habit or leaves is of little use in guessing at the affinity of Filices, which must be determined exclusively by their fructifications. Lycopodiacea follow Filices, and in the character of the order a mark of interrogation is placed after their globular seeds; a doubt which is completely removed by Mr. Salisbury's paper, with faithful figures of their germination, in the last volume of the Linnéan Transactions, whence they appear to be a strong connecting link of cryptogamous and dicotyledonous plants: for, though their plumula springs up perpendicularly opposite to their radicle, out of the middle of two equal regular Cotyledons of a different shape from the next leaves, a lateral solid process, exactly like the internal part of the embryo in cryptogamous plants, is attached just under the Cotyledons, and remains within the seed-coat. Isoetes is placed under Lycopodiacea, which we should rather have joined to the next order, Marsileaceae. Equisetacea and Charadeæ terminate the cryptogamous plants.

Of the Endogenous phænogamous plants, 1. Fluviatiles of Ventenat follow Charadea. 2. Aroidea. 3. Typhacea, justly separated in our opinion from Aroidea; and here Dodonæus's genus of Platanaria (Sparganium Natans, L.) is restored and distinguished by its ovate peltate stigma. 4. Cyperaceae, divided into two sections, Caricea and Cyperaceae: of the former, all the species with three stigmata are detached from Carex by the name of Thrasus, perhaps derived from τερσώς the foliage being arid, and not so juicy as in some of the Grasses. 5. Graminea. To the genera of this extensive natural class, which we yet know but partially by our own examination, the author prefixes a key, the wards of which are made by the similarity of the Locusta, the shape of their rachis, the number of flowers in each glume, and the shape of the glumes, or chaff, as it is vulgarly denominated. In the natural arrangement, he makes the eight following sections, or orders; Hordeacea, Chloridea, Bromea, Arundinacea, Avenacea, Agrostidea, Stipaceæ, and Panicea; of which, we think, the names alone indicate the propriety, though we have some doubts of the series. 6. Restiacea. 7. Junceæ : prefixed to these we are surprized at finding Acorns, a genus which Dryander always supposed to be allied to Aroidea, and Mr. Robert Brown says it has the same embryo. Narthecium Ossifragum, under Adanson's name of Abama, is also left among Junceæ, to which it has so little affinity that more than half the genera of Monocotyledones intervene in a truly natural series; for Narthecium unquestionably is very near akin to Pleca, Tofieldia, Aletris, Lophiola, Anigozanthus, Conostylis, Phlebocar, Hæmodorum, Gyrotheca, Argolasia, Dilutris,

and

and Wachendorfia, which form one if not two intermediate orders between Hypoxidea and Veratrea. 8. Colchicaceae. We have long suspected that two distinct, species of Colchicum exist in Great Britain. In that which grows north of the river Trent, especially between Doncaster and Ferrybridge, continuing through the lime-stone parts of Wharfdale, the leaves are longer, petals narrower, without any tessellation of color, and anthers much paler, than in the southern plants. so common near Bath and Bristol. 9. Liliacea. 10. Asphodelea. Having lately examined all these plants most closely, we must observe that none of them have the smallest immediate relationship to Asphodelus. Anthericum Serotinum, L., here referred to Phalangium, has a bulbous root, laterally beaked at its base, as in Colchicum, and must with Gagea join Colchicacea. As for the Linnéan genera of Scilla and Ornithogalum, we feel confident that hereafter they will constitute orders by themselves, distinguished from one another by the longer and more complete persistence of the petals in the latter; among which Mr. Gray justly detaches Ornithogalum Nutans, L., calling it Honorius. He also excludes Moly and Cepa from that most heterogeneous assemblage, Allium. 16. Asparagi. It is here observed that the young shoots are diuretic, and communicate a violet scent to the urine:' but, instead of that charming perfume, we have invariably. perceived a most disagreeable foetid odour. 17. Smilaceæ. Here the author inserts Paris, possibly misled by Mr. Robert Brown's Prodromus: but it has totally different petals, as well as seeds, and ought to be placed even in a different class of Monocotyledones, Tetræ, with Esdra, Trillium, and a new genus from Nepal. 18. Tamacea. 19. Amaryllidea. 20. Irideæ. 21. Orchidea. Many of Richard's and Mr. Robert Brown's genera are adopted in, this order, or rather class; and. Apuleius's name of Entaticus is given to Satyrium Albidum,.. which Mr. Robert Brown has joined with Habenaria. 22. Alismacea: 23. Butomaceæ. 24. Juncagineæ. 25. Hydro-.. charidea.

}

The Exogenous phænogamous plants, or Dicotyledones, now: succeed; and the author says, "This series contains full twothirds of the whole number of natural families, but the arrangement of these families among themselves has not yet been. determined, and therefore they are arranged for the present by De Candolle into four artificial sub-series, resting upon the differences observable in their petals and stamens;' namely, Monochlamydea, Corolliflora, Calyciflora, Thalamiflora. Here are placed 1. Abietidea. 2. Cupressidea. 3. Taxideæ. 4. Salicina of these difficult trees, or shrubs, Dr. Hull's divi

sions are adopted, first by their sessile or pedicelled ovaries, then by their short or long styles, and, lastly, those of which the female aments remain unknown. 5. Betulidea. 6. Cory lidea. 7. Myriceæ. 8. Ulmacea. 9. Urticea, where Xanthium is introduced. 10. Euphorbiaceæ. 11. Aristolochia 12. Santalaceæ. 18. Eleagnea. 14. Thymelea. 15. Poly gonea, under which Rumex Digynus is detached as sui generis, with Hill's name of Oxyria. 16. Chenopodea. 17. Amaran-· thidea. 18. Plantaginea. Here all the 2-locular fruited species with one seed in each cell are separated by Dioscorides's name of Arnoglossum: Plantago Coronopus, L., having more cells and seeds, is called Asterogeum; and Littorella Lacustris alone constitutes a 19th order, Littorellidea. 20. Staticina. 21. Primulacea. Here Irasekia, Naumburgia, and Numularia are removed from Lysimachia, leaving a mark of doubt after Primula Farinosa, which has a totally different calyx from Primula. 22. Rhinanthacea. 23. Orobancheæ. 24. Lenti bularia; these surely ought to have followed Primulacea, as they do in the artificial arrangement prefixed to the sub-series. We met with a nearly regular or pelorized flower of Pinguicula last summer near Buxton, which was 4-androus, and another flower was 5-androus, but not so regular, the lobes opposite to the fifth stamen being much larger. 25. Scrofu larinæ. 26. Solanec. 27. Gentianes. In this order, many genera are detached from the officinal species, as Pneumonanthæ, Ericala, Erythræa, and to these is added 'Franquevillia. R. A. Salisbury. MSS.:' but its characters, as well as those of the other genera, are by the author, of whom it may be truly said, "Nullius addictus in verba jurare magistri." 28. Polemoniacea. 29. Vincæ. 30. Convolvulacea; here Mr. Robert Brown is quoted as the author of Calystegia, but that genus was proposed by Solander, and described in his MSS. a very long time since. Under this order, also, Cuscuta Major is separated from the Minor by Gesner's name of Cassytha, its filaments being devoid of scales: but Cuscuta and Cassytha being in fact only different pronunciations of the same Arabic word Keschout, and the latter being applied by Linné to a genus which has flowers very similar to those of Laurus, if Cuscuta Minor be really sui generis, it must have another name. 31. Boraginea; here again we find several necessary reforms of genera, though more remain to be added. 32. Labiata. 33. Pyrenacea; the only genus of this order, indigenous in Great Britain, is Verbena, which has not a pyrenaceous fruit, and Persoon's name of Verbenacea is now more generally adopted for the order. 34. Oleinæ. 35. Fraxinea. We doubt. greatly the propriety of separating these two orders of M. Link

from

from one another, or even from Jasmineæ, having found the dif ferences brought forwards by Mr. Robert Brown not so constant; for the seeds of Jasminum are often horizontal, or even somewhat pendulous, and frequently two in a cell. A parallel anomaly to that which occurs in the æstivation of their flowers is found also in the order of Cinchonea, where Cinchona has a contorted but Exostema a valved æstivation. 36. Ericina; here Chamacistus and Uva Ursi of the old botanists are very properly restored; the former differing from Rhododendron in its strictly regular corolla, and the latter from Arbutus in its fruit, which is not a berry, but a true Pyrena, composed in those which we have examined of from six to ten seeds grown so firmly together as to be inseparable. Of other genera, Erica Daboccia has been referred to Menziesia by Jussieu but this little shrub is totally distinct in habit, inflorescence, fruit, and seeds, from that genus. Andromeda Cærulea, referred also to Menziesia by Swartz, is here separated under Mr. Salisbury's name of Phyllodoce. Erica Cinerea, L., is also detached with the last-mentioned botanist's name of Eremo callis, to which Mr. Gray joins Erica Tetralix too hastily, for even these differ materially in fruit and seeds. Erica Vagans is made another genus, and called Gypsocallis, Erica Vulgaris has the name of Calluna, and Pyrola Uniflora that of Moneses; all from Mr. Salisbury's MSS. At the end of this order, Monotropa is added; its supposed affinity, first published by Nuttal, having been confirmed by Mr. J. Lindley. These are both high authorities, but the structure of the anthers is so different from those of all Ericea and Andromedea, and they so closely resemble the anthers of many Personata, L that we cannot yet adopt their opinion. The situation of the embryo will perhaps determine this point. 37. Vaccinieæ. 38. Campanulacea. 39. Composita. Under the third division of this numerous class, Inula Crithmifolia is very justly separated: but we cannot account for the author's name Eritheis. 40. Dipsacea, where Xenocrates's name of Galedragon is applied to Dipsacus Pilosus. 41. Valerianeæ. 42. Rubiacea. 43. Caprifoliacea. 44. Loranthidea; under this order we are surprized to find Adoxa, the author nevertheless remarking that it is an anomalous plant, scarcely referable to this or any other order.' Its seeds will perhaps ascertain its real affinity. 45. Umbellifera. 46. Saxifragea. Here Robertsonia of Haworth is adopted for the London Pride and its congeners; Neckar's objectionable name of Hydatica for Saxi fraga Stellaris; and that of Dr. Kingstone is commemorated in Saxifraga Hirculus, he having discovered it growing wild on Knutsford-moor. 47. Crassulacea. 48. Portulacea, 49. Paronychidea

1

ronychidea are divided into three sections, one containing Scleranthus, and Arenaria Peploides, L., under Rafinesque's name of Adenarium: but this plant surely has no immediate affinity to Scleranthus. The second section contains Illecebrum, the author regarding with Gartner Verticillatum as the type, Corrigiola, Herniaria, and Polycarpon. The third section has only Larbrea of Aug. St. Hilaire. 50. Grossularea. 51. Cucurbitaceae. 52. Salicariæ. 53. Tamaricina. 54. Cerato

phylla. 55. Halorageæ. 56. Onagraria. 57. Pomace; where Pyrus remains divided into the three Linnéan sections of Pyrus, Malus, and Sorbus, followed by Crataegus and Mespilus. 58. Rosacea. It is to be regretted that Mr. Gray had not studied Mr. J. Lindley's Monograph, who reduces the number of British species to 12, making Doniana a variety of Sabini. Mr.G.'s Villosa, Scabriuscula, Heterophylla, Pulchella, varieties of Tomentosa; his Gracilis, Nuda, Casia, Sarmentacea, Bractescens, Dumetorum, Collina, and Surculosa, varieties of Canina; and, lastly, Micrantha and Borreri, varieties of Rubiginosa. 59. Agrimoniacea. 60. Dryadea. 61. Ulmariæ. 62. Drupaceæ. 63. Leguminosa. 64. Celastrinæ 65. Rhamneæ. 66. Geraniea, where Oxalis and Impatiens each form a section. 67. Hypericinea, of which Androsæmum is restored. 68. Acerineæ. 69. Tiliacea. 70. Malvacea. 71. Lineæ. These vegetables, we think, shew the inutility of laying down rules à priori for combining vegetables naturally: for Linea have no affinity to Malvacea, being oligandrous, with a totally different insertion of their floral envelopes and stamina, as well as different fruits: but they introduce very naturally the next order. 72. Caryophylle. Jussieu has removed these plants from several others, to which he confesses that they have a close affinity, on account of the supposed hypogynous insertion of their petals and filaments : but many of them are perigynous in Jussieu's strictest sense of the term, and they are all so in ours; since, although their alternate filaments are often attached to the petals, their bases run down projecting in front, and terminate in the same common circle of the receptacle or disc, which always adheres more to the calyx than to the stipes of the pericarpium, and is dilated either into a marginal ring or manifest saucer, exactly as in Scleranthea. We therefore do not hesitate to consider it as a legitimate perigynous receptacle, occasionally stipitated like that of Passiflorea; and, in this view, how many apparent interruptions of the chain vanish, by which the Creator has linked all vegetables! Thus Larbrea of Aug. St. Hilaire is no longer to be regarded as essentially anomalous to the immediate relatives near which all preceding writers

had