerous and various Order, of which the Proteaceæ make a conspicuous part, consisting of Protea, Banksia, Lambertia, Embothrium, &c. See Botany of New Holland, t. 710. Scabiosa, Engl. Bot. t. 659; Plantago, t. 1558, 1559, remarkable for its capsula circumscissa, a membranous capsule, separating by a complete circular fissure into two parts, as in the next genus, Centunculus, t. 531; Rubia, t. 851, and others of its natural order, of whose stipulation we have spoken p. 219, are found here, and the curious Epimedium, t. 438.

2. Digynia. Buffonia, t. 1313.

Cuscuta, placed here by Linnæus, is best removed to the next class.

3. Tetragynia. Ilex, t. 496, a genus sometimes furnished with a few barren flowers, and therefore removed by Hudson to the 23d class, of which it only serves to show the disadvantage; Potamogeton, t. 168, 376, and Ruppia, t. 136, are examples of this Order. They all have sessile stigmas.

Class 5. Pentandria. Stamens 5. A very large class.—Orders 6.

1. Monogynia. One of the largest and most important Orders of the whole system. The genera are enumerated first artificially, according to the corolla being of one petal or more, or wanting; inferior or superior; with naked or covered seeds; but stand in the system according to their affinities, and compose some natural orders; as Asperifoliæ, rough-leaved plants, which have a monopetalous inferior corolla, and four naked seeds, with always more or less of spinous bristles or callous asperities on their foliage; see Borago, Engl. Bot. t. 36, Lycopsis, t. 938, and Echium, t. 181. Next comes that most elegant tribe of spring plants denominated Preciæ by Linnæus, Primula, t. 46, Cyclamen, t. 548, the charming alpine Aretia, and Androsace, Curt. Mag. t. 743. These are followed

by another Linnæan order, nearly akin, called Rotaceæ, from the wheel-shaped corolla, Hottonia, Engl. Bot. t. 364, Lysimachia, t. 761.—Convolvulus and Campanula, two large well-known genera, come afterwards; then Lobelia, t. 140, Impatiens, t. 937, and Viola, t. 619, 620, brought hither from the abolished Linnæan order Syngenesia Monogamia. The Luridæ follow, so called from their frequently dark, gloomy aspect, indicative of their narcotic and very dangerous qualities; as Datura, t. 1288, Hyoscyamus, t. 591, Atropa, t. 592, and Nicotiana, or Tobacco. In a subsequent part we meet with the Vine, Currant and Ivy, and the Order finishes with some of the natural family of Contortæ, so called from their oblique or twisted corolla, and which are many of them very fine plants, as Vinca, t. 514, 917. They often abound with milky juice, generally highly acrid; but Dr. Afzelius met with a shrub of this order at Sierra Leone, the milk of whose fruit was so sweet, as well as copious, as to be used instead of cream for tea. This is certainly what no one could have guessed from analogy. Gardenia is erroneously reckoned a contorta by Linnæus.

2. Digynia begins with the remainder of the Contortæ; then follow some incomplete flowers, as Chenopodium, t. 1033, Beta, t. 285, and afterwards the fine alpine genus of Gentiana, t. 20, 493, 896, famous for its extreme bitterness and consequent stomachic virtues.

The rest of the Order consists of the very natural Umbelliferous family, characterized by having five superior petals, and a pair of naked seeds, suspended vertically when ripe from the summit of a slender hair-like receptacle. Of the inflorescence of this tribe, and the difficulties attending their generic distinctions, we have spoken p. 309. In Eryngium, t. 718 and 57, the umbel is condensed into a capitulum, or conical scaly head, showing an approach towards the compound flowers, and accompanied, as Jussieu observes, by the habit of a Thistle. Lagoecia is justly referred to this natural order by the same writer, though it has only a solitary seed and style. The Umbelliferæ are mostly herbaceous; the qualities of such as grow on dry ground are aromatic, while the aquatic species are among the most deadly of poisons; according to the remark of Linnæus, who detected the cause of a dreadful disorder among horned cattle in Lapland, in their eating young leaves of Cicuta virosa, Engl. Bot. t. 479, under water.

Botanists in general shrink from the study of the Umbelliferæ, nor have these plants much beauty in the eyes of amateurs; but they will repay the trouble of a careful observation. The late M. Cusson of Montpellier bestowed more pains upon them than any other botanist has ever done; but the world has, as yet, been favoured with only a part of his remarks. His labours met with a most ungrateful check, in the unkindness, and still more mortifying stupidity, of his wife, who, on his absence from home, is recorded to have destroyed his whole herbarium, scraping off the dried specimens, for the sake of the paper on which they were pasted!

3. Trigynia is illustrated by the Elder, the Sumach or Rhus, Viburnum, &c., also Corrigicla, Engl. Bot. t. 668, and Tamarix, t. 1318, of which last one species, has 10 stamens.

4. Tetragynia has only Evolvulus, nearly allied to Convolvulus, and the elegant and curious Parnassia, t. 82.

5. Pentagynia contains Statice, t. 226, 102, and 328, a beautiful maritime genus, with a kind of everlasting calyx. The Flora Græca has many fine species. Linum or Flax follows; also the curious exotic Aldrovanda, Dicks. Dr. Pl. 30; Drosera, Engl. Bot. t. 8679; the numerous succulent genus Crassula; and the alpine Sibbaldia, t. 897, of the natural order of Rosaceæ.

6. Polygynia. Myosurus, t. 435, a remarkable instance of few stamens (though they often exceed five) to a multitude of pistils.

Class 6. Hexandria. Stamens 6. Orders 6.

1. Monogynia. This, as usual, is the most numerous. The Liliaceous family, with or without a spatha, called by Linnæus the nobles of the vegetable kingdom, constitute its most splendid ornament. The beautiful White Lily is commonly chosen by popular writers to exemplify the stamens and pistils. The less ostentatious genus of Juncus or Rush, which soon follows, is more nearly allied to the Lilies than a young botanist would suppose. Near it stand several genera which have little affinity to each other, and of these Capura is a mistake, having been made out of a specimen of Daphne indica, which chanced to have but six stamens.

2. Digynia has but few genera. The valuable Oryza, Rice, of which there now seems to be more than one species, is the most remarkable. It is a grass with six stamens.

3. Trigynia. See Rumex, Engl. Bot. t. 1533, 127, &c., some species of which has separated flowers; Tofieldia, t. 536; and Colchicum, t. 133 and t. 1432.

4. Tetragynia. Petiveria alliacea, a plant the number of whose stamens is not very constant, and whose specific name is supposed to allude, not only to its garlic scent, but also to the caustic humour of the botanist whom it commemorates.

5. Hexagynia. An order in Schreber and Willdenow, contains Wendlandia populifolia of the latter; with Damasonium of the former, a genus consisting of the Linnæan Stratiotes alismoides, Exot. Bot. t. 15.

6. Polygynia. Alisma only—Engl. Bot. t. 837, 775, &c.

Class 7. Heptandria. Stamens 7. Orders 4.

1. Monogynia. Trientalis, Engl. Bot. t. 15, a favourite plant of Linnæus; and Æsculus, the Horse Chesnut. Several genera are removed to this order by late writers.

2. Digynia. Limeum, an African genus, only.

3. Tetragynia. Saururus, a Virginian plant. Aponogeton, placed here by Linnæus, is now properly removed to Dodecandria. It is an East Indian and Cape aquatic genus, bearing above the water white fragrant flowers in a peculiar spike, which is either solitary or double.

4. Heptagynia. Septas, a Cape plant, very nearly akin to Crassula, to which Thunberg refers it. If its character in Linnæus be constant with respect to number, it is very remarkable, having the calyx in 7 deep segments, 7 petals, 7 germens, and consequently 7 capsules.

Class 8. Octandria. Stamens 8. Orders 4.

1. Monogynia. A very various and rich order, consisting of the well-known Tropæolum or Nasturtium, whose original Latin name, given from the flavour of the plant, like Garden Cresses, is now become its English one in every body's mouth. The elegant and fanciful Linnæan appellation, equivalent to a trophy plant, alludes to its use for decorating bowers, and the resemblance of its peltate leaves to shields, as well as of its flowers to golden helmets, pierced through and through, and stained with blood. See Linn. Hort. Cliff. 143.—Epilobium, Engl. Bot. t. 838, 795, &c., with its allies, makes a beautiful part of this order; but above all are conspicuous the favourite Fuchsia, the chiefly American genus Vaccinium, t. 456, 319, &c.; the immense and most elegant genus Erica, so abundant in southern Africa, but not known in America; and the fragrant Daphne, t. 1381, of which last the Levant possesses many charming species. Acer, the Maple, is removed hither in Fl. Brit. from the 23d class.

2. Digynia has a few plants, but little known; among them are Galenia africana, and Moehringia muscosa.

3. Trigynia. Polygonum, t. 436, 509, 941, is a genus whose species differ in the number of their stamens and styles, and yet none can be more natural. Here therefore the Linnæan system claims our indulgence. Paullinia and Cardiospermum are more constant.

4. Tetragynia. Here we find the curious Paris, t. 7, and Adoxa, t. 453. Of the former I have lately received a new species, gathered by my liberal friend Buchanan among the mountains of Nepal.

Class 9. Enneandria. Stamens 9. Orders 3.

1. Monogynia. Of this the precious genus Laurus, including the Cinnamon, Bay, Sassafras, Camphor, and many other noble plants, is an example.

2. Trigynia has only Rheum, the Rhubarb, nearly related to Rumex.

3. Hexagynia. Butomus umbellatus, Engl. Bot. t. 651, a great ornament to our rivers and pools.

Class 10. Decandria. Stamens 10. Orders 5.

1. Monogynia. A numerous and fine assemblage, beginning with a tribe of flowers more or less correctly papilionaceous and leguminous, which differ very materially from the rest of that natural order in having ten stout, firm, separate stamens. See Cassia, Curt. Mag. t. 107, 633, and Sophora, t. 167; also Exot. Bot. t. 2527, and Annals of Botany, v. 1. 501.

The Ruta, Rue, and its allies, now become very numerous, follow. See Tracts on Nat. Hist. 287. Dictamnus, vulgarly called Fraxinella, is one of them. Dionæa muscipula, see p. 174, stands in this artificial order, as do the beautiful Kalmia, Rhododendron, Andromeda, Arbutus and Pyrola, Engl. Bot. t. 213, &c.

2. Digynia. Saxifraga, remarkable for having the germen inferior, half inferior, and superior, in different species, a very rare example. See Engl Bot. t. 167, 440, 663, 1009, 500, 501. Dianthus, the Pink or Carnation tribe, and some of its very distinct natural order, Caryophylleæ, conclude the Decandria Digynia.

3. Trigynia. The Caryphylleæ are here continued, as Cucubalus, t. 1577, Silene, t. 465, 1398, Arenaria, t. 189, 512, very prolific and intricate genera in the Levant. Malpighia and Banisteria, beautiful plants of the Maple family, which next occur, have no affinity to the foregoing.

4. Pentagynia. Abounds in more Caryophylleæ, as Lychnis, t. 573, and Cerastium, t. 789, 790. Cotyledon, t. 325, Sedum, t. 1319, and Oxalis, t. 762, are placed here. Some of the last genus have the filaments united at their base, and therefore should belong to the 16th class,—another defect in the artificial system.

5. Decagynia. Consists of only Neurada, with Phytolacca; the latter an irregular genus as to stamens and styles, which therefore afford good marks to discriminate the species.

Class 11. Dodecandria. Stamens 12 to 19. Orders 6.

1. Monogynia. A rather numerous and very various order, with scarcely any natural affinity between the genera. Some of them have twelve, others, fifteen or more stamens, which should be mentioned in their characters. Asarum, Engl. Bot. t. 1083, and the handsome Lythrum Salicaria, t. 1061, also the American Snow-drop-tree, Halesia, not rare in our gardens, may serve as examples of this order. Sterculia is very properly removed hither from Gynandria by Schreber and Willdenow, as its stamens are not inserted above the germen.

2. Digynia consists of Heliocarpus, a very rare American tree with a singularly fringed or radiated fruit; and Agrimonia, Engl. Bot. t. 1335. The latter might as well have been placed in the next class, with which it agrees in natural order.

3. Trigynia is chiefly occupied by Reseda, the Mignonette, t. 320, 321, and Euphorbia, t. 256, 883, &c., one of the most well defined and natural genera, of which the punicea, Ic. Pict. t. 3, is a splendid exotic species.

4. Tetragynia, in Schreber and Willdenow, consists of Calligonum, a genus illustrated by L'Heritier in the Transactions of Linn. Society, v. 1; and Aponogeton, already mentioned p. 420.

5. Pentagynia has Glinus, an insignificant genus; and Blackwellia, a doubtful one.

6. Dodecagynia is exemplified in Sempervivum, the Houseleek, Engl. Bot. t. 1320, whose styles vary from 12 to 18 or 20. Sempervivum sediforme, Jacq. Hort. Vind. t. 81, is a Sedum with a superabundance of parts in the fructification. Linnaeus confounded it with S. rupestre.

Class 12. Icosandria. Stamens 20 or more, inserted into the Calyx. Orders 3.

1. Monogynia consists of fine trees, bearing for the most part stone fruits, as the Peach. Plum, Cherry, &c., though the leaves and other parts are bitter, acrid, and, as we have already mentioned, sometimes very dangerous, owing to a peculiar essential oil, known by its bitter-almond flavour. See specimens of this family in Engl. Bot. t. 1383, 706, 841, 842. The Myrtle tribe is another natural order, comprehended chiefly under Icosandria Monogynia, abounding in a fragrant and wholesome aromatic oil. These are plentiful in New Holland. See Tr. of Linn. Soc. v. 3. 255, also Exot. Bot. t. 42, 59, and 84. Caryophyllus aromaticus, the Clove, should on every account be removed hither.

2. Pentagynia. In this order it is most convenient to include such plants as have from two to five styles, and occasionally, from accidental luxuriance only, one or two more. An example of it is the very natural family of the Pomaceæ, as Pyrus, the Apple, Pear, &c. Engl. Bot. t. 179, 350, 337; and Mespilus, t. 1523, Exot. Bot. t. 18, 85. In this family some species of the same genus have five, others three, two, or only one style, and a corresponding number of seeds. Spiræa, nearly allied to it, stands here, most of its species having five styles, though some have a much greater number; see Engl. Bot. t. 284, 960. Mesembryanthemum, a vast and brilliant exotic genus, of a succulent habit, abounding in alkaline salt, and a few genera naturally allied to it, make up the rest of the order.

3. Polygynia. An entirely natural order of genuine Rosaceous flowers, except possibly Calycanthus. Here we find Rosa, Engl. Bot. t. 187, 9902; Rubus, t. 826, 827, 716; Fragaria, t. 1524; Potentilla, t. 88, 89, 862; Tormentilla, t. 863, 864; Geum, t. 106; Dryas, t. 451; and Comarum, t. 172: all elegant plants, agreeing in the astringent qualities of their roots, bark and foliage, and in their generally eatable, always innocent, fruit. The vegetable kingdom does not afford a more satisfactory example of a natural order, composed of natural genera, than this; and Linnæus has well illustrated it in the Flora Lapponica. His genus Tormentilla, differing from Potentilla in number of petals and segments of the calyx, though retained by Jussieu, may perhaps be scarcely distinct; yet there is a difference in their habit, which has induced me to leave it for further consideration. Haller united them both with Fragaria and Comarum, which the character and habit of the latter totally forbid, and Gærtner has well suggested a mark from the smoothness of the seeds in Fragaria, (as well as Comarum) to strengthen that of its pulpy receptacle. Whatever difficulties may attend these genera, how admirably does the fruit serve us in Rosa, Rubus, Dryas and Geum, to discriminate those whose leaves, flowers, and habit all stamp them as distinct! A student cannot do better than to study this order and these genera, as an introduction to the knowledge of more obscure ones; and the beautiful plants which compose it, mostly familiar to every body, are easily obtained.

Class 13. Polyandria. Stamens numerous, inserted into the Receptacle.

1. Monogynia. The genera of this order are artificially distributed according to the number of their petals, but not so arranged in the body of the system. They form a numerous and various assemblage of handsome plants, but many are of a suspected quality. Among them are the Poppy, the Caper-shrub, the Sanguinaria canadensis, Curt. Mag. t. 162, remarkable for its orange juice, like our Celandine, Engl. Bot. t. 1581; also the beautiful genus Cistus with its copious but short-lived flowers, some of which (Engl. Bot. t. 1321) have irritable stamens; the splendid aquatic tribe of Nymphæa, &c., t. 159, 160. But the precious Nutmeg and the Tea are perhaps erroneously placed here by Linnæus, as well as the Clove; while on the other hand Cleome more properly belongs to this part of the system than to the 15th Class.

2. Digynia has principally the Pæonia, t. 1513, variable in number of pistils, and Fothergilla alnifolia, an American shrub.

3. Trigynia. Delphinium the Larkspur, and Aconitum the Monk's hood, two variable and uncertain genera as to number of pistils.

4. Tetragynia. Tetracera ought, by its name, to have constantly four pistils, but the rest of this order are very doubtful. Caryocar, whose large rugged woody nuts contain the most exquisite kernel ever brought to our tables, and which is the same plant with Gærtner's and Schreber's Rhizobolus, as the excellent Willdenow rightly judged, is not certain in number; and still less the Cimicifuga; whilst Wahlbomia is probably a Tetracera: see Willdenow.

5. Pentagynia contains chiefly Aquilegia the Columbine, and Nigella—both strictly allied to genera in the third order. Reaumuria indeed is here well placed. Some Nigellæ have ten styles.

6. Hexagynia consists of Stratiotes, Engl. Bot. t. 379; and Brasenia, a new genus of Schreber's with which I am not acquainted.

I would recommend an union of the last five orders, for the same reason that influenced me in the preceding class. They now only serve to keep natural genera asunder, the species of which not only differ among themselves as to number of pistils, but each species is often variable besides. The genera are so few that no inconvenience could arise on that account. I conceive such reforms, founded in experience not in theory, serve to strengthen the system, by greatly facilitating its application to practice.

7. Polygynia. An order for the most part natural, comprehending some fine exotic trees, as Dillenia, Exot. Bot. t. 2, 3, 92 and 93; Liriodendron, the Tulip-tree; the noble Magnolia, &c.; a tribe concerning whose genera our periodical writers are falling into great mistakes. To these succeed a family of plants, either herbaceous or climbing, of great elegance, but of acrid and dangerous qualities, as Anemone, in a single state the most lovely, in a double one the most splendid, ornament of our parterres in the spring; Atragene and Clematis, so graceful for bowers; Thalictrum, Adonis, Ranunculus, Trollius, Helleborus and Caltha, all conspicuous in our gardens or meadows, which, with a few less familiar, close this class.

Nothing can be more injudicious than uniting these two last classes, as some inexperienced authors have done. They are immutably distinct in nature and characters, whether we call the part which immediately bears the stamens in the Icosandria a calyx, with most botanists, or a receptacle, with Mr. Salisbury in the 8th vol. of the Linnaean Society's Transactions, where, among many things which I wish had been omitted, are some good remarks concerning the distinction between calyx and corolla. This the writer in question considers as decided in doubtful cases by the latter sometimes bearing the stamens, which the former, in his opinion, never really does.

Class 14. Didynamia. Stamens 2 long and 2 short. Orders 2, each on the whole very natural.

1. Gymnospermia, Seeds naked, in the bottom of the calyx, 4, except in Phryma, which has a solitary seed.—Corolla monopetalous and irregular, a little inflated at the base, and holding honey, without any particular nectary. Stamens in 2 pairs, incurved, with the style between them, so that the impregnation rarely fails. The plants of this order are mostly aromatic, and none, I believe, poisonous. The calyx is either in 5 nearly equal segments, or 2-lipped. Most of the genera afford excellent essential characters, taken frequently from the corolla, or from some other part. Thus, Perilla has 2 styles, of which it is an unique example in this class.

Mentha a corolla whose segments are nearly equal, and spreading stamens. Engl. Bot. t. 4468.

Lavandula the Lavender, and Westringia, Tracts on Natural History, 277, t. 3, have a corolla resupinata, reversed or laid on its back.

Teucrium a deeply divided upper lip, allowing the stamens and style to project between its lobes. Engl. Bot. t. 680.

Ajuga scarcely any upper lip at all, t. 77 and 489.

Lamium has the mouth toothed on each side, t. 768.

Prunella, t. 961, has forked filaments; Cleonia 4 stigmas; Prasium a pulpy coat to its seeds. These instances will suffice as clear examples of natural genera, distinguished by an essential technical character, in a most natural order.

2. Angiospermia. Seeds in a capsule, and generally very numerous. The plants of this order have the greatest possible affinity with some families in Pentandria Monogynia. Some species even vary from one class to the other, as Bignonia radicans, Curt. Mag. t. 485, and Antirrhinum Linaria, Engl. Bot. t. 653, 260, in which the irregular corolla becomes regular, and the 4 unequal stamens are changed to 5 equal ones; nor does this depend, as has been asserted, on the action of any extraneous pollen upon the stigmas of the parent plant, neither are the seeds always abortive. No method of arrangement, natural or artificial, could provide against such anomalies as these, and therefore imperfections must be expected in every system.

Class 15. Tetradynamia. Stamens 4 long and 2 short. Orders 2, perfectly natural. Flowers cruciform.

1. Siliculosa. Fruit a roundish pod, or pouch. In some genera it is entire, as Draba, Engl. Bot. t. 586, and the Honesty or Satin flower Lunaria: in others notched, as Thlaspi, t. 1659, and Iberis, t. 52; which last genus is unique in its natural order in having unequal petals. Crambe, t. 924; Isatis, t. 97; and Bunias, t. 231; certainly belong to this Order, though placed by Linnæus in the next.

2. Siliquosa. Fruit a very long pod. Some genera have a calyx clausus, its leaves slightly cohering by their sides, as Raphanus, t. 856; Cheiranthus, t. 462; Hesperis, t. 731; Brassica, t. 637; &c. Others have a spreading or gaping calyx, as Cardamine, t. 1000; Sisymbrium, t. 855; and especially Sinapis, t. 969 and t. 1677.

Cleome is a very irregular genus, allied in habit, and even in the number of stamens of several species, to the Polyandria Monogynia. Its fruit, moreover, is a capsule of one cell, not the real two-celled pod of this Order. Most of its species are fœtid and very poisonous, whereas scarcely any plants properly belonging to this Class are remarkably noxious, for I have great doubts concerning the disease called Raphania, attributed by Linnæus to the seeds of Raphanus Raphanistrum.

The Cruciform plants are vulgarly called antiscorbutic, and supposed to be of an alkalescent nature. Their essential oil, which is generally obtainable in very small quantities by distillation, smells like volatile alkali, and is of a very acrid quality. Hence the fœtid scent of water in which cabbages, or other plants of this tribe, have been boiled.

Class 16. Monadelphia. Stamens united by their filaments into one tube. Orders 8, distinguished by the number of their stamens.

1. Triandria is exemplified by Sisyrinchium, Ic. Pict. t. 9, and Ferraria, Curt. Mag. t. 144, 532, both erroneously placed by Linnæus in Gynandria. Also the singular Cape plant Aphyteia, consisting of a large fiower and succulent fruit, springing immediately from the root, without stem or leaves. On this plant Linnæus published a dissertation in 1775. Tamarindus has lately beer removed hither from the third Class, perhaps justly.

2. Pentandria. Erodium, Engl. Bot. t. 902, separated, with great propriety, from Geranium by L'Heritier; Hermannia, a pretty Cape genus, Curt. Mag. t. 307; and a few other plants, more or less akin to the Mallow tribe, compose this Order; to which also strictly belong some species of Linum, Geranium, &c. Passiflora, removed from Gynandria, belongs most unquestionably to Pentandria Trigynia, and by no means to this Class.

3. Heptandria consists only of Pelargonium of L'Heritier, an excellent genus, comprising most of the Cape Geraniums, and marked by its irregular flower, 7 stamens, and tubular nectary.

4. Octandria contains Ailonia, Curt. Mag. t. 173, named in honour of the excellent and universally respected author of the Hortus Kewensis. Pistia is, I believe justly, placed here by Schreber and Willdenow.

5. Decandria. Geranium, properly so called, Engl. Bot. t. 404, 405, 272, &c., is the principal genus here. The late Professor Cavanilles, however, in his Dissertationes Botanicæ, referred to this Order a vast number of genera, never before suspected to belong to it, as Bannisteria, Malpighia, Turræa, Melia, &c., on account of some fancied union of their filaments, perhaps through the medium of a tubular nectary; which principle is absolutely inadmissible; for we might just as well refer to Monodelphia every plant whose filaments are connected by insertion into a tubular corolla. Some species of Oxalis, see p. 424, belong to this Order; as do several papilionaceous genera, of which we shall speak under the next Class.

6. Endecandria contains only the splendid South-American genus Brownea, the number of whose stamens is different in different species.

7. Dodecandria, Stamens mostly 15, is composed of some fine plants allied to the Mallows, as Pterospermum, t. 620, Pentapetes, &c.

8. Polyandria, a very numerous and magnificent Order, comprises, among other things, the true Columniferæ or Malvaceæ, as Malva, Engl. Bot. t. 671, 754, Althæa, t. 147, Hibiscus, Spicil. Bot. t. 8, Gossypium the Cotton-tree, Alcea the Hollyhock, &c. Stately and beautiful plants of this Order, though not Malvaceæ, are Carolinea, whose angular seeds are sold in our shops by the name of Brasil nuts; Gustavia, named after the late King of Sweden, a great patron of botany and of Linnæus; Camellia, Curt. Mag. t. 42, whose splendid varieties have of late become favourites with collectors; Stuartia, Exot. Bot. t. 110; and Barringtonia, the original Commersonia, Sonnerat Voy. à la Nouv. Guinée, t. 8, 9.

Class 17. Diadelphia. Stamens united by their filaments into 2 parcels, both sometimes cohering at the base. Orders 4, distinguished by the number of their .—Flowers almost universally papilionaceous.

1. Pentandria. The only genus in this Order is Monnieria, Lamarck, t. 506, a rare little South American plant, whose natural order is uncertain. It has a ringent corolla, ternate leaves, a simple bristly pubescence, and is besprinkled with resinous dots.

2. Hexandria. Saraca, in this Order, is as little known as the Monnieria, except that it undoubtedly belongs to the leguminous family. It seems most allied to Brownea, Jonesia, Afzelia, &c. Fumaria, the only genus besides, is remarkable for the great variety of forms in its seed-vessel, whence botanists who make genera from technical characters, without regard to natural principles, have injudiciously subdivided it. See Engl. Bot. t. 588590, 943, 1471.

3. Octandria. Polygala, t. 76, is the principal genus here. America and the Cape of Good Hope abound in beautiful species of it, and New Holland affords some new genera, long confounded with this. Dalbergia is perhaps as well placed in the next Order.

4. Decandria is by far the most numerous as well as natural, Order of this Class, consequently the genera are difficult to characterize. They compose the family of proper Papilionaceæ or Leguminosæ, the Pea, Vetch, Broom, &c. Their stamens are most usually 9 in one set, with a single one separate.

The genera are arranged in sections variously characterized.

* Stamens all united, that is, all in one set. The plants of this section are really not diadelphous but monadelphous. See Spartium, Engl. Bot. t. 1339. Some of them, as Lupinus, and Ulex, t. 742, 743, have indeed the tenth stamen evidently distinguished from the rest, though incorporated with them by its lower part. Others have a longitudinal slit in the upper side of the tube, or the latter easily separates there, as Ononis, t. 682, without any indication of a separate stamen. Here therefore the Linnæan System swerves from its strict artificial laws, in compliancy with the decisive natural character which marks the plants in question. We easily perceive that character, and have only to ascertain whether any papilionaceous plant we may have to examine has 10 stamens, all alike separate and distinct, in which case it belongs to the 10th Class, or whether they are in any way combined, which refers it to the 11th.

** Stigma downy, without the character of the preceding section, for this and all the following are truly diadelphous. Very nice, but accurate, marks distinguish the genera, which are sufficiently natural. The style and stigma afford the discriminative characteristics of Orobusi, t. 1153; Pisum, t. 1046; Lathyrus, t. 670, 1108; Vicia, t. 334, 481483; and no less decisively in Ervum, t. 970, 1223, which last genus, notwithstanding the remark in Jussieu 360, "stigma nou barbatum" (taken probably from no genuine species), most evidently belongs to this section, as was first remarked in the Flora Britannica; and it is clearly distinguished from all the other genera of the section by the capitate stigma hairy all over; nor is any genus in the whole Class more natural, when the hitherto mistaken species are removed to their proper places. See Fl. Brit.

*** Legume imperfectly divided into two cells, always, as in all the following, without the character of the preceding sections. This is composed of the singular Biserrula, known by its doubly serrated fruit, of which there is only one species; the Phaca, Jacq. Ic. Har. t. 151; and the vast genus of Astragalus, Engl. Bot. t. 274, &c., lately illustrated in a splendid work by an able French botanist, Decandolle.

**** Legume with scarcely more than one seed. Of this Psoralea, Curt. Mag. t. 665; the curious Stylosanthes of Swartz; the Hallia of Thunberg; and our own Trifolium, Engl. Bot. t. 1770, 10481050, are examples. The last genus, one of the most natural as to habit and qualities, is extremely untractable with respect to botanical characters. Some species, t. 1047, 1340, 1769, have many seeds in each pod; some have not even the capitate inflorescence made a part of the generic definition. The difficulty is lessened by establishing Melilotus as a genus, with Jussieu: but the whole requires to be well reconsidered; for, if possible, so great a laxity of definition, with such glaring exceptions, should not disgrace any system.

***** Legume composed of single-valved joints, which are rarely solitary. Hedysarum, t. 96, is the most important genus of this section, and is known by its obtuse or rectangular keel. Hippocrepis, t. 31; Ornithopus, t. 369; and Scorpiurus, known in gardens by the name of Caterpillar, from its worm-like pod, are further examples. Smithia, Ait. Hort. Kew. t. 13, is remarkable for having the joints of the legume connected by means of the style, as by a thread; the stamens in 2 equal divisions, with 5 anthers to each; and a two-lipped calyx. Hedysarum vespertilionis, Jacq. Ic. Rar. t. 566, in some points approaches this genus, and more certain species are possibly latent among the numerous unsettled papilionaceous plants of India.

****** Legume of one cell, with several seeds. To this belong the genus Melilotus, if separated from Trifolium, the Indigofera, several species of which are so valuable for dyeing blue; the handsome Robinia, Curt. Mag. t. 311; Cytisus, t. 176, &c.; and Clitoria[2], Ins. of Georgia, t. 18: also Lotus, Engl. Bot. t. 925, and Medicago, t. 1616; which last is justly transferred by Willdenow from the foregoing section to this.

Papilionaceous plants are rarely noxious to the larger tribes of animals, though some species of Galega intoxicate fish. The seeds of Cytisus Laburnum have of late been found violently emetic, and those of Lathyrus sativus have been supposed at Florence to soften the bones, and cause death; we know of no other similar instances in this Class, which is one of the most abundant in valuable esculent plants. The negroes have a notion that the beautiful little scarlet and black seeds of Abrus precatorius, so frequently used for necklaces, are extremely poisonous, insomuch that half of one is sufficient to kill a man. This is totally incredible. Linnæus however asserts rather too absolutely, that "among all the leguminous or papilionaceous tribe there is no deleterious plant to be found."

Class 18. Polyadelphia. Stamens united by their filaments into more than 2 parcels. Orders 3, distinguished by the number or insertion of their stamens, which last particular Linnæus here overlooked.

No part of the Linnæan system has been less accurately defined or understood than the Orders of the 18th Class. Willdenow, aware of this, has made some improvements, but they appear to me not sufficient, and I venture to propose the following arrangement.

1. Dodecandria. Stamens, or rather Anthers, from 12 to 20, or 25, their filaments unconnected with the calyx. Of this the first example that presents itself is Theobroma, the Chocolate tree, Merian. Surin. t. 26, 63, Lamarck Encycl. t. 635. The flowers have not been seen fresh in Europe, and we only know them from drawings made in the West Indies, one of which, preserved in the Linnæan herbarium, is my authority for the following descriptions. The filaments are inserted between the long tapering segments of a 5-cleft nectary, on its outside, and each bears at its summit 4 sessile, obtuse, spreading anthers. Aublet's figure of this genus, which Schreber and Willdenow seem to have followed, represents but 2. The fruit is perhaps most properly a berry with a hard coat, whose seeds, when roasted, make chocolate. Bubroma of Schreber, Guazuma Lamarck, t. 637, confounded by Linnæus with the preceding genus, has similar filaments, but each bears 5 anthers; Jussieu and Cavanilles say 3. The fruit is a woody capsule, with 10 rows of perforations. Abroma, Jacq. Hort. Vind. v. 3. t. 1. Miller Illustr. t. 63, has 5 parcels of anthers, nearly sessile on the outside of the nectary, between its obtuse, reflexed, notched lobes. It is difficult to say how many anthers compose each parcel, for the different accounts on record are totally irreconcileable. We have found 3; the drawing sent to Linnæus represents 6; and Miller has a much greater number. Perhaps they may vary. In this uncertainty the genus in question is best placed with its natural allies in this order, with a reference to it in italics at the end of Polyadelphia Polyandria. Its fruit is a membranous winged capsule, opening at the top. Monsonia, Curt. Mag. t. 73, Lamarck, t. 638, removed by Schreber and Willdenow to Monadelphia, rather, I think, belongs to this class where Linnæus placed it. The 5 filaments, bearing each 3 long-stalked anthers, are merely inserted into a short membranous cup, or nectary, for so the analogy of the 3 preceding genera induces us to call it; and if we refer Monsonia to Monadelphia, we fall into the error of Cavanilles mentioned p. 439. Lastly, Citrus, the Orange, Lemon, &c., Lamarck, t. 639, most unquestionably belongs to this Order. Its stamens are about 19 or 20, combined variously and unequally in several distinct parcels; but those parcels are inserted into a proper receptacle, by no means into the calyx, as the character of the Class Icosandria indispensably requires. Even the number of the anthers of Citrus accords better with most plants in Dodecandria than in Icosandria, notwithstanding the title of the latter.

2. Icosandria. Stamens numerous, their filaments inserted (in several parcels) into the calyx.—To this Order Professor Willdenow properly refers Melaleuca, Exot. Bot. t. 3436, 55, 56, which had previously stood in Polyandria, botanists having only considered number and not insertion in the Orders of Polyadelphia, whence a double mistake has arisen, concerning Citrus on the one hand, and Melaleuca on the other.

3. Polyandria. Stamens very numerous, unconnected with the calyx. This Order consists of several genera. The most remarkable is Hypericum, Engl. Bot. t. 109, 12251227 &c., whose stamens are united into 3 or 5 parcels, corresponding with the number of its styles. Munchhausia is a Lagerstromia, nor does it appear to be polyadelphous at all. Linnæus seems to have intended bringing Thea into this Order.

Class 19. Syngenesia. Anthers united into a tube. Flowers compound. Orders 5.

This being truly a natural Class, its Orders are most of them equally so, though some are liable to exceptions, as will presently be explained.

1. Polygamia æ. In this each floret, taken separately, is perfect or united, being furnished with its own perfect stamens and pistil, and capable of bringing its seed to maturity without the assistance of any other floret. The Order consists of 3 sections.

* Florets all ligulate, or strap-shaped, called by Tournefort semiflosculous. These flowers are generally yellow, sometimes blue, very rarely reddish. They expand in a morning, and close towards noon or in cloudy weather. Their herbage is commonly milky and bitter. Leontodon, Engl. Bot. t. 510; Tragopogon, t. 434, 638; Hieracium, t. 349, &c.; and Cichorium, t. 539, exemplify this very natural section.

** Flowers globose, generally uniform and regular, their florets all tubular, 5-cleft, and spreading. Carduus, t. 107, 675, 973976; Onopordum, t. 977; and Arctium, t. 1228, well exemplify this. Carlina, t. 1144, does not so exactly agree with the above definition, having a flat disk; but its affinity to the other genera is indubitable. Its flattened disk and radiating coloured calyx seem contrived to imitate the radiated flowers of the following Order.

*** Flowers discoid, their florets all tubular, regular, crowded and parallel, forming a surface nearly flat, or exactly conical. Their colour is most generally yellow, in some cases pink. Santolina, t. 141; and Bidens, t. 1113, t. 1114, are genuine examples of this section: Eupatorium, t. 428, and the exotic Stæhelina, Dicks. Dr. Pl. 13, approach to the preceding one. There is however the most absolute difference between these two sections, collectively, and the first; while, on the other hand, they have considerable affinity with some of the following Orders, as will be hereafter explained.

2. Polygamia superflua. Florets of the disk perfect or united; those of the margin furnished with pistils only; but all producing perfect seed.

* Discoid, the florets of the margin being obsolete or inconspicuous, from the smallness or peculiar form of the corolla; as Artemisia, Engl. Bot. t. 338, 978, 1230; Tanacetum, t. 1229; Conyza, t. 1195; and Gnaphalium, t. 267, 1157. In the last the marginal florets are mostly 5-cleft and tubular like the rest, only wanting stamens. Caution is requisite to detect the difference between this section and the preceding Order.

** Ligulate, 2-lipped, of which Perdicium, a rare exotic genus, is the only instance.

*** Radiant, the marginal florets ligulate, forming spreading conspicuous rays; as Bellis the Daisy, t. 424; Aster, t. 87, a very numerous genus in America; Chrysanthemum, t. 601, 540; Inula, t. 1546, &c. This section seems, at first sight, a combination of the first and third sections of the former Order, but this is chiefly in the form of its corollas. It is rather an approach of that third section towards what is equivalent to becoming double in other tribes. Accordingly, the Chamomile, Anthemis nobilis, t. 980; Chrysanthemum Leucanthemum, t. 601; and some others, occasionally have their whole disk changed to ligulate white florets, destitute of stamens, and consequently abortive. Such are actually called double flowers in this Class, and very properly. Many exotic species so circumstanced are met with in gardens. A few very strange anomalies occur in this section, as already mentioned, p. 306, one Sigesbeckia having but 3 stamens, instead of 5, the otherwise universal number in the Class; and Tussilago hybrida, t. 430, as well as paradoxa of Retzius, having distinct anthers. Nature therefore, even in this most natural Class, is not quite without exceptions.

3. Polygamia frustranea. Florets of the disk, as in the preceding, perfect or united; those of the margin neuter, or destitute of pistils as well as of stamens ; only some few genera having the rudiments of pistils in their radiant florets.

This Order is, still more evidently than the last, analogous to double flowers of other Classes. Accordingly, Coreopsis is the very same genus as Bidens, only furnished with unproductive radiant florets, C. bidens of Linnæus is the same species as his B. cernua; C. coronata is his B. frondosa; and C. leucantha, B. pilosa. Some species of Coreopsis indeed have never been found without rays. Linnæus expresses his difficulties on this subject in Phil. Bot. sect. 209, but seems inclined to unite the two genera. A similar ambiguity occurs between Gorteria and Atractylis, Relhania (of the last Order) and Athanasia, and in some degree between Centaurea, Engl. Bot. t. 278, 1673, t. 56, &c., and Carduus or Serratula; only the scales of the calyx of Centaurea generally keep that genus distinct.

I should be much inclined to abolish this Order. Those of its genera which have rudiments of pistils in their radiant florets, as Rudbeckia and Helianthus, would very commodiously range with their near relations in Polygamia superflua, nor are we sure that such radiant florets are in all circumstances abortive, neither can a student often know whether they are so or not. It does not follow, from what has just been observed, that the presence of radiant florets, whether abortive or not, can never afford a generic character, provided there be no corresponding genus without them. This must be determined by experience and observation. They are indeed to be considered as a very secondary mark, the most essential in this Class being derived from the receptacle, crown of the seed, and calyx. These Gærtner has illustrated with the greatest accuracy and skill, but even these must not be blindly followed to the destruction of natural genera.

4. Polygamia necessaria. Florets of the disk furnished with stamens only, those of the margin, or radius, only with pistils; so that both are necessary to each other. This is well seen in the common Garden Marigold, Calendula, in whose calyx, when ripening seed, the naked and barren disk is conspicuous. Othonna, Curt. Mag. t. 306, 768, Arctotis, Osteospermum and Silphium, not rare in gardens, are further examples of this Order, which I believe is constant and founded in nature. We have no British specimens either of it or the following. Filago, at least as far as our Flora is concerned, belongs to Gnaphalium. See Engl. Bot. t. 946, 1193, &c.

5. Polygamia segregata. Several flowers, either simple or compound, but with united tubular anthers, and with a partial calyx, all included in one general calyx. Of these the Globe-thistle, Echinops, and Stoebe, with Seriphium and Corymbium, (which two last require to be removed hither from the abolished Linnæan Order Syngenesia Monogamia,) have only 1 floret in each partial calyx; Jungia has 3, Elephantopus 4, others more. In every case the partial calyx is distinguished from the chaffy seed-crown observable in several genera of the other Orders, (though the latter is indeed analogous to a calyx,) either by being inferior, or by the presence of a seed-crown, or feathery down, besides. See Lamarck, t. 718—723, where the plants in question are well represented.

Class 20. Gynandria, Stamens inserted either upon the style or germen. Orders 9 in Linnæus, but some alterations concerning them are necessary.

This is one of those Classes abolished by the celebrated Thunberg, and by several less intelligent writers who have followed him. The reasons which led to this measure appear to have been that Linnæus has erroneously placed in Gynandria several plants which have not the requisite character; hence that character itself has been judged ambiguous, or not founded in nature, and the system has been supposed to be simplified by overlooking it. This appears to me a great mistake. The character of the Class, taken as above, is as evident, constant and genuine as that of any other in the system. No doubt can arise, if we be careful to observe that the stamens actually grow out of the germen or style, and not merely out of any part that supports the germen; as will appear by examples.

1. Monandria. Stamen, or sessile Anther, 1 only. This contains all the beautiful and curious natural family of the Orchideæ, or Orchis tribe, except only Cypripedium, which belongs to the next Order, I am induced to consider the bulk of this family as monandrous, upon a careful review of Professor Swartz's representation of the subject, in his excellent treatise, just come to my hands in English. See Tracts relative to Botany translated from different Languages (by Mr. König), printed for Phillips and Fardon, 1805. I have already, p. 272 mentioned the glutinous nature of the pollen of these plants. This forms yellow elastic masses, often stalked, in each cell of the anther, and the cells are either parallel and close together, or removed from each other to the opposite sides of the style: which serves to connect them, just as the filament does in many Scitamineous plants, alike therefore decided to be monandrous. Such a decision with regard to those also is justified by the analogy of other species, whose cells being approximated or conjoined, properly constitute but one anther. The grand and absolute subdivision of the Orchideæ is justly founded by Dr. Swartz, after Haller, on the structure of the anther, whether it be, as just described, parallel, like that of Orchis, Engl. Bot. t. 22; Ophrys, t. 65; Diuris, Exot. Bot, t. 9, &c.; or vertical, consisting of a moveable lid on the top of the style, like Dendrobium, t. 1012; or Malaxis, Engl. Bot. t. 72. The style of the Orchideæ has been called a column, but I think that term now altogether superfluous. It is really a style, and the stigma is a moist shining space, generally concave, and situated, for the most part, in front of the style beneath the anther. In Orchis bifolia, t. 22, and others, it is just above the orifice of the spur. Concerning the nectary of these plants there has been much diversity of opinion. The calcar, spur, in Orchis, and some other genera, is acknowledged to be such, and holds abundance of honey. This spur is judged by Swartz, as well as Linnæus, a decisive generic mark of distinction, and it commonly is so; but some Indian species brought by Dr. Buchanan prove it not to be absolute. The remarkable and often highly ornamented lip, considered by Swartz as the only corolla, for he takes all the other leaves of the flower for a calyx, has, by Linnæus and others, been thought, either a part of the nectary, or, where no spur is present, the only nectary. Nor is this opinion so ill-founded as many botanists suppose; for the front of the lip evidently secretes honey in Ophrys (or Epipactis) ovata, t. 1548, and probably in others not yet attended to. Nevertheless, this lip might, like the petals of lilies, be deemed a nectariferous corolla, were it certain that all the other leaves were truly a calyx. But the 2 inner are so remarkably different from the 3 outer ones in Ophrys, t. 64, 65, 71, 383, and above all, in Stelis, Exot. Bot. t. 75, that I am most inclined to take the former for the corolla, the latter being, according to all appearance, a calyx. An insensible gradation from one to the other, of which we have pointed out other instances in treating of this subject already, occurs in Diuris, t. 8, 9; while in some Orchideæ the leaves all partake more of the habit of a calyx, and in others of a corolla. Even the lip in Thelymitra, t. 29, assumes the exact form, colour, and texture, of the rest of the flower; which proves that a dissimilarity between any of these parts is not always to be expected in the family under consideration. Vahl appears by the preface to his Enumeratio Plantarum to have removed the Scitamineæ to Gynandria, because the stamen of Canna adheres to the style. This, if constant, could only concern that genus, for the rest of the Order are in no sense gynandrous.

2. Diandria. To this Order Cypripedium, Engl. Bot. t. 1, must be referred, having a pair of very distinct double-celled anthers. See Tr. of Linn. Soc. v. 1, t. 2, 3. Here we find Forstera, so well illustrated by Professor Swartz in Sims and König's Annals of Botany, v. 1. 291, t. 6; of which genus Phyllachne, t. 5 of the same volume, is justly there reckoned a species. Of the same natural order with Forstera is Stylidium, but that having 4 anthers, belongs to the fourth Order of the present Class. Gunnera, placed by Linnæus in Gynandria Diandria, is not yet sufficiently well understood.

3. Triandria. Salacia, if Linnæus's description be right, is properly placed here; but Jussieu doubts it, nor does my dried specimen serve to remove the uncertainty. Stilago proves to be merely the barren plant of Antidesma alexiteria, and belongs to Dioecia; as Sisyrinchium and Ferraria do to Monadelphia, the tubular united stamens of the two last having been mistaken for a solid style. Rhopium of Schreber (Meborea of Aublet, t. 323,) seems therefore the only certain genus of the Order under consideration; unless Lamarck be right in referring to it Jacquin's Strumpfia, upon which I have not materials to form any opinion. The original discoverer attributes to this plant 5 stamens with united anthers; hence it found a place in the Syngenesia Monogamia of Linnæus. Lamarck merits attention, as he appears to have had an authentic specimen. See his t. 731.

4. Tetrandria. Nepenthes, of whose extraordinary kind of leaf mention is made p. 197, is the only genus of this Order in Linnæus, but very erroneously placed here, for it belongs to Dioecia Monadelphia. The Order however must be retained for the sake of Stylidium, a New Holland genus, related, as above mentioned, to Forstera. This is my Ventenatia, Exot. Bot. t. 66, 67; but another genus having previously, without my knowledge, received the latter denomination, that of Stylidium, under which I had, some time ago, communicated this genus to the French botanists[3], and which they have adopted, becomes established. See La Billardiere's excellent work on New Holland plants, where several species of it are figured.

5. Pentandria. The original genera of this Order, Ayenia, Gluta, and Passiftora, Exot. Bot. t. 28, most unquestionably have nothing to do with it, their stamens being inserted below the germen, merely on a columnar receptacle. The learned Schreber therefore justly removed them to the 5th Class.

But this Order may receive a reinforcement from the Linnæan Pentandria Digynia. Several of the Contortæ have long been thought to belong to Gynandria; see Pergularia, Ic. Pict. t. 16, and Andr. Repos. t. 184. In this genus, as well as Cynanchum and Asclepias, the pollen is produced in 5 pair of glutinous masses, exactly like the pollen of Orchideæ, from 6 glands inserted upon the stigma, so that no plants can be more certainly gynandrous. Some obscurity arises from each mass of pollen being received into a kind of bag or cell, formed by a peculiar valvular apparatus that encircles the organs of impregnation, and bears a great resemblance to filaments or stamens. The pollen however is, in the above genera, neither attached to, nor secreted by these cells or valves, but by the 5 glands, each of which is double or two-lobed, and all of them seated on that thick abrupt angular body which performs the functions of a stigma. Nor is it worth while to dispute whether this whole body be a stigma or not, with regard to the question under consideration, for it is borne by the styles, above the germen, and itself bears the anthers. I humbly conceive, however, with Linnæus and Jacquin, that as part of it, at least, receives the pollen, stigma is full as good a name for this body as Haller's term dolium, a tub! Still less is it worth while to controvert with Kölreuter the propriety of the term pollen, because the substance in question is not actually a dry powder, any more than in the Orchis tribe, or in Mirabilis, Exot. Bot. t. 23. That term is technically used for the matter which renders the seeds fertile, including its vehicle, whether the latter be capsular or glutinous, in short, whatever the appearance or texture of the whole may be. Another question remains, more immediately to our present purpose, whether these plants have 5 stamens or 10? Jacquin, who has well illustrated several of them in his Miscell. Austr. v. 1. t. 1—4, and Rottböll in a dissertation on the subject, contend for the latter, Rottböll wrote to Haller, that "finding Linnæus deaf to all that had been said, he sent him his treatise, to see whether he would persist in falsifying nature." Thus sordid underlings foment the animosities and flatter the failings of their superiors! Linnaeus judiciously suspended his opinion, and, after all, proves to be most correct. The analogies of the Orchideæ and Scitamineæ very clearly decide that each gland with its double masses of naked pollen can only be considered as one anther of 2 cells or lobes. Even Periploca græca, though not gynandrous, confirms this. Each lobe of its anthers stands, as in many Scitamineæ, on the outermost edge of the filament; thus meeting that on the adjoining filament, and in appearance constituting with it a 2-lobed anther, as the lobe of the Scitamineæ, where there is but one filament, meets its corresponding lobe by embracing the style.

6. Hexandria. Aristolochia, Engl. Bot. t. 398, a curious genus, of which there are many exotic species, is the only example of this, Pistia being removed to Monodelphia Octandria.

7. Octandria. The Scopolia of Linnæus, which originally constituted this Order, proves to be a Daphne; see Plant. Ic. ex Herb. Linn. t. 34. Cytinus however, Cavan. Ic. t. 171, a singular parasitical plant on the roots of Cistus in the south of Europe, has properly been brought hither from the Order Dodecandria, of which it originally formed the only example. The observations of Dr. Sibthorp and Mr. Ferd. Bauer confirm those of other botanists, that the anthers are 8, not 16, and that they are truly inserted upon the style.

8. Decandria is now abolished. Of the two genera which constituted it, Kleinhovia belongs to the Class Dodecandria, having 15 stamens, see Cavan Monadelph. t. 146; and Helicteres to Decandria Monogynia.

9. Dodecandria is likewise abolished.

10. Polyandria is in a similar predicament, for I am not aware of any genus that can be admitted into it. Xylopia goes with the greatest propriety to its natural allies in Polyandria Polygynia, Annona, &c., its short stamens being inserted into the receptacle below the germen. Grewia, as well as Schreber's Microcos if a good genus, belong to Polyandria Monogynia, the organs of impregnation being merely elevated on a common stalk, like those of Passiflora and Ayenia. Ambrosinia, Arum, and Calla, are all justly removed by Schreber to Monoecia, though I think, for reasons hereafter given, they are more commodiously and naturally placed in the Order Polyandria of that Class, than in the Order Monandria. Dracontium and Pothos, of the same natural family, having perfect or united flowers, the former with 7 stamens to each, the latter with 4, are undoubtedly to be referred to their corresponding Classes, Heptandria and Tetrandria. Zostera, the only remaining genus of Gynandria Polyandria in Linnæus, I have long ago ventured to remove to Monandria Monogynia; See Engl. Bot.

Class 21. -Monecia. Stamens and pistils in separate flovers, but both growing on the same individual plant. Orders 9 or 10.

Several reformers of the Linnæan system have also abolished this Class and the two following, by way of rendering that system more simple. Ten years' additional experience since the preface to the 7th volume of English Botany was written, have but confirmed my opinion on this subject. If any plants ought to be removed from these Classes, they must be such as have the structure of all the accessory parts of the flower exactly alike, (the essential parts, or stamens and pistils only, differing,) in both barren and fertile flowers; and especially such as have in one flower perfect organs of one kind, accompanied by rudiments of the other kind, for these rudiments are liable occasionally to become perfect. By this means dioecious species of a genus, as in Lychnis, Valeriana, Rumex, &c., would no longer be a reproach or inconvenience to the system. But, on the other hand, some difficulty would occasionally arise to a student, in deciding whether there were any real difference of structure between these accessory parts or not, and it might puzzle an adept to determine the question. For instance, whether the nectary in Salix, different in the barren and fertile flowers of some species, should lead us to keep that genus in Dioecia, though in other species the nectary is precisely alike in both the kinds, and occasionally an abortive germen occurs in the barren flowers, as stamens do, more rarely, in some fertile ones. Considering all this, I should refer Salix to Diandria Monogynia.

With respect to those Monoecious or Dioecious genera whose barren flowers are decidedly unlike the fertile ones, the former being in a catkin, the latter not, as Corylus, Quercus, &c., I conceive nothing more pernicious or troublesome can be attempted than to remove them to the Classes of united flowers. They meet with no allies there, but, on the contrary, form so natural an assemblage by themselves, as to be unanimously kept separate by the authors of every natural system that has appeared. But even if this were not the case, there is a most important reason for keeping them as they are, which regards the artificial system more particularly, and of which its author was well aware; they are of all plants most uncertain in the number of their stamens. Now this uncertainty is of little moment, when we have them primarily distinguished and set apart from other plants by their Monoecious or Dioecious character; because the genera being few, and the Orders constructed widely as to number of Stamens, we find little difficulty in determining any genus, which would be by no means the case if we had them confounded with the mass of the system. Even the species of the same genus, as well as individuals of each species, differ among themselves. How unwise and unscientific then is it, to take as a primary mark of discrimination, what nature has evidently made of less consequence here than in any other case! It is somewhat like attempting a natural system, and founding its primary divisions on the artificial circumstance of number of stamens.

I proceed to give some illustrations of the Orders in Monoecia.

1. Monandria. Zannichellia, Mill. Illustr. t. 77, and Aegopricon, Plant. Ic. ex Herb. Linn. t. 42, are genuine examples of this Class and Order, having a different structure in the accessory parts of their barren and fertile flowers. Artocarpus, the celebrated Bread-fruit, may likewise be esteemed so on account of a partial calyx in the barren flowers. The other amentaceous genera may most intelligibly perhaps be referred to the Order Polyandria. Chara is now removed to the first Class in the System; see Engl. Bot. t. 336.

2. Diandria. Anguria can remain here only till the proposed reformation takes place, having no difference of structure in its flowers. Lemna, so imperfectly known when Linnæus wrote, is now well understood, and, having frequently united flowers, belongs to the second Class; see Engl. Bot. t. 926, 1095, 1233.

3. Triandria. The great genus of Carex, t. 1051, 928, 993995, &c., and some other grassy plants, are found here. Typha, t. 1455t. 1457, is less clear in its structure; Sparganium, t. 744, 745, 273, is sufficiently so. Tragia, Hernandia and Phyllanthus are properly placed in this Class and Order.

4. Tetrandria. Littorella, t. 468; the valuable genera Betula, t. 1508, and Buxus, t. 1341; also the Nettle Urtica, t. 1236; are good examples of this. Morus the Mulberry, of the same natural order as the Nettle, has scarcely any difference of structure in the accessory organs of the flowers. This tree however is remarkable for being often inclined to become even dioecious in its constitution, one individual bearing most fruit when accompanied by another whose barren flowers are more effective than its own. Empleurum, Exot. Bot. t. 63, is one of those ambiguous genera which are but imperfectly monoecious.

5. Pentandria. Xanthium, Ambrosia, Nephelium, Parthenium, Iva and Clibadium all partake, more or less accurately, of the nature of compound flowers, but their anthers not being united, they could not be referred to the Class Syngenesia; particularly Xanthium and Nephelium, whose fertile flowers have no resemblance to that Class. Amaranthus, an extensive dung-hill genus in warm countries, analogous to our Chenopodium, follows next Leea is the same with Aquilicia, and belongs to Pentandria Monogynia, the former name being retained for the sake of the highly meritorious botanist and cultivator whom it commemorates. The Gourd tribe, Cucurbita, Cucumis, Bryonia, Engl. Bot. t. 439, might be brought hither from the abolished Order Syngenesia, unless it should be thought better to consider them as polyadelphous, to which I am most inclined.

6. Hexandria. Zizania, Tr. of Linn. Soc. v. 1. t. 13; and Pharus, Browne's Jamaica, t. 38, both grasses, compose this Order, to which Schreber has added Epibaterium and Pometia of Forster, as well as the splendid Guettarda, Hort. Mal. v. 4. t. 48. The latter varies from 6 to 9 in the parts of the flower, and constitutes the Order Heptandria in Linnæus, according to his usual principle, of placing such irregular plants, as much as possible, in small Classes or Orders, that they might be the more easily found.

7. Polyandria. Stamens more than 7. Ceratophyllum, Engl. Bot. t. 947, 679; Myriophyllum, t. 83, 218; and the handsome Sagittaria, t. 84, stand here at present, but the accessory parts in their two kinds of flowers are alike. Begonia, Exot. Bot. t. 101, has the number of its petals, though various in several species, always sufficiently different in the barren and fertile flowers to fix it here. The most indubitable plants of this Order are amentaceous, Quercus, Engl. Bot. t. 1342; Fagus, t. 886; Corylus, t. 723; Carpinus, Juglans, Platanus, &c.—Arum, t. 1298, Calla and Ambrosinia, all brought hither from the 20th Class, seem to me perfectly intelligible as simple monoecious flowers, the barren one, with many stamens, being superior or interior with respect to the fertile, like the generality of monoecious as well as all compound flowers, and not inferior, or, as in every simple one, exterior.

8. Monadelphia. The Fir, Pinus, so magnificently illustrated by Mr. Lambert, is very distinct in its two kinds of flowers. Each barren one consists of a naked tuft of monadelphous stamens, accompanied only by a few bracteas at the base. The fertile ones are catkins, with similar braeteas, each scale bearing on its upper side a pair of winged seeds, and on its under a leaf-like style and acute stigma; as Jussieu first, rightly I believe, suggested, though some botanists have understood these parts otherwise. Acalypha, Croton, Jatropha, Ricinus, and several others of the natural order of Euphorbiæ, acrid milky plants, form a conspicuous and legitimate part of Monoecia Monadelphia. Omphalea is justly associated with them by Schreber, though placed by Linnæus in the Order Triandria, and this alteration is the more fortunate, as one of its species is diandrous. Sterculia is best removed to the 11th Class, next to Kleinhovia.

9. Polyadelphia. If the system should be preserved in its present state, without regard to agreement or difference in the accessory parts of the barren and fertile flowers, I conceive this Order might be established for the reception of the Gourd tribe, as already hinted under the 5th Order. Their filaments are united, in 3 sets, a character much more intelligible and constant than the casual and irregular connection of their anthers, which led Linnæus to reckon them syngenesious; for they only afford an additional proof that union of anthers is, in simple flowers, neither a good natural nor artificial guide. If the monoecious and dioecious Classes be reformed according to the plan to which I have so often adverted, these plants should go to the Class Polyadelphia.

10. Gynandria is scarcely tenable, being paradoxical in its character, and the two Linnæan genera which compose it, Andrachne and Agyneia, seem most properly, even as the system stands at present, to belong to the 8th Order, to great part of which they are, moreover, naturally related.

Class 22. Dioecia. Stamens and Pistils in separate flowers, situated on two separate plants. Orders 8.

The foregoing remarks on the Orders of Monoecia apply also to those of this Class. I shall therefore only briefly mention some genera properly illustrative of each Order, more particularly specifying such as require to be placed elsewhere.

1. Monandria. Brosimum of Swartz, and Ascarina of Forster, seem, by their descriptions, to be well placed here. Pandanus (Athrodactylis of Forster) is more doubtful, not having any partial calyx or corolla to divide the stamens into separate blossoms, so that the whole may be taken either for a polyandrous or a monadelphous flower, as well as for an assemblage of monandrous ones. Najas is a good and immutable example of this Order. Of Thunberg's Phelypæa I have not materials to form a judgment.

2. Diandria. The wonderful Valisneria, already described p. 335, is a decisive example of this. Cecropia also seems unexceptionable. Of Salix, see Engl. Bot. v. 20 and 21, &c., I have already spoken, p. 471. The scales of its barren and fertile catkins are alike; its nectaries various.

3. Triandria. Elegia and Restio, hard rushy plants chiefly of the Cape of Good Hope and New Holland, appear to bewithout any difference in the accessory parts of their flowers, which is certainly the case with Empetrum, Engl. Bot. t. 526, Ruscus, t. 560, brought hither from Dioecia Syngenesia, Osyris, Excœcaria and Maba; Caturus only seeming differently constructed in this particular; but I have not been able to examine the three last.

4. Tetrandria. Trophis, Batis, and Hippophae, t. 425, are good examples of this, though Mr. Viborg is recorded by Schreber to have occasionally found united flowers intermixed with the barren ones in the last-mentioned genus. If this be usual, Hippophae must be removed to Polygamia Dioecia. The rest of the Order appear to have the accessory parts alike in both flowers, as Viscum, t. 1470.

5. Pentandria. Humulus, t. 427, is almost the only certain instance here. Spinacia, Acnida and Cannabis would be unexceptionable, but they are less absolutely dioecious, being sometimes monoecious; see p. 331. The rest of the Order is at best doubtful; nor can the pretended amentum of the barren-flowered Pistacia entitle it to a permanent place in this Class, for its fructification is truly a panicle. Clutia, more properly Cluytia, may possibly remain here. It has no business in the Order Gynandria.

6. Hexandria. No difference of structure is discernible between the barren and fertile flowers of any genus in this Order; witness Tamus, t. 91, though something to the contrary is mentioned in the Genera Plantarum of Linnæus.

7. Polyandria. Under this Order I would certainly comprehend all dioecious plants that have from 8 to any greater number of stamens, according to the example set by Linnæus himself in the last Class. The genera are exceedingly variable in this respect; and if all those the accessory parts of whose flowers are uniform were taken away, the remainder would be so few, that it is hard to say whether any would remain at all. Instances of the Order as it now stands are Populus, t. 1618, 1619; Hydrocharis, t. 808; Mercurialis, t. 559. The fertile flowers of the latter have, in some cases, a nectary or corolla of two slender leaves, not found in the barren ones, which may entitle it to a permanent place here, Carica will also probably remain. Rhodiola can scarcely be kept distinct from Sedum. Coriaria and Ailanthus, having often united flowers, are best in the 10th Class, as Euclea in the llth. I find no genera truly icosandrous here, though Schreber esteems Flacourtia and Hedycarya to be so.

8. Monadelphia. Taxus, t. 746, and perhaps Juniperus, t. 1100, also the exotic Ephedra, are legitimate examples of this Order. Spurious ones are Nepenthes, Myristica the Nutmeg, and Schreber's Xanthe, all placed by him in the now abolished Order Syngenesia, and which can only take shelter here while the Class remains as it is, for they have no difference of structure in the accessory parts of their flowers.

Class 23. Polygamia. Stamens and Pistils separate in some flowers, united in others, either on the same plant or on two or three distinct ones; such difference in the essential organs being moreover accompanied with a diversity in the accessory parts of the flowers. Orders 3.

1. Monoecia. United flowers accompanied with barren or fertile, or both, all on one plant. Atriplex, Engl. Bot. t. 261, 232, &c., is an instance of this, having the barren flowers of 5 regular spreading segments, the united ones of 2 compressed valves, which, becoming greatly enlarged, protect the seed. In several species however the flowers are none of them united, each having only stamens or only pistils. Throughout the rest of the Order, as it stands in Linnæus and Schreber, I can find no genus that has the requisite character. Some of the grasses indeed have awns to one kind of flower only, but that part is too uncertain to establish a character upon; and this family is so natural in itself, and so liable to variations in the perfecting of its flowers or florets, that there can be no doubt of the propriety of classing its genera simply by the number of their stamens and styles, which are very constant.

2. Dioecia. The different flowers on two different plants. I can scarcely find a certain instance of this, except Hippophae, already mentioned under Monoecia Tetrandria.

3. Trioecia. Of the only two genera which have ever been placed here, Ceratonia, Cavan. Ic. t. 113, belongs to Pentandria Monogynia. Ficus is so celebrated for the diversity of its flowers, as connected with the history of vegetable impregnation, see p. 336, that we are glad to take advantage of a trifling difference in the calyx of the two florets, (the barren one being most frequently three-cleft, the fertile five-cleft,) to keep it here.

All things being considered, this Class may be thought scarcely worth retaining. Yet as we know two or three genera entitled to a place in it, upon principles which the analogy of the two preceding Classes shows to be sound, we cannot tell but others may exist in the unexplored parts of the globe. For this reason, and for the uniformity of the system, I would venture to preserve it. If the 21st and 22d Classes should hereafter be reformed by some judicious and experienced hand, according to the principle I have suggested, of retaining in them such genera only as have a permanent difference in the accessory as well as the essential parts of their flowers, their bulk being by such a reformation much diminished, it might be advisable to reduce them to one Class, in which the slender remains of Polygamia might commodiously be included, and the title of such a Class should be Diclinia, expressing the two distinct seats or stations of the organs of fructification.

Class 24. Cryptogamia. Stamens and Pistils either not well ascertained, or not to be numbered with any certainty. Orders 5.

1. Filices. Ferns. The parts of their flowers are almost entirely unknown. The fructification, taken collectively, and proved to be such by the production of prolific seeds, grows either on the back, summit, or near the base of the frond. Some are called annulatæ, annulated, their capsules being bound with an elastic transverse ring; others thecatæ, or more properly exannulatæ, from the want of such an appendage, of which some of the latter have nevertheless a spurious vestige. All the former, and some of the latter, are dorsiferous, bearing fruit on the back of the frond, and of these the fructification is either naked, or else covered with a membranous involucrum. The genera are distinguished by Linnæeus according to the shape and situation of the spots, or assemblages of capsules, besides which I have first found it necessary to take into consideration the absence or presence of the involucrum, and especially the direction in which it bursts. See Tracts relating to Nat. Hist. 215. t. 1.

Polypodium, Engl. Bot. t. 1149, has no involuerum; Aspidium, t. 14581461, has a single, and Scolopendrium, t. 1150, a double one. Osmunda, t. 209, has been remarked by Professor Swartz to have a spurious ring. It is one of those ferns the lobes of whose frond are metamorphosed, as it were, into spikes of capsules. Botrychium of Swartz, more distinctly spiked, and having no vestige of a ring, is separated by him from Osmunda. See one species of it in Engl. Bot. t. 318. Ophioglossum, t. 108, and Equisetum, t. 915, t. 929, are other examples of spiked ferns. Each seed of the latter is embraced by 4 filaments, judged by Hedwig to be the stamens. Supposed ferns with radical fructifications are Pilularia, t. 521, and Isoetes, t. 1084; but the former might possibly be referred to Monoecia Polyandria, and the latter to Monoecia Monandria as the system at present stands. Lycopodium, t. 224, 1148, &c., is a fern, at least in my opinion, with axillary fructification.

2. Musci, Mosses. These are really herbs[4] with distinct leaves and frequently as distinct a stem. Their conical membranous corolla is called a calyptra, or veil, its summit being the stigma, This veil clothes the capsule, which, before the seed ripens, is elevated on a fruit-stalk. The capsule is of one cell and one valve, opening by a Vertical lid[5]. Seeds very numerous and minute. The barren flowers of mosses consist of an indefinite number of nearly cylindrical, almost sessile anthers; the fertile flowers of one, rarely more, perfect pistils, accompanied by several barren pistils. Both stamens and pistils are intermixed with numerous succulent jointed threads, which perhaps answer the purpose of a calyx or corolla, as far as protection is concerned. Some few species of moss have the stamens and pistils associated in the same flower, but they are generally separate. Hypnum, Engl. Bot. t. 1424, 1425, has a peculiar scaly sheath, or perichætium, at the base of its fruit-stalk, composed of leaves very different from the foliage of the plant. This is considered as a sort of calyx, see p. 251, and as such is allowed to enter into the generic character; but there is some reason to esteem it rather of the nature of bracteas. The capsule of Splachnum, Engl. Bot. t. 144, &c., stands on a peculiar fleshy base, called apophysis.

Micheli in his Genera Plantarum, published in 1729, tab. 59, has well reprerented the parts above described, though he mistook their use, being quite ignorant of the fecundation of plants. Dillenius took the one flower precisely for the other, and yet absurdly called capsula what he believed to be anthera. Linnæus, who had previously formed just ideas on the subject, as appears from his manuscript Tour to Lapland, too implicitly submitted his own judgment to that of Dillenius, and adopted his hypothesis, at the same time correcting, as he thought, his phraseology. Hence the whole glare of the blunder of Dillenius has fallen on Linnæus; for while we read in the Linnæan definitions of mosses every where the word anthera, and in those of Dillenius, usually accompanying them, capsula; few persons, who have lately been instructed by Hedwig that the part in question is really a capsule, take the trouble to recollect that Dillenius so grossly misused that word. Various ideas have been started on this subject by Haller, Necker, and others, which could only claim attention while it remained in great obscurity. The excellent Hedwig has entirely the merit of an original discoverer in this branch of physiology. He examined all that had been done before his time, detected the truth, raised mosses from seed, and established their characters on the principles we have already explained.

The Linnæan genera of Mosses are chiefly founded on the situation of the capsule, whether lateral or terminal, with some other circumstances. They are too few, and not strictly natural. Hedwig first brought into notice the structure of the fringe, peristomium, which in most mosses borders the orifice of the capsule. This is either simple or double, and consists either of separate teeth, or of a plaited and jagged membrane. The external fringe is mostly of the former kind, the inner, when present, of the latter. The number of teeth, remarkably constant in each genus and species, is either 4, 8, 16, 32 or 64. On these therefore Hedwig and his followers have placed great dependance, only perhaps going into too great refinements relative to the internal fringe, which is more difficult to examine, and less certain, than the outer. Their great error has been laying down certain principles as absolute in forming genera, without observing whether all such genera were natural. Such mistakes are very excusable in persons not conversant with botany on a general scale, and whose minute and indefatigable attention to the detail of their subject, more than compensates the want of what is easily supplied by more experienced systematics. Thus Barbula of Hedwig is separated from Tortula, Engl. Bot. t. 1663, and Fissidens from Dicranum, t. 1272, 1273, on account of a difference of form or situation in the barren flowers, which is evidently of no moment, and merely divides genera that ought to be united. The same may be said of genera founded on the union of the stamens and pistils in one flower. On this subject I have been more diffuse in a paper on Mnium, in Tr. of Linn. Soc. v. 7, 254, to which I beg leave to refer those who are desirous to study it further. Various and abundant specimens of this tribe of plants, showing the various structure of the fringe, lid and other parts, may be seen in the latter volumes of English Botany more especially.

Mosses are found in the hottest and coldest climates. They are extremely tenacious of life, and, after being long dried, easily recover their health and vigour by moisture. Their beautiful structure cannot be too much admired. Their species are numerous, and in some cases difficult to determine, particularly in the genera Tortula and Orthotrichum; nor is the generic character of the latter so easy or certain as most others. Schreber, Dickson, Swartz, Bridel, Weber, Mohr and Turner are great names in this department of Botany, besides those of whom we have already spoken.

3. Hepaticæ. Liverworts. Of these the herbage is commonly frondose, the fructification originating from what is at the same time both leaf and stem. This character, however, proves less absolute than one founded on their capsules, which differ essentially from those of the preceding Order in having nothing like a lid or operculum. The corolla or veil of some of the genera is like that of Mosses, but usually bursts at the top. The barren flowers are unlike the organized stamens of the last-mentioned plants, being either undefined powdery heads, as in Jungermannia, see Hedwig's Theoria', t. 15, or of some peculiar conformation, as in Marchantia, Engl. Bot. t. 210, where they are imbedded in a disk like the seeds of Lichens, in a manner so contrary to all analogy, that botanists can scarcely agree which are the barren and which the fertile flowers of this genus. The four-valved capsule of Jungermannia, with the veil bursting at its summit to let the fruit-stalk pass, may be seen in Engl. Bot. t. 185, 186, which are both frondose species, like J. epiphylla, t. 771 whose calyx as well as corolla are evident; and t. 605608, which have apparently distinct leaves, like Mosses. Anthoceros, t. 1537, 1538, is a curious genus of the Hepaticæ. Linnæus comprehended this Order under the following one, to which it is, most assuredly, far less akin than to the foregoing.

4. Algæ. Flags. In this Order the herbage is frondose, sometimes a mere crust, sometimes of a leathery or gelatinous texture. The seeds are imbedded, either in the frond itself, or in some peculiar receptacle. The barren flowers are but imperfectly known. Here we find that great natural Order, comprehended by Linnæus under one genus by the name of Lichen, the fructification of which, for the most part, consists of a smooth round disk, flat, convex, or concave, with or without an adventitious border, in the substance of which disk the seeds are lodged. In some others they are placed in powdery warts, or in fibrous receptacles. The barren flowers are supposed to be powdery also, very much like those of Jungermannia. See Engl. Bot. t. 126, and various other parts of that work, where a great number of species are figured. The whole tribe has been much investigated, and attempted to be divided into natural genera founded on habit, by Dr. Hoffmann of Goettingen, whose figures are perfect in their kind. But a more complete scheme for reducing this family to systematic order has been receritly made known to the world by Dr. Acharius, a learned Swede, who in his Prodromus, and Methodus Lichenum, has divided it into genera founded on the receptacle of the seeds alone. Hence those genera, though more technical, are less natural than Hoffmann's; but they will, most likely, prove the foundation of all that can in future be done on the subject, and the works of Acharius form a new æra in cryptogamic botany. It is only perhaps to be regretted that he has been somewhat too prodigal of new terms, which when not wanted are always a burthen to science, and rather obscure than illustrate it. Thus Hedwig used the term sporangium for a seed-vessel, pericarpium, in which the learner would seek in vain for any distinction, or new idea. A student might very justly complain if, in a science necessarily so overburthened with words, he were required to call the same part by a different name in every different family. I would gladly therefore retain the word frons in preference to the thallus of Acharius, receptaculum for his apothecium, pedicellus for his bacillum or podetium, and semina for his sporæ, because I see no improvement in the change. When this or any other writer strikes out new ideas, and discriminates parts hitherto mistaken or unknown, we thankfully receive from him new terms to express his discoveries. Thus the cyphella of Acharius is a peculiar sort of pit or pore on the under side of the frond in that section of Lichens called Sticta, see Engl. Bot. t. 1103, 1104; his lirellæ are the black letter-like receptacles of the genus Opegrapha, t. 17531756; his tricæ the analogous parts, resembling a coiled horse-hair, in Gyrophora, the Umbilicaria of Hoffmann, t. 522. These terms are necessary and instructive, and are chosen with that accuracy and taste for which Dr. Acharius is conspicuous.

The aquatic or submersed Algæ form a distinct and peculiar tribe. Some of these abound in fresh water, others in the sea, whence the latter are commonly denominated sea-weeds. The chief genera are Ulva, t. 419, 420, 1276, well defined by its seeds being dispersed under the cuticle throughout the membranous or gelatinous substance of the frond; Fucus, t. 10661069, &c., whose seeds are collected together in tubercles or swellings, of various forms and sizes; and Conferva, of which the 24th and 25th volumes of Engl. Bot., more especially, show various specimens. This last genus is commonly known by its capillary, and, for the most part, jointed frond. The seeds of some species are lodged in external capsules or tubercles; of others in the joints of the frond; and hence the ingenious Dr. Roth has formed a genus of the former, called Ceramium. His Rivularia, Engl. Bot. t. 17971799, is perhaps more satisfactorily separated from Conferva, as we trust is Vaucheria, t. 1765, 1766, a fresh-water genus named after M. Vaucher of Geneva, who has published an elaborate and faithful microscopical work on Fresh-water Confervas. The submersed Algæ in general are merely fixed by the roots, their nourishment being imbibed by their surface. Many of them float without being attached to any thing. The genus Fucus has received more botanical attention than the rest of this tribe, and the works of Gmelin, Esper, Stackhouse and Velley have ascertained many species, which the labours of Dr. Goodenough, Mr. Woodward and Mr. Turner have reduced to systematic order. Still a more perfect combination of the skill of the painter and the botanist is to be desired, relative to the genus in question, and this is about to be supplied by the Historia Facorum of the writer last mentioned, and his friend Mr. W. J. Hooker.

5. Fungi. Mushrooms. These cannot properly be said to have any herbage. Their substance is fleshy, generally of quick growth and short duration, differing in firmness, from a watery pulp to a leathery or even woody texture. By some naturalists they have been thought of an animal nature, chiefly because of their fœtid scent in decay, and because little white bodies like eggs are found in them at that period. But these are truly the eggs of flies, laid there by the parent insect, and destined to produce a brood of maggots, to feed on the decaying fungus, as on a dead carcase. Ellis's beautiful discoveries, relative to corals and their inhabiting polypes, led to the strange analogical hypothesis that these insects formed the fungus, which Munchausen and others have asserted. Some have thought fungi were composed of the sap of corrupted wood, transmuted into a new sort of being, an idea as unphilosophical as the former, and unsupported by any semblance of truth.

Dryander, Schæffer and Hedwig have, on much better grounds, asserted their vegetable nature, detected their seeds, and in many cases explained their parts of fructification. In fact, they propagate their species as regularly as any other organized beings, though, like others, subject to varieties. Their sequestered and obscure habitations, their short duration, their mutability of form and substance, render them indeed more difficult of investigation than common plants, but there is no reason to suppose them less perfect, or less accurately defined. Splendid and accurate works, illustrative of this Order, have been given to the world by Schæffer, Bulliard and Sowerby, which are the more useful as the generality of fungi cannot well be preserved. The most distinguished writer upon them, indeed the only good systematic one, is Persoon, who has moreover supplied us with some exquisite figures. His Synopsis Methodica Fungorum helps us to the following arrangement.

1. Angiocarpi, such as bear seeds internally. These are either hard, like Sphæria, Sowerb. Fung. t. 159, 160; or membranous, tough and leathery, like Lycoperdon, t. 331, 332; Cyathus (Nidularia) t. 28, 29; or Batarrea (Lycoperdon) t. 390.

2. Gymnocarpi, such as bear seeds imbedded in an appropriate, dilated, exposed membrane, denominated hymenium, like Helvella, t. 39, in which that part is smooth and even; Boletus, t. 34, 87, 134, in which it is porous; and the vast genus Agaricus, t. 1, 2, &c., in which it consists of parallel plates called lamellæ, or gills.

Persoon has been commendably sparing of new terms. Besides hymenium above explained, he has scarcely introduced any other than peridium, for the round membranous dry case of the seeds in some of the 1st section. The term pileus, a hat, is used by all authors for the head of those fungi that compose the 2d section.

Appendix. Palmæ. The natural order of Palms was so little understood when Linnæus formed his systematical arrangement of plants, and so few of their flowers had been scientifically examined, that he was under the necessity of leaving this order as an appendix to his system, till it could be better investigated. To its peculiar habit and physiology we have adverted in several of the foregoing pages, see p. 5759, 62, 133, &c.

Late observations show Palms to have for the most part 6 stamens, rarely 3 or 9, with 3 or 6 petals, and 1 or 3 styles; which last are sometimes in the same flower with the stamens, sometimes in a separate one, but both flowers always agree in general structure. Their fruit is generally a drupa. They are akin to the liliaceous tribe, and Linnæus happily terms them the princes of the vegetable kingdom. His most numerous remarks concerning them occur in his Prælectiones in Ordines Naturales Plantarum, published by Professor Giseke at Hamburgh in 1792, from private lectures and conversations of Linnæus. This work however is necessarily full of errors and mistakes, not only from its mode of compilation and the intricacy of the subject, but because Linnæus had only partially studied certain parts of that subject, and was undecided in his sentiments upon those parts. It was a singular instance of indulgent liberality in him to allow his disciples Fabricius and Giseke to make notes, for their own use, of what he considered himself as scarcely competent to lay in a finished form before the public. We are obliged to the editor for preserving these valuable though crude materials, and he has shown ability in digesting and elucidating them. I should scarcely, for my own part, have thought it right to furnish still more crude and imperfect guesses and opinions, from manuscripts which their illustrious author had purposely, as it appears, withheld from his auditors, lest he should lead them into error. This will explain a note in Professor Giseke's preface, p. 19, which however was printed before his request came to my knowledge; for two very intelligent friends, through whom it was meant to be conveyed, judged it unreasonable to be made, as well as improper to be complied with, and therefore suppressed the message.

I have only to add a few practical remarks on the preparation and use of an Herbarium or Hortus Siccus. The advantages of preserving specimens of plants, as far as it can be done, for examination at all times and seasons, is abundantly obvious. Notwithtanding the multitude of books filled with descriptions and figures of plants, and however ample or perfect such may he, they can teach no more than their authors observed; but when we have the works of Nature before us, we can investigate them for ourselves, pursuing any train of inquiry to its utmost extent, nor are we liable to be misled by the errors or misconceptions of others. A good practical botanist must be educated among the wild scenes of nature, while a finished theoretical one requires the additional assistance of gardens and books, to which must be superadded the frequent use of a good herbarium. When plants are well dried, the original forms and positions of even their minutest parts, though not their colours, may at any time be restored by immersion in hot water. By this means the productions of the most distant and various countries, such as no garden could possibly supply, are brought together at once under our eyes, at any season of the year. If these be assisted with drawings and descriptions, nothing less than an actual survey of the whole vegetable world, in a state of nature, could excell such a store of information.

Some persons recommend the preservation of specimens in weak spirits of wine, and this mode is by far the most eligible for such as are very juicy. But it totally destroys their colours, and often renders their parts less fit for examination than the above-mentioned mode. It is besides incommodious for frequent study, and a very expensive and bulky way of making an herbarium.

The greater part of plants dry with facility between the leaves of books, or other paper, the smoother the better. If there be plenty of paper, they often dry best without shifting; but if the specimens are crowded, they must be taken out frequently, and the paper dried before they are replaced. The great point to be attended to is that the process should meet with no check. Several vegetables are so tenacious of their vital principle, that they will grow between papers, the consequence of which is a destruction of their proper habit and colours. It is necessary to destroy the life of such, either by immersion in boiling water, or by the application of a hot iron, such as is used for linen, after which they are easily dried. I cannot however approve of the practice of applying such an iron, as some persons do, with great labour and perseverance, till the plants are quite dry, and all their parts incorporated into a smooth flat mass. This renders them unfit for subsequent examination, and destroys their natural habit, the most important thing to be preserved. Even in spreading plants between papers, we should refrain from that precise and artificial disposition of their branches, leaves, and other parts, which takes away from their natural aspect, except for the purpose of displaying the internal parts of some one or two of their flowers, for ready observation.

After all we can do, plants dry very variously. The blue colours of their flowers generally fade, nor are reds always permanent. Yellows are much more so, but very few white flowers retain their natural aspect. The Snowdrop and Parnassia, if well dried, continue white. Some greens are much more permanent than others; for there are some natural families whose leaves as well as flowers turn almost black by drying, as Melampyrum, Bartsia, and their allies, several Willows, and most of the Orchideæ. The Heaths and Firs in general cast off their leaves between papers, which appears to be an effort of the living principle, for it is prevented by immersion of the fresh specimen in boiling water. Nandina domestica, a Japanese shrub, lately introduced among us by Lady Hume and Mr. Evans of Stepney, is very remarkable in this respect. Every leaflet of its very compound leaves separates from its stalk in drying, and even those stalks all fall to pieces at their joints.

Dried specimens are best preserved by being fastened, with weak carpenter's glue, to paper, so that they may be turned over without damage. Thick and heavy stalks require the additional support of a few transverse strips of paper, to bind them more firmly down. A half sheet, of a convenient folio size, should be allotted to each species, and all the species of a genus may be placed in one or more whole sheets. On the latter the name of the genus should externally be written, while the name of every species, with its place of growth, time of gathering, the finder's name, or ary other concise piece of information, may be inscribed on its appropriate paper. This is the plan of the Linnaean Herbarium, in which every species, which its original possessor had before him when he wrote his great work the Species Plantarum, is numbered both in pencil and in ink, as well as named, the former kind of numbers having been temporary till the book to which they refer was printed, after which they were confirmed with a pen, and a copy of the book, now also in my hands, was marked in reference to them. Here therefore we do not depend on the opinion merely, even of Linnæus, for we have always before our eyes the very object which was under his inspection. We have similar indications of the plants described in his subsequent works, the herbarium being most defective in those of his 2d Mantissa, his least accurate publication. We often find remarks there, made from specimens acquired after the Species Plantarum was published. These the herbarium occasionally shows to be of a different species from the original one, and it thus enables us to correct such errors.

The specimens thus pasted, are conveniently kept in lockers, or on the shelves of a proper cabinet. Linnæus in the Philosophia Botanica exhibits a figure of one, divided into appropriate spaces for each class, which he supposed would hold his whole collection. But he lived to fill two more of equal size, and his herbarium has been perhaps doubled since his death by the acquisitions of his son and of its present possessor.

One great and mortifying impediment to the perfect preservation of an herbarium arises from the attacks of insects. A little beetle called Ptinus Fur is, more especially, the pest of collectors, laying its eggs in the germens or receptacles of flowers, and others of the more solid parts, which are speedily devoured by the maggots when hatched, and by their devastations paper and plants are alike involved in ruin. The most bitter and acrid tribes, as Euphorbia, Gentiana, Primus, the Syngenesious class, and especicially Willows, are preferred by these vermin. The last-mentioned family can scarcely be thoroughly dried before it is devoured. Ferns are scarcely ever attacked, and grasses but seldom.—To remedy this inconvenience I have found a solution of corrosive sublimate of mercury in rectified spirits of wine, about two drams to a pint, with a little camphor, perfectly efficacious. It is easily applied with a camel-hair pencil when the specimens are perfectly dry, not before; and if they are not too tender, it is best done before they are pasted, as the spirit extracts a yellow dye from many plants, and stains the paper. A few drops of this solution should be mixed with the glue used for pasting. This application not only destroys or keeps off all vermin, but it greatly revives the colours of most plants, giving the collection a most pleasing air of freshness and neatness. After several years' experience, I can find no inconvenience from it whatever, nor do I see that any dried plants can long be preserved without it.

The herbarium is best kept in a dry room without a constant fire. Linnæus had a stone building for his museum, remote from his dwelling-house, into which, I have been told, neither fire nor candle was ever admitted, yet nothing can be more free than his collection from the injuries of dampness, or other causes of decay.

  1. See Linn. Sp. Pl. 186, and Curt. Lond. fasc. 6. t. 31.
  2. From χλειω, to close or shut up, in allusion to the situation of the wings and keel.
  3. I was not aware of Loureiro's Stylidium, a plant, according to his description, of the 7th Class; Fl. Cochinch. v. 1. 221; but this can scarcely interfere with ours, being probably, as it grows about Canton, some well-known shrub that happened to have a 7-cleft flower. It should seem to belong to the Rubaceæ, notwithstanding some points in the description.
  4. Hedwig's term musci frondosi is incorrect.
  5. This part in Phascum only does not separate from the capsule.