THE JOURNAL OF IHE LINNEAN SOCIETY. BOTANY. VOL, XXL. Q LONDON: SOLD AT THE SOCIETY'S APARTMENTS, BURLINGTON HOUSE, AND BY LONGMANS, GREEN, AND CO., AND WILLIAMS AND NORGATE. 1886, - Dates of Publication of the several Numbers included in this Volume, No. 132 . , 133 j pp. 1-202, published April 30, 1884. » 194, ,, 208-258, » August 18, 1884. » 195, ,, 259-353, » December 12, 1884. » 190, ,, 353-406, » April 14, 1885. » 197, ,, 407-522, » April 29, 1885. , 523-578, » August 21, 1885. » 189, ,, 573-633, s September 19, 1885. » 140, ,, €33-690, » January 25, 1886. 9 » 13 , PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET. LIST OF PAPERS. Bakzn, JOHN GILBERT, F.R.S., F.L.S., Royal Gardens, Kew. Pago Further Contributions to the Flora of Central Madagascar .... 317 Further Contributions to the Flora of Central Madagascar.— Second and Final Part ................eeeeeee enne 407 BALL, Jonn, F.R.S., M.R.LA,, F.L.S. Contributions to the Flora of North Patagonia and the adjoining Territory .... eese III 203 Bower, FREDERICK ORPEN, M.A., F.L.S. On Apospory in Ferns (with special reference to Mr. Charles T. Druery’s Observations). (Plates XI. & XII., and 6 wood- Cuts. ). eese eee heh thee esee 360 CLARKE, CHARLES Baron, F.R.S., F.L.S. On the Indian Species of Cyperus; with Remarks on some others that specially illustrate the Subdivisions of the Genus, (Plates L-IV.) .....sseeeeeeeeeee III hte 1 - Notes on the Flora of Parasnath, a Mountain of North-western Bengal, in a Letter from C. B. Clarke to, and with an Intro- ductory Note by, Sir J. D. Hooxer, K.C.S.L, F.R.S. ...... 252 Botanic Notes from Darjeeling to Tonglo and Sundukphoo .... 384 CRouBi£E, Rey. JAuEs M. F.L.S. On the Algo-Lichen Hypothesis. (Plates VIII. & IX.) ...... 259 Drvery, CHARLES THomas, F.L.S. Observations on a Singular Mode of Development in the Lady- Fern (Athyrium Filix-fæmina) oo. eee es 354 Further Notes on a Singular Mode of Reproduction in Athyrium Filiz-fomina, var. clarissima, (With 2 woodcuts.) ........ 358 iv Page Dyer, Witi1aM T. TurseLTON, C.M.G., F.R.S., Assist. Director Royal Gardens, Kew. Report on the Botany of Mr. H. O. Forbes's Expedition to Timor- Laut; with a List of Determinations of the Plants collected, by Prof. OLIVER, F.R.S.....isseeeeeeee n 37C FonnBrs, Henry O., A.L.S., F.R.G.S., F.Z.S. Report on the Botany of his Expedition to Timor-Laut. (See under W. T. THISELTON DYER.) ........... sese eee 370 On the Contrivances for insuring Self-fertilization in some Tro- pical Orchids. (Plates XVI. & XVII.) GARDNER, J. SrAnKIE, F.L.S., F.G.S. Eocene Ferns from the Basalts of Ireland and Scotland. (Plate XXVI.) Groves, Henry, F.L.S. The Coast Flora of Japygia, South Italy HrNsrow, Rev. GrEonaz, M.A., F.L.S., F.G.S. On Vernation and the Methods of Development of Foliage as protective against Radiation. (With 15 woodcuts.)........ 624 Horxrs, Epwarp Monzrr, F.L.S. Remarks on Cinchona Ledgeriana as a Species................ 374 Hooker, Sir JosrPu D., K.C.S.I., Director Royal Gardens, Kew. Notes on the Flora of Parasnath, a Mountain of North-western Bengal, in a Letter from C. B. CLARKE, F.R.S., F.L.S., to, and with an Introductory Note by, Sir J. D. Hooker, F.R.S. .... 252 List of Plants collected by Mr. Joseph Thomson, F.R.G.S., on the Mountains of Eastern Equatorial Africa, by Prof. DaNrEL OLIvER, F.R.S.; with Observations on their Distribution by Sir J. D. Hooker, F.R.S. 392 Ce 9 s | | n n oc t |] | 9n» t] s ] o | n | t9 | n €" JosHUA, WiLLiAM, F.L.S., F.R.M.S. Burmese Desmidieæ, with Descriptions of new Species occurring in the neighbourhood of Rangoon. (Plates XXIL-XXV.).. 634 KipsrowN, Rosert, F.G.S. On the Occurrence of Lycopodites (Sigillaria) Vanuremi, Góp- pert, in Britain, with Remarks on its Affinities, (Plate XVIIL) oo... eee Cote tect es hee 560 v Masters, Maxwe 1, T., M.D., F.R.S., F.L.S. Supplementary Notes on Restiaceæ ......... 0. cece cece ees 574 Mitten, Wirum, A.L.S. Notes on the European and North-American Species of Mosses of the Genus Fissidens oo... ci cece ccc etna 550 Moore, Spencer Le Mancnaxr, F.L.S. Studies in Vegetable Biology.—I. Observations on the Con- tinuity of Protoplasm. (Plates XIX.-XXL).............. 595 Studies in Vegetable Biology.—II. On Rosanoff's Crystals in the Endosperm-Cells of Manihot Glaziovii, Müll. Arg. (With 8 woodcuts.) 0... eee ec cee ete e ete teeeee 621 OrivEn, Prof. Danrer, F.R.S.. Royal Gardens, Kew. List of the Determinations of Plants from Timor-Laut. (See Report on, by W. T. THISELTON DYER.) ..... DEIN 370 List of Plants collected by Mr. Joseph Thomson, F.R.G.S., on the Mountains of Eastern Equatorial Africa ; with Observa- tions on their Distribution by Sir J. D. Hooxzn, F.R.S..... 392 PLowrieut, CHARLES Baccz, F.L.S. Remarks on the Reproduction of the Hetereecious Uredines.... 368 REPORT on the Botany of Mr. H. O. Forbess Expedition to Timor- Laut, by W. T. Ta1srLton Dyer, F.R.S., Secretary to the British-Association Committee for the Exploration of the Island; with a List of Determinations of the Plants collected, by Prof. OLIVER, F.R.S. ... esee 370 Riptey, Henry N., M.A., F.L.S., Assistant Botanical Depart- ment, British Museum. The Orchids of Madagascar, (Plate XV.)..... uiii 456 ROLFE, ROBERT ALLEN, Assistant, Herbarium, Royal Gardens, Kew. On Hyalocalyx, a new Genus of Turneraces from Madagascar. (Plate VIL)..... cece cece eee Hmmm hn : 256 On the Flora of the Philippine Islands, and its probable Deri- D vation. (Plate X.)........sussssssesserererrsrrrreret 283 Roper, Freeman C. S., F.L.S. . s Note on Ranunculus Lingua, Linn. (Plates XIII. & XIV.).... 380 ScnaAnsenuipr, Dr. Jurus, Assistant in the Botanic Garden of the Royal Hungarian University at Kolosvár (Klausenburg . Notes on Afghanistan Alge. (Plate V.) ...... nnn 241 vi Page Scorr, DvkiNriELD H., M.A., F.L.S. On the Occurrence of Articulated Laticiferous Vessels in Hevea. (With 4 woodcuts.) oo. ce ccc cece mnn 566 THomson, JosEpy, F.R.G.S. Plants collected by, in Mountains of Eastern Equatorial Africa. (See under Prof. OLIVER.) 66... ee eee ete ees 302 ' WHITE, OHARLES FREDERICK, F.LS. On some Pollen from Funereal Garlands found in an Egyptian Tomb, cirea B.c. 1000. (Plate VL) oo... cece eee ees 251 vii iXPLANATION OF THE PLATES. PraTR I.- . . . II ANATOMICAL DETAILS of rhizome, culm, inflorescence, &c., of Indian HL Species of Cyperus, in illustration of Mr, O. B. Clarke's paper IV. thereon. V. AramawisTAN ALG, delineated by Dr. J. Schaarschmidt. VI. Ancient Eayprian and recent pollen and anther of Papaver rheas, sketched by Mr. C. F. White. VII. HYALOCALYX sETIFERUS, and floral details. VIII. } HYPOTHALLINE condition of Lichens, demonstrating the Rev. J. M. IX. Crombie's Algo-Lichen Hypothesis. X. VILLARIA PHILIPPINENSIS, and dissections of flower. XI. } ArosPony ix Ferns. Sectional views of sporangia &c., elucidating XII. Mr. F. O. Bower's observations thereon. XIII.) RawvxcuLvs Linava, submerged and aerial leaves of, representing xv} Mr. F. C. S. Roper’s specimens. XV. Tue Enpemic Mapacascar Orcnins, Bicornella gracilis amd B. parviflora, XVI. l Dracrams illustrating structural parts of self-fertilizing Tropical XVII. Orchids, described by Mr. H. O. Forbes. XVIII. LvcoroprrEs VaxvxEwi, fossil plant. XIX. . . XX I or PRnoropPLAsw. Diagrammatic sketches in support of XXI. Mr. 8. Le Marchant Moore's Observations. XXIII. | Burmese Drswrprgg. New species and varieties described by Mr. W. XXIV. | Joshua. | XXV. ` | XXVI. Fossit. Eocexe Frnss, described by Mr. J. Starkie Gardner. ERRATA. Page 265, fifth line from bottom, read “ Mucidines.” For the word “ cellulose," read “ cellular," pages 271, 276, 278, 280, and 983. Page 475, seventh line from top, for “ P. MULTIFLORA " read “ P. MINUTIFLORA”; and on same page, third line from bottom, delete ** (nomen nudum) ” Page 514, seventh line from top, delete * C. gibbosa, Ridley, l. c." Page 632, from a certain similarity in appearance, the woodcuts fig. 10 and fig. 15 have accidentally got transposed. Tug JOURNAL OF THE LINNEAN SOCIETY. On the Indian Species of Cyperus; with Remarks on some others that specially ilustrate the Subdivisions of the Genus. By C. B. Craxkz, F.R.S., F.L.S. [Read March 6, 1884.] (Prarzs I.-IV.) Duniwe the last two months I was at Kew I was employed in revising and re-sorting the Indian species of Cyperus in that Her- barium. In the first half of the genus I was able to look up the Synonymy and note the geography pretty fully, and to compare a considerable number of the non-Indian species. In the latter half of the genus, though I was able to sort and examine the Indian material, I was too much pressed by time to be able to write down much. Nevertheless, on leaving for India, Sir J. D. Hooker expressed a wish that I would publish my work on Indian Cyperus in some form, as it might assist another day in the elabo- ration of the genus for the ‘ Flora of British India.’ Since arriving at Calcutta I have been able, through the kind- ness of Dr. G. King, to compare the Calcutta Herbarium, which is immensely rich in Indian material; and I have been able to nearly complete the paper, so far as the description of the Indian Species is concerned; but in the latter half of the genus I have been unable to touch the synonymy (there is not here even a copy of Bentham’s ‘Flora Australiensis, so that I have been obliged to employ MS. names of F. Mueller probably long ago LINN. JOURN.-—BOTANY, VOL. XXI. B 2 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. superseded) ; nor have I been able to give much account of geo- graphy beyond the limits of India, nor to compare (and illustrate subdivisions with) many African or American species, except a few very well-known ones. The present paper, therefore, in these * respects, presents a gradual tailing off. The Society may fairly ask me, why, then, publish so unequal a paper? And my reply is, because it contains a good deal of work, and because many years must elapse before I can hope to make it more complete. Of course I may never be able to add any thing to it. All papers, at least of a systematic kind, prepared.in Asia, Africa, or America, must be, as literary work, very poor perform- ances in the eyes of botanists in the herbaria of London, Paris, and Geneva; but it has not been, I believe, hitherto the system of the Linnean Society to reject, on the score of incompleteness, any paper that contains additions to the sum of scientific know- ledge. And the most highly-finished systematic papers of Euro- pean botanists are often, though in form perfect, extremely in- complete really: completion and finality are alike in such work unattainable. In this paper Sect. I. is the result of taking each part of a Cyperus, culm, inflorescence, &c., and comparing it in all the Indian species ; Sect. II. contains a discussion of some difficult species and disputed genera ; Sect. III. is a systematic arrangement, with descriptions of the Indian species, with short citations of some non-Indian species that more particularly illustrate the subdivi- sions and groups. SECTION I. (1) THe Rurzome. There are two classes of Cyperus, viz. (1) strictly annual, with fibrous roots ; (2) biennial or perennial, or with a horizontal rhi- zome. In this second case the rhizome may be (in the same species) very short, with very short nodes and cespitose culms, or the stolons may be elongate, developing into creeping rhizomes with long joints and distant solitary culms. Of the second class, a few species appear to flower sometimes in the first season, and examples of such in the herbarium cannot be distinguished from annuals. The regular annual species, once-flowering, such as Cyperus pumilus, I have never known MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 8 to produce stolons or seen with a woody rhizome ; though it is not improbable that if prevented flowering, by a jungle-fire or otherwise, just when shooting for bloom, they might take on a different habit, and live to flower in the next season. At each node of the rhizome there is normally a scale, often 1-2 centim. long, usually slate-coloured or blaek-brown, almost scarious in texture, lanceolate or sometimes almost ovate, some- times almost linear. In the figures which I give I have removed these, partly because the essential features of the rhizome can be more distinctly shown, partly because I am not artist enough to draw them properly, and can only attempt pictures virtually “diagrammatic.” Figs. 20 and 21 represent two states of the rhizome of C. levi- gatus, Linn., which may be seen in one plant at Suez: on the dry sandy bank the nodes of the rhizome are very short, very tough, the plant rigid, with white spikelets usually (fig. 20); but directly the same rhizome touches the brackish pools adjacent, it com- mences to run freely (as in fig. 21), the rhizome becomes supple, the spikes become generally chestnut-purple. When the same rhizome floats out on the sweet-water canal, the culms become dwarfed, with often only 1 or 2-3 spikelets. Boeckeler has founded two new species (C. viridulus and C. rep- tans) on Peruvian plants which I have seen, and which differ from C. levigatus only in habit, and in habit scarcely more than C. levigatus varies in itself. 1 doubt whether they may not be merely geographic forms of C. levigatus. At all events, both from the example of C. levigatus and from many other well- known species, I think it is to be inferred that a difference in rhizome alone (supposing the plant not annual) is insufficient ground for specific separation. In C. umbellatus, Benth., in the type form alone, we have cæspitose fibrous roots, or a short thick woody rhizome, or long slender stolons becoming ultimately tough black slender rhizomes. I propose to separate specifically all the annuals from the others. The only difficulty arising is in the case of C. Haspan. This has typically an elongate rhizome, the result of stolons (fig. 23); but we frequently have, as Boeckeler says, the rhizome exceedingly short (fig. 24); and beyond this we have flowering and fruiting examples, apparently annuals, with no trace of a rhizome. I must allow these, though I separate specifically as C. flavidus, Retz., a large portion of Boeckeler's C. Haspan, which indeed he B 4 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. has subsequently made the new species C. microcarpus, Boeck. C. flavidus, I believe, never exists more than six months; it is one of the universal annuals in late dibbled rice, the culms are often clustered with tangled fibrous roots, and appear sometimes to grow together (fig. 25); but I have never seen any thing like a rhizome in this species. But though there are these (and other) difficulties in employ- ing the general characters of the rhizome for the discrimination of species, the particular characters of the rhizome are in many species exceedingly constant for that species, and in some cases afford the best specific character we have. In C. rotundus, Linn. (fig. 16), the culm arises from a nodule of - the rhizome, the remainder of the rhizome being slender wiry ; the nodule is formed by a number of shortened joints of the rhizome, secreting much starch, whence the strain on the plant caused by fruiting is met. In C. longus (fig. 14) the rhizome is nearly uniformly thick; there are no very prominent nodules whence the culms spring, nor any very slender wiry portions between. I confidently refer Aitchison’s n. 684 (marked by Boeckeler C. longus) to C. rotundus. A species much confounded with the preceding is C. escu- lentus, Linn. (fig. 15); the stolons are very slender, herbaceous, not wiry. In this species (and in many others, as C. procerus) they appear to burst from the base of a flowering culm; whereas in C. longus, C. Haspan, &e. the rhizome creeps and the flowering culms spring from a subterminal bud on it. But both these modes of growth may be seen in the same plant of C. arti- culatus, C. scariosus (fig. 22), and others. C. esculentus and many other species produce tubers on the rhizomes. Another plant much confounded with C. rotundus is C. jemi- nicus (fig. 17, 18), which is very easily separable by the rhizome, the base of each eulm being enclosed in black horny prominent lanceolate cuspidate scales. Fig. 18 shows one culm in flower, the next not started from the circle of black scales; while fig. 17 shows the same with the black scales rubbed off from the base of the culm, a common state in herbaria, as the scales are ulti- mately caducous; and if a herbarium specimen shows no young circle of black scales, it is usually marked C. rotundus. C. usi- tatus, Burchell, is very near C. jeminicus ; but in it the rhizome is wiry elongate, so that the thickened “bulb” is usually 1—4 centim. distant from the base of the flowering culm. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 5 It must not be inferred, however, that the characters of the rhizome are generally useful in this genus for separating closely allied species ; the contrary is the case, e. g. in C. umbellatus and its numerous subspecies and allied species the rhizome is nearly the same; while in the long series of forms included by Boeckeler under C. polystachyus he says the root is fibrous, or there is a short rhizome, or the rhizome is elongate, woody ; it is, I believe, always perennial. Again, in the series of plants allied to C. ele- gans, Linn., the rhizome is generally much the same. This leads us to the question how far we can utilize the rhizome for characterizing groups higher than species; and the answer is, as so often in botany, Sometimes we can, sometimes we cannot. In C. conglomeratus and its allies the main roots are densely . clothed on all sides with rootlets so as to appear thick and woolly ; the character of roots (thick, woolly) defines a very natural group, to which belongs C. proteinolepis, Steud., removed by Boeckeler to another group. The forms of the C. conglomeratus group of Boeckeler are indeed so closely allied that Boissier (probably rightly) unites, under the old C. conglomeratus, several of them admitted as species by Boeckeler. The species of the group Corymbosi of Boeckeler (C. rotundus, C. longus, &c.) are connected naturally by the stolons or ereeping rhizomes. C. tenuiculmis, Boeck. (=C. lucidulus. Klein), should be placed next C. compressus; and C. sphacelatus, because of its fibrous root, I think should not be arranged here, nor the American C. letus, C. dissolutus. The whole of the Exaltati (in which I include the Papyri) are without stolons: the stout culms are generally solitary; but as the culm dies down after flowering there often appear lateral suberect shoots from close to the base ; such shoots also occur in several stoloniferous species, and in some with creeping rhizomes. They may occur, accidentally probably, in nearly every species of the genus; and though they are generally easily distinguished from stolons, in C. procerus there is a very gradual transition from one to the other. The rhizome is thus a character to which we must pay a very great degree of attention in some parts of the genus, and little in others, whether for specific or subsectional sorting. The roots of the same species become manifestly thicker, more spongy, when the specimen grows in a very wet place. The species that grow in damp woods, like C. elegans, have usually 6 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. very slender tough black roots; while the species that actually float in tanks (C. cephalotes, C. platystylis) have very thick roots. The woolly-rooted species are confined to Arabia and the neigh- bouring desert countries, and assist probably in collecting water, or perhaps give a better hold on the sand. (2) THe CULM. The culm is erect, never branched until the umbel, and 0-1 metre (a very few species attaining 1-2 metres) high. Itis naked except for -5—4 its length near the base, where it is sheathed by the bases of the leaves; in a few species, especially in C. Mundtii and C. distichophyllus, the culm is enveloped half its length by the leaf-sheaths, as also sometimes in C. Eragrostis. Great variation in the length of the culm (accompanied by varia- tion in the development of the umbel) occurs in many spe- cies, and depauperated examples are common. C. Eragrostis, Vahl, as collected by Schrenk at Lake Alakul (called C. flaves- cens, B. rubromarginatus, Schrenk) is only 2-3 centim. high, each culm bearing 1-3 spikelets only: a stunted (not depauperated) form of C. compressus is not uncommon in which the culms are 1 centim., the umbel contracted but of numerous large spikelets. C. latispicatus, Boeck., is very variable in this respect; and I have collected (in Khasia) a form of this absolutely culmless, the well- developed umbel being exactly radical. The culm is generally much longer than the umbel-rays; but C. radians, Nees et Meyer, and C. Griffithii, Steudel, are at once distinguished by having the umbel-rays 2-3 (or more) times as long as the culm. The stoutness of the culm ranges from 1-2 centim. down to 1-2 millim. in diam. ; yet it is a much more useful character than the length, each species varying in this respect between evident (though troublesome-to-define) limits. Indeed the stoutness of the culm is one of the main characteristics, both with Kunth and with Boeckeler, for forming natural subsections of the species. The group Exaltati (including Papyri) is eminently natural, as also the slender-culmed Aristati. The stout character of the culm is maintained in very short-cuimed examples of Exaltati, and the slender character of the culm in very long-culmed examples of Aristati. The culm is trigonous; but in some species this character is modified until the culm appears perfectly terete ; in others itis MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 7 intensified, and we have a triquetrous stem. The culm is com- monly nearly round below, becoming trigonous upwards, most sharply so under the umbel. The culm is frequently more or less unequally trigonous, and sometimes excessively so, until it is described as complanate, as occurs in C. Haspan (type) and C. malaccensis. In C. articulatus (and C. nodosus, scarcely spe- cifically distinct) the culm is terete, quite smooth, continuous when fresh; but after a day or two in drying-paper the pith thickens at regular intervals, giving the culms a falsely articulated appearance. This character, not uncommon in Eleocharis and Scirpus, is not distinctly brought out in any other species of Cyperus. The degree of sharpness of the angles of the culm is marked in certain species, as in C. turgidulus, in which the culm is 3-winged, and is in specific descriptions usually treated as an important point, but I fear can only be relied on as a subsidiary character. C. tegetiformis, Roxb. (C. enodis, Boeck.), has been separated from C. corymbosus, Rottb. (C. diphyllus, Retz.), by the more acutely-angled culm (especialiy close under the umbel); but I doubt whether the two forms can be separated by that or by any other character. Many species (both among those with slender and those with moderately thick culms) have the faces between the three angles striate ; and these striations in some species are strengthened so as to obscure the angles, until we finally geta terete, striated, or fluted culm. Such culms are often accompanied by rigid leaves (with inrolled margins), and are characteristic of dry-country species, as the Conglomerati of Arabia and many species of the Great Thirst-Land in South Africa. The culms in Oyperus are generally smooth, or only microsco- pically scabrous on the angles near the top. In C. platyphyllus, Roem. et Sch., a very large species, the angles of the culm are serrulate, cutting the hand. The species of Cyperus are in general quite glabrous; in a few species there is a pubescence on the rhachis of the spikelets, and in a very few species there are very slender hairs on the leaves. The pubescence only extends to the culm in the following :— in C. viscosus (and one or two allied species) the eulm is puberu- lous, or with microscopic papillæ giving it almost a sticky cha- racter; in C. retrofractus, Asa Gray, the culm is manifestly pilose, 8 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. (3) THE LEAVES. The uppermost leaf is generally narrow-linear, with a midrib, often scabrous on the margins towards the tip with forward- pointing serrulations; with many longitudinal nerves, certain of which are in some species especially prominent ; flat, green, in the moist-country species; straw-coloured, rigid, inrolled in the dry-country species; the lower leaves are by degrees shorter, becoming lanceolate or even triangular, the sheath often short- ened also, until the lowest leaf may be a nearly free ovate- lanceolate scale, hardly different from the scales on the stolons. The length of the leaves is an obvious character, dear to syste- matists, and employed by Boeckeler even as a subsectional cha- racter; but in my opinion it is hardly characteristic in any one species, and it should be employed cautiously as a subsidiary character only. Beginning with C. Haspan, in typical forms the leaves are very short, often reduced to mere short-lanceolate appendages of the sheaths; but there are other forms with the leaves more developed, and in very numerous. examples they are longer than the culm. (This is true even after every example possible has been separated off into C. flavidus, Retz.) In C. latispicatus, Boeck., the uppermost leaf is often as long as the culm, but generally most of the lower leaves are reduced to sheaths ; C. aphyllus, Boeck., is the same thing, all the leaves being reduced. Boeckeler places C. enodis, Boeck., C. diphyllus, Retz., and C. corymbosus, Rottb., in a sect. Subaphylli, from which the Foliati beginning with C. Schimperianus, Steud., are separate. The fact is that in C. corymbosus, Rottb., the leaves vary from O to the length of the culm ; and that the greater portion of the Indian material sorted under C. corymbosus, Rottb. (on the character of length of leaves), by Boeckeler is C. fegetum, Roxb., a species from which C. Schimperianus, Steud., cannot be separated. (I state this broadly, as C. fegetum is separable from C. corymbosus defi- nitely by the structure of the spikelets.) In C. zegetum also the leaves may be as long as the culm or subobsolete, and in the nearly allied C. scariosus, R. Br., the variation is nearly as great. After collecting C. tegetum and C. corymbosus with my own hands for years, I am of opinion that the leaves are nearly or quite valueless in sorting these and their allies. C. diphyllus, Retz. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 9 is as near as may be typical C. corymbosus, Rottb.; the longer- leaved forms of this (added to much of C. tegetum) make up C. corymbosus, Boeck. ; short-leaved forms are doubtfully distinct specifically from C. tegetiformis, Roxb. (C. enodis, Boeck.), in which the culm (dried) is often obscurely septate. In C. articulatus itself the leaves very generally are subob- solete, but accidentally I have seen leaves 1-2 decim. long even in this species. The roughness of the leaf-margins is a character of very small value: most species exhibit it to some extent; and in the same species the roughness may be evident in some leaves, obscure in other less developed-leaves. The roughness also extends in many species to the upper part of the midrib beneath. The midrib is very prominent in some species, so much so as to be described by Boeckeler as “ winged ” in C. lucidus, R. Br. As in the case of the culms, the breadth and stoutness or tex- ture of the leaves is a much more constant character, both in species and in subsections, than their length. The rigid, thick, ineurved, straw-coloured leaves of the Conglomerati have been mentioned. In the group of Elegantes the leaves are unusually broad, flat, green, prominently many-nerved. In the whole sec- tion Mariscus the numerous grass-like leaves are similar. In the Exaltati the leaves are stout, with mostly compressed sheaths ; and in C. platyphyllus attain 2-3 centim. in breadth. In several Species, from various groups, the leaves are of coarser looser tissue than usual; so that when dried they appear full of trans- verse small septa. This is a very constant and convenient cha- racter in particular species, as C. dilutus, Vahl, C. canescens, Vahl, C. virens, Asa Gray, C. Lechleri, Steud. ; more obscure in C. procerus, Rottb., and many others. The leaves are very generally glabrous ; but hairy or pube- rulous in the species mentioned above as having hairy or pube- rulous culms. Also in C. strigosus, and a few other American species, the leaves are strigose upwards, with such very fine hairs, however, that they are hardly noticeable until searched for. (4) Tue Bnacrs. The bracts to the umbel are the cauline leaves of the plant, and are always similar to the radical leaves. This is not true merely of species, but of individuals : in a leafless specimen of C. Haspan or C. corymbosus the bracts will be short, much shorter 10 MR. C. B. CLABKE ON INDIAN SPECIES OF CYPERUS. than the umbel; but in leafy examples of C. Haspan or C. corym- bosus the bracts will be long, far overtopping the umbel. The most nearly leafless species in the whole genus is C. articulatus; and in this the bracts are 1-2 centim. long only. In breadth, scabrousness (or pubescence), nervation, and texture the bracts invariably resemble the leaves; in many cases therefore, where a critical species as C. Haspan is propped up by characters taken from the bracts, there is no real combination of characters, and the delimitation of the species is not materially strengthened. In C. vaginatus, R. Br., of which I have seen very little material, the leaves are reduced to short ovate acute terminations of the sheaths, the bracts are 2-3 centim. long, linear, rigid: this may be a partial exception to the rule above laid down; but I should expect among a large quantity of C. vaginatus to find examples with the leaves developed to 3—6 centim. long, and in such case I feel confident they would closely resemble the bracts. The bracts are in most species apparently umbelled, but they are always really alternate ; and in many species they are mani- festly alternate, so that the umbel is said to be imperfect or eorymbose: Linneus named C. alternifolius from this character. C. jeminicus (when the rhizome is wanting) can be separated from C. rotundus (to which I find C. jeminicus without rhizome very generally referred in herbaria) by the lowest bract being clearly remote (2-3 millim., sometimes by 1 centim.) from the one next above it—a trifling, but so far as seen invariable, mark. The bracts are usually divaricate or erect-patent; but in several species the lowest bract is nearly erect, sometimes quite 80, 80 as to appear a continuation of the culm, and the umbel appears lateral, as in C. levigatus, C. stramineus, C. pauper, and in several of the Conglomerati. The character is useful in recog- nizing some of these species rapidly; but it is of small value, for, of many species with spreading bracts, in smaller specimens with 2 or 3 bracts the lowest bract becomes nearly erect, as in C. flavescens depauperated. From the axil of each bract springs a ray: the number of rays is thus never less than the number of bracts ; it is in general more, because, from the shortening of the uppermost nodes of the axis, the bracts are suppressed though the rays belonging to them are developed. This law can be supposed true, however, only theoretically in the case of con- gested or capitate umbels, which often appear as simple heads with numerous bracts. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 11 The bracts are usually 3-6 in number, occasionally 3-10. In C. equalis and C. Papyrus, where the number of rays may reach 100, the number of bracts is not increased at all in proportion. (5) TuE UMBEL. In Cyperus the inflorescence is always an umbel, 7. e., as we have seen, a congested corymb, which may be reduced to a dense capitulum, or may be compound or supradecompound ; in the compound umbels the umbellules are similar to the main umbel, but often less congested, till in many species they lose the um- bellate appearance altogether, and can only be described as open corymbs, often with patent or divaricate branches. The corymb is always very much depressed, so that the nodes between its upper branches are obsolete, while the branches themselyes gradually increase in length from the uppermost, which is 0, to the lowest, which may be 0-3 decim. The main umbel thus has more or less unequal branches with a sessile spike in its centre, and the secondary, tertiary, &c. umbels take the same form whieh gives Cyperus its distinetive appearance. Perhaps the most aberrant species in the genus in habit are thus C. Papyrus and C. equalis, in which we have a great number (up to 100) primary rays (instead of the usual 3-10), and these all very nearly equal in length ; and in these two species the central sessile spike is depauperate amid the crowded bases of the rays or altogether suppressed. But throughout the genus the struc- ture of the umbel is most essentially and minutely the same; and great differences in its aspect are caused by mere variations in the length (and angle of divergence) of the primary, secondary, &c. rays. This identity of real character has always been a source of difficulty in the verbal definition of species and subsections of Cyperus, in which all authors are agreed in regarding the umbel as a character of great weight. The development of the umbel varies enormously in nearly all the compound species ; 2. e. those species which, when fully de- veloped, have a compound or decompound umbel, produce in numerous examples only a simple or small umbel. Thus C. rotun- dus, Linn., has ordinarily a compound (often lax) umbel with very numerous spikelets ; but small examples often have a simple umbel or even a singlespike ; and a form which I have called form ** Sal- sola" is common near the sea (as at Calcutta), in which each 12 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. culm produces 1-5 spikelets only. The number of spikelets on one culm of C. platystylis varies from 1300 down to 5 in the series of examples in the Calcutta herbarium. In C. pilosus the umbel varies from a capitulum 3 centim. in diameter up to a long-rayed umbel 3 decim. in diameter, and the spikes themselves vary so greatly in density that numerous species have been founded there- on. The degree of development of the umbel would therefore be a character of little value, but that there is a great number of species in which the umbel is apparently simple, and never (or most rarely) in these varies so as to be clearly compound. Yet in the species best defined as to their umbels variation occurs, so that both Kunth and Boeckeler avoid as much as possible ** umbel simple," “umbel compound ;" though as a description of the general appearance and average character of the inflorescence such terms are short and convenient. Thus in C. pumilus, Linn. (i. e. C. nitens, Vahl) the umbel has nearly always the definite appearance of being simple, with rays carrying heads at their extremities (or, as in all such inflorescence, the umbel may be reduced to a capitulum). In numerous examples the simplicity of the umbel appears absolute; but in those having heads of numerous spikelets it is easily seen that these heads are not simple spikes, but have their axis divided. In very fine speci- mens a green bract 1 centim. long will sometimes be found under the heads; the presence of such a bract (homologous with the main bracts of the umbel) is here (as I believe in all species) a sure indication that the apparent head is really an umbel. In a specimen I have now before me, C. pumilus has distinctly a compound umbel, the secondary rays attaining 15 millim. in length. In this genus the distinction in the number of subdivisions of the axis before we come to the flower, a distinction held of such high value by Eichler &c. in Kobresia and Carez, is, from the nature of the inflorescence, utterly illusory. In C. polystachyus, var. ferruginea (fig. 27), the umbellule is laxly corymbose, the lower branches manifest, bearing 4-2-1 spikelets ; and from the picture it is evident that the upper solitary spikelets may or may not at some arbitrary point be considered as spikes of one spikelet. Other primary rays on this same plant carry apparently simple spikes which are clearly homologous, being = fig. 27 with the two lower branches suppressed. These are treated as simple spikes in ordinary descriptions; and indeed in ordinary descrip- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 13 tions arrangements like that in fig. 27 are described as “rays with 3 spikes, terminal spike elongate, lateral short or auriculate : " this is a frequent occurrence in Cyperus, typical in C. Luzule, C. lucidus, and many others, and occurring occasionally in C. umbellatus, Benth., and other species where it is not admitted by the book description. It is so characteristic of the very nature of Cyperus inflorescence, that I should not be surprised at its occurrence anywhere. C. polystachyus is a well-known species excellently adapted for a discussion of the inflorescence of Cyperus. In even mode- rately-well developed examples I find the umbel invariably com- pound ; the umbellules are often congested into dense heads, but their axis is then evidently divided. In the vulgar form of C. polystachyus, common near the sea in nearly all the continents and islands of the tropical world, the umbel is usually (as in C. rotundus and some other species when near the sea) greatly con- tracted; and this form, the commonest in collections, has been treated as C. polystachyus type; while a multitude of “species” have been created by herbarium botanists out of the inland forms with more or less open umbel. In nearly every example, whether with open or close umbel, the smaller few-flowered umbellules may appear perfectly simple spikes until they are compared with all the other spikes on the same plant. But in C. filicinus, Vahl (— C. Nuttallii, Eddy), esteemed, not without reason, a var. of C. polystachyus by Boeckeler, every spike in every specimen I have is in appearance perfectly simple ; fig. 26 exhibits one of the largest. The spikelets themselves are here considerably larger than in ordinary C. polystachyus; but it occurs in many species with various umbels that, besides the standard form, there occurs another with the umbel divided one time less, and the spikelets considerably larger. l In this fig. 27 I have shown the two lowest glumes of the spike- let A and B, alternate, smaller than those above them, and both (or sometimes one only) empty ; the lowest empty glume (A) is (in other species) sometimes truncate, sometimes with a long caudate point bracteiform : this is a useful character in specific diserimi- nation. In other species too, as a first indication that the spike is beginning to become compound, there is a green caudate small bract close below A and on the same side of the spikelet (it is therefore I call it a bract, and not a glume). This is (at least I hope I have, after much labour in distinguishing the bract from 14 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. the glume, come to that view) a point of little or no value in defining species. The spikes (as explained above under C. pu- milus, and as shown by the discussion of fig. 27) pass so insensibly, in the same species and in the same plant, from simple to com- pound, that the presence of this bract is even in species like C. inundatus, Roxb., where it has been fixed on as an important character, highly uncertain. The characters of the spike itself are of high differential im- portance, and also in high degree constitute the features of the whole umbel. We have first the spikes of few spikelets, as in C. elegans, the spikelets strictly digitate. In some of this set (as C. diffusus, Willd., C. multistriatus, Boeck.) we have “ pedicelled spikelets,” which I consider, however, always as spikelets sessile in spikes of one spikelet. From this we have “ stellate spikes;” and so by degrees the rhachis more elongate, until we have cylindric spikes with a rhachis 1-5centim. long: in these the spikelets may be densely crowded or distant, but many species (as C. exaltatus) are very variable in this respect. When the spikes are long, bowing with subremote spikelets, they are often described as racemose, which view is somewhat supported sometimes by the remote empty - (or deciduous) lower glumes; but I think in all these cases the spikelets should be considered sessile and the ultimate divisions of the inflorescence “ spikes." While the inflorescence is thus flexible in character, it must remain a chief character in all species; but exactly the same kind of inflorescence meets us in various (and remote) sections of the genus, so that reliance on external general character of the inflo- rescence, without also examination of the nut and stigmas, has led botanists into numerous errors for the last century. C. cepha- lotes, C. pygmeus, and C. dubius all have a single capitulum, the leaves and braets much alike; they were variously confounded, then mixed, then reseparated under new names by Roxburgh, Wallich, and others. Boeckeler has united with C. pygmaeus also Isolepis Micheliana, Roem. et Sch., which I agree with Kunth (as attempted to be proved below) to be of a different genus. C. procerus and C. puncticulatus, than which no two species are more essentially distinct, are confounded in the description of Roxburgh and in the collections of Wallich, and the confusion has (by the issue of specimens correctly matched from Kew) been widely dis- seminated. Roxburgh, after figuring C. amenus, Heyne MS., guessed, on comparing Rottboell’s figure of C. alopecuroides, that MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 15 (from the similarity in umbel) it must represent the same plant, despite the fact that Rottboell gives a totally different nut. The error has been continued to Boeckeler, though it is possible that Boeckeler has never seen the true C. alopecuroides, Rottb., and that he has only seen C. alopecuroides, Roxb., and allowed it occa- sionally to have a dorsally compressed nut, in order to include Rottboell’s. However that may be, Boeckeler's C. alopecuroides is (as he says) very uearly the same as C. dives, Delile; and both may, I think, be best considered mere varieties of C. exaltatus, Retz., while C. alopecuroides, Rottb., is very closely allied to C. Monti. A more difficult case is the union of C. inundatus, Roxb., with C. procerus, Rottb., which I also believe an error, founded on trusting too absolutely the general aspect of the umbel. The character of the umbel is put forward prominently, both by Kunth and Boeckeler, in grouping the numerous species of the sub- genus Lucyperus into sections: a matter so difficult that we may fairly catch at any straw that can possibly aid us. Some groups, as the Exaltati (including therein the Papyri), are allied by an inde- finable similarity of umbel. But the character of the umbel, even in hands of the experience of Boeckeler, will not always enable us to place a species next to its neighbour. Boeckeler places C. jeminicus, Rottb. (= C. tuberosus, Vahl), among the Corymbosi, whereas I take the plant to be very near to (specifically distinct from) C. usitatus, Burchell, which Boeckeler has in a very different place. ©. tenuiculmis, Boeck.=C. Zollingeri, Steud.— C. luci- dulus, Klein, is placed also by Boeckeler among the Corymbosi, but has been rightly placed both by J. D. Hooker and Hance (under other names) close to C. compressus. Many of Boeckeler's sections include great variety of inflorescence ; and my argument is, not that I am prepared to propose a much better grouping than his, but that the value of the umbel as a character in forming groups must be regarded as doubtful until backed up, and as un- equal in different parts of the genus. l In the type section of the subgenus Mariscus the character of the inflorescence appears more constant, even in variable and composite species. In C. wmbellatus, Benth., there is one (appa- rently) spike on each primary ray. This character 1s constant through an enormous range of subspecies, though occasionally I have seen one or two smaller spikes added on one ray. One spike on each ray seems also the invariable rule in C. retrofractus, C. ovularis, Torrey, C. flavus, and the whole of this series ; while 16 ME. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. in the series C. Andersonii, C. Seemannianus, &c., digitate spikes on each ray appear equally constant; but of these species I have seen few examples; of C. umbellatus, Benth., thousands (in a literal sense). The rays and radioles of the umbel are terete, triangular, or compressed, much as the culms; very generally glabrous, but hairy in the few species with hairy culms. The rhachis of the spikes is generally angular, deeply grooved, and occasionally minutely scabrous. In C. pilosus the scabrousness of the angles of the rhachis of the spikes becomes “ pilose ;" 2. e. the rough points come very close together, and take the form of very short brownish hairs. This distinguishes the species from all others readily except from C. procerus, in which the rhachis of the spikes is some- times glabrous, usually scabrous, sometimes * pilose," or at least undistinguishable from that of C. pilosus. (6) Tux RHACHILLA OF THE SPIKELET. The examination of tbe rhachilla, especially after some of the lower glumes have fallen oif, is a convenient way of observing the structure of the spikelet. A glance at the rhachilla of Isolepis Micheliana will show how far the plant must be removed from Cyperus, in which the notches are on two opposite sides of the compressed (or subquadrate) rhachilla ; and a similar glance will show Cyperus Ranko, Steudel, to be an Isolepis. The great variety in the rhachilla is seen by comparing a few common species, as C. puncticulatus, Vahl (fig. 30), C. Monti, Linn. (fig. 29), C. luci- dulus, Klein (fig. 28), C. auricomus, Sieber (fig. 31), C. flexuosus, Vahl, C. umbellatus, Benth. ; and the differences are not merely manifest, but constant; and are treated as of importance, for species and for groups, by Kunth and Boeckeler. The bases of the glumes are decurrent on the rhachilla, forming two lines of * wings" toit on either face. These wings are in some species (as in many of the subgenus Pycreus) obsolete ; in others, as in C. compressus, they unite to form narrow continuous scarious wings running down the whole length of the spikelet; and such wings are in many species broader, sometimes purple-spotted. In other species the wing is not continuous (at least not in subequal breadth); frequently the base, sometimes nearly the whole, of the nut, lodged in the notch of the rhachilla, is held on each side by the widened wing of the rhachilla. In most species the wing separates off from the base of the glume, long before the nut is MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 17 ripe, by a transverse line; and in a limited number of species the wing, at nearly the same early period, separates from the rhachilla on which it is decurrent, and thus drops off freely: i. e., in the words of cyperologists, the wing is soluble, not persistent. The soluble wing is often discoloured, yellow or reddish, and thus pro- minent, and is the character which separates the Papyriof Kunth and Boeckeler from the Exaltati. In the Exaltati, however, I find sometimes a very small linear (but discoloured unmistakable) soluble wing. Also in C. tegetwm, and some neighbouring species, the solubility of the wing appears to vary greatly in the same Species, and the wing differs altogether in degree and character in species placed close together by authors. The solubility of the wing of the rhachilla is a useful character, but one concerning which I can draw no hard-and-fast line; it connects the Papyri, a series of species naturally connected by many other characters ; it only throws doubt on the next group, the Corymbosi, as to whether they have been rightly placed together. The form of the rhachilla (much compressed or quadrangular) is not of sectional value, 7. e. a subquadrangular rhachilla occurs in species from various sections of the genus, as in C. globosus, C. Monti, &c.; but it is very useful in separating some jumbled species, e. g. the quadrangular rhachilla of C. alopecuroides, Rottb., distinguishes it from C. alopecuroides of Roxb. and Boeckeler, and the shape and markings of the notches in C. alopecuroides, Rottb., support its otherwise evident affinity to C. Monti. The shape of the notches of the rhachilla, which vary from quadrangular (or even broader than long) up to narrow-oblong and linear-obovate, is à character rarely used, for it is (in the main) involved in the character “ glumes crowded” or “glumes remote ;” but, in the case of critical species, it may be well used, not as a mere repeti- tion, but as capable of much greater definiteness and of more additional detail than the ordinary “ glumes subremote.” The later groups of Cyperus—viz. Leptostachyi, Mariscus, Dicli- diwm—differ from all the rest, not so much in the number of the flowers or the persistence of the wing of the rhachilla, as in the per- sistence of the glumes, which do not fall off by a clean disarticula- tion from their decurrent bases. In the Leptostachyi the spike- lets fall entire from the papilliform disks on which they are seated before the glumes fall from the rhachilla, usually leaving the two lowest barren glumes still attached, or leaving only the papilliform disk bare; while the rhachilla is very narrow. Mariscus 1s nearly LINN. JOURN.—BOTANY, VOL. XXI. c 18 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. the same; but the rhachilla is usually thickened, often almost spongy. In Diclidium the spikelets finally break up into one- seeded joints, nearly in the manner of Rottbellie among Grasses. (7) Tae GLUMES. In Cyperus perhaps more attention than is deserved has already been paid to the glumes, more especially to their being mucronate or obtuse, to their colour, and to the greenness of the dorsal nerves. As to the mucronation, there is no point more illusory, even in the same species and example. The green nerves of the back unite and are just excurrent as a mucro close beneath the apex of the glume, or are not excurrent; and the two cases fade one into the other, as in the common species C. pumilus, Linn. (i. e. C. nitens, Vahl), of which the varieties called mutica, patens, trun- catula are without the mucro. The same variety in mucronation occurs in the Conglomerati, and almost throughout the genus; though in particular species, as in C. rubicundus, Vahl, the mu- cronation appears characteristic. Still more untrustworthy is the mucronation as a sectional character: C. amabilis, Vahl, one of the most characteristic of the Aristati, is often absolutely devoid of aristation ; while C. lucidulus, Klein (correctly placed next C. compressus by Hooker f., and Hance), has been removed entirely out of its proper place by Boeckeler, apparently solely because its glumes are not acute cuspidate. The colour of the glumes is an even more dangerous character to lean upon: it still remains as diagnostic in C. melanostachyus, C. atronitens, &c., and many other species of which (with their allies) we know little; but in the widely-known C. polystachyus, C. glo- bosus, C. levigatus, &c., colour goes for nothing as a specific cha- racter; we have, indeed, the whole range of colour from straw- colour to black in the self-same species. There are indeed few speeies amply illustrated in our herbaria of which the range of colour is not considerable. Nevertheless, in a broad way, a par- ticular tint characterizes most species. C. flavescens is nearly always yellowish ; C. dilutus is nearly always a reddish brown, &c. The occurrence in Cyperus of glumes which are 3 —7-nerved, green on the back, yellow, red, purple, &c. on the sides, is so common that the green nerve is hardly worth mention in specific descriptions; and from accidental circumstances the usually green-backed species sometimes have no green tint. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 19 The difference between round-backed and carinate glumes makes the difference between terete and compressed spikelets. In many species there are no nerves except near the keel of the glume. In other species the glumes are plicate-striate through- out nearly their whole breadth. This character is useful, not merely in separating some critical species as C. esculentus, but in defining some sections, as the Nivei. The open cellular texture of the glumes of C. hyalinus, Vahl, is strikingly unlike any other species of the genus. The glumes are generally quite smooth; in a few species the keel upwards is scabrous under a magnifier. More important points in the glume are the dorsal compression in sect. Juncellus, accompanying the dorsally-compressed nut, and its insertion, already considered under the rhachilla. (8) THE STAMENS. (a) Number. The number of stamens is 3, 2, or 1; in most large species 3, in many small ones (whether the style be bifid or trifid)2-1. The sect. Luzuloideæ (C. vegetus, C. virens, C. Luzulæ) are almost the only large or medium-sized species in which the stamen is 1 only. In many species the stamens are 2 in well- developed, 1 in poorly-developed specimens, or mainly so. In Pycreus perhaps two thirds of the species have 2 stamens (or 1), but this goes with the size mainly of the species. In the small species of Eucyperus (as in sect. Aristati, and in C. flavidus and small C. Haspan) the stamens are also 2-1; in Juncellus the large species (as C. Monti, C. inundatus) have 3 stamens, the small (as C. levigatus, C. pannonicus) 2-1. The number of stamens rarely varies 3 or 2 in one species; so that we have in the number of stamens a character often useful as a specific, sometimes as à sub- sectional distinction. (b) Filaments. These are always more or less ligulate ; they are exceedingly broad, somewhat widened upwards in the Conglome- rati and in C. cephalotes and C. platystylis. They are usually distinetly ligulate, and only obseurely so in the small-flowered Species with small stamens. The length of the filaments and exsertion of the anthers are Sometimes described; but there is much uncertainty in the value of the character, and the filaments (apparently from an obscure tendency to unisexuality) sometimes romain short in c 20 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. nearly all the flowers of a spikelet or of a plant, so that errors may easily arise here. The tissue of the filaments is of lax oblong cells, and in many species (as C. globosus) this is so lax as to give a scabrous-papil- lose character. The filaments look here rough to the naked eye, but there is hardly a species in which the microscope will not — reveal a structure essentially the same. The persistency of the filaments is very characteristic of some species (as of C. globosus), where they remain attached, without anthers, after the glumes have fallen. The anthers themselves are always normal, oblong or linear- oblong, yellow, red, or white, with purple spots. The anthers are in the majority of species simple, muticous, or with a minute apiculus at the apex; but in many species the anther has a lanceolate scarlet crest, very constant in size for each species ; in some minute and not always to be made out, in others prominent, in C. platyphyllus (fig. 82) two thirds the length of the anther. Iu C. levigatus (fig. 33) the crest is much smaller; in C. platystylis (fig. 7) and several others it is small, truncate ; in C. Haspan (fig. 34) the crest is obsolete, only a few of its linear acute papille sometimes developing. The anther-crest is wanting in a majority of species ; but where present it is, from its constancy, a very valuable character in the separation of species. In C. inundatus, Roxb., there is a depressed rudimentary crest exactly as in the closely-allied C. Monti. In C. procerus, Rottb. on the contrary, the anther is absolutely ecristate and similar to that of the allied C. pilosus. The crest of the anther also has a subsectional, but no sec- tional or subgeneric value. A small hispid crest is charac- teristic of the whole group Elegantes. C. elatus, Linn., is by its anther-crest closely joined to C. platyphyllus, and C. pannonicus to C. levigatus. There is no clearly cristate anther in any species of the subgenera Pycreus, Mariscus cum Leptostachyo, Diclidium. Nevertheless the species with a highly-developed crest are scat- tered in Juncellus and in remote sections of Eucyperus, so that the character cannot be used as a primary one in the delimita- tion of the larger divisions. (9) THE STYLE. The style is trifid or bifid, or in one species, O. cephalotes, sub- entire. The style is in each species, section, and subgenus of MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 21 Cyperus either 3-fid or 2-fid, without exception; and therefore the primary subgenera of Cyperus have always been founded on this character. This was the view of Kunth. Regarding the character as absolute, Nees, Reichenbach, and others have elevated Pycreus to generic rank. But Boeckeler has blown upon the character a little by uniting certain species into composite species, each of which may have either a 3-fid style with triangular nut, or 2-fid style with complanate nut. These species are C. alope- curotdes, Rottb., cum C. dives, Delile; C. inundatus, Roxb., cum C. procero, Rottb. ; C. pygmeus, Linn., cum Zsolepide Michelianá, Roem. et Sch. In each of these cases I consider two remote species have, from general aspect, been mixed, and that the more carefully they are examined, the more essentially and structurally they are found to be distinct. This I have attempted to explain fully below. In C. stoloniferus, Retz., Boeckeler says, “stylo elongato (quan- doque abbreviato) profunde bi- v. trifido aut indiviso.” I find the style always deeply trifid. To examine the style, a flower should be dissected just before expansion (this gives you the stamens by the same dissection); it is dangerous to trust to older flowers, as the brittle style-branches often break off exactly at their point of separation, in which case, even with the microscope, it is difficult to see the scar. The casual cases of suppression of a style-branch that occur are excessively few, I should say in my experience not lin 1000 for the genus. The stamens (ecristate &c.) are certain proof that Boeckeler is in error in placing this species with C. levigatus &c. Itis not far from C. rotundus, but perhaps nearer C. arenarius. In nearly all the critical species it is less trouble to examine the nut than the style (é.e. in mere sorting and naming). The nuts of C. alopecuroides, Rottb., and C. dives, Delile, are as unlike (not merely by two rounded or three sharp angles) as any two in the genus. D. The style-branches are usually long-linear, shortly or distinctly exserted from the glume, but sometimes very much exserted, 80 that the spicula are said to becomose. This character is the only point left to keep C. diandrus distinct from C. rivularis; and some (even of the American botanists) do not think the two distinct, There is a strongly comose Cyperus, resembling C. Monti in all other points but this, among Griffith’s collections ; and there is a fragment, collected by V. Ball in Chota Nagpore, which does not differ much from C. pilosus except in being comose. I 22 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. do not like to propose new specific names and give long descrip- tions of these, when I see the considerable variation in length of style which occurs in many fully-known species. The style-branches are unusually short and broad, much papillose-floeculose (reminding of Fimbristylis) in the happily- named C. platystylis, R. Br.; and the style is something similar in C. egyptiacus. (10) Tue Nvr. The form, size, and surface-appearance of the nut are characters of primary importance. The first distinction in form is between the trigonous and flat- tened nut, according as the style is trifid or bifid. The character from the style has been generally preferred, being more definite ; for in Eucyperus the face of the nut is frequently flattened against the rhachilla, whilst the nut is unequally trigonous, the dorsal angle being but slightly developed; and in Juncellus the nut is often convex on the back, so that the distinction founded on the nut, though clearly enough visible in most cases, fades away in some instances. C. stoloniferus, Retz., is placed by Boeckeler in Juncellus; but the nut is not less trigonous than it is in C. arenarius, C. Auchert, and others of the Conglomerati, which he places in Eucyperus. (The var. 6 and y in Boeckeler do not (mihi) belong.) I think the species altogether a Hucyperus. C. pygmeus, Rottb., contra, is placed by Boeckeler in Eucyperus, as having a "triangular" nut: this I do not find so at all (fig. 10). I agree with Kunth and Bentham that the species is a Juncellus. In the cases of O. alopecuroides, Rottb., and O. inundatus, Roxb. (which I have considered separately below), the form of the nut is definite in each species (or var., as Boeckeler con- siders). The flattened nut may have an edge (Pycreus) or a face (Jun- cellus) next the rhachilla; and this distinction is very strong. Juncellus is much more clearly separate from Pycreus than from Eucyperus. The shape of the nut, as narrow-oblong, broad-obovoid, trun- cate, shortly rostrate, &c., affords an excellent specific character, constant (speaking in a broad way) in every species. In most of the really natural sections a similarity of nut 18 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 23 found. In the C. auricomus group the nut is peculiarly umbo- nate; in the Elegantes group the nut is ellipsoid, acute at both ends Sharply trigonous, somewhat large; in the Conglomerati (and some of the allied Arenarii) the nut is unequal, dorsally com- pressed. So much is this the case that, where great differences in the nut are found in one of our gronps, it may be suspected we have not discovered the true affinity. The nut is smooth throughout the genus; an imperfect tuber- culation occurs in C. Haspan (and a few others), but only casually in some specimens, not throughout the species. The colour of the nut is a good character, but the nut is in a great majority dark; and care of course must be taken not to describe colour from unripe or imperfect nuts, which abound in some species (as C. pygm«eus). The size of the nut is measured in comparison with the length of the glume; it is said, therefore, to be “very large” in C. dif- formis (where it is nearly as long as the glume), though, speaking absolutely, it is small The nut in C. Haspan is very small, in C. Iria large; and these species, from Linneus downwards, are nevertheless mixed in herbaria. I find, however, a considerable range of size (not of form) in the nuts of many species. The plant which I call C. exaltatus, Retz., var. Oatesii, is very likely distinguishable as a species; but as I have been able to hit on no tangible character except the size of the nut (to separate it from C. exaltatus) Y leave 1t as à variety. In Cyperus, as in some other genera of Cyperaces, botanists have exhausted their vocabulary in describing the surface of the nut as shining, reticulate, velate, puncticulate, transversely marked, &c. &c., according to the general appearance it presents under a pocket lens, without (as I conceive) getting clear hold of some important differences of structure met with. 2. The outermost layer of cells of the nut in Cyperus is without colour, usually hyaline, sometimes opaque, white, permanent or flaking off (detergibiles), not very thick or lax (as often m Tso- lepis), and either subquadrate (figs. 31-39) or oblong longitudi- nally (fig. 40) (2 or 3 times as long as broad). There appears nothing intermediate between these two cases. The only longi- tudinal-celled species of Cyperus are about a dozen, all in Pycreus, and forming a natural group, of which C. flavescens 18 a well- known example. This character is so absolute that I am able to 24 MR. €. B. CLARKE ON INDIAN SPECIES OF CYPERUS. say with much confidence that the Australian fragment (on which Mr. Bentham doubtfully extended C. flavescens to Australia) 1s not O. flavescens, and that O. flavescens has not yet been col- lected east of Cabul. Its area, corrected by this character, becomes very natural; for the South-American OC. Olfersianus (with its allies or subspecies from Mexico and the Andes) has the outermost layer of cells of the nut subquadrate, and separates off cleanly. These species are usually described (Boeckeler &c.) as having the nut “transversely striate” or “transversely undulate-rugose.” The outer cells are in this group hyaline, and more prominent at the narrow ends of the cells, in many cases actually wearing away from the middle of the cells; hence the nut, under a pocket lens, takes the aspect in fig. 41. The “transverse striation”’ is, however, a very deceptive character, being prominent in some nuts, utterly obscure in others from the same plant (all characters de- rived from degree of flaking-off are untrustworthy throughout the Order, as I attempted to show for Carex in my ‘ List of Andover Plants’ 17 years ago). The character derived from the oblong (not quadrate) cells is absolute. In the quadrate cells there are minute differences, which are mostly very constant for the same species, and even for the same variety. In many of the smaller nuts the outermost layer of cells is thin, hyaline, not detergible ; consequently it cannot be detected without a microscope and a dark field. The nut is usually described by cyperologists in this case as lucida. Sometimes the outer cells are opaque, white, permanent, as fig. 35. Such a nut is seen in C. bromoides, Willd., and is de- scribed (by Boeckeler) as “ transverse lineato-granulata fusco- : nigra nitidula.” In C. unioloides and C. angulatus the outermost cells of the nut are similar, but nearly always broken at the centre, so that the nut is described as “ porose " (fig.38). Icon- sider (as does Mr. Bentham) C. bromoides, C. unioloides, and C. angulatus one species ; but I can nevertheless from one nut only generally tell whether the specimen came from Australia, India, the Cape, or South America. The outer cells of the nut (rarely) are somewhat inflated, lax, and yet do not break up at the centre. Such nuts are described as scabrous-papilose and by various names. The character occurs sometimes in C. globosus (as see fig. 86), but is not uni- versal for that species, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 25 A very common form is the “ puncticulate nut;” here most of the outermost cells have a minute microscopic puncture (a very small pore, if you like) at their centre, through which the darker body of the nut is seen (fig. 37), as in C. pumilus, var. punctata. But a great number of the nuts described as punctate are not punctate. In those species in which the outer cells are specially impunctate but slightly inflated, their middle points, being ele- vated above their margins, reflect the light strongly, and, under a pocket lens, the nut looks covered with rows of white points; these are more particularly the punctate nuts of cyperologists, and they are particularly devoid of punetations; something be- tween fig. 35 and fig. 36 represents them. Lastly, the outer cells are more detergible, so that only their margins remain (fig. 39). The dark body of the nut is covered with a reticulate white veil; this is the “ reticulated nut" pro- perly so called; but this name is applied loosely to any nut of which the quadrate outer cells are to be seen with a pocket lens. The outer cells often flake off entirely in places, giving the nut a patchy appearance. This is more strictly the velate nut of authors ; but the term velate is applied to any nut where the outer cells have been rubbed off partially, which causes the presence of the outer cells to be recognizable under a pocket lens; and it means, with cyperologists, no more than that the outer layer of cells is distinctly recognizable. But in either sense the term velate is (in my opinion) of no (or the very smallest) value. SECTION II. Parag. 1. On the Subgenus Anosporum. The abundant Bengal species Cyperus cephalotes, Vahl, Was proposed as a genus by Nees in hb. Wight as Ungeria, a name preoccupied ; he therefore published the genus in ' Linnea’ under the name Anosporum, and placed it in the tribe Hypolytree. Boeckeler published the same species (in 1858) as the genus Trentepohlia, which he placed among the Ficinem; but in his Cyperacee Boeckeler admits the prior name Anosporum, and removes the genus next Oyperus. Finally, Bentham (in ‘ Flora Australiensis ") not merely reduces the plant to Cyperus, but does not think it deserving of a section or even subsection of its own ; he places it among the Juncelli. u l The cause of so great discordance of opinion arises on the 26 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. question whether the species has any perigynium or not. The final view of Boeckeler is, “ Perigynium e squamis constat tribus, quarum una quam relique multo minor; squamse crassi- uscule stramines, basi inter se et pedicello, parte superiore, angulis fructus connate ibique evanescentes, v. caryopsin usque ad apicem marginantes et cum styli basi in rostrum complanatum vel teretiusculum connate.” Turning at once to my pictures, fig. 1 represents the pistil just before fertilization ; there is no trace of perigynium at this time. The ovary is very thin-walled, and its cavity with the ovule is visible through the walls, and is shown in my picture. The narrow-conic base of the style at this time is extremely firm and opaque (quite unlike all other species of Cyperus). Fig. 2 represents the same ovary shortly after fertilization ; even before the ovary has sensibly swollen the axis has under- gone a rapid prolongation and produced the pedicel. This pedicel appears to spring directly out of the torus; nor do I see that it differs from the base of a carpel in Anonaces for instance. Fig. 3 represents the ripe nut from within (the side pressed against the rhachilla). Figs. 4 and 5 represent two horizontal sections taken one near the base, one near the top of the pedicel ; and fig. 6 is the same nut seen from without, the stamens having been removed. In these drawings the closer cells represent the dark-brown body of the nut; the larger paler cells the straw- coloured corky cells of the * squamz," which are really hardly larger than those of the nut, but are exaggerated a little in the drawing to make a contrast which is really due to colour. On the inner face of the nut (fig. 3) the tissue of the nut is perfectly continuous with that of the straw-coloured squame ; there is a perfectly gradual transition of colour from one to the other, both at the base and on the pale-angles. On the back of the nut, however (fig. 6), as it becomes ripe, a chink appears between the base of the nut and the corky scales, which widens into a chasm as the nut gets perfectly ripe. At the same time a small ellipsoidal portion of the corky pedicel splits out by two deep narrow chinks (which, however, never, I believe, proceed further than as shown in cross-section fig. 5); and thus is pro- duced the “ third smaller scale " of Boeckeler, firmly connected to the last at its apex with the smaller angle of the nut. These chinks, imperfect dehiscences, undoubtedly occur in the nut with undeviating regularity. The theory of Boeckeler is that MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS, 27 they are the outcome of three scales, which in the younger state ofthe pistil are connate with and absorbed into it, but which betray the true secret of their origin. This is a theory that cannot be maintained without far more evidence: the position of the three squam: opposite the stamens does suit their being considered homologous with the squame in some other Cype- racee. But in the dehiscence of fruit in general, how is it pos- sible to argue back from a cireumscissile pyxis (as in Hyoscya- mus), and to say that the line of dehiscence shows the true base of the ovary ? The style is continuous with the ovary in this species, and persistent on the ripe nut. The style-base is at first terete, but finally unequally 3-winged, strongly serrulate on the margins of the wings. These wings can often be traced under the micro- scope nearly to the summit of the style. I have attempted to draw, in fig. 3, a very common appearance the wing exhibits (especially in dry specimens) as if it were a narrow bract clinging to the style. Boeckeler rightly states the wings “connate; ” they are altogether inseparable, and I see not the slightest ground for regarding these wings as the upper portions of the perigynial scales connate with the style; they much more recall the style- base of some species of Fimbristylis. l l I think Bentham is perfectly right in replacing this species in Cyperus. On the other hand, I place in a subgenus distinct from Juncellus, &c., a plant admitted by Arnott, Griffith, Kurz, and F. Mueller as worthy of generic separation. Its differences from other Cyperi are considerable: we have first the remark- able * pedicel" of the nut, secondly the persistent style, thirdly sub-undivided stigma. The firm conical base of the style, so prominent even in the young stage of the ovary, is another (or part of the second) character. By these characters the subgenus Anosporum is separated from Juncellus as from all other subdi- visions of Cyperus. . I have treated of Anosporwm here as though of one species, Cyperus cephalotes, Vahl. Of the other species placed in Ano- sporum by Boeckeler, I have only one here, viz. Cyperus platystylis, R. Br., = Anosporum pallidum, Boeck., = Cyperus pallidus, Hey ne, =C. caducus, Steud. The nut of this is shown from the inside in fig. 7, from the back in fig. 8, and in horizontal section in fig. 9. "The cells are drawn very much too large in proportion to the scale of the nut, and the horizontal section is more magnified. 28 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. The generic character given Anosporum by Boeckeler can only be made to include this species by enormous inferences from com- parison with Cyperus cephalotes. There is no more development of pedicel than in many other Cyperi: the style is deciduous, _ with three branches, essentially as of ordinary Cyperi; but it is somewhat short, thick, and more papillose than usual in Cyperus (Fimbristylis-like), whence R. Brown took his good specific name. As to the corky alteration of the cells, it is very similar to that in C. cephalotes ; but it does not separate by fissures into three scales (as in C. cephalotes), though there is a kind of indication of sepa- ration if the nut is compared side by side with that of C. cepha- lotes. On the back of the nut the straw-coloured alteration of the cells proceeds over the whole surface (Benth. et Hook. f. Gen. Pl. iii. p. 1044); but the change only extends to the outermost layer of cells in the centre of the back, while it comprises the whole of the margins of the nut. The change in the nature of the cells where it extends only one layer deep can be easily exa- mined, and is exactly similar to that which takes place in the outermost layer of cells of the nuts of various Cyperacee. It does not appear to me that Anosporum can be maintained either as a section or a subsection if this species be placed in it. It has neither the pedicel, the permanent style, nor the entire stigma. The inflorescence is entirely remote. Steudel, no other- wise than Bentham, is right in placing it somewhere near Kunth's Alternifoli, of which it has the umbel, digitate spikelets, and subexalate rhachilla. _ I would venture the suggestion that the resemblance of these two species in the nut is accidental to their present habitat, and not a mark of common cousinship. The nuts of most Cyperi (even after years in a dry herbarium) sink in water, and remain sunk ; when they drop into the rice-sea, they sink to the mud at the bottom, where they germinate. But C. cephalotes (C. natans, Buch.-Ham.) and C. platystylis (C. fluitans, Buch.-Ham.) are tank- floaters; their roots are entangled in the mass of rotten Pistia, Salvinia, &c. in overgrown tanks. Their nuts float in water (and remain floating, as I find experimentally), so that they can ger- minate in their proper nidus. Parag. 2. On Cyperus pygmeus and Isolepis Micheliana. Cyperus pygmaeus, Rottb., and Isolepis pygmea are treated by Kunth (Enum. pp. 18, 203) as two plants, belonging to remote MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 29 genera, undoubtedly distinct ; though from the synonymy quoted it appears that they had been confounded, by reason of superficial resemblances. Boeckeler, in ‘ Linnea,’ xxxv. p. 493,(cf. Boeck. in ‘Flora,’ 1871, p. 158), unites the two under one species, Oyperus pygmaeus ; he esteems Isolepis Micheliana a form merely “ squamis trifariam imbricatis.” The number of stamens which he gives for C. pyg- meus and I. Micheliana is a character of no value; the number of stamens in fairly developed C. pygmaeus is two; but the number is not constant in one spikelet of either species. The glumes of I. Micheliana are, however, not trifarious, but on the regular 3 arrangement. Benth. et Hook. f. Gen. Pl. iii. p. 1044, follow Boeckeler. I have no doubt that the two plants are amply distinct, and that Kunth's view is correct. If we examine a good series ofthe Cyperus with ripe nuts, we shall find that though superficially the glumes appear somewhat irre- gularly placed on the rhachilla, that is not really so. The irregu- larity disappears under the microscope. The spikelets are densely crowded, hence curved and sometimes twisted as in many other species of Cyperus; but if we examine the upper two-third part of a ripe spikelet, the glumes are accurately biseriate, and the rhachilla * alternatim excavata ” as in other typical Cyperi. In T. Micheliana, on the other hand, the glumes of the upper two-third part of a spikelet with ripe nuts are seen to be on the 8 arrangement with lozenge-shaped scars after they have fallen from the rhachilla. The distinction is so marked that I imagine confusion to have arisen only from examining herbarium spe- cimens of immature heads with a lens. Such specimens with unripe nuts are difficult to separate, even by dissection, because the best characters in support of the foregoing (to separate the two plants) depend on the nut. The nut of C. pygmeus (fig. 10) is, when fully developed, chestnut-coloured, the outer cells are the quadrate cells common in Cyperus; the outermost layer is hyaline punctate, but not very thick or lax; so that when the nut is viewed under the micro- Scope it has no hyaline margin all round it. The nut of Isolepis Micheliana (fig. 11) is yellow, or a very pale yellow-brown ; the two outer layers of cells are hyaline, inflated, lax; so that when the nut is viewed under the microseope it appears surrounded on all sides by a broad hyaline margin. 30 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. This structure of nut is very common in Isolepis, unknown in Cyperus. These differences in the nut invariably accompany the differ- ence in the arrangement of the glumes (biseriate or 2); so that thereby Isolepis Micheliana and Cyperus pygmaus are completely carried apart to their several genera. These differences are supported by a number of trivial cha- racters. In Cyperus pygmeus the glumes are ovate-lanceolate or oblong-lanceolate, the filaments ligulate, the anthers somewhat large linear-oblong, the style-branches longer than the nut, the style itself nearly free from papille, never 3-fid with me. In Isolepis Micheliana the glumes are often somewhat widened at the shoulder, oblong (subobovate) lanceolate; the filaments slender, the anthers small oblong, the style-branches shorter than the nut, the style itself papillose, usually 2-fid, occasionally 3-fid. In plants that fruit superabundantly as these, a large number of the nuts are never perfected. In Cyperus pygmaeus a quantity of pale-coloured nuts occur which will be found to contain no perfected seed. These are often not only pale, but marcescent, with very hyaline outermost cells; and they must not be mis- taken for the ripe seeds of the Isolepis. Parag. 3. On Cyperus alopecuroides and C. dives. Cyperus alopecuroides, Rottb., isa large species with very many spikelets spicate in a large umbel, with a 2-fid style, nut com- pressed contrary to the rhachilla, and is arranged in Sect. Jun- cellus by Kunth, &c. C. dives, Delile, is very similar to it in size, bracts, and inflo- rescence; but has a 3-fid style and strongly trigonous nut. It has been placed next C. exaltatus, Retz., and should be considered, I think, a form only of that species. Boeckeler, in ‘ Linnea,’ xxxvi. p. 822, unites C. alopecuroides, Rottb., and C. dives. Benth. et Hook. f. Gen. Pl. iii. p. 1004, say :— “C. alopecuroides, Rottb. (C. dives, Delile, Pl. Eg. t. 4), habitu C. (Papyro) exaltato, Retz., accedit, sed nux omnino Juncelli, variat tamen stylo 2-fido v. presertim in speciminibus Africanis 3-fido." In C. alopecuroides, Rottb., the style is always 2-fid, the nut always compressed, and moreover the rhachilla (fig. 12) is very MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 31 stout, with quadrangular excavations (altogether of Juncellus), without the vestige of a wing. In C. dives (fig. 13) the rhachilla is narrow, with very narrow, almost soluble, yellow wings, as in C. exaltatus (and its varieties). In C. alopecuroides the packing of the flat-backed glumes is very similar to that in C. Monti: in C. dives the boat-shaped glumes are as in C. ewaltatus, differing somewhat in colour. I esteem C. alopecuroides, Rottb., as very distinct from C. dives. C. alopecuroides, Boeck., is mainly founded on C. amenus, Koenig et Roxb. MS. Roxburgh subsequently assumed this to be C. alopecuroides, Rottb., despite the difference in style and nuts; he therefore dropped the name C. amenus altogether, and pub- lished it as C. alopecuroides, Rottb. This is really very near C. exaltatus, Retz.; the true C. alopecuroides, Rottb., Roxburgh never saw. Parag. 4. On C. inundatus and C. procerus. The most critical of all species is C. inundatus, Roxb., a species with bifid style, which so closely resembles C. procerus, Rottb. (with trifid style), that the two have been mixed together by Wallich, nor are they easily separable except by the style and nut, the bifid style having a nut compressed contrary to the rha- chilla, the trifid style having a distinctly trigonous nut: these two kinds of nut do not occur on one plant. I for some time considered that this was an exceptional Cyperus, in which the style was 2-fid or 3-fid in the same species; but I am satisfied, from the full series of specimens in the Calcutta Herbarium, that they are two well distinct species. l I observed first that the compressed nut of C. inundatus is very mueh wider than the trigonous nut of C. procerus, and has a very different base. In the species of Cyperus, however, with an unequally trigonous nut there is no tendency in the nut to get wider as it gets less trigonous. Moreover the base of the nut of C. inundatus has a different shape and structure from the nut of C. procerus. Secondly, the rhachilla in C. inundatus 18 thick, with quadrate excavations and central strim exactly as in C. Monti; and the nut is so like that of C. Monti, that the species are not easily separated but by the inflorescence. In C. procerus the rhachilla is much slenderer, with oblong excavations, nar- rowly distinctly hyaline-winged. The glumes of C. de are very dull-coloured, flat on the back, strongly striated, the 32 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. margins greatly incurved when dry. The glumes of C. procerus are more brightly somewhat yellow- or red-stained, less flat- backed, less strongly plicate-striate, less incurved when dry. The umbel in C. procerus is compound, in C. inundatus decom- pound in every example seen : from the analogy of other species it is probable that poor specimens of C. inundatus would have a compound umbel; but no example of C. procerus (an abundarit plant) has the umbel decompound. €. procerus has the anthers absolutely muticous, like those of C. pilosus; but C. inundatus has an obsolete crest to the anther exactly as has C. Monti. I have dwelt on these minute points because C. inundatus is the only species of Cyperus in which there remains to me any question that a much-compressed and clearly trigonous nut may occur in one species. If this species be allowed to be distinct from C. procerus, then we have an absolute distinction to separate Pycreus and Juncellus from Eucyperus. Parag. 5. On the Genus Kyllinga. Benth. et Hook. f. Gen. Pl. iii. p. 1039, separate the 1-flowered (1-nutted) species of Cyperus from Kyllinga by the fact that the upper (empty or male) glume is protruded from the flowering (nut-bearing) glume. This isin most cases a working character; but Cyperus umbellatus, Benth., shows such a range of variation in this character, that it is impossible to put it forward as the chief diagnostic one: in one of the most abundant forms of C. umbellatus, Benth., the upper (barren) glume does not overtop the nut-bearing glume. In another frequent form, hardly distin- guishable from this as a variety (C. pictus, Wall.), the spikelets are exceedingly similar, but the barren glume is clearly protruded ; and from this form we pass on to the linear-spikeleted forms of C. umbellatus by insensible gradations. Boeckeler, in his ‘Conspectus Generum,’ mentions the two points which separate Kyllinga from Cyperus (though he does not give contrasted differenti), viz. “ Spikelets one-flowered upheld on discs, and style bifid.” Kyllinga can be separated easily from each subgenus of Cype- rus, but not (except by alternative cross-referenced diagnostic points) from the genus as a whole. I should therefore have pre- ferred to have made it a subgenus of Cyperus. The subgenus of Mariscus includes (for me) all those species (Leptostachyi of Boeckeler and others) in which the glumes are MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 83 permanent ; or at all events do not fall from the spikelet before the spikelet has fallen from the rhachis. The spikelets are in these species very generally * discis impositz" in the language of Boeckeler ; i. e. the rhachilla (usually above the two lowest persis- tent barren glumes) separates by a clean scar, leaving a small disc where it separated. In all the numerous preceding species of Pycreus, Juncellus, Eucyperus the glumes fall from the rhachilla (while the rhachilla remains attached) by a clean-cut line sepa- rating the glume from its basal portion, which is decurrent down the rhachilla. In all the Mariscus set the style is trifid; while in Kyllinga the style is bifid. This I take to be the real difference (carrying with it of course a difference in the nut). Boeckeler has, in defiance of his own character, placed three species at the end of Kyllinga having a trifid style (hence a triangular nut): these I have not seen, but I suppose they are only small-flowered Marisci. Kyllinga differs therefore from Mariscus by its 2-fid style, from Pycreus by its 1-flowered spikelets. [The number of fertile flowers to the spikelet is no use as a diagnostic character: in Cyperus flavus there are sometimes five or six nuts to the spikelet, though it is admittedly a typical Mariscus.] SEOTION IIL (Conspectus generis.) CYPERUS, Linn. Spieule co-1-nucigere ; flores, nisi 2-1 basi, 1-2 apice, omnes hermaphroditi. Glume distiche. Sete bypogyne 0. Stamina 8-1.—Inflorescentia umbellata interdum capitato-umbellata. Series A. | Glume decidusm ; i. e. antequam spicule rhachilla a rhachi (spice axi) sejuneta fuerit, glume (a basi spicule gradatim) ope concisure rect decidunt. Subgenus 1. Awosronuw. Stylus subindivisus. 2. Pycreus. Stylus 2-fidus. Nucis compresse margo rhachille adjectus. » 3. Juncentus. Stylus 2-fidus. facies rhachille adpressa. » 4 Eucyperus. Stylus 3-fidus. sequaliter) trigona aut subrotunda. Nux stipitata. 9 Nucis compresse Nux (interdum in- LINN. JOURN.—BOTANY, VOL. XXI. D 94 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Series B. Glumz persistentes ; 7. e. spicule rhachilla a rhachi (omnino vel per articulos) sejungitur, antequam glume a rhachilla de- cidunt.— Stylus trifidus. Subgenus 5. DrcrrpruM. Spicule rhachilla per articulos dis- juncta. » 6. Manrscus (char. amplif.). Spicule rhachilla con- tinua basi, sxpissime (an semper?) supra 2 imas glumas vacuas, sejuncta. N.B. Marisci, in speciebus nonnullis, spicule exstant 1-nuci- gere; ab his Kyllinga stylo 2-fido. Courtoisia stylobasi pyra- midato persistente, cognosci possunt. Subgenus 1. ANOSPORUM. [See Sect. II. parag. 1 supra for the numerous characters by which this plant (I here treat C. cephalotes as the only species of the subgenus) stands apart from Cyperus. | 1. C. cePeHALOTES (Vahl, Enum. ii. p. 311); umbella dense capitata; stylo indiviso aut obscure 3-lobo ; nuce stipitata, cel- lulis angulorum suberosis stramineis.—Spreny. Syst. i. p. 218, syn. excl.; Kunth, Enum. ii. p. 28; Benth.! Fl. Austral. vu. p. 263. C. monocephalus, Roxb. Hort. Beng. p. 5 ; Fl. Ind. i. p. 188 (non Roxb. Ic. ined. in h. Kew.). C. leucocephalus, Wight MS. in h. propr., non Retz. C. monocephaloides, Roxb, ! Ic. ined. in h. Kew. C. Hookerianus, Thwaites! Enum. Pl. Zeyl. p. 342, non Arn. C. kyllingizoides, Willd. h. 1309, plag. 1 (fide Boeck.), non Vahl. ? C. monogynus, Boeck. in Linnea, xxxv. p. 565. C. natans, Buch.-Ham. MS. in h. propr. Anosporum monocephalum, Nees! in Linnea, ix. p. 287; in Wight, Contrib. p. 92; Endl. Gen. Pl. p.117; Arn. in Edinb. New Phil. Journ. xvii. (1834) p. 263 ; Griff. Notul. iii. p. 103 ; Boeck. in Linnea, xxxvi. p. 411; F. Muell. Fragm. Phyt. Austral. viii. p. 272. A. cephalotes, Kurz! in Journ. Asiat. Soc. xlv, (1876) pars 2, p. 159. Ungeria monocephala, Nees! in Wight h. propr. n. 1855. Trentepohlia bifoliata, Beck. in Bot. Zeit. 1858, p. 249. Cyperus sp. n. 21 bis, Herb. Ind. Or., Hook. f.et T. Thoms. Kyllingia, Wall.! List n. 3441, litt. A, D partim. In udis radicans, aut szpe in lacubus natans, perennis ; sto- lones tenues lenti. Culmi 1—4 dm., sursum trigoni leves. Folia MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 85 2-5, cum $ parte culmi equilonga, plana, graminea, sursum scabra. Bractee 3-5, 2-20 em. longs, divaricatz, foliis similes. Um- bella in diam. 5-20 mm., 10-70-spieuloss»; radii 3-5, obsoleti. Spicule long 5-10 mm., late 3—4 mm., compresse, sspe curvate, 10-24-flore ; rhachilla validior, compressa ; ale anguste, persis- tentes; gluma ima parva, vacua. Glume longs 2-3 mm., late ovate subacute, naviculari-concave, subeoriaces, in dorso virides, aliquando minute scabride, in lateribus stramines aut sspe cas- tanee. Stamina 3 (fide Bentham haud raro 2), hypogyna, libera: filamenta late ligulata, levia ; anthere lineari-oblong:e, flavide ; connectivum vix brevissime excurrens, scabrum, rubrum. Peri- gonium nullum, aut saltem nullo tempore a pistillo (nisi rup- tura) separabile. Stylus nuce longior; basi dilatatus in nucem sensim decurrens, cum hac continuus, persistens. Nux ovoidea, curvata, fusca, contra rhacheolam compressam, ob nervum dor- salem tenuissimum, in:qualiter 3-gona ; anguli fructus tempore, presertim stipitis, spongioso-incrassati; stylobasis laxe triquetra, in angulis scabra. India et Malaya, in regione tropicali; ab Assam usque ad Zey- laniam, China, et Australiam borealem. Bengalia: vulgaris (Wallich, hh. Kew., Linn. Soc. ; Wallich n. 8441 A, D, 4. Calcutta, dc.). l Assam: vulgaris (Griffith n. 1609, h. Kew; Simons, n. 26, h. Calcutta); Gowhatty (Nuttall, h. Mus. Brit.) ; &c. 4. Madras Peninsula (Wight n. 1855, hh. propr., Kew, Mus. Brit. ; Wight n. 2919, h. Calcutta). oa Zeylania (Thwaites, C. P. n. 3221, hh. Kew, Mus. Brit., Cat- cuita). Burma (Griffith n. 6172, hh. Kew, Calcutta). Pegu (Kurz n. 2670, hh. Kew, Calcutta; n. 646 bis, cutta); Bhamo (J. Anderson, h. Calcutta). Amoy (Jenkins, h. Mueller). Borneo: Banjermassing (Motley n. 1261, A. Kew). Batavia (J. B., h. Mus. Brit.). Australia: Sinus Rockingham (Dallachy, ^. Kew). h. Cal- Subgenus 2. PycREUS. Pycreus, Beauv, Fl. d'Owar.ii. p. 48, t. 86, fig. i ” Linnea, ix. p. 483.—Torreya, Rafinesque m Journ. de Phys. . Ixxxix. p. 105. i 2 ssa, fere Stylus 2-fidus; rami linearese Nux plus minus compresss, p 2 86 MR, C, B. CLARKE ON INDIAN SPECIES OF CYPERUS. symmetrica; margo rhachille adjectus.—Umbellule nude vel a bracteis parvulis suffultz. Spicule multi-(10-60-)flore, persis- tentes. Rhachille ale anguste, persistentes, hyaline, inter- dum purpureo-maculate. ^ Filamenta 1-2-3, ligulata, parum dilatata, sepissime (interdum conspicue) papilloso-scabrata ; an- there mutice aut vix apicatz, nec cristate. Glabre. Species plures annus; paucissime stolonifere aut late repentes; omnes typice foliaces, exempla pauca foliis omni- bus ad vaginas depauperatis reperiuntur, mihi varietates mere. Forme pygmez (imo acaules), interdum 1—4 cm. longz, in plu- ribus speciebus bene notis vis: sunt. This is the most natural and the most easily separable sub- genus of all in Cyperus; and few differences of opinion have occurred as to the position of any of its species. I have removed, however, two (C. diaphanus, Schrad., aud C. Barteri, Boeck.) into Juncellus. Of these, probably, Boeckeler had not the ripe nut. A single detached nut of Pycreus can generally be known from one of Juncellus by the fact that its two halves, made by a plane through its broadest part, are similar (fig. 42); whereas in Jun- cellus they are not (fig. 44), owing to the upper part of the nut being always more or less curved inwards towards the rhachilla. .But the nut of Pycreus is not rarely somewhat unsymmetric in the other plane (fig. 44), owing to the margin of the nut on one side being pressed against the rhachilla; this is à marked cha- racter in some species, as in C. sulcinux below. * Cellule extime nucis oblonge (figs. 40, 41). T Species in Indid Orientali indigene. 2. C. ELAVESCENS (Linn. Sp. Pl. p. 68, nec Linn. h. propr.) umbella specie simplice aut subcapitata; bracteis 3—4, divaricatis, umbella longioribus ; spiculis multifloris, lineari-oblongis, com- pressis, subturgidis ; glumis ovatis, obtusis, flavescenti-stramineis ; nuce cum 3-3 parte glume equilonga obscurius transversim notata.— Lam. 1U. t. 88. fig. 1; Roem. et Sch. Syst. ii. p. 191; Host, Gram. ii. t. 72; Bertol. Fl. Ital. i. p. 261; Sibth. Fl. Grec. t. 47, Reichb. Ic. Fl. Germ. t. 278. fig. 662-664; Poit. et Turp. Fl. Paris. t. 74 (non visa); Fl. Dan. t. 1682 ; Sturm, Fl. Deutsch. xiii, t. 52 (opt.) ; Anders. Cyp. p. 1, t. 1. fig. 1; Ledeb, Fl. Ross. i. p. 289 (var. B excl); Gren. et Godr. Fl. Fr. iii. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS, 37 p. 862; Parl. Fl. Ital. ii. p. 26; Asa Gray, Man. p. 552; Torrey in Ann. Lyceum New York, iii. p. 250; Boeck. in Linnea, xxxv. p. 438; Boiss. Fl. Or. v. p. 364 (ex Afgh. excl.); Coss. Expl. Algér., Bot. ii. p. 254 (nec Schk. Handb. t. 7, nee Benth. Fl. Austral. vii. p. 259, quoad ex Australia). C. xanthinus, Pres! in Oken Isis, xxi. p. 271, fide Steud. C. pozformis, Pursh. Fl. Amer. Sept. i.p. 50; Kunth, Enum. ii. p. 113; cf. Torrey in Ann. Lyceum New York, iii. p. 250. C. Gussoni, Gasp. ! in Guss. Syn. Fl. Sicul. ii. Add. p. 779. C. setaceus, Wulfen. Fl. Norica, p. 49, non Retz. Pycreus flavescens, Reichb. Fl. Germ. p. 72; Nees, Gen. Pl. Germ. ii. t. 22, fig. 14-16. — Moris Hist. iii. p. 239, sect. 8. t. 11. fig. 37. — Bauh. Theat. i. p. 89, eum ie. — Scheuz. Agr. p. 386, t. 8. fig. 13. Glabra, cespitosa, radicibus fibrosis. Culmi 1-5 dm., sursum trigoni, leves. Folia 2-4, cum 3 parte culmi subswquilonga, angusta, levia. Bracteæ longs 1-8 cm., foliis similes. Umbelle radii 2-5, longi 0-5 cm., plerumque breves; ochree breviter cylindriez, unidentate. Umbellule congestim corymbos:e, ebrac- teatz, radiolis brevissimis, 1-3-spiculosis, aut e spica simplice con- stantes. Spicule in unoquoque radio 3-10, fasciculate aut bre- viter spicatz, longæ 5-18 mm. late 3 mm., 10-40-flore ; rhachilla exalata ; gluma ima lanceolata, bractezformis, ceteris subbrevior. Glume spisse, subturgide, in dorso viridi-3-nervie, in lateribus enervoss, nitide aut fusce. Stamina 3; anthere anguste ob- longs, mutice. Stylus nuce brevior; rami 2, e gluma breviter exserti. Nux obovoidea, obtusa, apicata, modice compressa, nigro- castanea ; cellule extime longitudinaliter oblonge (3-4plo lon- giores quam late), emarcidz, hyaline, detergibiles ; ideoque nux plus minus transversim a lineis undulatis albidis interruptis notata .— Exemplum authenticum C. flavescentis, in h. Linn. proprio conservatum est C. rivularis: idcirco cl. Boott exempla Americana C. rivularis, in h. Mus. Brit. conservata, manu sua "C. flavescens, Linn.," nominavit. Regio Mediterranea, ab Holstein et Algiers usque ad Cabul ; America borealis calidior; Varr. usque ad Natal et Brasil extensis. Europa: Berlin (Reichenbach n. 151, hh. Kew, Mus. Brit) i Frankfort (Lawrence, h. Calcutta); Palermo (Todaro, h. Cal- cutta); Switzerland (W. Hooker, h. Calcutta); Portugal, Cintra (Welwitsch n. 340, hh. Kew, Mus. Brit.). 38 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Africa: Algiers (Lefranc n. 485, h. Mus. Brit.). Asia: Syria (Lowne in h. Calcutta); Persia (Haussknecht, hh. Kew, Mus. Brit.) ; Cabul, in valle Kurrum (Aitchison n. 964, hh. Kew, Mus. Brit., Calcutta). America: Kentucky, Lexington (Peter, h. Kew); Florida (Chapman, h. Calcutta); New Orleans (Drummond, hh. Kew, Calcutta). Sine habitat. Agrostotheca, Sieber n. 105, hh. Kew, Mus. Brit. Var. B. abyssinica; evolutior, ssepe lucida; nuce sspe con- spieue transversim notata. C abyssinieus, Hochst.! (sp.); Steud. Cyp. p. 4; Boeck. in Linnea, xxxv. p. 440, non Oliver. C. flavescens, A. Rich. Fl. Abyss. ii. p. 474. Mihi forma, vix varietas.—In Schimper, n. 122 in h. Kew, alterum exemplum nota (?) affixum est C. Eragrostis, Vahl. Alpes Maritimi: Mentone (Moggridge, h. Kew). Madeira: (Burchell n. 634, h. Kew; Mandon n. 258, hh. Kew, Mus. Brit.). Abyssinia: Adoa (Schimper n. 122 partim, k. Kew); Tigré (Schimper n. 297, h. Mus. Brit.). Africa Centralis: Bongo (Schweinfurth n. 1433, h. Kew); Gazelle Fl. (Schweinfurth nn. 1173, 1251, h. Kew); Gallabat (Schweinfurth no. 2011, hh. Kew, Mus. Brit.). Natal (Buchanan no. 309, h. Kew; Gerrard n. A89, h. Kew). Var. y. paraensis, Boeck. in Linnza, xxxv. p. 439; culmis fere a basi trigonis ; spiculis interdum angustioribus subfuscis. Cyperus paraensis (sp.), Steud. Cyp. p. 5. Pycreus paraensis (sp.), Nees in Mart. Brasil., Cyp. p. 8. In forma typica hujus varietatis, spicule anguste subfusce a C. flavescente recedunt; sed exempla optima Gardneri a C. Jlavescente typ. vix distingui possunt. Brasilia Australis: S. Catharine (5. Kew). Brasilia Tropica: Piauhy (Gardner n. 2383, hh. Kew, Mus. Brit.). Var. à. Fontanesii ; culmis tenuissimis, foliis filiformibus ; spi- culis in unoquoque culmo 2-6, quam C. flavescentis typici paullo angustioribus; glumis pallidis, in dorso viridibus, in lateribus albidis vix luteo-tinctis. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 89 C. Fontanesii (sp.), Kunth, Enum. ii. p. 5; Boeck, in Linnea, xxxv. p. 438. C. stramineus, Desf. MS. ` Causa glumarum minus spisse imbricatarum, bractearum diva- ricatarum, C. flavescenti quam C. stramineo affinior. Mihi videtur forma hortulana, invalida aut morbosa, in hypocaustis fortasse culta. Nux &c. omnino C. ffavescentis.—Exempla parvula (non culta) C. flavescentis, culmis 5-6 em. longis 1-5-spiculosis, a C. Fontanesii tamen longius distant culmis multo minus tenuibus, spiculis latioribus, flavescentibus. Patria ignota (in horto Kew culta, h. Calcutta). 3. C. srRAMINEUS (Nees! in Wight Contrib. p. 74); umbeila pauciradiata, congesta, spiculis laxius antice fastigiatis, bracteis 2, longiore jam fruetus tempore erecto; spiculis lineari-oblongis, multifloris, stramineis; glumis spissis, ovatis, subacutis.— Kunth, Enum. ii. p. 9. C. capillaris, Hochst.! in Pl. Hohen. n. 302, non Koen. C. coromandelinus, Boeck.! in Linnea, xxxv. p. 480, non Spreng. Cyperus, Wall.! List n. 3320 B partim. Cyperus, Wall.! List n. 3318 partim. Cyperus, Wall.! Cyp. indet. n. 1. Glabra, cespitosa ; radicibus fibrosis, a nota Hooker filii aro- maticis. Culmi 1-3 dm., tenuiores, curvati, sursum obscurius trigoni, leves. Folia 3-5, eum culmo sspe equilonga, angusta, levia. Bracteæ 2, rarius 3, major usque ad 8-10 em. longa, erecta, fere quasi culmi continuatio. Umbellæ radii 1-4, rarius 1-3 em. longi, ssepius brevissimi, 5-15-spiculose. Spicule longs 10-35 mm., late 2-3 mm.. 15-70-florz, leviter compresse ; rha- chilla fere exalata; gluma ima parva, oyata, bractezformis. Glume carinate, subturgide, in dorso viridescentes 1—3-nervise, in lateribus enervosæ, luteo-hyaline. Stamina 2 P filamenta per- sistentia, squamoso-papilloso-scabra ; authere lineari-oblongs, mutice, flave. Stylus nuce subbrevior; rami 2, e gluma plane exserti. Nux cum 3-4 parte glume equilonga, obovoidea, leviter compressa, obtusa, breviter apicata, nigra ; cellule extime® longi- tudinaliter oblong:e, emarcidze, albidz, detergibiles, ideoque nus transversim interrupte albo-notata.—A C. Jlavescente Canoa exempla hujus parvula monocephala) vix, nisi a bracteis P tempore suberectis, distinguenda—In Thwaites C. P. i , > bracteæ 3, subpatulæ sed glumæ subacutæ.—0. coromandeltnus, 40 MR. €. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Spreng. cum O. pectinato, Roxb., mere synonymus, "squamis apice recurvis &c." est C. Wighti?, Nees,— C. rubicundus, Vahl. India, alt. 0-1000 metr.: in planitie Bengalie orientalis fre- quens, aliunde rarius communicata. Bengalia: Khasia (J. D. Hooker et T. Thomson, h. Calcutta) ; Mymensingho (C. B. Clarke nn. 7956, 8023) ; Sylhet ( Wal- lich n. 8320 B, h. Calcutta) ; Chittagong (J. D. Hooker n. 158, h. Kew). Madras Peninsula: Mangalore (Hohenacker n. 802, hh. Kew, Mus. Brit.). Zeylania (Thwaites C. P. n. 3776, hh. Kew, Calcutta). Burma: Arracan (Kurz n. 676, h. Calcutta); Mergui (Griffith n. 166, 2. Kew n. 6169); Tavoy (Wallich n. 209, h. Linn. Soc., non Wall. List n. 209). | 4. C. LATISPICATUS (Boeck. in Flora, 1859, p. 441, in Linnea, xxxv. p. 467); umbella specie simplice aut capitata; spiculis magnis, compressis, multifloris; glumis spissis, ovatis, obtusis, castaneo-luteis aut castaneo-rubris; nuce cum 4 parte glume sequilonga, obovoidea. C. intermedius, var. indica, Boeck. in Linnea, xxxv. p. 451, non Steud. C. pseudo-bromoides, Boeck. (quoad ex Nepalense) in Linnea, xxxv. p. 464. | Cyperacea, Griff. Itin. Notes, p. 32, n. 506. Glabra, cespitosa ; radicibus fibrosis. Culmi 2-3 dm., sursum trigoni, leves. Folia brevia, sepe brevissima, apice minute scabra. Bracteæ 3—4, usque ad 8-12 cm. longs, divaricate, foliis similes. Umbelle radii 3-5, usque ad 9 em. longi, aut szpius 1-3, breves vel subnulli ; ochrez 1 cm., fusci, truncate. Umbellule congestim corymboss, ebracteatz, radiolis brevissimis, 2—3-spiculosis, aut e spica simplice constantes. Spice 8-9-spiculose, ebracteate. Spicule long: 2 cm., late 3—4 mm., 30-40-flore, lateribus paral- lelis ; rhachilla tetragona, fere exalata; gluma ima parva, obtusa. Glum:e carinate, subturgide, in dorso virides, 3-5-nervire, in late- ribus enervose, lucide. Filamenta 2, persistentia, squamoso- papillosa; anthere lineari-oblonge, mutice aut vix apicate, rubescentes. Stylus nuce subbrevior; rami 2, e gluma brevius exserti. Nux obtusa, breviter apicata, leviter compressa, castanea, lucida; cellule extime longitudinaliter oblong, emarcidz, hya- line vix detergibiles, ideoque nux non aut vix transversim notata. — Species a cl, Boeckeler ab exemplo manco haud bene evoluto MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 41 descripta; in herbariis cum C. wnioloide var. angulata sspe con- fusa, a Bentham cum C. flavescente recte collata, a quo differt spiculis latioribus, glumis plus minus castaneis.—In exemplis Khasianis (forma typica Boeckeleri) umbella sepius contracta: exempla in India Centrali lecta multo majora sunt, umbella longius radiata, spiculis longis 32 mm., latis 5-6 mm., 50—floris. India centralis et boreali-orientalis ; frequens. Khasi colles, alt. 1000 metr. (J. D. Hooker, hh. Kew, Calcutta ; Griffith n. 506, h. Kew n. 6200 partim; Griffith n. 506, A. Calcutta). Assam (Masters n. 198, k. Kew; Jenkins, h. Calcutta). Bengalia orientalis (Griffith n. 6200, h. Calcutta). Sikkim Terai (Kurz, h. Calcutta). Chota Nagpore: Hazaribagh alt. 600 metr. (C. B. Clarke n. 21208, hh. Kew, Calcutta). India centralis: Sumbulpore (A. Calcutta). Var. B. acaulis: culmo nullo; spica centrali in radice annua plane sessili; radiis 1—4, basi ochreatis, 1-7 cm. longis. Khasi colles, alt. 1200 metr.: Cherra(C. B. Clarke n. 15165, h. Kew). 7 E Var. y. aphylla; culmo 8-pollicari, ima basi a vaginis efoliatis vestito. C. aphyllus (sp.), Boeck. in Linnea, xxxv. p. 450. l Non visa; Boeckeleri descriptio in omni parte cum exemplis C. latispicati quadrat. In C. latispicati exemplis folia sæpe bre- vissima videntur. . » India Orientalis ; «comm. a W. Arnott in Reliq. Lehmann. — Boeck. 5. C. pauper (Hochst.; A. Rich. Fl. Abyss. ii. p. 478) ; culmo tenui 1-2 spiculas proferente ; involucri bracteis 2, altero sub- erecto; nuce subglobosa, nigra, cellulis extimis oblongis.—Steud. Cyp. p. 5 (culmi descriptione confusa); Boeck. in Linnea, xxxv. p. 451. C. angulatus, Strachey, Cat. Pl. Kumaon, p. /4, non Nees. l Annua, tenuis. Folia angustissima, cum culmo interdum æqui- longa. Involucri bractea altera, 4-6 cm. longa, quasi culmum producens, proventu suberecta, —Spicula (sepius solitaria) longa l em., lata 4 mm., late lanceolata, paullo-compressa, 8-16-flora ; rhachilla subexalata. Glume leviter imbricate, ovate, VIx acute, turgidz, obscure carinatz, luteo-brunnee, in dorso viridi-3-nervis, 42 ME. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. apice sepe castaneæ. Stamina 2. Stylus brevis; rami 2e gluma breviter exserti. Nux eum 3-4 parte glume equilonga, vix com- pressa; cellule extime albescentes, inconspicue. Abyssinia (Schimper n. 1602, hh. Kew, Mus. Brit. cum Fimbri- stylide intermixta ; Schimper n. 1311, hh. Kew, Mus. Brit. ; Schimper n. 1243, h. Kew admixta). Himalaya occidentalis (Munro, h. Kew): Kumaon, alt. 1800 metr. (Strachey et Winterbottom, h. Kew); Sikkim (J. D. Hooker, h. Calcutta). ++ Species in Indiá Orientali nondum reperta. 6. C. MACRANTHUS (Boeck. in Linnea, xxxv. p. 462); umbella pauciradiata, contracta aut capitata; spiculis magnis, compressis, nigro-castaneis; glumis ovatis, obtusis ; nuce cum 1 parte glum@ vix equilonga, cellulis extimis oblongis. C. permutatus, Boeck. in Linnea, xxxv. p. 477. Africa australis (Zeyher n. 1745, hh. Kew, Mus. Brit.); Macalisterberg (Burke, h. Calcutta) ; Boschberg (MacOwan n. 1365, k. Kew, n. 1862, h. Mus. Brit.) ; Natal (Wood, h. Kew ; Sutherland, h. Kew; &c.). 7. C. NIGRICANS (Steud. Cyp. p. 12); umbella contracta, 1-3-radiata aut capitata; glumis majusculis, nigris, in dorso viri- dibus; nuce obovoidea, vix compressa, cellulis extimis oblongis. —A. Rich. Fl. Abyss. ii. p. 476 ; Boeck. in Linnea, xxxv. p. 462. Abyssinia (Schimper nn. 1286, 1373, hh. Kew, Mus. Brit.). 8. C. LANCEOLATUS (Poir. Encyc. vii. p. 245) ; umbella 1-pauci- radiata, radiis brevibus ; spiculis multifloris, oblongis, subturgidis ; glumis ovatis obtusis, in latere castaneo-rubescentibus; nuce ellipsoidea, brunnescente, cellulis extimis oblongis.— Kunth, Enum. ii. p. 9; Boeck. in Linnea, xxxv. p. 442. C. intermedius, Steud. in Flora, 1842, p. 581; A. Rich. Fl. Abyss. ii. p. 483 (sub Eucypero perperam ordinata); Boeck. in Linnea, xxxv. p. 450, var. 8 excl. C. Eragrostis, Krauss in Flora, 1845, p. 754, non Vahl. Abyssinia (Schimper n. 1267, hh. Kew, Mus. Brit. ; Schimper n. 1243, h. Mus. Brit., h. Kew partim). Madagascar (Blackburn, h. Kew). This species is remote from C. Eragrostis by the surface-cells of the nut; it stands between C. flavescens and C. latispicatus, dif- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 43 fering from the latter in its smaller spikelets, from the former in its chestnut-red colour. 9. C. TRISTACHYUS (Boeck. in Linnea, xxxv. p. 455); umbelle radiis paucis, breviusculis; spieulis paucis, subspicatis; glumis ovatis, obtusis, fuscis; nuce obovoidea, compressa, cellulis extimis oblongis. America australis: Columbia, Merida (Moritz, hh. Kew, Mus. Brit.); Panama (Seemann, h. Calcutta) ; Venezuela ( Fendler n. 1586, 5. Kew); Bolivia, alt. 2700 metr. ( Mandon n. 1398, h. Kew); Respublica Argentina (Grisebach, h. Kew). This differs a little from C. lanceolatus in the more obovoid nut. 10. C. prceus (Liehm.! Mex. Halfgr. p. 12) ; umbellz radiis paucis, brevibus aut subnullis; spieulis lineari-oblongis, com- pressis, castaneo-rubris; nuce obovoidea compressa, cellulis ex- timis oblongis. C. picreus (err. typogr.), Boeck. in Linnea, xxxv. p. 455 in notd. Mexico, alt. 1500 metr. (Liebmann, h. Kew) (F. Mueller n. 1987 partim, k. Kew). In the few examples examined I found stamens 3, whereas in C. tristachyus stamens 2. ** Cellule extime nucis quadrate (figs. 35-39). T Umbella specie simplex aut capitata, rarius (in varietatibus C. polystachyi, C. atronitentis, C. Mundtii) composita, um- bellularum radiolis elongatis. i Species in Indiá Orientali indigene. 11. C. puminus (Linn. ! h. propr., Amon. Acad. iv. p- 302, Sp. Pl. p. 69, nec Rottb., nec Nees); umbella specie simplice aut congesta ; spiculis fasciculatis aut breviter spicatis aut congestim corymbosis, linearibus, pallidis aut fuscis ; glumis spissis, subre- curvo- mucronatis ; stamine sepius 1; nuce obovoidea, obtusa, api- cata cum 1 parte glume vix sequilonga.— Gertn. Pruct. i. p. 9, t. 2. fig. 2; Vahl, Enum. ii. p. 330; Roem. et Sch. Syst. 11. p. 1885 Kunth, Enum. ii. p. 4 pro minore parte ; Steud. Cyp. p. 3 potins Roxb. Fl. Ind. i. p. 196, syn. Rottb. excl. ; Benth. Fl. Austral. vii. p. 258. C. nitens, Vahl, Enum. i. p. 331; Kunth, Enum. ii. p. 33 Boeck. in Linnea, xxxv. p. 483, non Retz. 44 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. membranaceus, Vahl, Enum. ii. p. 330; Kunth, Enum. ii. p. 3.- C. pulvinatus, Nees et Meyen! in Wight Contrib. p. 74 ; Thwaites, Enum. Pl. Zeyl. p. 342. C. gymnoleptus, Steud.! Cyp. p. 3. C. patens, Hochst. ! in Pl. Hohen. n. 825. C. obstinatus, Steud. ! Cyp. p. 10. Pycreus pulvinatus, Nees in Nov. Act. Nat. Cur. 1843, Suppl. ü. . 53. Cyperus, Wall.! List n. 3312 litt. F, K, I, L, M; litt. B, C, D, E, partim. Cyperus, Wall. ! List n. 3339 (non C. tortuosus, Roxb.). Cyperus, Wall. ! List n. 3376 partim (admizt.). — Pluk. Alm. t. 191. fig. 8. Annua, glabra, radicibus tenuiter fibrosis. Culmi 1-25 em., sursum trigoni, leves. Folia 2-4, cum culmo sepe equilonga, angusta (lata 1-3 mm.), levia. Bractee 3-5, longs 1-15 cm., patulz, foliis similes. Umbella 1-6-radiata; radii usque ad 9 em. (ssepius 0-3 em.) longi; ochreæ truncate. Umbellule 5-30 spi- culose, ebracteate, simplices, vel composite, radiolis 0-2 mm. longis. Spicule undique patule, longe 5-18 mm., late 1-2 mm., compress, 8-44-florz, virides membranacesz, nitidæ aut fusco- brunnee ; rhachilla vix alata; gluma ima ceteris minor, lanceo- lata, bracteseformis; altera truncata, vacua. Glume fertiles ovate, longe 1-2 mm., naviculares, imbricate (in forma C. patente sp. Hochst. subremotz), ovate, apice emarginate, in dorso viridi- 3-5-nervie, nervis apice coalitis in forma mucronis brevis sub- recurvati excurrentibus; glume latera scariosa aut fusca, ecos- tata. Stamen in forma typica semper 1; filamentum sursum minutissime squamoso-papilosum, deciduum; anthera parva, oblonga, mutica, lutea. Stylus brevis (nuce brevior); rami 2, e gluma vix exserti. Nux parvula, compressa, nigro-castanea ; cellule extime subquadrate, emarcide, hyaline, inflate, sspe porose. In h. Linn. propr. planta a manu Linnei “ C. pumilus” notata in centro plagule affixa est; in altero latere fragmentum, ex hort. Fothergill a J. E. Smith (?) additum, est C. flavescens. Species a Pluk. Alm. t. 191. fig. 8, Gertn. Fruct. t. 2. fig. 2 stabilita est, que tabule speciei formam vulgarem optime exhibent; ab hac paullo recedunt forme sequentes :— Forma membranacea sp., Vahl, Enum. ii. p. 330, spicis magis laxis; forma in var. y. punctatam forsan transeuns, sed stamen 1. Forma patens sp., Hochst. (= Wall. List n. 3339), glumis sub- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 45 remotis, nuce sepe pallidiore ; forma in C. pumilum typ. sensim transeuns. Forma borneensis, a Burbidge in Borneo lecta in hh. Kew, Mus. Brit. conservata, elegans, lete albida, culmis elongatis, nuce (ob cellulas extimas laxe saccatas) specie papillosa aut scabro- punctata. In exemplo in Rangoon a R. Scott lecto, umbellularum radioli usque ad 5 mm. longi. In exemplo in Assam a Simons lecto, umbellula plane composita a bracteola 15 mm. longa folieformi suffulta. In omni India, alt. ‘0-1500 metr., ab Himalaya usque ad Zey- laniam et Malacca, communis; in Asia, Australi-orientali et Malaya usque ad Ins. Philippinenses, et (var. incl.) Australiam tropicalem. India: Kumaon (Strachey et Winterbottom n. 3, h. Kew); Dehra Doon (G. King, h. Calcutta); Sikkim (G. King, h. Calcutta); Assam (Jenkins et Masters, h. Kew n. 199); Ben- gal (Griffith, h. Kew n. 6213) ; Mons Aboo (G. King, h. Cal- cutta); Chota Nagpore, alt.600 metr. (C. B. Clarke n. 20419); Madras Peninsula (Wight n. 1809, h. Kew; Hohenacker nn. 825, 826, h. Kew; Hohenacker n. 825, h. Mus. Brit.) ; Zeylania (Thwaites C. P. n. 806, hh. Kew, Mus. Brit., Cal- cutta; Beckett n. 2541, h. Calcutta); Burma et Malacca (Griffith, h. Kew n. 6186). Borneo (Motley n. 274, h. Kew). Ins. Philippine (Cuming n. 559, hh. Kew, Mus. Brit.). Var. B. mutica; glume nervo medio vix aut brevissime excur- rente.— C. nitens, varr. y et 6, Boeck. in Linnea, ix. p. 484. C. patens, Vahl, Enum. ii. p. 334; Kunth, Enum. ii. p. 16. C. truneatulus, Steud.! Cyp. p. 10. C. squarrosus, Kotschy! Iter. Nub. n. 53, non. Linn. C. commntatus, Steud. ! Cyp. p. 10. Stamina 2,in exemplis paucis examinatis. C. patens, Vahl, differt a var. mutica, Boeck., glumis magis distantibus, a C. patente, Hochst., glumis fere muticis; huic fortasse affinior. l Africa tropicalis: Nubia (Kotschy n. 53, hh. Kew, Mus. Brit.) ; Abyssinia (Schimper n. 821, hh. Kew, Mus. Brit.) ; Sene- gambia (A. Kew n. 331). Ins. Socotra (Schweinfurth n. 592, h. Balfour propr. ; Balfour n. 467 bis, 4. propr.). 46 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Var. y. punctata (sp. Roxb.! h. propr., Fl. Ind. i. p. 193, tab. Pluk. excl, non Roxb. Ic. ined. in h. Kew); spiculis paullo laxius spicatis, staminibus 2.—Kunth, Enum. ii. p. 4, syn. excl. C. pumilus, Benth.! Fl. Austral. vii. p. 258. Cyperus, sp., Wall. List n. 3312 litt. A, H ; litt. B, C, E, partim. C. punctatus, Roxb. Ic. ined. in h. Kew mihi species ignota est ; nisi (ut vereor) e duabus speciebus composita fuerit. India (Roxburgh, h. Kew; Wallich n. 3312 A, h. Linn. Soc.) ; Monghyr (Bwchanan- Hamilton, Wallich n. 3312 E, partim, h. Linn. Soc.); Bengal (J. D. Hooker, h. Kew) ; Rangoon (Wallich n. 3312 H, hh. Linn. Soc., Mus. Brit., Calcutta; R. Scott, h. Calcutta) ; Tenasserim (Helfer, h. Kew n. 6209/2) ; Malaya (Kunstler n. 62, h. Calcutta) Australia tropicalis: Rockhampton (O’ Shaughnessy, h. Kew) ; Portus Denison (Fitzalan, h. Kew). 12. C. HYALINUS (Vahl, Enum. ii. p. 329) ; umbella simplice ; spiculis spicatis, admodum compressis, oblongis, pallidis; glumis mucronatis, in unoquoque latere lato conspicue 2—3-costatis ; sta- minibus 2; nuce late ellipsoidea, obtusa, cum 1 parte glum equilonga.— Kunth, Enum. i. p. 3; Boeck.! in Linnea, xxxv. p. 482. C. pumilus, Nees! in Wight Contrib. p. 74, syn. omn. excl. ; Kunth, Enum. i. p. 4, pro majore parte; Steud. Syn. p. 3 partim, non Linn. C. strictus, Wight! h. propr., non Roxb. Pycreus pumilus, Nees in Linnea, ix. p. 283. Cyperus, Wall.! List n. 3336 partim. Annua, glabra, radicibus tenuiter fibrosis. Culmi 5-28 cm., sursum trigoni, leves. Folia 2—4, cum culmo sæpe æquilonga, lata 5 mm., levia, flaccida. Bracteæ 5-6, long: 5—16 cm., latius- cule, foliis similes. Umbella 5-8-radıata ; radii 0-1 dm. longi; ochreæ longæ 15 mm., truncate. Spice longs 0-15 mm., 5-20- spiculosæ, ebracteate ; rhachis 4-alata. Spiculæ rectangulatim patule, longe 9 mm., late 4mm., 8-flore ; rhachille ale per- anguste, persistentes ; glume 2 ime parvule, quadrate. Glume imbricate, paulio distantes, naviculares, summopere compresse ; carina (cum 3-5 nervis adjectis) viridis, excurrens, sursum (in exemplis explanatis) serrulato-scabra; latera lucide alba, a cel- lulis laxis lete reticulata, a nervis conspicue striata. Anthere lineari-oblongx, mutice, lute. Stylus nuce multo brevior ; rami MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 47 2, e gluma plane exserti. Nux rhachi parallele compressa, quam C. pumili 6-plo gravior, castaneo-brunnea; cellule extime quad- rate, emarcidz, hyaline, inconspicue.—Species a C. pumilo plane discreta, cum hoc a Bentham Fl. Austr. vii. p. 258 conjuncta ; Bentham autem C. hyalinum verum non viderat. Madras Peninsula, rara: Monampatta in oryzeto (Wight in Wallich, n. 3336, litt. C partim, h. Linn. Soc.) (Wight n. 1808, hh. propr., Mus. Brit. ; Wight n. 2864, h. Calcutta); sine ha- bitat. (h. Linn. propr., sine nota Linnsi, cum nota lectoris “an Cyperus pumilus umbella composita partialis paniculata medium inter Zaspan et pumilum ”). 13. C. eronosus (Allioni, Fl. Pedem. Auctuarium, p. 49) ; um- bella composita, umbellulis congestis aut simplice pauciradiata ; spiculis breviter spicatis, multifloris, linearibus, compressis; glumis Spissis, ovatis, obtusis; staminibus 2; nuce ellipsoidea, apicata, cum 3 parte glume fere equilonga.— Roem. et Sch. Syst. i. p. 170 (quoad formam typicam), Mant. iii. p. 542 partim; Bertol. Fl. Ital. i. p. 256; Reichb. Iconogr. t. 229, Fl. Germ. t. 279; Boiss. Voy. Esp. ii. p. 626; Fl. Orient. v. p. 864; Gren. et Godr. Fl. Franc. iii. p. 962; Parl. Fl. Ital. ii. p. 24; Boeck. in Linnea, xxxv. p. 458 (non Boeck. in Flora, 1879, p. 546); Benth. Fl. Austral. vii. p. 260. C. confertus, Lam. Ill. i. p. 145. C. divaricatus, Lam. Ill. i. p. 145. C. flavidus, Vahl, Enum. ii. p. 334, syn. Retz. ezcl. ; Decaisne in Nouv. Ann. Mus. d’ Hist. Nat. iii. p. 359, nec Retz., nec Roxb. C. curvatus, Vahl, Enum. ii. p. 338. C. conglomeratus, Vahl, Enum. ii. p. 339, non Rottb. C. Lamarckianus, Roem. et Sch. Syst. Mant. ii. p. 108; Kunth, Enum. li. p. 9. l C. capillaris, Koen. ; Roxb. Fl. Ind. i. p. 194; Nees in Wight Contrib. p. 76; Dalz. and Gibs. Bomb. Fl. p. 283. C. fascicularis, Lam. et DC. Fl. Franç. i. p. 722, nec Rottb., nec Desf. C. lanceolatus, Presi in Rel. Haenk. i. p. 167, non Poir. , C. vulgaris, Sieber ; Kunth, Enum. ii. p. 4 ; Hassk. Pl. Jav. Rar. pr Led. Fl. Ross. i. p. 239; F. Muell. Fragm. Phyt. Austral. vin. p. 260; Benth. Fl. Hongk. p. 385, non Hochst. C. mucronatus, Steud. in Zoll. Verz. Ind. Archip. ii. p. 62, n. 1608, non Rottb. | C. teretifructus, Steud. in Zoll. Verz. Ind. Archip. ii. p. 62; Cyp. p. 3. C. jungendus et C. trachyrhachis, Steud. Cyp. p. 3. 48 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. flavescens, Thwaites! Enum. Pl. Zeyl. p. 342; Benth.! Fl. Austral. vii. p. 260, syn. excl., non Linn. C. Eragrostis et C. lucidus, Rottler! in h. propr. Pycreus capillaris, Nees in Linnea, ix. p. 283. Cyperus, Wall.! List n. 3318 litt. B, D; litt. A, C, partim. Glabra, annua, radice fibrosa. Culmi 1-9 dm., subtrigoni, leves. Folia 2-3, cum 3$ parte culmi subequilonga, angusta, levia. Bractee 2-4, usque ad 15 cm. longe, vix scabre, foliis similes. Umbella 1-5-radiata; radii 0-7 cm. longi; ochrez 5-10 mm., trun- cate. Umbellule 2-12 cm. longer, 10—50-spiculosz, ebracteate, congestim corymbosz, radiolis 0-2 mm. longis, paucispiculosis aut e spica simplice constantes. Spicule undique patule, longe 18 mm., late 3 mm., 30-40-flore, pallide, stramineo-brunnez aut nigro-castanee ; rhachilla subtetragona, alis angustis, persis- tentibus, rubro maculatis; glume 2 ime parvule, bractecformes. Glume florigerz longs» 2 mm., naviculares, in dorso 3—5-nervis, virides, rarius pallide, in marginibus hyaline. Filamenta persis- tentia, squamoso-scabra; anthere oblong, mutice, rubro-macu- late. Nux compressa, nigro-castanea ; cellule extime quadrate, emarcide, hyaline, paullo saccatz, non porose. Ab Europa australi (varr. incl.) usque ad Japoniam, Malayam, Australiam. Europa: Alpes maritime (Reichenbach n. 2309, h. Kew) ; Murcia (Bourgeau n. 1540, hh. Kew, Calcutta). Ins. Mascarene (Macgregor, h. Mus. Brit.). Mauritius (Sieber, Fl. Maurit. ii. n. 14, hh. Kew, Mus. Brit.). Iberia ( Wilhelms, h. Calcutta). India, usque ad 2000 metr. alt.: Cabul (Griffith, Kew n. 6190) ; Nachar (Stoliczka, h. Calcutta); Himalaya boreali-occi- dentalis (T. Thomson, Brandis, &c.). China: Macao (Vachell, h. Kew); Shanghai (Maingay n. 637, h. Calcutta). Ins. Formosa (Siebold, h. Kew). Japan (Siebold, h. Kew). Australia: New South Wales (Woolls, h. Kew.); Victoria, Flum. Upper Hume (F. Mueller, h. Kew). Ins. Nieobars (Kurz, h.! Calcutta). Forma khasiana; spiculis admodum compressis, longis 22 mm. latis 23 mm., 40 floris; glumis nitidis ferrugineis aut atris, in carina viridescentibus. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 49 Khasia colles, alt. 1000-1500 metr. (Griffith, h. Kew 6191; C. B. Clarke nn. 5425, 5758, 17479, 18475, &oc.). Forma cinnamomea ; spiculis minus compressis, sublanceolatis, longis 24 mm., latis 3 mm., 40-50-floris ; glumis concoloribus, cin- namomeo-brunneis, in carina vix viridescentibus. Assam (Mann n. 372, h. Calcutta). Var. B. Nilagirica (Hochst. ! Steud. Cyp. p. 2, sp.); spiculis admodum compressis, quam in C. globoso typico paullo angusti- oribus; glumis admodum spissis, in dorso viridibus, in lateribus atris, castaneis, ferrugineis aut pallidioribus.— Boeck. in Linnea, XXXV. p. 457. C. lividus (sp.), Heyne! MS. C. atroferrugineus, Steud.! Cyp. p. 2. C. Junghuhnii, Mig.! Fl. Ind. Bat. iii. p. 260. C. ater, Dalz. and Gibs.! Bomb. Fl. p. 283, non Vahl. Cyperus, Wall.! List n. 3318, A. Cyperus, Wall.! List n. 3310 [C. semiteres, Heyne]. Africa Australis (Burchell n. 2081, h. Kew). Zeylania (Gardner nn. 954; 955, h. Kew; Thwaites, C. P. n. 801, k. Calcutta). Montes Nilagirici (Hochstetter nn. 945, 945 a, h. Kew; Wight n. 2875, k. Calcutta; C. B. Clarke n. 10890 &e.). Bombay: Hewra (Dalzell, h. Calcutta). Montes Kurg (G. Thomson, h. Calcutta). Java: Tugu, alt. 3300 metr. prope Buitenzorg (Kurz, h. Cal- cutta). Amurland: Kengka Lacus (Maack, h. Kew). China borealis: Shantung (Maingay n. 152, h. Calcutta). Var. y. stricta (sp. Roxb. ! Fl. Ind. i. p. 200); caule stricto ; umbellæ radiis sæpe paullo elongatis ; spiculis minus compressis, quam C. globosi typici paullo angustioribus, haud raro paucifloris ; glumis obscure inflatis, pallide aut fusce stramineis.— Kunth, Enum. ii. p. 12 (an Lam. Il. i. p. 146 ?). C. tortuosus (sp. Rozb.! Fl. Ind. i. p. 197); Nees in Wight Contrib. p. 75; Kunth, Enum. ii. p. 16 (non Wall. List n. 3339). C. complanatus, Presl in Oken Isis, 1828, p. 270; Steud. Cyp. p. 3. partim; Franch. et Savat. Fl. Jap. ii. p. 100. Cyperus, Wall. ! List n. 3318, E, C partim. Cyperus, Wall.! List n. 3312, G. Cyperus, Wall.! List n. 3319, B partim. Punjab (Stewart nn. 293, 796, h. Calcutta). LINN. JOURN.— BOTANY, VOL. XXI. .rTrceHÓANTTTS 50 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Nepaul (Wallich n. 3385, h. Calcutta). Bhotan, alt. 1000 metr. (Gamble n. 9598, h. Calcutta). Assam (Griffith n. 1022, h. Calcutta). Chota Nagpore (Wood, h. Oulautta). Madras (Wight in Wallich n. 3318 B, h. Calcutta; Wight n. 2878, h. Calcutta). Forma fortuosa; spicularum rhachilla ssepe torta. Cabul (Grifith n. 31, h. Kew). Persia: Susiana (Haussknecht, h. Mus. Brit.). Tle de France (Roxburgh, h. Mus. Brit.; Balfour, h. Kew). Of this most interesting species, the above list of localities and of varieties and forms is poor and unfinished; because, when I wrote out the varieties in England, I accepted Boeckeler's view that the colour was the most important character, and I therefore put the blaek Khasi form (which is just as black sometimes as the Nilgherry one) with the var. (sp. Boeck.) nilagirica, as Boeckeler himself has done. But a longer consideration of the splendid series of material in the Caleutta Herbarium has convinced me that the Khasi black variety is as distinet from the Nilgherry black variety as is any form collected under the name globosus ; and I should now arrange the varieties in two main series, one, globosus proper, extending from the Mediterranean through Central Asia, North India, to China and Japan; the other, strictus, with narrower spikelets, in Java, South India, Persia, South Africa. The difference between C. strictus and C. nilagiricus is wholly one of colour. C. globosus, in both its varieties, exhibits every variety of colour, from pale straw through ruddy-brown to black, but not the purple-red tinge so common in Cyperus. It is very hard to say what is the essential character of the species: in most forms, the exactly parallel sides of the spikelet and very regularly-arranged glumes are characteristic, but in the beautiful form cinnamomea (Mann, n. 372) this fails us. It is difficult to say how C. lanceus and C. atronitens differ, or to show that they differ as much from C. globosus as the admitted forms of C. globosus differ between themselves. Many of our * species " of Cyperus, as C. atronitens, are founded on specimens all collected within one small area, which are exactly similar infer se, and form a sharply-defined species. By com- paring a set of these with some one local form of C. globosus, we — rere ee- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 51 are irresistibly led to the suspicion that further collections in other areas may lead us to regard C. atronitens &c. as varieties. Besides the “forms” mentioned above, a plant, Maingay n. 637, collected at Shanghai, with the glumes green and chestnut-brown, has exactly the external aspect of C. Eragrostis; but, on cutting up, is found to be unmistakably C. globosus. C. complanatus, Steud. Cyp. p. 3, is an amalgam of C. compla- natus, Presl, with Kotschy n. 358, whieh is C. rotundus, Linn. (mihi), and is C. stoloniferus, var. pallida, Boeck. 14. C. porystacnyus (Rottb. Descr. et Ic. p. 39, t. 11. fig. 1, nuce et pistillo excl.) ; umbella composita, specie ssepe simplice aut capitata; spiculis fasciculatis aut spicatis, numerosis, multi- floris, compressis, anguste oblongis, pallidis, fuscis aut rubescen- tibus; glumis spissis, ovatis, vix acutis; nuce oblonga, basi conica, apice truncata, compressa faciebus eonvexis.— Vahl, Enum. ii. p. 315 ; Roem. et Sch. Syst. ii. p. 175; Roxb. Fl. Ind. i. p. 193 ; Nees in Wight Contrib. p. 75; Kunth, Enum. ii. p. 13; Hook. Niger Fl. p. 549 ; Meyer, Prim. Esseg. p. 27; Hassk. Pl. Jav. Rar. p. 74; Dalz. and Gibs. Bomb. Fl. p. 281; Bertol. Fl. Ital. i. p. 257; Guss. Pl. Ins. Inarime, p. 847 ; Parl. Fl. Ital. ii. p. 25; Mig. Fl. Ind. Bat. iii. p. 258; F. Muell. Fragm. Phyt. Austral. vii. p. 265; Benth. Fl. Hongk. p. 385, Fl. Austral. vii. p. 261; Boeck. in Linnea, xxxv. p. 477 ; Boiss. Fl. Orient. v. p. 365; Coss. Expl. Algér., Bot. iii. p. 253. C. fascicularis, Lam. Ill. i. p. 144, t. 38. fig. 2; Desf.! FI. Atl. i. p. 44. C. cespitosus, Poir. Encyc. vii. p. 250; Kunth, Enum. ii. p. 109; Hcok. et Arn. ! Bot. Beechey Voy. p. 99. . subulatus, R. Br. Prodr. p. 217. . brizeeus, Presl, Rel. Haenk. i. p. 168, non Rich. . Sonderi, Schmidt, Fl. Ins. Vir. p. 162; Steud. Cyp. p- 8. . strigosus, Wight! h. propr. n. 1811, non Linn. . corymbosus, Rowb.! in h. propr. et Ic. ined. in h. Kew, non Rottb. . paniculatus, Hook. et Arn. Bot. Beechey Voy. p. 99, viz Rottb. . polystachyus, var. 6. ferruginea, Boeck.! in Flora, 1879, p. 548 (non Boeck. in Linnea, xxxv. p. 479). Pycreus polystachyus, Beauv. Fl. d’ in Linnea, ix. p. 283. Cyperus, Wall.! List n. 3340. Cyperus, Wall.! List n. 3333. Cyperus, Wall.! List n. 3320 litt. A partim. Cyperus, Wall.! List n. 3332 litt. B partim. aaaaaaa Owar. ii. p. 48, t. 86. fig. 2; Nees E 2 52 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Glabra, ezspitosa, radicibus fibrosis; aut interdum (in forma typica) umbella capitata; rhizoma horizontale, lignosum, nigrum, tenue. Culmi 3-8 dm. (interdum in arena maritima subnulli), sursum trigoni, leves. Folia 3-7, culmum superantia aut hoc multo breviora, linearia, basi paullo dilatata, vix scabra. Bracteæ 3-8, usque ad 2-4 dm. longe, patentes, foliis similes. Umbella (in forma typica) specie simplex aut sspe capitata; radii 1-7, longi 0-3 cm.; ochree truncate, unidentate. Spicule in uno- quoque radio 10-40, erecta, fastigiatz (in var. laxius spicatz) 20-50-flore, longe 2 cm., late 2 mm.; rhachille ale anguste persistentes, juxta basin glume dilatate; gluma ima minuta, lanceolata, bractezeformis. Glume imbricate, carinatz, in dorso viridi-3-5-nervie, in lateribus nec nervose nec nitide. Sta- mina 2; filamenta parum dilatata, levia vel microscopice parce papillosa ; anther oblonger, mutice, lutez, sepe rubro-maculate. Stylus cum nuce subequilongus ; rami 2 e gluma longiuscule exserti. Nux cum 42 parte glume equilonga, subsymmetrica, apicata, nigra; cellule extime quadrate, emarcide, persistentes, albide, sspe porose.—Umbellula in unoquoque radio revera semper corymbosa; radioli imi ochreolati, breves, 1-3-spiculosi aut in varietatibus multispiculosi; spica terminalis, multispi- culosa. In totâ orbe terrarum, in tropicis et subtropicis, prsesertim in insulis et maritimis vulgaris (varr. 0, ı in Americam borealem temperatam longe prolatis). Europa: Ins. Ischia (Hubbard, h. Kew). Africa: Algeria (Lefranc n. 487, hh. Kew, Mus. Brit); Flum. Nun (Mann n. 530, h. Kew); Africa Centralis (Schweinfurth n.1152,h. Kew) ; Djur (Schweinfurth n. 1462, h. Kew); Congo flum. (Chr. Smith nn. 14, 25, h. Mus. Brit.) ; Caput Bone Spei (Burchell nn. 4433, 7567, 8643, h. Kew; Miller, Poeppe, h. Calcutta); Natal (Krauss n. 218, A. Kew) ; Zanzibar (Hildebrandt n. 1069, hh. Kew, Mus. Brit.) ; Madagascar (Pervillé n. 477, h. Kew); Mau- ritius (Sieber n. 10, hh. Kew, Mus. Brit.) ; Ins. Seychelles (Pervillé nn. 90, 484, h. Kew); Ins. Rodriguez (Balfour, h. Mus. Brit.). India australis et orientalis: Zeylania (Thwaites, C. P. 800, hh. Kew, Mus. Brit.); Madras Peninsula (Wight n. 1811, h. Kew; Wight in Wallich n. 3340 B, ^. Calcutta) ; Bengal, Khoolna (C. B. Clarke n. 21770) ; Sylhet (J. D. Hooker MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 58 et T. Thomson, h. Kew); Burma (Griffith nn. 6192, 6205, h. Calcutta) ; Mergui (Griffith, h. Kew n. 6184); Malacca (Griffith, h. Kew n. 6205) ; Singapore (Kurz n. 3018, h. Calcutta) ; Ins. Andaman et Ins. Nicobar (Kurz, h. Cal- cutta). China: Shantung (Maingay n. 155, h. Kew); Whampoa (Hance n. 1230, h. Kew); Canton (Sampson n. 258, h. Kew) ; Shanghai (Maingay n. 638, h. Calcutta); Hongkong (C. Wright n. 555, h. Kew). Malaya: Java, Buitenzorg (Kurz n. 1854, h. Calcutta); Borneo (Motley n. 07, h. Kew; Beccari n. 8685, h. Kew); Labuan (Barber n. 67, h. Calcutta). Australia: Paterson flumen (A. Brown n. 5907, h. Mus. Brit.) ; Portus Jackson (R. Brown n. 5908, hh. Mus. Brit., Kew) ; Botany Bay (Sir J. Banks, h. Mus. Brit.) ; Queens- land (Am. Dietrich n. 632, h. Kew). Polynesia: Isle of Pines (Milne n. 169, h. Kew); Aneiteum (Macgillivray, h. Mus. Brit); Ins. Sandwich (Beechey, h. Kew; Barclay n. 1911, h. Mus. Brit.; Seemann n. 2250, h. Kew; Hillebrand n. 522, h. Kew). America meridionalis et centralis, cum India occidentali: Mar- tinique (Sieber n. 13, h. Kew); Cuba (C. Wright n. 3354, hh. Kew, Mus. Brit.); Guadeloupe (L Hérminier, ^. Mus. Brit.) ; S. Vincent (Guilding n. 36,h. Mus. Brit.) ; Nicaragua (Tate n. 508, hh. Kew, Mus. Brit); Guiana Francorum (Sagot nn. 637, 643, h. Mus. Brit.) ; Pernambuco (Gardner n. 124, k. Kew); Brasil (Martius n. 1086, h. Kew; Bur- chell nn. 1209, 1210, 1489, 9919, h. Kew; Glocker nn. 552, 600, 4. Mus. Brit. ; Swainson, h. Calcutta) ; Rio Negro, Barra (Spruce, h. Calcutta) ; Monte Video (Baile, h. Mus. Brit.). Var. B. laxiflora, Benth.! Fl. Austral. vil. i" 261; culmis foliis- que elongatis, laxis; umbella plane composita, radiis usque a l dm. longis. i | C. aquatilis, F. Muell.! Fragm. Phyt. Austral. viii. p. 270 (species). C. subulatus, Sieber, Agrost. n. 145, viz R. Br. Australia: Sinus Rockingham (Dallachy, h. Kew). Singapore (Lobb, h. Kew). Zeylania (Col. Walker, h. Kew). Madras australis: Montes Anamallay (Beddome, h. propr.). Malaya (Kunstler n. 43, h. Calcutta). 54 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Var. y. minor ; umbella laxiuscula; spiculis minus compressis, pallidis, fusco-rubro notatis. C. pilosus, Baker! Fl. Maurit. p. 413, non Vahl. An C. minor, Steud. Cyp. p. 4? Mauritius (Grey, h. Kew; Bouton, h. Kew). Var. 6. ferruginea, Boeck. in Linnwa, xxxv. p. 479; umbella laxiore sepius plane composita; spiculis rubescentibus aut sub- castaneis, staminibus 2 vel 3. . ferrugineus (sp.), Poir. Encycl. vii. p. 261; Kunth, Enum. ii. p. 11. . olivaceus, Vahl, Enum. à. p. 332; Kunth, Enum. ii. p. 14. . intactus, Vahl, Enum. ii. p. 332; Kunth, Enum. ii. p. 14. . plenus, Heyne; Kunth, Enum. ii. p. 12. . Hookerianus, Arn.! in Wight n. 2055, non Thwaites. . Thouarsii, Kunth, Enum. ii. p. 12; Boeck. in Linnea, xxxv. p. 481. . foliosus, Kunth, Enum. ii. p. 11. . globosus, Boeck. in Flora, 1879, p. 546, non Allioni. . strictus, Rottler MS.; Heyne partim, non Roxb. Cyperus, Wall.! List n. 3329, K. Cyperus, Wall. ! List n. 3331. Cyperus, Wall.! List n. 3332, litt. B partim. Africa: Djur (Sehweinfurth n. 1462, h. Kew); Zanzibar (Hildebrandt n. 1071, h. Mus. Brit.). Mauritius (Bouton, h. Kew; Bojer, h. Kew; Sieber n. 5, hh. Kew, Mus. Brit.). Ins. Mascarene (Macgregor, h. Mus. Brit.). India: Madras (G. Thomson n. 70, h. Kew); Zeylania (Wigh n. 2055, h. Kew; Thwaites n. 800, h. Calcutta); Pegu (Kurz n. 2684, h. Calcutta). Singapore (Kurz n. 3012, h. Calcutta). Ins. Andaman (Kurz, h. Calcutta). Ins. Nicobar (Kurz, h. Calcutta). Ins. Sandwich: Oahu (Seemann, h. Calcutta ; Hillebrand n. 564, h. Kew). Var. e. micans; umbella laxa, composita (interdum magna); spiculis majoribus, rubro-brunneis aut luteo-brunneis. C. micans, Kunth, Enum. ii. p. 12 (species). C. polystachyus, var. ferruginea, partim, Boeck. in Linnea, xxxv. p. 479. In forma extrema spicule laxiuscule spicatz, longe 4 cem., latee 3-4 mm., 40-flore. Natal (Krauss n. 189, hh. Kew, Mus. Brit.; Buchanan nn. 103, 111, 319, 4. Kew; Sutherland, h. Calcutta; ge.). Zanzibar (Hildebrandt n. 1071, h. Kew). qaaaeaanaanaana MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 55 Var. ¢. holosericea; spiculis pallidis, sepe paucifloris, quam in var. filiciná angustioribus. C. holosericeus (sp.), Link, Hort. Berol. i. p. 317 ; Kunth, Enum. ii. . 15. C. inicrodontus, Torrey ! in. Ann. Lyceum New York, iii. p. 255; Bot. U.S. and Mex. Bound. Surv. p. 226. C. Gatesii, Torrey in Ann. Lyceum New York, iii. p. 255. C. texensis, Steud. Cyp. p. 9. C. brizæus, Schweinitz MS., non Rich. C. incrassatus, Beyr. MS. C. polystachyus, var. leptostachya, Boeck. in Linnea, xxxv. p. 478. Spieule debiles, laxiuscule spicatz vel congeste.—C. miero- dontus, var. (0. texensis eadem est ac Drummond n. 454.— C. brizeus, Schwein., es& exemplum parvulum spiculis congestis. —Hee var. a C. polystachyo typ. vix distingui potest. America borealis subtropica: Florida (Chapman, h. Calcutta); New Orleans (Drummond, h. Calcutta) ; Texas (Drummond n. 454, hh. Kew, Mus. Brit., Calcutta; Elihu Hall n. 676, h. Kew). Var. n. paniculata; umbella ssepius laxa; spiculis angustis, linearibus, flavidis, castaneis aut rubescentibus. C. paniculatus (sp.), Rottb. Descr. et Ic. p. 405 Vahl, Enum. ii. p. 333. C. scopellatus, Rich. in Act. Soc. Hist. Nat. Paris, i. p. 106. C. olidus, Vahl, Enum. ii. p. 333. C. microdontus, Sauv. Fl. Cub. p. 171, vix Torrey. C. fugax, Liebm. ! Mex. Halvgr. p. 8; Boeck. in Linnea, xxxv. p. 474. Pycreus Olfersianus, Nees! in Mart. Brasil., Cyp. p. 8, non Cyperus Olfersianus, Kunth. . America tropicalis cum India occidentali: Cuba (C. Wright, h. Kew) ; Mexico (Liebmann, h. Kew) ; Ecuador, Chanduy in litore maris Pacifici (Spruce n. 6415, hh. Kew, Calcutta) ; Brasil, Alagoas (Gardner n. 1436, hh. Kew, Mus. Brit.). Var. 0. filicina ; umbella simplice; spiculis (quam C. poly- stachyi typici) paullo latioribus; glumis sepe luteo-fuscis, sub- nitidis, haud raro apice mucronatis. . C. filicinus (sp.), Vahl, Enum. ii. p. 332; Kunth, Enum. ii. p. 14. C. esespitosus, Torrey, Cut. Pl. New York, p. 89; Spreng. Syst. i. p. 224, non Poir. C. Torreyanus, Roem. et Sch. Syst. Mant. ii. p. 101. . C. Nuttallii, Eddy in Spreng. Neue Entd. i. p 240; Torrey m Ann, Lyceum New York, iii. p. 252; Fl. New York, ii. p. 339, t. 135. C. brizeus, Pursh. Fl. Amer. Sept. i. p. 51, nec Rich., nec Schwein. 56 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. Louisiane, Steud. Cyp. p. 10 (ex descr.). C. polystachyus, var. macrostachya, Boeck. in Linnea, xxxv. p. 479. Torreya cespitosa, Rafin. in Journ. de Phys. lxxxix. p. 105 (fide Kunth). America borealis: Canada (Goldie, h. Calcutta) ; New Jersey (Torrey n. 69, h. Kew; Asa Gray, h. Calcutta) ; Virginia (Greville, h. Calcutta) ; Florida (Chapman, h. Kew; Curtiss n. 3050, h. Mus. Brit.); New Orleans (Drummond n. 3816, hh. Kew, Mus. Brit.) ; Mexico (F. Müller n. 2155, h. Kew). Ins. Bahama (Brace n. 498, h. Kew). This var. (of which I suppose var. «. Cleaverii is merely a starved form) may perhaps advantageously be treated as a species : it has always a simple umbel, whereas in other varieties and examples of C. polystachyus the umbellules appear to me really with divided axes. Var.«. Cleaverii : culmis unicam spiculam, ei C. filicini similem, proferentibus. C. Cleaverii (sp.), Torrey! in Ann. Lyceum New York, iii. p. 258 in notà; Steud. Cyp. p. 13. C. minimus ?, Nutt. Gen. i. p. 35, non Thunb. C. tenellus, Torrey in Ann. Lyceum New York, iii. p. 257, non Linn. New Jersey (Cleaver, h. Carey in h. Kew). 15. C. suncrnux ; umbella simplice; spiculis breviter spicatis aut fasciculatis, multifloris, compressis, linearibus, pallidis; glu- mis ovatis, obtusis; nuce oblonga, compressa, faciebus concavis. Annua, tenuior. Culmi 5-40 cm., apice trigoni, leves. Folia 2-3, perangusta, cum culmo equilonga aut sepius multo bre- viora, levia. Bracteæ 3-8, anguste, modo 12 cm. longs, um- bellam longe superantes, modo umbella breviores. Radii 3-8 usque ad 1 dm. longi, graciles, aut umbella parvula subcapitata ; ochre usque ad 1 em. longer, anguste, subtruncate. Spicule in unoquoque radio 6-12, longe 1-3 cm., late vix 2 mm., 10-50- florz, erectz aut patule; rhachilla vix alata; gluma ima minuta, bracteeformis. Glumez subremote imbricate, in dorso carinate, 3-5-nervie, fusez, in lateribus enervoss, pallide aut obscure rubro-notate Stamina 2; anthere lineari-oblongs, mutice, flavee, apice rubro. Stylus brevis; rami 2, breves, e gluma parum exserti, Nux $ partem glume paullo superans, apice truncata brevissime apicata, in utraque facie sulcato-concava, subsymmetrica aut in margine axin versus spectante obliqua, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. ^57 interdum sulcata; nigra, cellule extime quadrate, emarcide, albidz non conspicue. Bengalia borealis et orientalis, alt. 0-1000 metr., frequens (C. B. Clarke nn. 17295, 24860, 26479, 27072, 27081, 27156, &c. ; Kurz, h. Calcutta). Assam (Griffith n. 1601, h. Kew). Pegu (Kurz n. 659, h. Calcutta ; Eng. Oates, h. Calcutta). Tenasserim (Helfer, h. Kew n. 6209 / 4). Madras australis, Montes Anamallay, alt. 1000 metr. ( Beddome, h. propr.). Borneo borealis (Burbidge, hh. Kew, Mus. Brit.). Ins. Philippine (Moseley, h. Kew). 16. C. Eracrostis (Vahl, Enum. ii. p. 322) ; umbella specie simplice, pauciradiata, contracta; spiculis fasciculatis, lineari- oblongis, compressis, subturgidis ; glumis ovatis, obtusis, in late- ribus rubescentibus; nuce cum 3-2 parte glume æquilonga, ob- ovoidea, crassiuscula.— Kunth, Enum, ii.p.7; Benth.! Fl. Hongk. p. 385; Mig. Fl. Ind. Bat. iii. p. 256; F. Muell. Fragm. Phyt. Austral. viii. p. 260; Boeck. in Linnea, xxxv. p. 443 (varr. B partim, e excl.). C. cruentus, Retz. Obs. v. p. 13; Roxb. Fl. Ind. i. p. 196, non Rottb. C. nitidus, Lam. Ill. i. p. 145, non Boeck. C. albidus, Lam. Ill. i. p. 146. . C. sanguinolentus, Vahl, Enum. ii. p. 351; Nees in Wight Contrib. p. 75; Thwaites, Enum. Pl. Zeyl. p. 342; Herb. Ind. or., Hook. f. et T. Thoms. n. 43. C. ater, Vahl, Enum. ii. p. 335; Kunth, Enum. ii. p. 7, non Dalz. and Gibs. C. areolatus, R. Br. Prod. p. 216. C. erythraeus, Roem. et Sch. Syst. Mant. ii. p. 4/7. ——— C. pumilus, Rottb. Descr. et Ic. p. 29, t. 9. fig. 4, non Linn. — C. flavescens, 8. rubro-marginatus, Schrenk! Enum. Pl. Nov. i. p. 3; Led. Fl. Ross. i. p. 239. o. M; C. atratus, Steud. in Zoll. Verz. Ind. Archip. p. 62, Cyp. p. 12; Mig. Fl. Ind. Bat. iii. p. 259 ; Boeck.! in Linnea, xxxv. p. 446. C. concolor, Steud.! Cyp. p. 6. C. melanocephalus, Mig. Fl. Ind. Bat. iii. p. 259, e descr. C. grossarius, Koenig ; Heyne MS. in h. propr. Cyperus, Wall.! List n. 3335. Cyperus, Wall.! List n. 3319, A, B partim, C, D, E, F, G, H. Cyperus, Wall. ! Cyp. Indet, n. 265, h. propr. Cyperus, n. 15, Herb. Ind. or., Hook. f. et T. Thoms. 58 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Annua. Culmi basi decumbentes, radicantes, sepe divisi, non lignosi, 1-5 dm. longi, apice trigoni, leves; vagine foliorum longs, sepe more C. Mundtiz, 3 partem culmi integentes. Folia 3-6, cum culmo subequilonga aut breviora, linearia basi paullum dilatata, in marginibus minutissime scabra. Bractes 2-5, usque ad 5-15 em. longe, divaricate, folis similes. Umbelle radii 3-9, sepius pauci, usque ad 5-8 em. longi, sepius multo breviores aut subnulli; ochre cylindrice, inconspicue, subbidentate. Umbellule obscurius composite, congestim corymbose, ebracteo- late, radiolis imis brevissimis, paucispiculosis aut e spica simplice constantes. Spicule in unoquoque radio 3-12, longe 8-25 mm., latz 4 mm., 6-24-flore; rhachilla fere exalata; gluma ima par- vula, triangulari-lanceolata. Glume compresses, vix naviculares, marginibus imbricatis aut (in sieco) subliberis, in dorso virides, sub-3-nervie, in lateribus rubescentes aut rubro-brunnesx nec flavescentes. Stamina 3, rarius 2; filamenta squamoso-scabra, persistentia ; anthera lineari-oblonga, mutica, flavida. Stylus nuce longior, rami 2, cum stylo subequilongi, e gluma exserti. Nux lata, subtumida, obtusa, apicata, nigro-fusca; cellule extime quadrate, emarcide, albidz, detergibiles.—In forma depauperata, culmi 3-1-spiculigeri, bractea erecta quasi culmum producente. Asia australi-orientalis, alt. 0-3000 metr., presertim in mon- tanis eommunis; usque ad Amurland, Lacum Alakul, Abys- siniam et Australiam extensa. Lacus Alakul (Schrenk, h. Kew). Baltisthan, alt. 2500 metr. (C. B. Clarke n. 30021, h. Kew). India: Sikkim, Lachoong, alt. 3000 metr. (J. D. Hooker, h. Kew); Punjab (Jacquemont nn. 432, 437, h. Kew) ; Cabul (Griffith n. 6188 partim, k. Kew; n. 6188, h. Calcutta) ; Kumaon (Strachey et Winterbottom n. 5, h. Kew) ; Nepaul (Wallich n. 3319 H, h. Calcutta); Himalayas Orientalis (Griffith nn. 6187, 6201, hh. Kew, Calcutta) ; Khasia, alt. 1200 metr. (C. B. Clarken. 5349) ; Bengal, Sahebgunj (Kurz, : inh. Calcutta) ; Mons Aboo (King, h. Calcutta) ; Montes nila- girici (Hohenacker n. 946, h. Kew; n. 946 partim, A. Mus. Brit.) ; Zeylania (Thwaites, C. P. n. 802, hh. Kew, Calcutta); Chittagong (J. D. Hooker et T. Thomson n. 57, h. Kew). Burma: Ava (Wallich n. 3335, hh. Kew, Linn. Soc., Calcutta) ; Pegu (Kurz n. 654, h. Calcutta) ; Tavoy (Wallich, h. Kew). Java: Buitenzorg (Kurz, h. Calcutta). China: Hongkong (Wilford n. 312, À. Kew) ; Kianang (Sir MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 59 G. Staunton, h. Mus. Brit.) ; Amoy (Hance n. 1387, h. Kew); Formosa (Oldham n. 586, hh. Kew, Mus. Brit.) ; Shautung (Maingay n. 156, h. Calcutta). Japan (Oldham n. 902, h. Kew). Australia: Portus Jackson (R. Brown n. 5896, k. Mus. Brit.). Ins. S. Helena (Burchell n. 7, h. Kew, an a navibus Anglicanis ab India allata ?). Var. B. cyrtostachys; spiculis paullo minoribus, sepius 6-10- floris, magis fuscis. C. cyrtostachys (sp.), Mig. Fl. Ind. Bat. iii. p. 257. C. levis, F. Muell. ! in h. propr., non R. Br. C. Eragrostis, Benth. ! Fl. Austral. vii. p. 258. C. Eragrostis, var. 8. microstachya quoad exempla Australica et var. 8. cyrtolepis, Boeck. in Linnea, xxxv. p. 445. Varietas in formam typicam sensim transeuns; exempla in Peninsula Malaccensi lecta inter C. Eragrosti typicum hujusque varietatem cyrtostachidem intermedia sunt. Java. Borneo (Motley n. 989, h. Kew). Australia: Queensland ; New South Wales; Richmond Flumen (C. Moore n. 151, h. Kew); Victoria, Mons Aberdeen et Buffalo Range (F. Mueller, h. Kew); Flumm. Mitta Mitta et Hume (F. Mueller, h. Kew) ; South Australia. Var. y. neurotropis; spiculis interdum (quam in C. Eragrostis typico) minoribus ; glumis breviter ovatis. C. neurotropis (sp.), A. Rich. Fl. Abyss. ii. p. 476. C. Eragrostis, var. microstachya, Boeck. in Linnea, xxxv. p. 445, quoad exemplum Africanum. . . , Nux in n. 765 Schimper parva, albida videtur: sed in omnibus formis C. Eragrostis exempla cum nucibus (imperfectis P) mino- ribus albis interdum obvia sunt. l Abyssinia (Schimper n. 1424, h. Mus. Brit.): Gapdia (Schim- per n. 765, hh. Kew, Mus. Brit.). Var. à. micronus; spiculis angustioribus, usque ad 36-floris, Concoloribus, cinnamomeo-brunneis ; glumis magis spissis, mar- ginibus in medio conspicue rhomboideo-angulatis ; stylis longius exsertis, spiculas comosas efficientibus. C. Eragrostis, var. minor, Kunth, Enum. ii. p. 7, in obs. C. jeminicus, 4. Heyne! nec Rottb. Cyperus, Wall.! List n. 3354, B. India (Wallich n. 3354 B, hh. Kew, Linn. Soc., Calcutta) ; Bengal, Mymensingh (C. B. Clarke n. 7957, h. Calcutta) ; Assam, Gowhatty (Simons, h. Calcutta). 60 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. The nut is perhaps hardly smaller than in C. Eragrostis type, but I have written up this name in European herbaria and cannot alter it. 17. C. untotorpEs (R. Br.! Prod. p. 216); umbella simplice aut subeapitata; spiculis breviter spicatis, magnis, compressis, lanceolatis, pallide brunneis; glumis spissis, carinatis, acutis; nuce cum 4-4 parte glume zquilonga, obovoidea.— Kunth, Enum. n. p. 112; Benth. Fl. Austral. vu. p. 260. C. lanceus, F. Muell. Fragm. Phyt. Austral. viii. p. 259, non Thunb. C. luteolus, Boeck.! in Flora, 1875, p. 82. Radix fibrosa. Culmi 4-8 dm., sursum trigoni, leves. Folia 2-4, cum $ parte culmi squilonga, in marginibus antice spi- nuloso-scabra. Bractes3-4,long: (interdum usque ad 3-4 dm.), antice spinuloso-scabre, foliis similes. Umbelle radii 1-5, usque ad 5 em. longi, sepius breves aut brevissimi; ochree 1 cm., anguste, truncate. Spice longe 0-1 cm., ebracteate, 4-12- spiculose. Spicule longe 8-18 mm., late 4-7 mm., 10—22-flore, a bracteola lineari longa 3-7 mm. interdum suffulte ; rhachilla fere exalata; gluma ima parva, oblonga. Glume ovate, com- presse, subturgide, in dorso viridi-3-nervie, in lateribus ener- voss, lucide luteo-brunnee, leves, in marginibus hyaline in sicco erispato-undulatse. Filamenta 3, persistentia; anthere lineari- oblonge, mutice. Stylus cum nuce equilongus; rami 2, e gluma breviter exserti. Nux leviter compressa, nigra; cellule extime quadrate, emarcide, opace, albe, porose. Australia, tropicalis et temperata ; varr. in India, Capite Bone Spei et America calidiore etiam extensis. Australia (R. Brown n. 5900, hh. Kew, Mus. Brit.); Queens- land (Am. Dietrich n. 612, h. Mus. Brit.); Queensland, Brisbane fl. (F. Mueller, h. Kew); Victoria, Hume flum. (F. Mueller, h. Kew). Var. f. angulata; nuce subobtuse obovoidea, in altero margine interdum subacuta, cellulis extimis late porosis (si mavis nuce albo-retieulata). C. angulatus (sp.), Nees! in Wight Contrib. p. 73; Boeck.! in Linnea, xxxv. p. 465. C. spinuliferus, Boeck. ! in Linnea, xxxv. p. 465. Pycreus angulatus, Nees in Linnea, ix. p. 283. Cyperus, Wall.! List n. 3324 (omnino). Cyperacea, Griff. Itin. Notes, p. 90, n. 1327. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 61 In Wallich n. 3324 A pro maxima parte exempla nimis juvenia sunt spiculis 3-4 mm. latis. In exemplis maturatis spicule usque ad 6-7 mm. latz videntur (exempla numerosa a Khasia et Nepaul) In exemplis nilagiricis, umbella subcongesta, spicule fere castanese. Nux in omnibus omnino eadem ; mihi forme nec varietates æstimandæ sunt. India, haud communis: Nepaul (Wallich n. 3324 A, hh. Linn. Soc., Kew, Mus. Brit., Calcutta); Khasi colles (Griffith n. 1327, h. Kew; Kurz n. 371, h. Calcutta); Khasi colles, alt. 1000 metr. (Hooker f. et T. Thomson, hh. Kew, Calcutta); Bengal, Dinajpur in lacubus natans (C. B. Clarke n. 26442); Montes Nilgiri (Wight n. 2875, h. Kew; Wight n. 2874, h. Calcutta ; King, h. Calcutta) ; Pycara (Beddome, h. propr.). Burma: Ava (Wallich n. 3324 B, h. Linn. Soc.). Var. y. capensis ; nuce obovoidea, in utroque margine a linea extrusa brunnea acute carinata, nigra, cellulis extimis albescenti- bus, obscuris. C. angulatus, var. capensis, Boeck. in Linnea, xxxv. p. 465. C. bromoides, Kunth, Enum. ii. p. 8 partim. C. pseudo-bromoides, Boeck. in Linnea, xxxv. p. 464, quoad exempla Drége. Caput Bone Spei (Drege, h. Kew, exempla plura); Transkei (R. Baur n. 871, h. Kew). Var. 6. bromoides (sp.), Link, Jahrb. iii. p. 88; nuce cum ] parte glume vix squilonga, apice subconoidea, marginibus obtusis, cellulis extimis albis, opacis, obscure scabridis, nec porosis nec reticulatis.— Kunth, Enum. ii. p. 8; Boeck. in Linnea, xxxv. p. 463. C. pseudo-bromoides, Boeck. in Linnea, xxxv. p. 464, quoad exempla americana. America meridionalis, cum India occidentali, et Mexico : Mexico (Berlandier n. 758, h. Mus. Brit.; Bourgeau n. 2731 partim, 2. Kew); Cuba (C. Wright n. 214, h. Kew); Guate- mala (Salvin, h. Kew); Venezuela, Tovar (Fendler n. 1590, h. Kew); Brasil (Burchell n. 8158, h. Kew); Paraguay, Villa-Rica (Balansa n. 405, h. Kew). tt Species in Indid Orientali itondum reperta. 18. C. ngvcorzprs (Carey MS. in h. propr); umbella congesta ; spiculis 4-6-floris, albescentibus; glumis obtusis, nervo sub apice breviter excurrente, in utroque latere hyalino 2-nervosis. 62 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. divergens, Chapm. Fl. South U.S. p. 512, non H. B. K. C. pumilo similis. Glum: fere C. hyalint sed multo minores. Stamina 3. Nux immatura. United States: Florida, prope Quincy (Chapman n. 7, in h. Carey in h. Kew). 19. C. ArzELIr (Boeck. in Linnea, xxxv. p. 475); umbella capitata; involucri bracteis 2-1, altero proventu suberecto quasi culmum producente; glumis ovatis, obtusis, stramineis; nuce anguste obovoidea, subtruncata, nigro-castanea, cellulis extimis quadratis. C. flavescens, Oliver in Trans. Linn. Soc. xxix. p. 164, non Linn. C. Afzelit, var. (9. capillifolia, Boeck. in Flora, 1879, p. 547, e descriptione videtur eadem ; hujus autem typus, Schweinfurth n. 2195 in h. Kew est species umbella composita, stylo trifido, &e. Africa: Sierra Leone (Afzelius, h. Mus. Brit.); in regione nilotica superiore Madi (Grant, h. Kew); Djur (Schwein- Surth n. 1948, h. Kew). 20. C. OLFERSIANUS (Kunth, Enum. ii. p. 10); umbella specie simplice, 1-pauciradiata ; spiculis plurifloris, oblongis, subtur- gidis; glumis ovatis, obtusis, stramineis; nuce late oblonga, utrinque angustata, pallide fusca, cellulis extimis quadratis.— Boeck. in Linnea, xxxv. p. 489 ; non Pycreus Olfersianus, Nees. . propinquus, Steud. Cyp. p. 4. . helvus, Liebm.! Mex. Halfgr. p. 9. . ambiguus, Liebm.! Mex. Halfgr. p. 11, non Steud. . correctus, Steud. Cyp. p. 7. . Maximiliani, Schrad.; Steud. Cyp. p. 4. . fasciculatus, Elliot, Bot. i.p. 63; Kunth, Enum. ii. p. 113. . variegatus, Griseb.! Fl. Brit. West Ind. p. 562; Sauvalle! Fl. Cub. p. 171, non H. B. K. C. Elliotianus, Roem. et Sch. Syst. ii. Mant. p. 100; Torrey in Ann. Lyceum New York, iii. p. 254. Pycreus infirmus, Nees in Mart. Brasil., Cyp. p. 8, fide Boeck. (sed repe- rire nequivi). P. propinquus, Nees! in Mart. Brasil., Cyp. p. 7. P. Maximiliani, Nees in Mart. Brasil., Cyp. p. 7. P. Elliotianus, Nees in Linnea, ix. p. 283. America: a Mexico et Cuba usque ad Monte Video frequens. Insule Indie Occidentalis: Cuba (Wright n. 706, h. Kew); Jamaica (forma = C. ambiguus, Liebm.) (Purdie, h. Kew); S. Domingo (Jacquemont, h. Kew). aaaaaaa MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 63 Mexico: Huatusco, alt. 1400 metr. (Liebmann, h. Kew; F. Mueller n. 1765, h. Kew). Guatemala (formaz C. ambiguus, Liebm.) (G. Bernoulli nn. 418,575, h. Kew). Brasil (Burchell nn. 2298, 2418, 4374, 6493, h. Kew; Glaziou n. 6761, h. Kew; Gardner n. 714, hh. Kew, Mus. Brit.; Glocker n. 202, hh. Kew, Mus. Brit.); Rio Janeiro (Glaziou n. 9339, h. Kew). Paraguay (Balansa n. 402, h. Kew) ; in ripis fl. Uruguay (For n. 249, h. Kew). Not easily distinguishable from C. flavescens, Linn., unless by the cells of the surface of the nut. C. Elliottianus, Roem. et Sch., founded on one very poor specimen, is referred (doubtfully) by Nees and Boeckeler to C. flavescens ; but itis described as having two, not three, bracts ; C. Olfersianus has often two bracts only, while C. flavescens has very rarely less than three. 21. C. Borvint (Boeck. in Linnea, xxxv. p. 481); umbellis (in exemplis visis) 3-1-radiatis, subcapitatis; spiculis paucis, fasci- culatis, compressis; glumis ovatis, minutissime mucronatis, in lateribus nitidis, purpureis; nuce ellipsoidea, compressa, cellulis extimis quadratis.— Bojer MS. in h. Kew. C. atropurpureus, Bojer MS. in h. Kew. Nux C. polystachyi. Stamina3. Spicule purpureo-sanguines lis C. rotundi similes. Madagascar (Blackburn, h. Kew). Bourbon (Balfour, h. Kew). 22. C. Munprit (Kunth, Enum. ii. p. 17) ; culmo decumbente aut sarmentoso, a vaginis foliorum alte intecto ; umbella simplice aut composita, bracteis breviusculis ; spiculis gubdigitatis, densi- floris, castaneo-rubris aut brunneis; nuce anguste ellipsoidea. —Boeck. in Linnea, xxxv. p. 448, in Flora, 1879, p. 545. C. densifolius, Steud.! Cyp. p. 6. C. cruentus, Baker! Fl. Maurit. p. 408, non Retz. C. reptans, Bojer ! MS. C. turfosus, Salzm.! MS. . C. Eragrostis, Willk. et Lange, Fl. Hisp. i. p. 138 (2), non Vahl. Pycreus Mundtii, Nees in Linnea, ix. p. 283, x. p. 131. Species Cypero Eragrostis, Vahl, simillima; differt nuce multo angustiore. Africa, fere tota, cum Madagascar. Tangier (Salzmann, hh. Kew, Calcutta). 64. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Libysche Wüste (Ascherson n. 530, h. Kew). Apud rivos White Nile et Gazelle (Schweinfurth nn. 1116, 1119, 5. Kew). Cape Town (Rehmann n. 1779, h. Mus. Brit.). Natal (Krauss n. 415, h. Kew). Delagoa Bay (Forbes, h. Kew). Madagascar (Grey, h. Kew; Blackburn, h. Kew; Thompson, h. Mus. Brit.). Var. f. distichophylla; culmis magis elongatis, debilibus, usque ad 24 folia proferentibus. C. distichophyllus (sp.), Steud. in Schimper n. 745, Cyp. p. 11; Boeck. in Linnea, xxxv. p. 488. C. Eragrostis, A. Rich. Fl. Abyss. ii. p. 475, non Vahl. Nux omnino C. Mundtiityp. Spicule modo indurate (Schimper nn. 745, 760), modo tenuiores (Schimper n. 344). Abyssinia (Schimper nn. 944, 745, 760, hh. Kew, Mus. Brit.). 23. C. ana MMICUS (Kunze; Kunth, Enum. ii. p. 6) ; umbella congesta; glumis remotiusculis, pallide fuscis aut rubescentibus ; styli ramis e gluma longiuscule exsertis.—0C. Gay, Hist. Chile, vi. p. 162, t. 70. fig. 1; Boeck. in Linnea, xxxv. p. 449. C. gramineus et C. parvulus (Steud. Cyp. pp. 4, 5). [Cyperus Ranko, Steud. ! Cyp. p. 315, est Isolepis. ] Stamina 2. Nucis cellule extime quadrate, porose. Chili: Osorno (Lechler n. 264, h. Kew); Valdivia flumen (Bridges n. 888, h. Kew); Antuco (Poeppig n. 18, h. Mus. Brit.). 24. C. ARGENTINUS; foliis plurimis, culmos longe superanti- bus; umbella simplice ; spiculis spicatis, latiusculis, castaneo- rubris aut colore eluto fere stramineis ; nuce obovoidea, turgida, cellulis extimis quadratis. Radix fibrosa. Culmi 10-15 cm. Involucri bracteæ 3-4, usque ad 1-2 dm. longs. Spicule longs 8 mm., late 3 mm., multum compresse, 8-14-flore. Glume obtuse, in dorso viridi-3-5-nervis. Stamina 2. Nux cum 3 parte glume subaequilonga. Respublica Argentina (Lorentz n. 1075, h. Kew). It is possible that this is the species described as C. Lozentzia- nus by Boeckeler (in Linnea, xxxv. p. 357); but the Lorentz number does not agree, nor does the description, satisfactorily. 25. C. LaavxETTO (Steud.! Cyp. p. 5); umbelle radiis 1-3, admodum inzqualibus; spiculis paucis, subspicatis, linearibus, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 65 fusco-castaneis; nuce obovoidea, compressa, cellulis extimis quadratis. C. melanostachyus, Palmer in h. Parry n. 904, non H. B. K. Stamina 2. Nux 3 parte glame brevior (non rugulosa), nigra cellulis extimis albidis, sepe porosis.—Differt a O. Olfersiano glumis (castaneis nec luteis) nuce magis obovoidea. 26. C. DIANDRUS (Torrey in Ann. Lyceum New York, iii. p. 251, var. excl.); umbella contracta ; spiculis lineari-oblongis, fusco- brunneis; glumis laxiuseule imbricatis; stylo brevi, ramis lon- gissime exsertis; nuce ellipsoidea, compressa.— Kunth, Enum. ii. p. 6; Boeck. in Linnea, xxxv. p. 447. Species a pluribus auctoribus cum C. rivulari conjuncta ; aquo differt nonnisi “ spicis comosis,” t. e. stylorum ramis e gluma lon- gissime exsertis. America borealis, in oris occidentalibus: Canada (Macoun, h. Kew); New Jersey (Asa Gray, h. Carey in h. Kew); Phila- delphia (Torrey, h. Kew); Louisiana (Torrey, h. Mus. Brit.). 27. C. RIVULARIS (Kunth, Enum. ii. p. 6); umbella simplice ; spiculis lineari-oblongis, obtusiusculis, brunneis aut subcastaneis ; glumis subrigide chartaceis, ovatis, obtusis ; stylo brevi, ramis e gluma breviter exsertis.— Boeck. in Linnea, xxxv. p. 452. C. diandrus, 8. castaneus, Torrey in Ann. Lyceum New York, iii. p. 252 ; Ser. Watson, Bot. California, ii. p. 214. C. flavescens, 8. castaneus, Pursh. Fl. Am. Sept. i. p. 52. C. Bigelovii, Steud. Cyp. p. 5. C. flavescens, Linn. ! in h. propr.; Boott | MS. in h. Mus. Brit. C. tenellus, Presl, Rel. Haenk. i. p. 176. Species cum C. flavescente maxime confusa, ab hac ope cellularum extimarum nucis quadratarum riter internota ; a C. Olfersiano quoad colorem diversa; hae autem nota in varietatibus elabitur. Ameriea borealis temperata; sat vulgaris: Canada (Macoun, h. Kew; Goldie, h. Calcutta); Tennessee (Rugel, h. Mus. Brit.) ; S. Louis (Drummond, h. Kew); Washington (Cooley n. 81, 4. Carey in h. Kew); California (fide Sereno Watson). Var. f. eluta ; spiculis majoribus, colore plus minus eluto, inter- dum fere stramineis. Pennsylvania (Moser, h. Kew ; Walton, h. Calcutta, d'c-). Var. y. depauperata ; culmo 3-1 spiculas proferente. C. diandrus, var., Carey, MS. in h. propr. Michigan : prope S. Josephi (4. Carey in h. Kew). LINN, JOURN.— BOTANY, VOL. XXI. F 66 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Var. è. acutata ; spiculis lanceolatis subacutis. Mexico (Schmitz in h. Auerswald n. 371, h. Mus. Brit. ; Ber- landier n. 871, h. Mus. Brit.). 28. C. nANCEUS (Thunb. Prod. p. 18; Fl. Cap. i. p. 388) ; sto- lonibus crassis; umbella specie simplice ; spiculis fasciculatis ; glumis obtusis, castaneis aut brunneis in carina viridibus ; nuce obovoidea, compressa, cum 1 parte glume vix equilonga.— Kunth, Enum. ii. p. 8 (var. B excl. ?). C. melanopus, Boeck. ! in Flora, 1879, p. 545. C. nitidus, Boeck. in Linnea, xxxv. p. 461 partim (i. e. plantis et syn. Asiaticis excl. ), nec Lam. Kunthii var. B. macrostachyam, Boeckeler (in Linnea, xxxv. p. 462), ad suum C. macranthum ducit, cujus nux (cellulis exti- mis oblongis) a nuce C. lancei longe distat. Exempla in h. Kew, sub nomine (3. macrostachys, Kunth, conservata, sunt C. lanceus, Thunb. Africa australis (Burchell nn. 55, 676, 6862, h. Kew); Uiten- hage (Zeyher nn. 634, 715, h. Kew); Mons Table (Ecklon n. 885, hh. Kew, Mus. Brit); Natal (Krauss n. 205, h. Kew). Africa centralis: Flumen White Nile (Schweinfurth n. 1219, h. Kew ; &c.). Madagascar (Parker n. 12, h. Kew; Baron n. 882, h. Kew). Var. ? B. Grantii ; glaucescens, involucri bracteis elongatis. C. nitidus, Oliver in Trans. Linn. Soc. xxix. p. 164. Mihi videtur C. globoso, var. nilagirico affinior, verisimiliter species nova; sed exemplum nimis juvene; styli rami an sint 2 vel 3 discernere nequivi. Uniyambene, lat. austral. 6° (Grant, h. Kew). 29. C. ATRONITENS (Hochst. in h. Schimper n. 312); umbella capitata, pauciradiata vel evoluta composita; glumis brevibus, ovatis, obtusis, atronitentibus in dorso viridibus; nuce ellip- soidea, compressa, cum $ parte glume zquilonga.— 4. Rich. Fl. Abyss. ii. p. 476; Boeck. in Linnea, xxxv. p. 456. A. melanocephalus, R. Br.! in Salt Abyss., App. iv. p. 62, nomen | nudum. C. elegantulus, Steud. in Flora, 1842, p. 583. C. ischnocornus, Steud. Cyp. p. 11. Umbella sepius plus minus contracta; in Schimper n. 505, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 67 radii plures, usque ad 6 cm. longi, apice umbellulati, spiculis numerosis. Africa; late dispersa: Abyssinia (Schimper nn. 25, 812, 505, 574, 845, 1245, hh. Kew, Mus. Brit. ; Rohr, h. Kew; Dillon et Petit, h. Kew); Fernando Po (Mann n. 1470, h. Kew); Natal (Buchanan nn. 50, 807, h. Kew). 30. C. ArRoNERVA TUS (Boeck. in Linnea, xxxv. p. 858); cum elongato a vaginis foliorum alte intecto; umbella subcapitata ; glumis longis, viridibus, per totam fere latitudinem longitudi- naliter atronervatis. Stamina 3.—Cypero Mundtii affinis. Abyssinia (Schimper nn. 1244, 1287, hh. Kew, Mus. Brit.). 31. C. curysantuus (Boeck. in Linnea, xxxv. p. 476) ; um- bella simplice; spiculis breviter spicatis, lineari-oblongis, luteis brunnescentibus ; glumis ovatis, sub apice mucronulatis; nuce anguste ellipsoidea, modice compressa, 2 partem glume super- ante, cellulis extimis quadratis. Africa australis (Drége, h. Kew; h. Harvey n. 80, h. Kew); Natal (Grant, h. Calcutta). C. permutatus, Boeck. in Linnea, xxxv. p. 476, which is C. lan- ceus y? mucronatus, Kunth, is said by Boeckeler to differ in the minute nut (not one fourth the length of the glume). So far as my memory serves, I believe that I considered it — C. chrysanthus ; but I have no note of it, and can verify nothing here. 32. C. metanostacuyus (H. B. K. Nov. Gen. et Sp. i. p. 207); umbelle radiis 1—4, brevibus aut subnullis; glumis ovatis, ob- tusis, castaneis in dorso viridibus; nuce cum 3 parte glume equilonga, ellipsoidea, subobtusa.— Kunth, Enum. ii. p. 10 (planta domingense excl); Liebm. Mexic. Halfgr. pp. 13, 14; Boeck. in Linnea, xxxv. p. 455. C. variegatus, H. B. K. Nov. Gen. et Sp. i. p. 208. . C. adustus, Presl, Rel. Haenk. i. p. 167 ; Kunth, Enum. n. p. 10. America calidior, alt. 1500-3000 metr.: Mexico (Coulter n. 1613, À. Kew; Bourgeau n. 431, h. Kew; F. Mueller n. 1985, h. Kew); Neo-Granada (Holton n. 121, hh. Kew, Calcutta); Ecuador (Spruce n. 5905, hh. Kew, Mus. Brit ° Calcutta), prope Quito (Jameson n. 270, h. Kew); Bolivia, alt. 3100 metr. (Mandon n. 1400, h. Kew). 38. C. cimicinus (Presl, Rel. Haenk. i. p. 166); um bella con- , 68 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. tracta; glumis parvis, ovatis obtusis, nigro-castaneis in carina viridibus; nuce cum 3 parte glume equilonga, ellipsoidea, utrinque angustata.— Boeck. in Linnea, xxxv. p. 454. A C. melanostachyo recedit glumis parvis admodum obtusis: a C. atronitente vix (nonnisi patria) differt. Neo-Granada (Purdie, h. Kew); Mexico, in convalle metropo- litana (Schmitz, h. Mus. Brit.). 34. C. MEGAPOTAMICUS (Kunth, Enum. ii. p.10, syn. excl); culmis elongatis aut filiformibus, monocephalis; spiculis parvis, lineari-oblongis, fuseis aut subrubescentibus; glumis ovatis ob- tusis ; nuce parvula, ellipsoidea.— Boeck. in Linnea, xxxv. p. 499. Pycreus megapotamieus, Nees in Mart. Brasil., Cyp. p. 6. Species a C. Olfersiano parum remota. Brasilia (Sellow, h. Kew); Pampas (Tweedie n. 615, k. Kew); Paraguay (Madon n. 672, h. Kew). Tt Umbella plane composita, umbellularum radiolis plus minus evolutis. i Species in Indiá Orientali indigene. 95. C. puncricunatus (Vahl, Enum. ii. p. 348); umbella com- posita; involueri bracteis longis, umbellarum partialium sub- nullis ; spiculis majuseulis, laxe spicatis, lineari-oblongis, admo- dum compressis, rubro-stramineis; nuce majuscula, obovoidea, multum compressa.— Kunth, Enum. ii. p. 15; Dalz. and Gibs. Bomb. Fl. p. 283; Thwaites, Enum. Pl. Zeyl. p. 842; Boeck. in Linnea, xxxv. p. 472 (synon. Roxburghianis ex omnibus auctori- bus delendis). C. inundatus, Nees in Wight Contrib. p. 76, non Roxb. C. procerus, Roxb. Fl. Ind. i. p. 203 partim, Ic. ined. in h. Calcutta. Cyperus (innominatus), Roxb. Ic. ined. n. 725 in h. Kew. Cyperus, Wail.! List n. 3336 A, C partim, D, F. Cyperus, Wall.! List n. 3355 B. Radix fibrosa, annua. Culmus solitarius, 3-8 dm., erectus, basi crassiusculus, sursum triqueter, levis. Folia 3-5, cum culmo sepe equilonga, lata 8-10 mm., carinata, levia. Bractes 3-5, usque ad 3—4 dm. longs, patule, foliis similes. Umbelle radii 4-10, longi 5-15 em.; ochrew 10-15 mm., truncate, 1-dentate ; umbelle partiales corymbosim 3-4-radiatz, bracteis nullis aut minimis; spice longe 3-5 cem., 4-18-spiculose, rhachi glabra: Spicule longe 1-2 cm., late 4 mm., 7-30-flore ; rhachilla fere Í MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 69 exalata; gluma ima parvula, oblonga, braetemformis. Glumæ leviter imbricatz, ovate, obtusa, compressz, in dorso virides sub- 3-nervie, in lateribus latis enervose, lucide. Stamina 2 vel 3 ; anther lineari-oblonge, muticæ, rubra. Stylus nuce brevior ; rami 2, e gluma breviter exserti. Nux cum 4 parte glume vix equilonga, truncata, nigra; cellule extime emarcide, hyalina, non porose.—C. inundatus, Roxb. in Sect. Juncello, C. procerus, Rottb. in Sect. Lucypero ordinati sunt. India australis: Madras Peninsula (Wallich n. 3336 litt. A, D, F, 7. propr.; Wight n. 2871, h. Calcutta; Wight n. 1818, h. Kew). Zeylania (Thwaites C. P.n. 3316, hh. Kew, Mus. Brit. ; Thwaites, C. P. n. 3751, hh, Calcutta, Mus. Brit.; Beckett, h. Kew). Var. B. quinquagintiflora; spiculis elongatis, usque ad 50-floris ; antheris plane apiculatis. C. macrostachyus, Vahl, Enum. ii. p. 349 partim, non Lam. Madras Peninsula (Wallich n. 3336, litt. B, A. propr.). 36. C. HocusrETTERI (Nees; Krauss in Flora, 1845, p. 755, in nota); umbella composita; involucri bracteis longis, umbellula- rum sepe lineari-caudatis ; spiculis magnis, laxius spicatis, nitidis ; glumis obtusis, fructus tempore vix imbricatis ; nuce cum i parte glume equilonga, compressa, late ellipsoidea aut obovoidea.— Steud. Cyp. p. 11; Boeck. in Linnea, xxxv. p. 471. C. retusus, A. Rich. Fl. Abyss. ii. p. 485, non Nees. C. albomarginatus, Steud. Cyp. p. 10. C. flavicomus, Torrey, Bot. U.S. & Mex. Bound. p. 226; Benth.! Fl. Austral. vii. p. 261, non Mich. C. patuliflorus, Boeck. ! in Linnea, xxxv. p. 473. 7 C. ornatissimus, F. Mueil.! Fragm. Phyt. Austral. viii. p. 265. Pyereus albo-marginatus, Nees in Mart. Brasil., Cyp. p. 9, Glabra, annua, radicibus fibrosis. Culmus solitarius, 5-8 dm., erectus, sursum triqueter, levis. Folia plura, cum 3 parte cu m sæpe æquilonga, lata 1 cm., minute scabra, nis: summum sepe li vaginas fere reducta. Bractee 4-5, usque ad 2-3 dm. longs, foliis similes. Umbella 1-3 dm. in diam., nunquam decompo- sita; radii 4-12, usque ad 1-2 dm. longi; ochrese 1-2 in trun- catæ, fuscæ. Umbellule 1-7-radiatæ; bracteolæ sepe 1- em ; radioli 1-3 em. Spice usque ad 3-6 em. longs, NN Spieule longs 2 cm., late 3-6 mm., distantes, patulæ, " > flore ; rhachilla fere exalata ; gluma ima parva, oblonga, obovo dea; bracteola lineari-subulata glumam imam fertilem multum 70 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS, superans. Glume longs 3 mm., late elliptice, compresse vix carinate, in dorso 5—7-nervia, in lateribus luteæ aut brunnes aut rubescentes, in marginibus conspicue lucide hyaline. Stamina 3 vel 2; anthere lineari-oblonge, mutice. Stylus nuce multo bre- vior; rami 2, e gluma plane exserti. Nux obtusa, magna (quam C. flavicomi 3-4plo gravior) nigra; cellule extime quadrata, emarcide, albido-hyaline, porose.—Forma typica est Kotschy n. 52, parva, umbella 4—5-radiata. Exemplum majus, Schimper n. 1199, spiculis eastaneo-brunneis gaudet. Schweinfurth n. 2018 est C. Hochstetteri, var. tenuis, Boeck. in Flora, 1879, p. 546; exemplum in h. Kew est juvene, minime tenue. Forma “ orna- tissima," sp. F. Mueller, umbella lata, spiculis luteo-hyalinis nitida est; cum hac exempla indica congruunt. Schweinfurth n. 2082 est C. patuliflorus, Boeck. (teste auctore ipso), sed cum forma vulgari C. Hochstetteri omnino congruit. Forma “ albo- marginata," sp. Nees, in America meridionali indigena est; differt a formis gerontogeis nervo medio glume sub apice vix ac ne vix excurrente. Africa, India, Australia, America australis, et Mexico, in regione tropicali. Africa: Senegambia (Roger, h. Kew); Djur et Matamma (Schweinfurth nn. 2012, 2082,.2018, 1939 pro maxima parte, h. Kew; Schweinfurth n. 2012, h. Mus. Brit); Nubia (Schimper n. 52, hh. Kew, Mus. Brit.); Abyssinia (Schimper n. 1199, hh. Kew, Mus. Brit.). India: Concan et Malabar (G. Thomson, hh. Kew, Mus. Brit. ; Stocks, h. Calcutta ; Beddome, h. propr.). Burma: Rangoon (R. Scott, Kurz, h. Calcutta). Australia borealis: Sturt’s Creek (F. Mueller, h. Kew); Arn- heim Land (.F. Mueller, h. Kew). America meridionalis: Ecuador, in ora maris Pacifici (Spruce n. 6417, hh. Kew, Calcutta). Mexico (Dugés n. 1551, h. Kew, spiculis lete rubris); New Mexico (Wright n. 1965, h. Kew). Var. f. pinguior; in omnibus partibus pinguior; caule in- crassato, vaginis magnis, laxis ; umbella densa; spiculis magnis, numerosis, adgregatis; glumis usque ad 4—5 mm. longis. Quamvis umbella, glume nucesque iis C. Hochstetteri typici ali- quanto majores videntur, revera tamen arcte congruunt: spe- cies distincta esse vix potest. Abyssinia (Schimper n. 1551, hh. Kew, Mus. Brit.). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 71 Var. y. russa; glumis russis (rubro-castaneis), paullo minus remotis (quoad cetera cum Schimper n. 1199 congruens). C. expansus, Bojer, Hort. Maurit. p. 380, non Poir. C. tremulus, Baker, Fl. Maurit. p. 409 partim, vir Poir. Mauritius (Grey, h. Kew). Madagasear (Blackburn, h. Kew ; Hilsenberg et Bojer, h. Mus. Brit.). ££ Species in Indiá Orientali nondum reperte. 97. C. ruavicomus (Vahl, Enum. ii. p. 360 partim, non Mich.); umbella composita; involucri bracteis longis, umbellularum sepe lineari-caudatis; spiculis spicatis, lineari-oblongis, lutescenti- brunneis; glumis admodum obtusis, hyalino-marginatis ; nuce ellipsoidea, compressa, nigra, cellulis extimis quadratis non po- rosis.— Torrey in Ann. Lyceum New York, ii. p. 253; Kunth, Enum. ii. p. 15; Boeck. in Linnea, xxxv. p. 470. C. elegans, Walt. Fl. Carol. p. 70, non Linn. C. flavicomus, Mich. Fl. Bor.-Am. i. p. 27, “spiculis 7-floris ” (Pursh. Fl. i. p. 53), est C. strigosus, Linn.—C. Slavicomus, Vahl, partim e C. strigoso, partim e C. flavicomo nostro, descriptus est. ' America borealis: a Pennsylvania usque ad Carolina (Elliott, h. Kew; Hale, h. Carey in h. Kew); Carolina australis (Curtiss, h. Carey in h. Kew). uu America australis (Burchell n. 8567, h. Kew); Brasil, Piauhy (Gardner n. 2384, h. Kew). 38. C. mREMULUS (Poir. Encyc. vii. p. 264); umbella decom- posita; involucri bracteis longis, umbellularum foliaceis ; spiculis Spicatis, lineari-oblongis, fusco-luteis ; glumis obtusis, albo-mar- ginatis; nuce oblonga, obtusa, cellulis extimis quadratis, porosis. —Kunth, Enum. ii. p.16 ; Boeck. in Linnea, xxxv. p. 469 ; Baker, Fl. Maurit. p. 409 partim. C. retusus, Nees; Kunth, Enum. ii. p. 115. Pycreus retusus, Nees in Linnea, ix. p. 285. l un Differt a C. Hochstetteri spiculis, floribus nucibusque minoribus, a C. flavicomo nucis cellulis porosis, ab utroque umbella decom- posita, nec semel composita. l Mauritius (Sieber n. 7, hh. Kew, Mus. Brit.; &c.). Subgenus 3. JUNCELLUS. Juncellus, Griseb. Fl. Brit. W. Ind. p. 502; Bent- J. Gen. Pl. iii. p. 1044 partim.—Pycrei sectio, Kunth, et Hook. Enum. ii. 72 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. p. 17.—Dichostylis, Nees in Linnea, ix. p. 289.—Pseudopycreus, Boeck. in Linnea, xxxv. p. 485 partim. Stylus 2-fidus ; rami lineares. Nux compressa, facies rhachille adpressa. Spicule multi-(10—60-)florz, persistentes. Rhachille ale anguste, persistentes. Filamenta 3-1; anthere cristate aut mutice. Glabrz, foliaceæ. Inflorescentia varia. This is Bentham's Juncellus without Anosporum. Boeckeler has included in his Pseudopyereus, Cyperus stoloniferus, which with me has a 3-fid style; and he has put C. pygmeus in Eucy- perus, Sect. Aristati, an arrangement to which he must have been led by the inflorescence ; but which seems to me very unnatural, and opposed to his definition of the group. The species here in- ciuded in Juncellus exhibit great variety in habit and inflorescence, but they are at all events easily discerned by the nut. The con- fusion I have got into about C. diaphanus is because the type specimen has the nut exceedingly young ; and without comparison I am not sure that my ripe Calcutta C. diaphanus is the same species. * Umbelle radii manifesti (cf. etiam C. diaphanum). 89. C. Mowrr (Linn. f. Suppl. p. 102); robusta; umbella composita, spiculis (sæpe dense) spieatis; glumis late ovatis, ob- tusis, in dorso subplanis late 7-9-nerviis ; nuce 4 partem glume superante.— Host, Gram. iv. p. 38, t. 67 ; Berl. Fl. Ital. i. p. 272; Reich. Ic. Fl. Germ. viii. t. 279; Sturm, Fl. Deutsch. xiii. t. 52; Ledeb. Fl. Ross. iv. p. 240; Gren. et Godr. Fl. Franç. iii. p. 361; Willk. et Lange, Fl. Hisp. i. p. 197 ; Parl. Fl. Ital. ii. p. 22; Boiss. Fl. Orient. v. p. 336. C. serotinus, Rottb. Descr. et Ic. p. 31; Kunth, Enum. ii. p.19; Boeck. in Linnea, xxxv. p. 492. C. longus, Linn. ! h. propr. C. puncticulatus, Aitch.! Punjab Pl. p. 189, in Journ. Linn. Soc. xix. p. 189, non Vahl. — Monti, Bonon. Prod. p. 12, t. 1. fig. 2. Stolones 2-4 mm. in diam., vaginati. Culmi solitarii, 3-10 dm., apice triquetri, leves. Folia 2-5, eum culmo sspe sequilonga, lata 7-10 mm., sursum antice scabra. Bractes 3-5, usque ad 2-5 dm. long, ereeto-patentes, foliis similes. Umbelle radii 3-6, usque ad 5-15 em. longi, admodum inequales, compressi, sursum triquetri ; ochreze 8-15 mm. truncate ; umbellule breviter corymbosi aut fere simplices ; radioli 3-10, breves, a bracteola | MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 73 | lineari-caudata suffulti, aut nudi; rhachis 5-30 mm., glabra aut | minute setuloso-scabra, nec pilosa. Spieule in unaquaque spica 5-40, undique spicate, longe 10-15 mm., late 4 mm., 12-30- flore, compresse turgidæ, lateribus parallelis; rhachilla tetra- gona, conspicue quadratim excavata, vix alata ; gluma ima ovato- lanceolata, interdum seti-caudata, ceteris longior. Glume non | carinatæ, rubræ, fuscæ aut pallescentes, per 4-2 latitudinem a | nervis striate ; margines anguste scarioso-hyalinw. Filamenta 3, anguste ligulata, levia, persistentia; anthere lineari-oblongse, mutice (obsolete cristatw), rubre. Stylus brevis; rami 2, sæ- pissime e gluma vix exserti (in paucis exemplis longius exserti). Nux rhachi contrarie admodum compressa, paullo curvata, in facie interiore contra rhachin adpressa, subconcava, in facie ex- teriore convexa, minime 2-suleata (nisi immatura ut in iconibus pluribus depicta), nigro-brunnea; cellule extime quadrate, emarcide, hyaline, persistentes, non porose, obscure. Ab Europa australi usque ad Himalayam, China et Amurland. Europa (Reichenbach n. 2311, hh. Kew, Mus. Brit.) ; Corsica ' Mabille n. 187, hh. Kew, Mus. Brit.); Lyon (Billot n. 86, h. Mus. Brit.; Bourgeau n. 32, h. Mus. Brit.); Tuscany (Billot n. 86 bis, h. Mus. Brit.); Verona (Reichenbach, h. Mus. Brit.) ; Carniola (Schwegrichen, h. Calcutta); Croatia (Schlosser n. 857, h. Mus. Brit.); Hungary (Frivaldsky, h. Mus. Brit.). Asia occidentalis: Caucasus; Astrakhan; Cabul (Griffith n. 6188 partim, k. Kew); Punjab (T. Thomson n. 1589, h. Kew; Aitchison n. 494, hh. Mus. Brit., Calcutta) ; Kashmir (Jacquemont n. 1138, h. Kew; W. S. Atkinson n. 24199, h. Calcutta); Chumba, alt. 1000 metr. (C. B. Olarke n. 24276, hh. Kew, Calcutta). Asia orientalis: Amurland, Kengka Lacus (Maack, h. Kew); China, Pekin (Hance n. 6531, h. Kew; Maaimowicz n. 484, À. Kew); Shantung (Maingay n. 157, hh. Kew, Calcutta). Var. f). (?) stylosa ; spiculis castaneo-rubris, a styli ramis longis, longissime exsertis, comosis. India Orientalis: Montes Khasia (Griffith, h. Kew). 40. C. rNuNDATUS (Roxb. Hort. Beng. p. 6, Fl. Ind. i. p. 201, Ic. ined. t. 110 in h. Kew, nec Nees, nec R. Br.) 3 robusta ; um- bella decomposita, majuscula, patente ; spiculis spicatis, distanti- bus, majusculis, compressis, linearibus ; glumis obtusis, late ellip- 74 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. ticis, in dorso subplanis conspicue striato-nervosis; nuce ma- juscula, rotundato-ellipsoidea, compressa. Cyperus, Wall. ! List n. 3342, B. Culmus 5-8 dm., solitarius, erectus, erassus, basi compressus, sursum acute triqueter. Folia 3-5, cum 2$ parte culmi sepe eequilonga, lata 8-14 mm., carinata, in marginibus parum scabra. Bracteæ 3-6, usque ad 5 dm. longs, 2 cm. late, foliis similes, basi auriculato-saccate, patentes. Umbella 3-5 dm. in diam., densa; radii c. 7, usque ad 5-20 em. longi; umbellule corym- bose iterum corymbosim subramose; ochree usque ad 2 cm. longe, fusce, truncate, obscure bilabiate ; umbellularum bractee lineari-caudate. Spice 2-4 em. longs, 10-16-spiculose ; rhachis glabra. Spicule rectangulatim divaricatz, usque ad 2 em. longs, 3-4 mm. late, 16-36-flore, fusce ; rhachilla subtetragona, robusta, vix alata; gluma ima ovato-lanceolata, interdum caudata, ceeteris longior. Glume spisse, majuscule, late, in marginibus hyaline ; margines in sicco sepe involute. Stamina 3; filamenta anguste ligulata, levia; anthers lineari-oblonge, mutice (obsolete cris- tate), flavee, sanguineo-maculate. Stylus brevis ; rami 2, e gluma breviter exserti. Nux majuscula, late rotundata (neque basi angustata), contra rhachillam compressa, in faciebus subconcava, brunnea ; cellule extime quadrate, hyaline, persistentes, obscure. —Species C. procero ab externo similis, cum hoe intermixta; Cypero Monti e visceribus simillima, ab hoc nonnisi inflorescentia laxiore vix distinguenda.—In ripis fluminis Ganges per æstus submersa (Roxburgh) ; in ripis luminis Woosung per estus sub- mersa (notula Maingayi). From Boeckeler's index it appears that his final treatment of C. inundatus, Roxb., is to be found on p. 505, which, after hours of hunting, I have never been able to find. Bengalia: Calcutta ( Buchanan- Hamilton, h. Wallich n. 3342 B in h. Linn. Soc. ; Kurz, h. Calcutta); in lacubus Jheels dictis (T. Anderson, h. Calcutta; Se.); Burisaul (C. B. Clarke n. 8460, h. Kew). China: Shanghai (Maingay n. 639, h. Calcutta). 41. C. ALOPECUROIDES (Rottb. Descr. et Ic. p. 38, t. 8. fig. 2); umbella composita aut decomposita, densispiculosa; spicis cylin- dricis, undique densissime spiculigeris; glumis ovatis, breviter mucronatis aut lanceolatis; nuce parva, ellipsoidea, compressa, pallide brunnea aut straminea—R. Br. Prod. p. 217; Nees in MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 75 Wight Contrib. p. 765 Kunth, Enum. ii. p. 19; Webb et Berth. Iles Canaries, Phyt. iii. p.861; Dalz. and Gibs. Bomb. Fi. p. 282 ; Boeck. in Linnea, xxxvi. p. 322, var. a excl.; Benth. Fl. Austral. vii. p. 264; Boiss. Fl. Orient. v. p. 367, syn. Delile excl.; non Roxb. C. compositus, R. Br. Prod. 217 ( fide Benth.). Cyperus, sp., Wall. ! List n. 3344. Radix fibrosa. Culmus (interdum 2-3) robustus, usque ad 1 metr. longus, sursum trigonus, levis. Folia 2-4, longa, interdum cum culmo equilonga, latiuscula, coriacea, in marginibus (sspe in carina quasi-alata) aculeato-scabra, rarius (in forma australi) levia. Umbella magna, spiculis innumerosis ; involucri bractex plures, usque ad 3-7 dm. longs, foliis similes. Umbelle radii 5-10, usque ad 5-10 em. longi; ochreæ 1-3 cm., a dentibus 2 longis lanceolatis (interdum longis 1-2 cm. foliaceis) terminate. Umbellule a bracteis foliaceis sepe suffultæ, haud raro iterum umbellatim divise. Spice plures, sessiles, longc 1-5 cm., cylin- driez. Spicule rectangulatim patentes, longe 8 mm., late 2-3 mm., lanceolate#, compresse, subtetragone, 12-24-flore ; gluma ima ovato-caudata, ceteris longior, persistens ; rhachilla non alata, pro magnitudine spicularum robusta, tetragona, pro- ventu alte excavata. Glume spisse, in sicco laxius imbricate, marginibus sepe involutis, in dorso plane, ecarinatz, pallide aut luteo-brunnee. Stamina 3; filamenta ligulata, levia; an- there lineari-oblongs, mutice. Stylus cum nuce equilongus ; rami 2, e gluma mediocriter exserti. Nux cum 3 parte glume equilonga, ellipsoidea, utrinque sequaliter angustata, facie inte- riore plana, exteriore leviter convexa, sepius (imperfecte matu- rata) pallida; matura griseo-brunnea.— Species a Boeckeler (et a Benth. et Hook. f. Gen. Pl. iii. p. 1044), cum C. divite, Delile, conjuncta, ideoque a Benth. et Hook. f. ad subgenus Juncellum, a Boeck. ad subgenus Eucyperum (sect. Exaltatos) relata. Mihi videntur 2 species inter se remote, viz. :— a. alopecuroides; stylo 2-fido; nuce compressa; rhachilla robuste tetragona, alte excavata, subexalata (Juncellus). B. dives; stylo 3-fido; nuce plane trigona; rhachilla tenui compressa, hyalino-alata (Eucyperus : Exaltati). Exemplum australiense a ceteris omnibus recedit spicis ovoideis, glumis minus dense stipatis, staminibus 2, antheris magnis. In exemplis africanis, glume late ovate aut obovate 76 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. acumine brevi; in exemplis Indicis glume ovato-lanceolate con- spicue plicato-striate videntur. Ins. Madeira ( Lowe, hh. Kew, Mus. Brit.). Ins. Teneriffe (Bourgeau n. 462, h. Kew). Ins. Virides ( Barker- Webb n. 100, k. Kew). Senegambia (Brunner, hh. Kew, Mus. Brit.). Egypt: Alexandria (Letourneuz n. 147, non n. 147 bis, h. Kew). Kordofan (Kotschy, h. Kew). Ins. Seychelles (Horne n. 221, h. Kew). India Orientalis: Punjab (T. Thomson, h. Calcutta); Peshawur (Griffith n. 199 a, h. Kew n. 6150); Moradabad (T. Thomson n. 347, h. Kew); Scinde (Stocks, k. Kew); Bombay (Jacque- mont nn. 422, 438, k. Kew); Madras (Heyne, h. Mus. Brit.; Koenig, h. Mus. Brit. ; Wight n. 2877, h. Calcutta). Zeylania (Thwaites C. P.n. 3560, hh. Kew, Calcutta). [India Occidentalis: Ins. Guadaloupe (Husnot n. 2, hh. Kew, Mus. Brit.) an allata ?.] The young specimens of this are very like those of C. dives, and I am not surprised that those who depend on hand-and-eye sorting, and have not looked at the style, have mixed them. In the mature C. alopecuroides the glumes fall early from the lower part of the spikelet ; the stout tetragonous deeply-notched rhachilla of Juncellus is then conspicuous, and the examples are in this state easily sorted out by hand and eye. When I named the English material, I had not learnt the value of this character ; but, as I looked at the stylein every specimen, I believe the above list of numbers and localities must be right. I should, however, add that I feel much less confident about the literature, which I have been unable to go over again out here; some of the quota- tions may belong, at least in part, to C. dives. 42. C. Barrert (Boeck. in Linnea, xxxv. p. 460); umbella simplice laxiuscula; spiculis paucis, fasciculatis, compressis, inflatis, albidis; glumis ovatis, obtusis, concavis, viridi-striatis ; nuce magna, contra rhachin compressa, obovoidea, in facie inte- riore concava, exteriore curvata, convexa. Niger flumen (Barter n. 1563, h. Kew); Nupe (Barter, h. Calcutta). Djur (Schweinfurth n. 1939, partim admixt., k. Kew). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 77 ** Umbella capitata. 43. C. nzviaATUS (Linn.! Mant. p. 79); rhizomate repente ; umbella capitata, specie laterali; involucri bracteis 2, altero quasi culmum producente; glumis spissis, concavis, obtusis, albis, pallidis aut castaneo-maculatis ; nuce obovoidea aut ellip- soidea, obtusa, cum 2 parte glume æquilonga.—Rottb. Descr. et Ic. p. 19, t. 16. fig. 1; Boeck. in Linnea, xxxv. p. 486 ; Benth. FI. Austral. vii. p. 263; Cosson, Expl. Algér., Glum. p. 250; Boiss. Fl. orient. v. p. 366; Ser. Watson, Bot. California, ii. p. 214. C. mucronatus, Rottb. Descr. et Ic. p. 19, t. 8. fig. 4; H. B. K. Nov. Gen. et Sp. i. p. 203; Roxb. Fl. Ind. i. p. 185; Nees in Wight Contrib. p. 72; Kunth, Enum. ii. p. 17; Webb et Berth. Iles Canaries, Phyt.ii. p. 360; Parl. Fl. Ital. ii. p. 21; C. Gay, Hist. Chile, vi. p. 163; Benth. Pl. Hartw. p. 27 ; Liebm. Mex. Halfgr. p. 15; Griseb. Fl. Brit. W. Ind. y. 362. C. lateralis, Forsk. Fl. Æg. Arab. p. 13; Roxb. Fl. Ind. i. p. 186, Ic. ined. t. 2014 in h. Kew; Nees in Wight Contrib. p. 73. C. pleuranthus, Nees in Wight Contrib. p. 73. C. leucostachys, Willd. ; Link, Jahrb. iii. p. 81. . monostachyus, Link! in Buch. Ins. Canaries, p. 138, non Rottb. . acuminatus, Rozb. h. propr., Ic. ined. t. 704 in h. Kew. . Cossyrensis, Tineo in Guss. Fl. Sicul. ii. p. 779. . viridulus e£ C. reptans, Boeck. in Linnea, xxxv. p. 485. . rivularis, Steud. in Lechler n. 1547, non Kunth. Pycreus mucronatus e£ P. lateralis, Nees in Linnea, ix. p. 283. P. lzvigatus, Nees in Linnea, x. p. 130. Cyperus, Wail.! List n. 3311. Rhizoma modo breve, culmis uniseriatim adgregatis, modo elongatum, a vaginis Jaxis castaneis ornatum, culmos distantes proferens. Culmi 1-5 dm., tenues aut crassiusculi, sursum tri- goni, leves ; vagine 2-3, longiuscule, truncata, basin arcte inte- gentes. Folia interdum cum culmo aquilonga, straminea, involuta, sepius brevia, haud raro omnia ad muerones vagina- rum fere reducta. Involucri bractex 2, altera erecta, 2-10 em. longa, foliis similis, basi interdum dilatata, altera multo brevior, patens. Spicule dense congestee, 10-30, aut in formis debilibus vel alpinis 10-1, long 1 cm., late 2 mm., 16-40-flore, compress, turgide. Glumz dense imbricate, albide, ssepe plus minus cas- taneo-maculatæ, in dorso pallidæ aut fuscæ, raro virides ; nervi 7-11 per 3 latitudinem glumæ extensi; rhachilla tetragona, ex- cavata, non alata. Stamina 3; filamenta late ligulata, fere levia; antheræ oblongæ, flavæ, a crista lanceolata, rubra, scabra, termi- aaaaa 78 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. nate. Stylus cum nuce subequilongus ; rami 2, e gluma breviter exserti. Nux in forma typica cum $ parte glume equilonga, in forma macra africana 4 parte glume brevior, in exemplis par- vulis sæpe vix gluma brevior, compressa, in facie interiore plana aut concava, obtusa, brunnea; cellule extime quadrate, albido, persistentes.—Forma andina reptans (sp. Boeck.) differt spiculis angustis, castaneis.— Forma Andina viridula (sp. Boeck. = C. rivu- laris, Steud.) differt folis viridulis (nee stramineis) gramineis.— Forma africana sacra differt spiculis magnis (longis 15 mm. latis 3 mm.) interdum tortis, glumis magis acutatis. Both Menzies and Hillebrand note that the finest mats in the Sandwich Isles are made of this species. Europa meridionalis, Africa cum Insulis, Asia occidentalis usque ad Indiam, Australia, Polynesia, America australis, et Mexico.—Species cosmopolitana. Europa: Nice (Bentham, h. Calcutta) ; Sicilia (Tineo, h. Kew; Citarda, hh. Mus. Brit., Calcutta). Africa: Ins. Madeira (Lowe n. 77, h. Mus. Brit.) ; Ins. Canary (Bourgeau n. 1568, h. Kew; Lemann, h. Calcutta); Ins. S. Vincent (Webb nn. 99, 114, 138, k. Kew); Ins. Cap Verd (Vogel n. 64, h. Kew; Melliss n. 160, h. Kew) ; Mogadore (J. D. Hooker, h. Kew) ; Niger flumen (Vogel, h. Kew) ; Libya CAscherson n. 2308, h. Kew); Kosser (Schweinfurth n. 2009, hh. Kew, Mus. Brit.) ; Abyssinia, alt. 1000 metr. (Schimper n. 2160, hh. Kew, Mus. Brit.) ; Mauritius (Sieber n. 15, hh. Kew, Mus. Brit.) ; Ile de France (Du Petit-Thouars, h. Mus. Brit.); Ins. Rodriguez (Balfour, hh. Kew, Mus. Brit.) ; Ins. Socotra (Balfour n. 27, h. propr.); Arabia Petrza (Macdonald, h. Kew) ; Egypt (Aucher-Eloy n. 3796, h. Kew); Suez (Schweinfurth, h. Kew; §c.); Alexandria (Aucher-Eloy, h. Calcutta) ; Ins. Rhodes (Postian, n. 655 h. Kew). [Forma macra. Mauritius (Blackburn, h. Kew); in fontibus Nili fl. (Grant, h. Kew); Somali-land (Hildebrandt n. 1474, hh. Kew, Mus. Brit.); Zambesi-land (Meller, h. Kew) ; Natal (Gerrard n. 488, h. Kew); Cape-Town (Harvey n. 188, h. Kew; Rehmann n. 1786 h. Mus. Brit.).] Asia: Syria (Bové nn. 27,30, h. Kew); Cabul (Griffith n. 1265, h. Kew n. 6215, n. 6215 h. Calcutta). India Orientalis, in regione occidentali et Madras: Punjab (Jacquemont n. 133, h. Kew; T. Thomson, h. Calcutta); Kashmir, Bhimbur (Stewart nn. 799, 871, h. Calcutta); MR, C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 79 Campbellpore (Stewart n. 56, h. Calcutta) ; Marwar (G. King, h. Calcutta) ; Madras Peninsula (Wight n. 2389, h. Kew ; Heyne in Wallich n. 3311, h. Caleutta). [Forma macra. Australia: Swan River (Drummond n. 7, h. Kew, n. 325 h. Mus. Brit.) ; Portus Gregory (Oldfield, h. Kew).] Ins. Sandwich (Hildebrandt n. 521, h. Kew; Menzies, h. Mus. Brit.). Chili (C. Gay, h. Kew). Lima (Seemann, h. Calcutta). La Plata: Tucuman (Zweedie, h. Kew). Mexico (Hartweg n. 240, h. Kew; W. G. Wright, h. Kew; Schaffner, h. Kew). Ins. Jamaica (Macnab, h. Kew; Latrobe, h. Mus. Brit.). [Forma viridula. America australis: Peru, Arica (Lechler n. 1547, hh. Kew, Calcutta); Lima (Seemann n. 871, h. Kew); Banda Or. (Tweedie, h. Kew).] [Forma reptans. America tropicalis: Andes Ecuador (Spruce nn. 5910, 5911, hh. Kew, Calcutta) ; Mexico (Schmitz n. 806, h. Mus. Brit.).] Var. f. junciformis; capitulo 6-1-spiculigero ; spiculis sepe elongatis multifloris ; glumis atro-castaneis, viridi-carinatis ; nuce ellipsoidea aut ovoidea cum 3-3 parte glume squilonga. C. junciformis, Desf.! Fl. Atlant. i. p. 42, t. 7. fig. 2; Cav. Ic. iii. p. 2, t. 204. fig. 1; Reich. Ic. Fl. Germ. t. 278. fig. 661; Boss. Fl. Orient. v. p. 367. C. distachyos, Allioni, Fl. Pedem. Auctuarium, p. 48, t. 2. fig. 5; P arl. Fl. Ital. i. p. 19; Willk. et Lange, Fl. Hisp. 1. p. 137; Ball in Journ. Linn. Soc. xvi. p. 700. C. mucronatus, Sibth. Fl. Grec. i. p. 34, t. 49. C. levigatis, var. picta, Boeck. in Linnea, xxxv. p. 487. C. mucronatus, var. y, Kunth, Enum. ii. p. 18. Per regionem mediterraneam, ubique sparsa. l ; Europa australis: Lusitania, Algarbia (Welwitsch n. 285, ) Mus. Brit.); Hispania, Murcia (Bourgeau n. 988, h. Kew) ; Cadiz (Clementi, h. Calcutta); Malaga (Salzmann sh. Calcutta; Boissier, h. Calcutta) ; Italia, Parma, Agrigento (h. Kew) } Apulia (Porta et Rigo n. 246, h. Kew); Attica (Heldreich n. 52, A. Kew). Insule Mediteranee ; Sicilia, Palermo (Todaro, h. Calcutta) ; Candia (Sieber, h. Kew); Girapetro (Desfontaines, h. Cal- 80 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. cutta); Melita (Duthie, h. Mus. Brit.); Ile de Scyra (h. Calcutta); Cyprus (Sintenis et Rigo n. 358, hh. Kew, Mus. Brit.). Africa borealis: Mogadore (J. D. Hooker, h. Kew): Barbaria (Desfontaines, h. Mus. Brit.); Algiers (Munby n. 26, hh. Kew, Calcutta; Balansa nn. 140, 786, hh. Kew, Mus. Brit.); Damietta (Sieber, h. Kew). Asia occidentalis: Syria (Zowne, h. Calcutta); Arabia (Schim- per n. 218, hh. Kew, Mus. Brit.) ; Persia (Haussknecht, h. Kew); Herat (Bunge n. 7, h. Kew); Cabul (Griffith n. 198, h. Kew 6199, n. 6199 Calcutta); Beloochistan (Duke n. 69, k. Calcutta); Scinde (Stocks n. 751, h. Calcutta); Punjab, Peshawur (Stewart, h. Calcutta). 44. C. PANNONICUS (Jacq.! Austr. v: App. p. 29, t. 6) ; radice fibrosa, annua; spiculis 3-8, congestis ; involucri bracteis 2, altera basi dilatata, suberecta; glumis dense stipatis, concavis, obtusis, stramineis castaneo-rubro-notatis ; nuce ellipsoidea, cum gluma fere equilonga, chlorito-brunnea.— Zinn. f. Suppl. p. 103; Host, Gram. iil. p. 47, t. 70; Sturm, Fl. Deutsch. xiii. t. 52; Reich. Ic. Fl. Germ. t. 278. fig. 660; Wulfen. Fl. Norica, p. 53; Griseb. Spicil. Fl. Rumel. p. 420 ; Boeck. in Linnea, xxxv. p. 487 ; Boiss. Fl. Orient. v. p. 367. C. mucronatus, var., Ledeb. Fl. Ross. iv. p. 240; Kunth, Enum. ii. . 18. Pycreus pannonicus, Reich. Fl. Germ. p. 72; Nees, Genera Pl. Germ. ii. t. 22. figs. 1-13. Hungaria (Reichenbach, h. Kew; Gowan, h. Kew; Bentham, h. Calcutta); Velencze (Tauscher, h. Calcutta); Plattensee (Bilimek, h. Mus. Brit.). — sine habitat. (Jacquin, h. Mus. Brit.; n. 14, h. Linn. propr.). Euxini littora (Sibthorp, h. Mus. Brit.). 49. C. DIAPHANUS (Schrad.; Roem. et Sch. Syst. i. Mant. p. 477); umbella 3-l-radiata, contracta aut capitata; spiculis paucis, fasciculatis, latiusculis, albidis; glumis ovatis, obtusis, in lateribus diaphanis estriatis; stylo bifido.— Kunth, Enum. ii. p. 9; Boeck. in Linnea, xxxv. p. 437. Radix tenuiter fibrosa. Culmi cespitosi, 10-15 cm., graciles. Folia gracilia, culmo longiora. Involucri bractee 3, usque ad 15 cm. longs, erecto-patentes, debiles ; umbelle radi 0-2 cm. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 81 Spicule in unoquoque radio 3-5, longe 8 mm.’ late 3 mm., compresse, anguste ovate, 6-12-flore. Glume longe 2-3 mm., arctius imbricatz, compresse, in dorso anguste viridi-3-nervia, in lateribus latis lucide hyalino-albide. Stamina 2; anthere breviuscule, oblonge, mutice, exsertæ. Stylus cum nuce imma- tura equilongus ; rami 2, e gluma breviter exserti. Nepaul (Wallick, h. Kew). The above was drawn up at Kew, and is (beyond doubt of mine) the same species as described by Boeckeler, also from a Nepaul Specimen of Wallich’s. The flowers in my specimen were too young. Boeckeler says his nut was immature; but, while de- Scribing it as obovoid, he does not separately say it was com- pressed parallel to the rhachis ; he, however, places the species at the head of Pycreus. In the Caleutta herbarium are complete ripe specimens of a Cyperus collected by Vicary at Saugor in Central India, which is undoubtedly a Juncellus. In all essential points, and in the peculiar diaphanous glumes, it agrees with my description (and recollection) of C. diaphanus. Either, therefore, we have here two distinct species, very like each other, diaphanus and Ficaryi; or diaphanus has been wrongly referred, and is really a Juncellus, of which the subjoined is a hot-cpuntry var. or form. m Var. (?) B. Vicaryi ; capitulo globoso, 10-20-spiculoso ; spiculis linearibus, 8—14-floris ; nuce obovoidea, eontra rhachin compressa, facie interiore concaya, exteriore angustissime carinata, concava, cum 3 parte glumæ squilonga, pallida, juniore superne quasi a rore glandulosa. Habitus, folia, braetes O. diaphani, sed culmus vix 1 dm. longus. India centralis : Saugor (Vicary, k. Kew). 46. C. trmosus (Maxim.! Prim. Fl. Amur. p. 294) ; umbella contracta aut capitata; spiculis lineari-oblongis, acutis, teneris ; glumis ovatis, obtusis, 3-nerviis; nuce oblonga, plano-convexa, luteo-brunnea. C. pygmzus, var., Boeck. in Linnea, xxxv. p. 494 in nota. Umbelle radii 3-6, 5-15 mm. longi. Amur flumen (Maximowicz, h. Kew). 47. C. prammus (Rottb. Descr. et Ic. p. 20, t. 14. figg. 4, 5); umbella capitata, admodum densa; spiculis sspe carvatis aut tortis; glumis biseriatis, ovato-lanceolatis (neque sursum latiori- LINN. JOURN.—BOTANY, VOL. XXI. G 82 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. bus) 5-7-nerviis; nuce ellipsoidea aut oblonga, utrinque angustata, cellulis extimis parum laxis.—JVees in Wight Contrib. p. 72; Kunth, Enum. ii. p. 18 ; Cav. Ic. vi. p. 65, t. 588. fig. 2; Benth. Fl. Austral. vii. p. 262, nota excl. ; Cosson, Expl. Algér., Glum. p. 252; Boiss. Fl. Orient. v. p. 868 ; Boeck. in Linnea, xxxv. p. £93, var. p excl. C. Mulenpulla, Roem. et Sch. Syst. Mant. ii. p. 99. C. pugioniformis, Dietr. Sp. Pl. ii. p. 230. C. Michelianus, Link, Hort. Berol. i. p. 303. C. hyalinolepis, Steud. Cyp. p. 316. C. monocephalus, Roxb. h. propr., et Ic. ined. in h. Kew, non Fl. Ind. C. squarrosus, Roxb. Fl. Ind. i. p. 194, Ic. ined. ii. t. 54. fig. 95, h. Cal- cutta, non Linn. Pycreus diffusus, Nees in Linnea, ix. p. 283. Dichostylis pygmea, Nees in Linnea, ix. p. 289. Cyperus, Wall. ! List n. 3325 B, C. Cyperus sp., n. 58, Herb. Ind. Or., Hook. f. et T. Thoms. Isolepis, Wall.! List, n. 3484 C partim. Glabra, annua. Culmi cespitosi (usque ad 120), longi 8-25 cm., rigidiusculi, sursum trigoni, leves. Folia 2-4, graminea, fcum culmo sepe :equilonga, plana, sursum minute scabra. Bracteæ 8-5, divaricate, usque ad 8-16 cm. longs, foliis similes. Capi- tulum 1-2 cm. in diam., densissime congestum, sepe 100-spicu- losum ; radii plures, obsoleti, nunquam evoluti. Spicule longs 6-10 mm., late 1-2 mm., com pressz, albo-virides, proventu cin- namomeo-fusce, 8—24-flore; rhachilla tenuissima, tenuissime bialata, undata. Glume spisss, persistentes, tenues, fructus tempore plane biseriatz, raro quodammodo irregulares, carinato- naviculares, basin versus sspe scariose nervis evanidis, a mar- ginibus hyalinis basi subdecurrentes. Stamina 2-1; filamenta anguste ligulata, fere levia; anthere lineari-oblongs, mutice, flave. Stylus sepe cum nuce subzquilongus ; rami 2, e gluma breviter exserti. Nux cum 3 parte glume zquilonga, in facie interiore (contra rhacheolam adpressa) subplana, in facie exteriore convexa curvata, interdum obscure angulata, brunnea; cellule extime quadrate, emarcide, hyaline, persistentes, non laxe; ideoque nux non (ut in Zsolepide Micheliana) a margine lucide hyalino sub lente circumdata. Ab Algeria usque ad Amurland, China, Australia. Africa borealis: Algiers (Kralik n. 90, h. Kew); Nubia (Kotschy n. 829, hh. Kew, Mus. Brit.); Egypt (Sieber, h. Kew; Schweinfurth n. 2010, h. Mus. Brit.). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 88 Asia occidentalis: Syria (Bové, h. Kew); Aleppo (Russell, h. Mus. Brit.) ; Bagdad (Haussknecht, h. Mus. Brit.). India, communis, presertim in planitie tropicali: Kashmir (Jaequemont n. 1087, h. Kew); Punjab (T. Thomson, h. Cal- cutta); Lucknow (T. Anderson, n. 14, h. Calcutta) ; Bengalia (Roxburgh, h. Mus. Brit.; Griffith n. 6185, h. Calcutta; C. B. Clarke n. 11740) ; Madras Peninsula (Wight n. 1807, h. Kew; Wight n. 2862, h. Calcutta ; G. Thomson, h. Oal- cutta); Ceylon (Thwaites C. P. n. 3947, h. Mus. Brit.). Burma (Wallich n. 3484 partim, h. propr.); Meaday (R. Scott, h. Calcutta) ; Pegu (Kurz n. 646, h. Calcutta). Amurland: Ussuri (Maack, h. Kew). China: Shanghai (Maingay n. 634, h. Calcutta). Australia subtropicalis (Bidwell n. 142, h. Kew); Victoria fl. (F. Mueller, h. Kew). Var. ? B. Aztecorum ; spiculis planis; glumis ovatis, distanter 8-nerviis ; nuce a basi lata ovoidea subconica, stramineo-brunnea, quasi-scabrida. C. pygmeus, Liebm. Mex. Halfgr. p. 15. Culmi folia inflorescentia C. pygmai. Mexico: Pital (Liebmann, h. Kew). Cuba: Havana (Liebmann, h. Kew). Subgenus 4. EUCYPERUS. Eucyperus, Boeck. in Linnea, xxxv. p. 493 (sect. Spicatis Leptostachyis, Marisco, Diclidio eaclusis).—Eucyperus (cum maxima parte Papyri), Benth. et Hook. f. Gen. PI. iii. pp. 1043, 1044.—Cyperus (C.), Kunth, Enum. u. p. 20 (pro maxima parte). l l l cub Stylus 3-fidus. Nux (interdum inequaliter) trigona aut sub- rotunda. 4 Species plurimæ. Habitus generis.—Spicule multiilore, luri-(4—6-)floree. i This Sabani contains the great bulk of the genus; Pycreus and Juncellus are comparatively small groups, of conven on, Sro split off from it by (what I bold) definite characters. Dic i m and Mariscus are two more groups of convenient size, which here cut off by (what I hope will be found) definite characters, There still remains an enormous mass of species, which Boecke er is only able to deal with by forming clusters of allied species, nowhere separated from each other by contrasted differential (nor raro 84 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. indeed workable) characters. It would be very convenient to split off more groups by definite characters, and it would still more facilitate a satisfactory dealing with the genus if we could dichotomize Eucyperus (as it here stands). This Benth. et Hook. f., Gen. Pl. iii. p. 1043, have endeavoured to do by admitting Papyrus and Eucyperus as subgenera. This, I regret to say, appears to me impossible to work with: the only distinction is that the wing of the rhachilla is more prominent in Papyrus than in Eucyperus. This is at best a poor and indefinite character; but, beyond that, the exceptions to it are so numerous (even after rearranging many species in Boeckeler) that it is really little help towards finding the place of a species. To enumerate but a few instances :—In C. glomeratus, Linn., and C. eleusinoides, Kunth, the wing is well developed, yet these species stand (and must stand) in Eucyperus. O. lucidulus, Klein, has an evident subsoluble wing, and has therefore been forced by Boeckeler into Papyrus; but its affinity is clearly with C. com- pressus, where Hook. f. et T. Thoms. placed it. Boeckeler has similarly divaricated C. jeminicus, Rottb., and C. usitatus, Burchell, in attempting to preserve the character of the wing of the rha- chilla as sectional. The character runs so irregularly, indeed, that if the subgenus Papyrus is maintained (on its present character in Benth. et Hook. f.) distinct from Eucyperus, the natural affinity of species will be broken into at very numerous places; and the line between Papyrus and Eucyperus, “the degree of development of the wing," will then be arbitrary, indefinable, and useless to work with. It is an additional but superfluous argu- ment to add that in many, as in C. tegetwm, the development of the wing varies very greatly in degree in the same species. Though, therefore, I am unable to propose any better dichotomization of Eucyperus than Benth. et Hook. f., and though I would most willingly have bolstered it up if I could have discovered any prop- ping subordinate characters, I here abandon it, and fall back on a system of mere groups, as in Kunth or Boeckeler, and attempt merely to improve these (as to the Indian species especially) in detail. The series proceeds “ generally," as in Boeckeler, from the groups with an inconspicuous wing to those with the wing of rhacheola conspicuous. By the removal of the series with per- sistent glumes, deciduous spikelets, and by some simplifications, I have considerably reduced the unwieldy number of groups in Kunth and Boeckeler. This, however, may be a doubtful advan- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 85 tage, as the more groups are made the more tangibly they may be defined. Sect. A. Aristati. Aristati, Kunth, Enum. ii. p. 20; Boeck. in Linnea, xxxv. p. 97. Annus, humiles, glabre, cespitose, foliate. ^ Umbella specie simplex aut capitata, in S. amabili interdum composita ; spicules numeros: fasciculate aut breviter densius spicate, compresse; rhachille ale anguste, insolubiles. Glume carinate, aristate aut mucronate, raro mutice. Stamina sepius 1-2 ; anthere parve, oblong», mutica, ut styli rami vix aut brevissime e gluma exserte. Nux castanea, brunnea aut lutea, neque (ut sepe in sect. Haspan) marmoraceo-albida. This is a natural small group, nor is there much difficulty or difference of opinion what species should be placed in it. But the name is unfortunate: not only do equally aristate glumes occur in most groups of Eucyperus, but in several of the Aristati the glume is merely minutely mucronate, and in C. amabilis often absolutely muticous.—Of Boeckeler’s species (which I have not been able to take up) in this section, Mr. Bentham has removed elsewhere C. flaccidus, R. Br., C. trinervis, R. Br.; and I have a note by me that C. seslerioides ought to stand next C. leuco- cephalus, Retz., below. * Spicule digitate. 48. C. AwanrLIS (Vahl, Enum. ii. p. 818); umbella capitata, simplice, composita aut decomposita; spiculis digitatis, nee brunneis; glumis truncatis, subemarginatis, mucrone brevi ve obsoleto; nuce anguste obovoidea obtusa, quam 2 pars glume paullo breviore.—Kunth, Enum. ii. p. 108. . C. microstachyus, Vahl, Enum, ii. p. 318; Kunth, Enum. ii. p. 108. C. aureus, H. B. K. Nov. Gen. et Sp. i. p. 205; Kunth, Enum. n. p. 21; Boeck. in Linnea, xxxv. p. 494. 205; Kunth, Enum. ii. C. aurantiacus, H. B. K. Nov. Gen. et Sp.i. p. ^ p. 30; Benth. in Hook. Niger Fl. p. 549; Nees in Mart. Brasil., Cyp. . 25. C. oligostachyus, H. B. K. Nov. Gen. et Sp. i. p. 204; Kunth, Enum. ii. p. 20. . C. brachyphyllus, Willd. ; Link, Jahrb. iii. p. 82. C. quitensis, Spreng. Syst. i. p. 224. C. lepidus, Hochst.!; Steud. Cyp. p. 14. 86 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. glareosus, Liebm.! Mex. Halfgr. p. 16. Cyperus sp. n. 24, Herb. Ind. Or., Hook. f. et T. Thoms. Radix annua, tenuiter fibrosa. Culmi cespitosi, pauci, longi 2-17 em., trigoni,leves. Folia 2-3, culmo (sepius multo) bre- viora. Involucri bractes plures, 1-10 cm., in umbellis bene evolutis umbella sepius breviores. Umbelle radii 4—15, usque ad 1-11 em. longi, tenues; umbella ssepius specie simplex aut capitata, sed umbellularum radioli interdum 3-6, usque ad 1-5 em. longi. Spicule in unoquoque radio 5-20, radiatim digitata, bracteatee (i. e. capitulum revera compositum) aut ebracteatz, usque ad 24 mm. longs, 1-2 mm. late, 16-28-flore aut rarius 40-50-flore ; rhachilla tenuis. Glume subdistantes, in sicco vix imbricate, ovate, in dorso viridi-8—5-7-nervis, in lateribus hya- line nitide brunnee. Stamen (in plante Indica) 1; anthera minima, oblonga, mutica, e gluma non exserta. Stylus nuce brevior; rami 3, e gluma brevissime exserti. Nux obtusissima, symmetrice subacute trigona, lucide brunnea; cellule extime quadrate, hyaline, persistentes nec poros. [ C. glareosus, Liebm., a Mexico, paullo recedit; in hoc spicule pallidiores, glume conspicue 3-nerviæ, mucro cum å parte glume sequilongus. | India orientalis, Africa tropicalis et australis cum insulis, et America tropicalis. India orientalis, boreali-occidentalis, et centralis: Chota Nagpore, alt. 1-600 metr. frequens (C. B. Clarke n. 24889) ; Moradabad (T. Thomson, hh. Kew, Calcutta, n. 393 h. Mus. Brit.) ; Naini Tal (Col. Davison, h. Kew); Kumaon (G. King, h. Calcutta). Insule Mascarenses &e. : Socotra (Balfour, h. Calcutta); Ma- yotta (Boivin, h. Kew); Nossibé (Pervillé n. 488, h. Kew); Madagascar (Gerard n. 95, h. Kew; Pervillé n. 521, h. Kew). Africa, centralis et australis: Nubia (Kotschy n. 189, hh. Kew, Mus. Brit.); Ador (Petherick, h. Kew); Djur et Bongo (Schweinfurth nn. 193, 2188, 2256, h. Kew); Quorra flumen (Vogel n. 198, 5. Kew); Africa australi-occidentalis, lat, austr. 23? (Chapman et Baines, h. Kew); Goldfields, lat. austr. 27° (Baines, h. Kew). America: Cuba (C. Wright n. 3356, hh. Kew, Mus. Brit.); Orizaba (F. Mueller n. 1986, h. Kew); Guatemala (Skinner, h. Kew); Surinam (Weigelt, h. Mus. Brit.=C. coercens, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 87 Reich.) ; Maracaybo (Sonder n. 1571, hh. Kew, Mus. Brit.); Venezuela (Moritz n. 683, h. Mus. Brit.; Funcken.701, h. Mus. Brit.); Guiana Anglorum (Schomburgk n. 221, hh. Kew, Mus. Brit.); Brasil (Burchell nn. 8488, 8528, 8639, h. Kew); Para (Spruce n. 88, h. Kew); Santarem (Spruce n. 745, h. Kew, h. Calcutta). 49. C. CASTANEUS (Willd. Sp. Pl. i. p. 278); umbella specie simplice (raro capitata), radiis longiusculis, spiculis digitatis ; glumis emarginatis, mucrone longiuseulo sub apice excurrente ; nuce oblonga, obtusa, quam 3 pars glume (mucrone excluso) lon- giore, pallida.—Aoxb. Fl. Ind. i. p. 195; Nees in Wight Contrib. p. 79; Kunth, Enum. ii. p. 21; Boeck. in Linnea, xxxv. p. 496 (Wail. n. 3376 B excl.); Benth. Fl. Austral. vii. p. 267. C. squarrosus, var. stenocarpa, F. Muell. Fragm. Phyt. Austral, viii. p. 262. C. pusillus, Wight ( h. propr.) MS. Cyperus sp., Wall. ! List n. 3323. Radix annua, tenuiter fibrosa. Culmi caspitosi, pauci, longi 3-10 cm., trigoni, leves. Folia 2-3, cum culmo :quilonga aut sepius breviora. Umbellæ radii 3-6, tenues, 1-5 em. longi (aut umbella capitata); involucri bractew 3-5, eum umbella æquilong® aut longiores ; ochreæ 3 mm., truncatæ. Spiculæ in unoquoque radio 4-16, longæ 1-3 cm., latæ 1-2 mm., 15-70-floræ, compresse; bracteæ usque ad 10-15 mm. longa, lineares (i. e. umbellula revera composita), aut breviores aut 0. Glumæ spissw, in dorso carinatæ virides, obscurius 3 5-nerviw, in lateribus (in forma typica) nitide castanes, modo brunnes modo pallide; mucro robustus, viridis, } parte glums subbrevior, paullo. recurvatus. Stamina 2-1; anther parve, breviter oblong», mutice, e gluma non exsertæ. Stylus nuce brevior; rami 3, e gluma breviuscule exserti. Nux oblonga, basi breviter angustata (vix obovoidea), trigona, paullo curvata, neque triquetra, pallide brunnea aut lutea; cellule extime quadrata, hyaline, persistentes.— Rox- burghii et Rottleri exempla sub nomine C. castaneo conservata sunt C. castaneus, Willd. ; in Roxb. Ic. ined. in h. Kew sub hoc nomine, gluma lanceolato-acutata, nux apice acuta, depicts sunt —tabula mihi dubia. . India Orientalis, cum Zeylania: Sikkim et Nepaul orient. (J: D. Hooker, h. Kew); Bengalia, in planitie Mymensingho (C. B. Clarke n. 17294); Pegu Maclelland, h. Calcutta); 88 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Mergui (Griffith n. 43, h. Kew); Chota Nagpore, Parasnath, alt. 500 metr. (C. B. Clarke n. 24896); Madras Peninsula (Wight n. 1821, h. Kew; Koenig, h. Mus. Brit.; Roxburgh, h. Mus. Brit. ; Wight in Wallich n. 3323, k. Calcutta) ; Zeylania (Thwaites, C. P. n. 8083, hh. Kew, Calcutta). Australia: Queensland (Leichardt, h. Kew). This is a critical, comparatively rare species, from which the wide-spread C. cuspidatus is with difficulty distinguishable. C. castaneus was originally founded on its chestnut colour, and in the typical form the glumes are chestnut-red to chestnut-black, a colour hardly found in all C. cuspidatus. Boeckeler, however, says the glumes are “ sanguineo-spadice: v. raro rufo-ferruginese nitide:" this is because Boeckeler worked with small material, of which some of the best would be Thwaites n. 803; but some of Thwaites’s no. 803 (which, with Thwaites and Boeckeler, I call true C. castaneus) is very pale, with a little obscure dull red about it, and not at all shining. The fact is that (as in most cases) colour fails us as a diagnostie mark, and is given up by Boeckeler, who keeps up the two species solely on the difference in the length and shape of the nut. This difference, though well marked in many specimens, becomes very obscure in others. An auxiliary character is that in C. cuspidatus the glume is very conspicuously 3-nerved, much less conspicuously in C. cas- taneus ; but this character is not to be depended on. The mucro of the glume is often more curved in C. cuspidatus, but not always. Ifthe two species are to be distinguished, it must be solely by the nut as Boeckeler has got it. Wall. List n. 3376 B, both at Caleutta and in h. propr., is most typical C. cuspidatus, var. angustifolia; Boeckeler refers this number to C. castaneus, but the Wallichian herbarium is so mixed that this proves nothing. 50. C. cusprpatus (H. B. K. Nov. Gen. et Sp. i. p. 104); umbella specie simplice rarius capitata, radiis longiusculis, spi- culis digitatis; glumis emarginatis, plane trinerviis, mucrone sub apice longe excurrente ; nuce anguste obovoidea, obtusa, fusco- atra, quam $ pars glume (mucrone excluso) longiore.— Avwnth, Enum. ii. p. 22; Nees in Mart. Brasil., Cyp. p. 24; Boeck. in Linnea, xxxv. p. 496 ; Benth. Fl. Austral. vii. p. 267. C. exiguus, Nees! in Mart. Brasil., Cyp. p. 22. C. Salzmanni, Steud.! Cyp. p. 1 . MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS, 89 C. castaneus, Hochst. in Hohenack. n. 824; Hance, h. propr.; Mig. FI. Ind. Bat. iii. p. 261, vix Willd. C. contractus et C. solutus, Steud. Cyp. pp. 13, 14. C. squarrosus, F. Muell. Fragm. Phyt. Austral. viii. p. 262, non Linn. Cyperus, JVall.! List n. 3376 A partim. Radix annua, tenuiter fibrosa. Culmi esspitosi, pauci, longi 3-12 cm., trigoni, leves. Folia 2-3, angusta, cum culmo æqui- longa aut sepius breviora. Umbellæ radii 2-6, usque ad 1-5 em. longi, tenues (vel raro umbella capitata); ochrew 3-6 mm. Spi- cule in unoquoque radio 5-20, long: 15 mm., lata 2 mm., 12-30- flore, compressx ; bractez usque ad 10-15 mm. longs (i. e. um- bellula revera composita), lineares, aut breviores aut 0. Glume approximate, imbricatæ, valide 3-nervie, nervis viridibus aut brunneis, in lateribus hyalino-castanew aut brunneæ, enervia ; mucro recurvatus, cum 1-$ parte glume æquilongus. Stamina 2 vell; anthere breves, oblonge, e gluma vix exserte. Stylus nuce brevior ; rami 3, e gluma breviuscule exserti. Nux trigona, non symmetrica, lateribus vix parallelis, cellule extimæ quadrata, hyaline, persistentes (cf. fig. 38, 39), minime tuberculato. Afriea, Asia, Australia, America ; in regione tropicali. Afriea: Niger flum. (Barter n. 1569, hh. Kew, Calcutta); Sierra Leone (Smeathmann, h. Mus. Brit.) ; Nilus flum., Madi (Grant, h. Kew). Asia :—India Or. in regione boreali a Kashmir usque ad Assam, frequens: Kashmir, alt. 1000 metr. (C. B. Clarke n. 31570, h. Kew); Sikkim (G. King, h. Calcutta); Lucknow (Bonavia, n. 236, h. Calcutta); Khasia Colles, alt. 800 metr. (J. D. Hooker n. 1818, k. Kew); Assam (Jenkins, h. Calcutta, $c); Chota Nagpore, alt. 600 metr. (C. B. Clarke); Bengalia orientalis (Griffith, h. Kew n. 6202, n. 6202 A. Calcutta) ; Pegu (Maclelland, h. Calcutta). Java (Zollinger n. 3739, h. Kew). China: Whampoa (Hance n. 7461, hh. Kew, Calcutta) ; Macao (Lay, h. Mus. Brit.); Hongkong (For bes, h. Mus. Brit.). Ins. Philippine (Cuming n. 676, hh. Kew, Mus. Brit.). Borneo (Motley n. 900, h. Kew; Burbidge, hh. Kew, Mus. Brit.). Australia: Queensland (fide Bentham). g America: Brasil (Burchell n. 957, h. Kew; Spruce nn. 101, 2041, hh. Kew, Mus. Brit., n. 2041 h. Calcutta) ; Guiana Anglorum (Schomburgk n. 824, hh. Kew, Mus, Brit.); Mexico (Liebmann, he Kew). 90 MR. C. B. CLARKE ON INDIAN SPECIES OF OYPERUS. [Forma angustifolia; spiculis brevioribus, sepe pallidis; nuce late obovoidea, quam 3 pars glume (mucrone excluso) sspe bre- viore, trigona, faciebus sæpe concavis. C. angustifolius, Buch.-Ham. sp. ; Nees in Wight Contrib. p. 79. C. pusillus, Wight MS. in h. propr. n. 2337, non Vahl. C. aristatus, Herb. Ind. Or., Hook. f. et T. Thomson, partim. Cyperus sp., Wall.! List n. 3376 B, A partim. India Orientalis, sat vulgaris: Nepaul (Wallich, h. Linn. Koc., n. 3312 C partim 4. Calcutta) ; Sikkim (G. King, h. Calcutta); Bengalia orientalis (Griffith, hh. Kew, Calcutta n. 6203 |1); Assam (Simons, h. Calcutta); Monghyr (.Buchanan- Hamilton in Wall. n. 3376 B, h. Calcutta) ; Madras Peninsula (Wight n. 2337, h. Kew). Zeylania (Beckett, h. Calcutta). I think this was made a species by those who did not know, or who forgot to compare C. cuspidatus. I do not think it is worthy of notice as a variety, but I would sink C. angustifolius among the synonyms of C. cuspidatus type. | Var. B. Burchellii; culmis nanis; spiculis numerosis, magnis. Spicule longs 25 mm., late 3—4 mm., 50-60-florze, pallide. Var. notabilis ; differt a C. cuspidato typ. ut C. pectiniformis B, Nees, differt a C. compresso. Brasil (Burchell n. 9642, h. Kew). 51. C. uncrnatus (Poir. Encycl. vii. p. 247) ; umbella contracta (aut capituliformi), radiis paucis, paucispiculosis ; spiculis digi- tatis, compressis ; glumis emarginatis, mucrone sub apice breviter excurrente ; nuce obovoidea, obtusa, cum 2 parte glume subequi- longa.— Kunth, Enum. ii. p. 21; Boeck. in Linnea, xxxv. p. 502- C. pectinatus, Hils, et Bojer! in h. propr., non Vahl. Folia evlmis ssepe longiora, uncinata. Spicule longs 1 em., late 2 mm., 8-16-flore, castaneæ aut viridi-brunnes.— Species a C. cuspidato :egre distinguenda. Madagascar (Lyell n. 84, h. Kew; Grey, Blackburn, h. Kew; Du Petit-Thouars, h. Mus. Brit.). 52. C. PARAGUAYENSIS; umbella capitata; glumis spissis, ob- tusis vix emarginatis, mucronatis ; nuce obovoidea, obtusa, quam 4 pars glume breviore, fusco-nigra. Radix fibrosa (perennis ?). Culmi longi 2-7 cm., basi incras- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 91 sati. Spicule longs 8 mm., late 2 mm., 12-28-flore, pallide aut fusco-brunnes ; rhachilla subrobusta, excavata. Glume cari- nate; mucro 3 parte glume paullo brevior.—Species C. pumilo similis, sed stylus trifidus. Confer C. wneinatulwm, Schrader ; Nees in Mart. Brasil., Cyp. p. 23 ; Boeck. in Linnea, xxxv. p. 497; in quo autem glume patule, fructifere remote, nux stramineo- pallida. Paraguay (Balansa nn. 419, 420, h. Kew). ** Snicule (sepius dense) spicata. 53. C. ARISTATUS (Rottb. Descr. et Ic. p. 23, t. 6. fig. 1); um- bella specie simplice vel capituliformi; glumis e basi ovata lan- ceolato-acuminatis, per totam fere latitudinem conspicue multi- nerviis; nuce oblonga vel anguste obovoidea obtusa, cum 3-3 parte glume xequilonga.— Kunth, Enum. ii. p. 28 (syn: Burmanni excl); Roxb. Fl. Ind. i. p. 190; Nees in Wight Contrib. p. 78 ; A. Rich. Fl. Abyss. ii. p. 478; Boeck. in Linnea, xxxv. p. 500 ; F. Muell. Fragm. Phyt. Austral. viii. p. 261 ; Sereno Watson, Bot. California, ii. p. 214. ` C. squarrosus, Linn. Amen. Acad. iv. p. 303; Sp. Pl. 66, h. propr. partim, nec Rottb.; Benth. Fl. Austral. vii. p. 268. C. intricatus, Linn. Mant. p. 182. C. inflexus, Muehl. Descr. Gram. p. 16; Torrey in Ann. Lyceum New York, iii. p. 273; C. Gay, Hist. Chile, v. p. 164; F. Muell. Fragm. Phyt. Austral. viii. p. 262. C. uncinatus, R. Br. Prod. p. 215; Pursh. Fl. Am. Sept. i. p. 50, non Poir. C. Purshii, Roem. et Sch. Syst. ii. p. 177. . C. Brownei, Roem. et Sch. Syst. ii. p. 228 ; Kunth, Enum. ii. p. 111. C. pygmaeus, Nutt. in Amer. Phil. Trans. new senes, v. p. 142, non Rottb. C. versicolor, Nees in Wight Contrib. p. 78. C. arenarius, Wight MS. in Wall. List n. 3374, non Retz. C. falciculosus, Liebm. Mex. Halfgr. p. 16. Scirpus lappaceus, Lam. Ill. i. p. 139. Isolepis echinulata, Kunth, Enum. ii. p. 205. Cyperus, Wall.! List n. 3375. Cyperus, Wall.! List n. 3374 (forma versicolor). Cyperus, Wall.! List n. 3376, plagula 2 partim (n. 3). Culmi, 2-14 cm. longi, cæspi- . iter fibrosa. Radix annua, tenuiter Folia 2-3, tosi, sursum trigoni, modo tetragoni modo bialati. 92 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. flaccide graminea, cum culmo sepe equilonga. Involueri bracteæ 2—5, usque ad 5-10 em. longs, umbella sepius multo longiores ; radii 2-6, usque ad 1-6 em. longi, aut umbella eapitata. Spicule in unoquoque radio multi rarius pauci, dense spicati, quoad mag- nitudinem magnopere variabiles ; in exemplis evolutis spicule in unoquoque radio 12, sublaxe spieatz, plane compressz, lucide, longe 12 mm., late 2-3 mm. ; in exemplis minus evolutis spicule minores, minus compresse, fusco-virides, globoso- aut cylindrico- congeste: umbellule subebracteate, sspe autem plane com- posite. Glume carinate, 7-ll-nervie, apice in mucronem attenuate, virides, fuscæ, pallide, aut rufescenti-brunne:s, rarius lutescentes, nervis sepius viridibus. Stamen 1; anthera parva, oblonga, e gluma non exserta. Stylus nuce brevior; rami 3, e gluma vix exserti. Nux trigona, fusco-brunnea ; cellule extime quadrate, hyaline, persistentes.—Sub hac specie, forme plures militant, scil. :— Forma princeps; umbella laxa, pluriradiata; spiculis magnis laxiuscule spicatis, compressis, pallidis aut lucidis, luteis vel sub- castaneis ; nuce plane obovoidea.—Forma in Himalaya et in Africa tropicali maxime reperta. Forma alpina; quasi forme precedentis forma alpina; culmis elongatis, tenuibus ; umbella 2-1-radiata ; bracteis 2, altera erecta quasi culmum producente ; spieulis paucis, magnis.—K ashmir. Forma campestris; umbella sepe 3-radiata; spiculis minimis, viridibus, densissime spicatis.—Bengal. Forma versicolor ; umbella pauciradiata aut capitata ; spiculis densissime spicatis; rhachillis sepe tortis; glumis longissime aristatis ; spiculis fructigeris sæpe rufescenti-brunneis.— Madras Peninsula. Forma inflexa; quasi inter formas principem et campestrem intermedia; umbella quam principis densiore tam campestris laxiore, spiculis quam principis minoribus tam campestris majori- bus, sepius fusca-viridibus interdum pallidis.—A merica; sed exempla gerontogea omnino eadem. Linnzus, quam in h. propr. tam in descriptionibus editis, 2 species confunderat; ejus descriptio (quoad spiculas ferrugineas) ad C. versicolorem, Nees, maxime spectat, ejus ic. citat. autem est species a C. aristato diversa. Rottbæll e Linnzano C. squarroso 2 species diversas bene extricavit et depinxit, alteri nomen C. squarrosum, Linn. alteri nomen C. aristatum, Rottb., tribuit. Quum que ita sint, nomenclaturam auctorum Boeckelerique MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 93 accepi; quamvis nomenclatura Benthami cum Linneo fortasse melius congruat. [Scirpus hamulosus, Steven in Mém. Soc. Natur. Mose. v. p. 356, = Isolepis hamulosa, Kunth, Enum. ii. p. 205, is reduced to Cyperus aristatus by Boeckeler. It is (to me) clearly an Zsolepis. But besides this, the material is so very distinct specifically from Cyperus aristatus (even if it be put in Cyperus), that I almost doubt whether the Berlin herbarium can have got the true plant.] Africa tropicalis; India; Australia; America fere tota. Africa: Niger flumen (Vogel n. 18, h. Kew); Nupe (Barter, h. Calcutta); Senegambia (h. Kew n. 328); Dammara-land (Hen, h. Mus. Brit.); Nilus flumen (Grant, h. Kew); Nubia (Kotschy n. 50, hh. Kew, Mus. Brit.) ; Abyssinia (Schimper nn. 437, 822, 855, 1208, 1735, 2155, hh. Kew, Mus. Brit. n. 822 h. Caleutta; Hildebrandt nn. 365, 366, h. Mus. Brit.). Ins. Socotra (Balfour n. 507, h. propr.). India Orientalis, alt. 0-2800 metr., ab Himalaya usque ad Zey- laniam, vulgaris: Tibet, Kargil (Stoliczka, h. Calcutta); Himalaya, Piti (Z. Thomson, h. Kew); Dalhousie, alt. 1500 metr. (C. B. Clarke n. 22845); Himalaya occidentalis (Brandis nn. 3326, 3327, h. Calcutta); Ghurwal (G. King, h. Calcutta); Sikkim, Lachoong, alt. 2800 metr. (J. D. Hooker, h. Kew); Peshawur (Jacquemont n. 1276, h. Kew); Poona (Jacquemont n. 815, h. Kew); India centralis (G. King, h. Calcutta); Madras Peninsula (Wight nn. 1819, 1820=C. versicolor, h. Kew). Zeylania (Thwaites, C. P. n. 966, hh. Kew, Mus. Brit., Cal- cutta). , Australia, tropicalis et centralis : Dampier’s Archipelago (Walcott, h. Kew); sinus Carpentaria (F. Mueller, h. Kew); Herbert’s Creek (Bowman, h. Kew). g America borealis, fere tota : Canada (Richardson, h. Kew); Winipeg Lacus (Drummond, h. Kew); Oregon (Geyer n. 445, h. Carey in h. Kew); Kentucky (Short, h. Mus. Brit.); Mas- sachusetts (Tuckerman, h. Mus. Brit.); Alabama (Buckley, h. Calcutta) ; Texas (Drummond n. 448, hh. Kew, Mus. Brit., Calcutta); California (Hartweg n. 2019, hh. Kew, Mus. Brit. ; Palmer n. 385, h. Mus. Brit.); Mexico Nova (C. Wright n. 1951, A. Kew, n. 702 hh. Kew, Mus. Brit.); Mexico, Vera Cruz (F. Mueller n. 2154, h. Kew). 94 MB. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. America centralis et meridionalis (exempla a Brasilia nondum a me visa): Yucatan (Schott nn. 77, 910, h. Mus. Brit.); Ecuador, Chanduy (Spruce n. 6416, hh. Kew, Calcutta); Chili (Germain, hh. Kew, Mus. Brit.); Paraguay (Balansa n. 418, h. Kew). Var. (?) B. Maingayi; glumis obovatis, in mucronem brevius- culum subito angustatis, in dorso viridibus conspicue 5—7-nerviis, in marginibus sursum latiuscule albescentibus, enerviis. Viridis. Culmi 1 dm. tenues. Bracteæ usque ad 5-8 cm., anguste, virides. Umbelle radii 1-4, usque ad 2 cm. longi, 5-9- spieulosi. Spicule plane spicatz, nec digitatz, 9-11-flore. China borealis: Shantung (Maingay n. 84, h. Calcutta). 54. C. SQUARROSUS (Linn. Amen. Acad. iv. p. 308, Sp. Pl. p. 66, h. propr. partim); umbella simplice aut capitata; spiculis spicatis, linearibus; glumis remotis, ovato-lanceolatis mucro- natis; nuce anguste oblonga, subobtusa, quam 3 pars glume giore.—Aottb. Descr. et Ic. p. 25, t. 6. fig. 3; Nees in Wight Contrib. p. 75; Kunth, Enum. ii. p. 22; Dalz. y Gibs. Bomb. Fl. p. 281; Boeck. in Linnea xxxv. p. 501. C. maderaspatanus, Willd. Sp. Pl. i. p. 278 (syn. Pluk. excl.); Benth. FI. Austral. vii. p. 268 in nota. C. hyalinus, Heyne in Wall. List n. 3313, non Vahl. Pycreus squarrosus, Nees in Linnea, ix. p. 283. Cyperus, Wall.! List n. 3313 A, B partim. Cyperus, Wall.! List n. 3312, D partim. Radix annua, fibrosa. Culmi cespitosi, longi 5-15 cm., trigoni, leves. Folia 2-3, graminea, culmo sspe longiora. Involucri bracteæ 3-5, usque ad 6-12 cm. longs, patentes. Umbelle radii 2-6, breves, rarius 2 cm. superantes, aut 0. Spicule in uno- quoque radio 6-20, spicate, divaricatz, long 8-12 mm., late vix 1 mm., compress, 10-18-flore, virides vel brunnescentes; spicæ a bracteola viridi, 3-4 mm. longs, sepe suffulte, rarius subcom- posite. Glume distantes, adpresss, in dorso valide viridi-3-5- nervie; margines latiusculi, hyalini, estriati; apex obtusus; mucro sub apice excurrens, cum 3 parte glume vix sequilongus. Stamen l. Stylus nuce multo brevior; rami 3, vix exserti. Nux anguste oblonga, trigona, obscure curvata, apice subito angustata, apiculata, fusco-brunnea; cellule extime quadratz, albidz, hyalinz.—A C. aristato spiculis anguste linearibus, glumis re- motis primo intuiter distinguenda. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 95 India Orientalis : Bengal (Masters, h. Kew) ; Madras Peninsula (Wight n. 1810, h. Kew); Mangalore (Hohenacker n. 823, h. Kew). Ins. Comorin (Bouton, hh. Kew, Mus. Brit.). Var. B. lancinux ; spiculis majoribus, longis 2 cm., latis 2-3 mm., 12-28-floris; glumis minus remotis, in lateribus plicato-striatis, mucrone fere terminali; nuce unsymmetrica, obliqua, in uno latere concava, apice conspicue lanceolata. l Exempla optima Griffithii a C. squarroso bene diversa viden- tur; exempla autem intermedia in herbario Griffithii ipsius alio- rumque reperiri possunt. Peninsula Malayana: Mergui (Griffith nn. 18, 337, h. Kew n. 6203). Sect. B. Compresst. Compressi, Kunth, Enum. ii. p. 23 (C. jeminico exel.); Boeck. in Linnea, xxxv. p. 505, partim. Foliacex, virides, nec cinnamomeo-brunnei (nisi C. rubicundus). Umbella pauciradiata, specie simplex aut fere simplex (nisi C. Griffithii); radii longi (nisi C. rubicundus). Spicule fascicu- late aut spicate, compresse (nisi C. Grifithii). Glume imbri- cate, sepius mucronate. Rhachilla fere exalata. Stamina 3. Nux late ellipsoidea aut subovoidea, majuscula, subsequaliter acute triquetra. l The group here called Compressi is natural, 4. e. united by à great number of characters, if we leave out C. Grifithi and C. rubicundus. O. Griffithii is kept here because it is so like C. radians that it has been united therewith specifically by Boeckeler; and though the structure of its spikelets (as well as its inflorescence) is totally remote from that characterizing the group, it has for the present to be arranged here as an anomalous species. C. amabilis in its colour and capitate inflorescence might perhaps be moved to the Sect. Arenarii, but the strongly mucronate glumes do not suit. ; and of Kunth has only three species in his group Compressi, an t these C. jeminicus, as Boeckeler rightly arranges it, belongs ° the Corymbosi. Boeckeler, on the other hand, has 33 species M Compressi; many of these (of the non-Indian ones) I have no taken up at all in this paper; but omitting these, I remove many of Boeckeler's Compressi, as follows :— 96 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. usitatus, Burchell, is very close to C. jeminicus, and belongs to the Corymbosi. C. proteinolepis, Steud., has woolly roots, and belongs to the Sect. Galeria ; indeed, it is hardly possible to extricate it speci- fically from C. glomeratus and C. curvulus. C. babakensis is, I think, only one of the endless varieties of C. pilosus ; and, if a subspecies, must go with C. pilosus. C. obtusiflorus, C. margaritaceus, and C. niveus differ from my type of Compressi in their stramineous or cinnamomeous tinge, in their capitate inflorescence, very closely packed concave obtuse glumes, and broad filaments—in a word they belong exactly to my type of Arenarii. I take C. compressus as the type of my Compressi, and (omitting the two doubtfully placed here, C. Griffithi? and C. rubicundus) the half-dozen species collected under Compressi are pretty clearly allied to it. But it is much more difficult to say why various other species should not be brought here also. The Com- pressi are one of the groups that touch the Corymbosi. Even C. sphacelatus itself, with its annual root, thin umbel, and green colour, seems to me exceedingly near C. compressus, much nearer to it indeed than to C. articulatus, which stands at the head of the Corymbosi. The only reason for not placing C. sphacelatus among the Compressi must be the wing of the rhachilla, which, though narrow and inconspicuous, is oblanceolate on each notch of the rhacheola in a manner more characteristic of the Corymbosi than of the Compressi. Still my best reason for not placing C. sphacelatus among the Compressi is that I find it among the Corymbosi. Among the Glomerati several differ from the Compressi by hardly any character except that the rays are shorter, the spikes longer in proportion, giving the inflorescence a denser character. C. strigosus even hardly differs in laxness of spikes, and the wing of the rhachilla is very narrow ; while the small group Glutinosi differ only by trivial characters. The difficulty thus is not to pick out a group of 6 or 8 species allied to C. compressus, but to show how such group differs from all other species of Eucyperus, which difficulty I leave virtually untouched. I think the stout, subequally triquetrous nut at least as important as any of the characters insisted on by Boeckeler. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 97 55. C. compressus (Linn.! Sp. Pl. p. 68, et h. propr.) ; viridis ; umbella specie simplice, laxa, spiculis lineari-oblongis; glumis ovatis, acuminatis, acutis, carinatis, multinerviis; styli ramis 3, e gluma vix exsertis; nuce obovoidea, obtusissima, triquetra facie- bus subconcavis.— Rottb. Descr. et Ic. p. 27, t. 9. fig. 3; Rob. Fl. Ind. i. p. 194; H. B. K. Nov. Gen. et Sp.i. p. 207; Torrey, in Ann. Lyceum New York, iii. p. 270; Meyer, in Mém. Acad. St. Pétersb. 6 ser. i. t. 8 (non visa); Kunth, Enum. ii. p. 23; Nees in Wight Contrib. p. 77, in Mart. Brasil, Cyp. p. 29; Hassk. Pl. Jav. Rar. y. 71 ; Dalz. & Gibs. Bomb. Fl.p. 282; Baker, Fl. Maurit. p. 410 ; Saunders § Baker, Ref. Bot. t. 240; Boeck. in Linnea, xxxv. p. 517; Boiss. Fl. Orient. v. p.972; non Jacq. C. brachiatus, Poir. Encyc. vii. p. 259. C. conglomeratus, Willd. Enum. Suppl. p. 5 ( fide Nees). C. pectiniformis, Roem. et Sch. Syst. Mant. ii. p. 128, syn. Roxb. excl. ; Nees in Wight Contrib. p. 77, syn. excl. C. Meyenii, Nees et Meyen in Pl. Meyen. p. 57. Cyperus, Wall.! List n. 3308. Cyperus, Wall.! List n. 3314 C. Cyperus, Griff. Itin. Notes, p. 12, n. 167. Cyperus, Griff. Itin. Notes, p. 362, n. 191. — Sloane, Jamaica, i. p. 117, t. 76. fig. 1. — Pluk. Alm. t. 417. fig. 2 (?). 2. Radix fibrosa. Culmi cespitosi, 1-35 cm. longi, trigonl, leves. Folia 2-3, viridia, cum culmo spe equilonga, sursum vix scabra. Umbella 7-2-radiata (interdum capitata); radii usque ad 6-16 em. longi; ochres 5-15 mm. longs, truncate. Involueri bracteze 9-5, usque ad 10-15 cm. longs, patentes. Umbellule 5—12-spi- culose, congestim corymbose, ssepius e spica simplice constant Spicule longæ 1-3 cm., late 4 mm., 12-40-floree, lateribus paral- lelis a mucronibus glumarum quasi serratis, inferiores MA caudato-bracteate ; rhachilla admodum compressa, subexa sta Glumæ naviculares, vix approximatæ, ad pressor ae á o" virides vix fusco-spadices ; nervi 9-13 per totam fere PM inem glumæ sparsi. Stamina 3; antheræ muticæ vix apiculatæ, T cluse ; filamenta subpersistentia. Nux 3 parte -— pau” o longior, non apicata, nigra; cellulæ extimæ quadrat, albi "M sistentes.—Quoad magnitudinem quam plante tam uu di Variabilis : in forma Americana vulgari spicule minores i m ^ 1a Orientali spiculæ interdum majores (forma “ pectinifi e! , " exemplo Jaequemontiano late lanceolate, 6-7 mm. late.— C. pe - H LINN. JOURN.—BOTANY, VOL. XXI. 98 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. tiniformis, Nees! var. B est exemplum caule 2 cm. longo, spiculis usque ad 45 cm. longis, 60-floris—Roem. et Sch. Syst. Mant. ii. p. 128 quam nomen tam descriptionem Rogburghii mutaverunt ; ideoque C. pectiniformis, Roem. et Sch., descriptio a C. pectinato, Roxb. abhorret. Africa, Asia, America, in zonis tropicis et australibus. Africa: Natal (Gerard n. 704, h. Kew); Zanzibar (Kirk, h. Kew; Hildebrandt n. 1072, hh. Kew, Mus. Brit.). Madagascar (Hildebrandt n. 2920, hh. Kew, Mus. Brit.). Mauritius (Sieber n. 12, hh. Kew, Mus. Brit.) Seychelles (Pervillé n. 91, h. Kew). Socotra (Balfour n. 875, h. propr.). Asia australi-orientalis ; in omni India, alt. 0-2000 metr., ab Himalaya usque ad Zeylaniam et Singapore, vulgaris: Cabul (Grifith n. 191, h. Kew n. 6183); Punjab (T. Thomson, h. Kew); Dehra Doon (G. King, h. Calcutta); Sikkim (Treutler, h. Kew); Oudh (R. Thompson n. 409, h. Calcutta) ; Assam (Griffith n. 1481, h. Mus. Brit.); Bengal (Griffith, h. Kew n. 6181); India centralis (G. King, À. Calcutta); Poonah (Woodrow, h. Kew); Madras Peninsula (Wight nn. 1814, 1815, ^. Kew); Mangalore (Hohenacker n. 822, h. Mus. Brit.) ; Zeylania (Thwaites, C. P. n. 812, hh. Kew, Mus. Brit., Calcutta); Burma (Griffth n. 6181, h. Calcutta); Pegu (Oates, h. Calcutta); Singapore (Kurz, h. Calcutta; Kunstler n. 53, h. Calcutta). [Forma pectiniformis—Lucknow (Bonavia n. 938, h. Cal- cutta); Sikkim (T. Anderson n. 1941, h. Calcutta); Khasia (Griffith n. 167, h. Calcutta); Bengal (Wallich n. 3308 B, A. Calcutta); Chota Nagpore (Wood, h. Calcutta); Arracan (Kurz, h. Calcutta); Pegu CMaclelland, h. Caleutta).} [Forma pectiniformis, B. Nees.—Pegu (Maclelland, h. Oal- cuta). Ins. Andaman (Kurz, hh. Kew, Calcutta). Java (Zollinger n. 448, h. Kew; Horsfield n. 1058, h. Mus. Brit.). Borneo (Motley n. 275, h. Kew). China: Whampoa (C. Wright n. 556, h. Kew) ; Canton (Samp- son n. 257, h. Kew). . Ins. Philippine (Cuming n. 546, hh. Kew, Mus. Brit.). America calida, frequens: Florida (Chapman, hh. Mus. Brit., Calcutta); New Orleans (Riddell, h. Kew; Drummond, h. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. ” 99 Calcutta); Carolina (Bose, h. Mus. Brit.=“C. amoenus, Bosc”); Ins. Bahama (Brace n. 365, k. Kew); Jamaica (Pur- die, h. Kew); Martinique (Husnot n. 3, hh. Kew, Mus. Brit.) ; Porto Rico (Wydler n. 348, h. Mus. Brit.) ; Nicaragua (Tate n. 498, hh. Kew, Mus. Brit.); Chagres (Fendler n. 352, h. Kew); Venezuela (Moritz n. 681, h. Kew); Trinidad (Fend- ler n. 896, h. Mus. Brit.); Guiana Anglica (Schomburgk n. 882, hh. Kew, Mus. Brit.); Neo-Granada (Schott n. 13, h. Mus. Brit.); Cartagena (Schott n. 25, h. Mus. Brit.); Ama- zon Superior (Traill nn. 1171, 1172, h. Kew); Santarem (Spruce nn. 85, h. Kew); Brasil (Burchell nn. 1367, 6457, 6927, 9596, 9694, k. Kew). 56. C. LvorpULUS (Klein; Link, Jahrb. iii. p. 86); luteo- viridis; umbella laxa, simpliee vel composita ; spiculis majusculis, Spicatis; glumis remotis; nuce magna, ovoidea aut ellipsoidea, acuta, triquetra.— Kunth, Enum. ii. p. 61. C. Zoliingeri, Steud, ! in Zoll. Verz. Ind. Arch. p. 62, Cyp. p. 17 ; Boeck. in Linnea, xxxvi. p. 352. C. tenuiculmis, Boeck.! in Linnea, xxxvi. p. 286, non in Flora, 1879, p. 554. C. Schweinfurthianus, Boeck.! in Flora, 1879, p. 553. C. compressus, var., Thwaites! Enum. Pl. Zeyl. p. 342; Herb. Ind. Or., Hook. f. et T. Thoms. C. rotundus, Mig. ! Fl. Ind. Bat. Suppl. pp. 260, 600. C. rotundus, var. carinalis, Benth.! Fl. Austral. vii. p. 280. C. Wightii, Hance! in Trimen Journ. Bot. 1878, p. 14, non Nees. Cyperus, Wall. ! List n. 3321. Cyperus, Wall. ! List n. 3367. Cyperacea indeterminata, Wall. h. n. 6. l Rhizoma breve, lignosum, horizontale, non crassum. Culmi solitarii aut 2—3ni, 2-6 dm. longi, tenues, trigoni, apice triquetri. Folia 3-4, angustiora, cum $ parte culmi sepe equilonga, parum scabra. Involucri bracteæ 3—4, usque ad 2-3 dm. longs, patule. Umbelle radii 4-8, inzequales, usque ad 2 dm. longi, rarius um- bella contracta, imo capitata, visa est ; radii tenues ; ochrew usque ad 2-4 cm. longi, acute, interdum foliigere. Umbellule stepius € spica simplice constantes, haud raro congestim cory mbose, rarius evolute, radiolis 1-3 cm. longis. Spices rhachis 1-2 cm. longa, basi subnuda, 4-16-spiculosa. Spicule longe 3 cm., late 4 mm., 20-floree, compress, apice acute, inferiores sepe eaudato- bracteata ; rhachilla angustissime hyalino-alata. Gume majus- H 100 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. culæ, ovate, subobtuse, compresse vix carinate, in dorso viridi- 5-7-nervie, in marginibus estriate scarioss, rubro-maculate ; margines primum imbricati, proventu autem liberi nucem am- plectentes. Stamina 8, subinclusa; anthere lineari-oblonge, mutice. Stylus nuce multo brevior; rami 3, e gluma breviter exserti. Nux cum $ parte glume equilonga, nigra; cellule ex- time quadrate, hyaline persistentes.—H:ee species (si non C. lucidulus, Link) est C. lucidulus, Klein, in Kunth, nomen ceteris veterius. Africa tropicalis, India, China, Malaya, and Australia. Africa: Gaboon flumen (Soyava n. 355, h. Kew); Niger flu- men (Barter n. 1573, h. Kew); Abbeokuta (Irving, h. Kew); Djur (Schweinfurth n. 2318, h. Kew). India orientalis, alt. 0-1600 metr. a Nepaul usque ad Sin- gapore: Nepaul (Wallich nn. 3321, 3367, hh. propr., Cal- cutta); Assam (Jenkins n. 696, h. Calcutta); Suddiya ( Grif- Jith, h. Calcutta); Khasia, Mansmai, alt. 1200 metr. (C. B. Clarke n. 16188); Bengalia or. (Griffith n. 1601, A. Kew, n. 6210, A. Calcutta); Pundua (J. D. Hooker n. 360, h. Kew); Bhotan (C. B. Clarke n. 18529, h. Kew); Ben- galia, Comilla, Sylhet, &c. (C. B. Clarke nn. 7124, 16198, 19146, 24995, &c.); Chittagong colles (Wood, h. Calcutta); Montes Anamallay, alt. 1600 metr. (Beddome, h. propr-); Zeylania (Thwaites, C. P. n. 800 partim A. Kew, n. 807 hh. Mus. Brit., Calcutta); Malacca (Griffith, h. Kew n. 6209); Siam (Lebeuf n. 589, h. Kew); Singapore (Kurz, h. Calcutta). China: Whampoa (Hance n. 19480, h. Calcutta); Hongkong (C. Wright n. 560, h. Kew). Java: Buitenzorg (Kurz n. 1885, k. Calcutta) (Zollinger n. 2689, h. Mus. Brit. ; Horsfield, h. Mus. Brit.). Bangka (Amand, h. Kew). Borneo (Barber n. 305, h. Kew ; Motley n. 90, h. Kew). Australia: sinus Rockingham (Dallachy, h. Kew ; O'Shaugh- nesy, h. Kew). 57. C. radians (Nees et Meyen in Linnea, ix. p. 285, nomen nudum, in Pl. Meyen. p. 63); caule brevi; umbella specie sim- plice radiis longissimis; spicis globosis, multispiculosis, pluri- bracteolatis ; Spiculis compressis ; glumis navicularibus, laxiuscule imbricatis, acutis submucronatis.—Boeck. in Linnea, XXXV. p. 515 partim (?); non Benth. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS, 101 C. radicans, Kunth, Enum. ii. p. 95 (errore typogr.). Radix fibrosa; culmus 2-3 em. Folia 2-3, usque ad 5-8 em. longa, 2-4 mm. lata, scabra. Umbelle radii 4-8, usque ad 8-12 em. longi; ochre 1-2 em. longs, castaneæ, truncate aut spa- thacee. Umbellule (capitula) 8-15-spiculose ; involucelli bracteæ 9-4, usque ad 8-12 mm. longs. Spieule longe 12 mm., late 9-4 mm., 10~-16-flore ; rhachilla subexalata; glume ime spicu- larum inconspicus. Glume ovate, in dorso late virides 5-9- nervos, in lateribus sanguineo-fusce obscurius striate. Stamina 2-3; anthere anguste oblonge, mutice, rubre. Stylus nuce brevior; rami 8, e gluma breviter exserti. Nux 3 partem glume superans, ellipsoidea, utrinque acuta, triquetra, nigra; cellule extime quadrate, hyaline, lucidæ, persistentes. In oris Burm: australis (Kurz, h. Calcutta). - 58. C. GRIFFITHIT (Steud. Cyp. p. 316); caule brevi; umbella composita, radiis longissimis ; spiculis 1-5nis fasciculatis, tereti- bus, turgide lanceolatis ; involucellis longis, erectis; glumis late ovatis, spinelloso-mueronatis, multistriatis. C. radians, Benth. Fl. Hongk. p. 386 ; Boeck. in Linnea, xxxv. p. 515, partim. C. macropus, Mig. Fl. Ind. Bat. Suppl. p. 599? Cyperus, Wail. ! List n. 3371, B partim. u Radix fibrosa. Culmi cespitosi, robustiores, modo abbreviati 1-5 em. longi, modo 10-15 em. Folia 2-5, culmum superantia, plana, carinata, fere levia. Umbelle radii 3-11, longissimi, in forma etiam subacauli usque ad 35 cm. aliquando elongati, in exemplis evolutis iterum umbellatim divisi; involucri bractew 8-7, late 8 mm., patenti-erecte, cum umbella subæquilonge ; ochreæ 1-2 cm. longæ, longiuscule 2-dentatæ ; umbellulæ 3-6- radiate, radioli usque ad 4-15 cm. longi, bractew MM folieformes, cum umbellulis subequilonge. Spice A ini 1- bracteatæ, sæpe l-pauci-spiculosæ; bracteolæ paullo is an c5 1-2 cm. longæ, corymbosæ (i. e. spice revera composi! D " cule radiatim digitate, longo 12 mm., late 4 mm., n i scu 5 compress, 6~-12-flore ; rhachilla excavata, subexalats 8 ume ime minores, vacus, bracteeformes, conspicuæ- i 5 motiuscule sed spisse, per margines (fructus tempore 5 " sese late arcte integentes, in dorso rotundate spe e » carinae, mucrone valido curvato excurrente, pallic e dur aut subcastanes ; nervi 10-18 per totam fere latitudinem glum 102 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. sparsi, sepius brunnei. Stamina (nisi in exemplis parvulis) 3 ; filamenta persistentia; anthere oblongs, mutice, rubre. Stylus nuce fere brevior; rami 3, e gluma breviter exserti. Nux ellip- soidea, subobtusa, acute triquetra, cum 3 parte glume equilonga, nigra; cellule extime quadratze, marcescentes, lucide albide, hya- line, non porose. Asia australi-orientalis. Malacca (Griffith, h. Kew; Maingay nu. 2987, 3191, h. Kew n. 1721). Tenasserim (Helfer, h. Kew n. 6209). Singapore (Wallich, h. Linn. Koc., n. 3371 B h. Calcutta). China: Macao (Hance n. 10135, h. Kew); Hongkong (C. Wright n. 565, h. Kew). Cochinchina (Lebeuf n. 832, h. Kew). Bangka (Teysmann, h. Kew). Borneo: Banjermassing (Motley nn. 55, 1135, h. Kew); Sara- wak (Beccari n. 3686 ; Barber, h. Kew). I am completely puzzled by this plant. When I took up the bundle of C. radians at Kew, it contained a large number of speci- mens, from which I compiled the description here given of C. Griffithii, and also extracted the select localities here quoted for C. Griffith. The plant differed from all other Cyperi I had then seen by the very short culms and enormously long rays. Nothing caused me to suspect more than one species in the bundle; I feel sure that nearly all of it was the plant here described as C. Grif- Jithii, though it is possible I may have hastily overlooked some specimens of true C. radians; the marked character of the short culm might seduce any one into hastily naming a specimen on that obvious point without looking further. But at Kew I was quite at a loss to imagine how Boeckeler had got the species among the Compressi; the spikelets are so terete that I can compare them in that respect with nothing except C. canescens, Vahl. In the Calcutta Herbarium, however, I find K urz's Burmese excellent specimens, also with short culm, very long rays, though with totally different spikelets, exceedingly like those of C. com- pressus. l might think it a trick of memory ; but, as the Cal- cutta Herbarium also has an excellent specimen of C. Grif- Jithii, I see by placing the two side by side that, though both have very short culm, very long rays, the spikelets are as different as well can be in Cyperus; and on further examination the two are found to differ widely in the glumes, inflorescence, bracts, in short at every point. | MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 108 lassume here in Calcutta two things—Ist, that Kunth's C. radicans is certainly Nees's radians; Kunth has marked it v. s. ; his description, * squamis navieularibus, multinerviis," as well as of the simple umbel, * capitulis subglobosis polystachyis," proves that he had before him Kurz's plant,and not the Kew-bundle C. radians. Secondly, it is clear that Boeckeler must have had this plant in his mind, and not the Kew-bundle plant, when he placed his C. radians in sect. Compressi. But from Boeckeler's quoting Griffith's Malacca collection for C. radians, it would appear that Boeckeler has, from the striking character of short culm, very long ray, rolled the two into one despite the other enormous differences. We have in Cyperus numerous instances of the same peculiar type of inflorescence turning up in very different parts of the . genus, as C. cephalotes, pygmeus, dubius, confused much by the old authors. At the same time it seems so very improbable that, over exactly the same area, two species both having the strange habit of C. radians should be found, that I have imagined the possibility of C. Griffithit being a monstrous or semi-bisexual state of C. radians. (Very curious differences in the structure of the spikelets accompany some such state among the sect. Ele- gantes.) But both C. radians amd C. Griffithii appear perfectly normal with ripe nuts; if they are one species, I do not know why C. canescens may not be a state of C. compressus. I give the matter up for the present. 59. C. Scawernrrzirt (Torrey in Ann. Lyceum Ne ew York, iii p. 276); bracteis erecto-adpressis, umbellam simplicem laxam paullo superantibus; spieulis majusculis, antice fastigiato-sprcarls glumis distantibus, ovatis, mucronatis, striatis; nuce magna, tri- quetra.— Boeck. in Linnea, xxxv. p. 518. Lee tri- Rhizoma repens, lignosum. Culmi 3-5 dm. long, ape T quetri spinuloso-scabri. Spiculæ longe 15 mme, maa 10-16-floræ. Chicago (Babcock, h. Calcutta). Prod. p. 20, FT. Ind. Occ.i. p.115); spicis breviter spicatis, in lateribus 60. C. conrEertus (Swartz, debilis, fusco-viridis ; umbella simplice A re 6-9-floris ; glumis carinatis, in dorso viridi-7-nerviis, fuscis vix luteis—Boeck. in Linnea, XXXV. P. 514. Columbia, &c. (Jide Boeckeler). 104 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 61. C. oxasER (Linn. Mant. p. 179); umbella specie simplice, congesta vel contracta; spiculis numerosis, capitato-spicatis, longis, multifloris, compressis; staminibus 3; nuce cum 1-2 parte glume :equilonga, obovoidea, trigona, apice brevissime rostrata.— Boeck. in Linnea, xxxv. p. 516. C. patulus, Kitaib. in Host, Gram. iii. p. 49, t. 74; Kunth, Enum. i. . 24. C. fuscus, var. virescens, Kunth, Enum. ii. p. 38, nec C. virescens, Hoffm. l C. erythræus, Kunth, Enum. ii. p. 27, non Schrad. ( fide Boeck.). Glume ovate, obtuse, submucronate, in dorso late virides 5-7-nerviæ, in lateribus anguste rubro-striatæ, in marginibus scariose. Stamina 3; filamenta fere levia; anthere lineari- oblong: flavide, a crista parva, subquadrata, coccinea cristate. Sicilia (Todaro, h. Calcutta). Ischia (Bentham, h. Calcutta). Tauria (h. Calcutta). Iberia ( Wilhelms, h. Calcutta). 62. C. RUBICUNDUS (Vahl, Enum. ii. p. 308); umbella capitata; spiculis majusculis, compressis, oblongis, rubieundis vel einna- momeis ; involucri bracteis 2-3, capitulum equantibus vel duplo superantibus ; nuce obovoidea aut late ellipsoidea, quam 3 pars glume breviore.— Kunth, Enum. ii. p. 49; Webb et Berth. Iles Canaries, Phyt. iii. p. 861, t. 240 (nuce nimis elongata, acuta) ; Boeck. in Linnea, xxxv. p. 507. C. pectinatus, Roxb. ! FI. Ind. i. p. 190, non Vahl. C. Teneriffe, Poir. Encyc. vii. p. 245; Roem. et Sch. Syst. ii. p. 209; Nees in Linnea, x. p. 131 (?). C. coromandelinus, Spreng. Syst. i. p. 217, non Boeck. C. Wightii, Nees! in Wight Contrib. p. 78; Kunth, Enum. ii. p. 99; Boeck. in Linnea, xxxv. p. 507. C. Serra, A. Rich.! Fl. Abyss. ii. p. 479. C. petrus, Hochst. in Hohenack. n. 1293. C. nitens, Rottler in h. propr., non Vahl. Cyperus, Wall.! List n. 3314 B, A partim. Radix annua, fibrosa. Culmi 1-2 dm. longi, cxspitosi, basi sæpe incrassati, sursum trigoni, leves. Folia 2-3, anguste gra- minea, sæpius culmo multo breviora, vel (in exemplo typico Neesiano C. Wightii) longiora; vaginæ swpe majuscule, laxe? Involueri braetes usque ad 2-5 cm. longs, sepius capitulum parum superantes, patule. Spicule 4-15, longe 23 mm., late MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 105 4-6 mm., vel in forma C. Wightii minores, admodum compresse, lateribus parallelis; rhachilla subexalata. Glume ovate, acumi- nate, mucronate, naviculares, carinatee, 9-13-nervie ; nervi rubi- cundi aut pallide brunnei per totam fere latitudinem glume sparsi; latera lete rubicunda, aut sepius pallidiora. Stamina 3; filamenta persistentia ; anthere oblonge fere multice. Stylus nuce longior; rami 8, e gluma paullo exserti. Nux in planta indiea late obovoidea, obtusa, triquetra, in planta canariensi paullo elongata ellipsoidea (nec conico-acutata), atro-fusca ; cellule extime quadrate, lucide albide, persistentes.— Hujus speciei nomen forsan antiquissimum certum est C. Teneriffe ; Vahl suum C. rubicundum a Porto Rico receperat ; C. rubicundus autem Kunth in Orbe Nova adhue non reperta est. Teneriffe (Bowrgeau nn. 463, 1566, h. Kew, n. 463, h. Mus. Brit. ; &e.). Africa australis (H. Barkly n. 5812, h. Kew). Africa tropica: Ukamba (Hildebrandt n. 2658, h. Kew). Abyssinia (Schimper nn. 603, 2340, nec 183, h. Kew, n. 2340 h. Mus. Brit.). Socotra (Balfour n. 106, k. propr.). Madras peninsula (Wight nn. 10, 1817, h. Kew); Carnatic (G. Thomson, h. Calcutta); montes Khoonda (Hohenacker n. 1293, hh. Kew, Mus. Brit.). Sect. C. Platystachyi. Platystachyi, Kunth, Enum. ii. p. 44 (C. arenario addito).— Compressi partim, Boeck. in Linnea, xxxv. p. 505 (sp. aliis additis).—Sorostachys, Steud. Cyp. p. 71 (genus). ae Rhizoma lignosum (nisi C. leucocephalus). Culmi foliati. Umbella congesta aut capitata. Spiculæ admodum compress®æ, spississimæ, stramineæ aut cinnamomeg. Glumæ ovatæ, sub- obtuse, multinerviæ. Stamina 3 (lin C. leucocephalo) ; filamenta mediocriter ligulata; antheræ lineari-oblongæ, mutice. Nux cum 3-3 parte glume equilonga, obovoidea, obtusa, trigona au triquetra, interdum inequalis a dorso compressa. edt This group consists of very similar species easily recognize y the subcapitate inflorescence and the colour. The nex group, Bobartia, differs in the woolly roots, the very broad ligu te filaments, and very rigid leaves. It might be more PAP throw these two small groups into one, distinguished e y by the pale colour and subcapitate inflorescence, and divide 1 106 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. into two subsections, one with woolly roots, one with non- woolly. 63. C. AnENARIUS (Retz. Obs. iv. p. 9); culmis in rhizomate repente diviso solitariis, distantibus, apice ipso subteretibus ; umbella capitata aut breviradiata ; involucri bracteis 2-3, altera multo longiore, suberecta; spiculis majusculis; latis, stramineo- brunneis; nuce obovoidea, a dorso compressa, inzquali, fusco- nigra.—JNVees in Wight Obs. p. 77; Kunth, Enum. ii. p. 46; Dalz. § Gibs. Bomb. Fl. p. 284; Boeck.in Linnea, xxxv. p. 536. C. persicus, Boiss. Diagn. 1 ser. xiii. p. 39. C. conglomeratus, var. y, Cosson, Expl. Algér., Bot. ii. p. 245; Boiss. Fl. Orient. v. p. 369, non Rottb. C. levigatus, Koenig et Roxb. MSS. in Mus. Brit., non Linn. Cyperus, Wall.! List n. 3314, A partim. Cyperus, Wall.! List n. 3535. Scirpus, Wall. ! List n. 3460. Rhizoma elongatum, repens, tenue, lignescens, a vaginis aureo- castaneis intectum, dichotome ramosum. Culmi remoti, solitarii, ima basi in rhizoma attenuati, 1-2 dm. longi, sepe crassiusculi. Folia 2-5, cum culmo fere zquilonga, crassiuscula, in sicco curvata, marginibus involutis. Involucri bractes sspissime 2; altera usque ad 4-8 cm. longa, quasi culmum producens, basi non dilatata (fructus tempore autem interdum divaricata), altera multo minor. Umbella 3-4-radiata, radiis usque ad 8-15 mm. longis, 3-7-spiculigeris, aut sepius subcapitata. Spicule longs 15 mm., late 5-8 mm., late lanceolate, compress», 10—16-flore ; rhachilla subexalata. Glume dense stipate, spisse, concave vix carinatz, late ellipticw, obtuse aut obsoleto-mucronats, stra- mine, per totam fere latitudinem a nervis 9-15 pallide brunneis longitudinaliter percurse, in marginibus stramineo- hyaline. Stamina 3; filamenta ligulata, persistentia ; anthere majuscule, lineari-oblongs, mutice, e gluma breviter exsertz. Stylus nuce multo brevior; rami 3, e gluma exserti. Nux cum 3-4 parte glumee zquilonga, obtusa, in facie interiore subconcava, in dorso convexo curvato obtuse angulata nee carinata; cellule extime quadratz, hyaline, persistentes. In oris Indie a Bengalia usque ad Zeylaniam et Sinum Persicum. Orissa (JV. S. Atkinson, h. Kew). Carnatic (Wight n. 2920, h. Calcutta; G. Thomson, h. Cal- cutta). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 107 Madras (Wight n. 1816, hh. Kew, Mus. Brit. n. 2389 h. Mus. Brit.). Zeylania (Thwaites C. P. n. 798, hh. Kew, Mus. Brit.). Bombay (Dalzell, h. Kew). Marwar: Mallarree (G. King, h. Calcutta). Seinde (Stocks, h. Kew; Dalzell, h. Calcutta). Sinus Persicus: ins. Karek (Kotschy n. 5, hh. Kew, Mus. Brit.). Persia australis (Aucher n. 5483, h. Kew, non n. 5483 h. Paris). Ava (Wallich n. 8535, h. Calcutta). 64. C. LEUCOCEPHALUS (Retz. Obs. v. p. 11); culmis cespitosis, basi bulboso-incrassatis ; umbella capitata, globosa, parva; invo- lueri bracteis 3, divaricato-deflexis ; spiculis lanceolatis, com- pressis, multifloris; glumis elliptico-oblongis, obtusis, albidis aut cinnamomeo-brunneis ; nuce cum 4 parte glume equilonga, ellip- soidea, trigona.— Boeck. in Linnea, xxxv. p. 590, non Hassk. C. pulchellus, R. Br. Prod. p. 213; Kunth, Enum. ii. p. 110; Benth. Fl. Austral. vii. p. 265. C. Schomburgkianus, Nees! im Hook. Journ. Bot. ii. (1840) p. 323, in Mart. Brasil., Cyp. p. 18; Boeck. in Linnea, xxxv. p. 589. C. sorostachys, Boeck. in Linnea, xxxv. p. 588. Sorostachys kyllingioides, Steud. Cyp. p. 71. Lipocarpha, Wall. ! List n. 3445, D. Cyperus sp. n. 25, Wall. h. propr. (n. 3536 bis). Cyperus sp. n. 46, Herb. Ind. Or., Hook. f. et T. Thoms. Radix fibrosa ; vel interdum (fide Boeckeler) rhizoma obliquum. Culmi 8-18 em. longi, graciliores, trigoni. Folia 2-8, cum culmo sæpe zquilonga, angusta. Involueri bractes usque ad 3-10 cm. longs. Capitulum 5-10 mm. in diam., densum. Spicule longs 5-9 mm., late 2-3 mm., 10-18-flore ; rhachilla subexalata. Glume concave, marginibus involutis, vix carinate, in dorso obscurius 1-3-nervie, in lateribus enervie, scariosæ. Stamen 1, sub- inclusum; filamentum scabrum ; anthera oblonga, mutica. Stylus brevis; rami 3, e gluma vix exserti. Nux obtusiuscula, nigra ; cellule extimæ quadratz, albide, emarcide, reticulate.—Spicule (cum glumis) iis C. nivei simillime sed multo minores. India orientalis, late sparsa nec vulgaris: Behar, Monghyr (Buchanan- Hamilton, h. Wall. propr-) Bengal, Burrakur (Kurz, h. Calcutta) ; Canara et Mysore (Law, hh. Kew, Mus. Brit.) ; Malabar (Stocks, h. Calcutta) ; Pegu (Kurz n. 647, h. Calcutta) ; Moulmein (R. Scott, h. Calcutta) ; Tenasserim 108 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. (Helfer n. 6247/1, hh. Kew, Calcutta) ; Amherst (Wallich, h. propr.). Ins, Philippine (Cuming n. 1417, hh. Kew, Mus. Brit.). Australia tropicalis (E. Brown n. 5917, hh. Kew, Mus. Brit.); Queensland, Rockhampton (Amalia Dietrich n. 600, h. Mus. Brit.). Brasil (Burchell nn. 8186, 8267, h. Kew). Guiana Anglica (Schomburgk n. 810, h. Kew). Kunth describes (Enum. ii. p. 44) C. coronarius from Bengal, which he says was Scirpus coronarius, Vahl, Enum. ii. p. 261, et herb., from which I infer Kunth had Vahl’s specimen; and he says the style was 3-fid (not 2-fid as Vahl), which he would only have said from a dissection of Vahl’s plant. There is no Bengal plant under this name in the English herbaria nor in Calcutta ; nor is there any Bengal plant that will answer the description except C. niveus, which common species Kunth knew too well to describe over again under the name O. coronarius. O. coronarius might be C. leucocephalus, which has sometimes (fide Boeckeler) * rhizoma obliquum." 65. C. nrvEus (Retz. Obs. v. p. 12) ; culmis in apice rhizomatis repentis obliqui dense uniseriatim stipatis, incrassatis, apice ipso trigonis; umbella capitata; spiculis majusculis, compressis, albidis aut cinnamomeis ; nuce late obovoidea, obtusissima, acute subeequaliter triquetra, faciebus 3 subconcavis.— Aoxb. Fl. Ind. i. p. 191; Nees in Wight Contrib. p. 785. Kunth, Enum. ii. p. 45; Boeck. in Linnea, xxxv. p. 530 (non in Flora, 1879, p. 550). Cyperus, Wall. ! List n. 3377. Rhizoma lignosum, tenuius. Culmi 1-5 dm. longi. Folia 3-6, cum 2 parte culmi æquilonga, vel abbreviata, marginibus in sicco involutis. Involucri bracteæ sæpissime 2, longæ 2-6 cm., deflexo- patentes, basi non dilatatæ. Spiculæ 3-15, usque ad 2 em. longs, 6 mm. late, 20-40-flore, admodum compressa, utrinque angus- tate; rhachilla exalata. Glume dense imbricate, spissæ, ovate, obtuss, compress, in dorso concoloriter 1-nervie, in lateribus plicatim multistriate. Stamina 3; filamenta anguste ligulata, persistentia; anthere lineari-oblonge, mutice, e gluma vix ex- serte. Stylus nuce brevior; rami 3, e gluma breviter exserti. Nux eum } parte glume equilonga, in apice truncata, sub- depressa, nigra; cellule extime quadrate, albidz, hyaline, de- tergibiles (7. e. * nux reticulatim albo-velata’’). ~- presan ne e UNDDENEE-eT 3 MR. C. B. CLARKE ON INDIAN SPECIES OF OYPERUS. 109 ln omni India orientali eum Burma; a Zelania nondum recepta. Montes Suleiman (Sanders n. 9, h. Calcutta). Beloochistan (Stocks n. 617, h. Kew). Kashmir (JW. S. Atkinson n. 24917, h. Kew). Punjab (Jacquemont n. 951, h. Kew; Stewart nn. 289, 383, 782, &e.). Kumaon, alt. 2000 metr. (Strachey et Winterbottom n. 4, h. Kew). Simla, alt. 2000 metr. (G. King, h. Calcutta, exemplum ma- crum). Kulu, alt. 2000 metr. prope Rotang Pass (Stoliczka, h. Oal- cutta). Panwi, alt. 2400 metr. prope Bursahir (Brandis, h. Calcutta). Kishtwar, usque ad 2500 metr. (Stoliczka, h. Calcutta). Nepaul ( Wallich u. 3377 B, k. Calcutta). Sikkim (J. D. Hooker, G. King h. Calcutta). Assam (Griffith n. 1610, hh. Kew, Mus. Brit., n. 1467, h. Mus. Brit. ; Fisher, h. Calcutta). Bengal orientalis et Burma (Grifft& nn. 6170, 6171, hh. Kew, Calcutta). l Bengalia septentrionalis occidentalisque, vulgaris (C. B. Clarke &e.). Rajmahl et Raneegunje (Kurz, h. Calcutta). Chota Nagpore (Wood, h. Calcutta). Jodhpoor (G. King, h. Calcutta). l Awrungabad (Hardwicke, h. Mus. Brit.). Madras peninsula (Wight n. 1818, h. Kew ; Campbell, h. Mus. Brit.). 66. C. ATKINSONI ; radicibus non lanosis ; eulmis ceespitosis, basi incrassatis, apice teretibus; umbella contracta ; spiculis nn digitatis, linearibus, densifloris, pallide brunneis; nuce angustius obovoidea, trigona, a tergo vix compressa. C. niveus, hh. Royle et Stocks, non Retz. C. effusus, 4. Dalzell, non Rottb. : Culmi in rhizomate lignoso 2-6ni, fasciculati, 10-16 cm. M inferne trigoni. Folia prope basin culmi plurima, oo eats. equilonga aut sepius multo breviora, rigida, 1n sico m M marginibus involutis, obscurius scabris. Involueri ne i 2 altera usque ad 5-8 cm. longa, junior suberecta, Umbelie r 110 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 3—4, usque ad 1-3 cm. longi, vel umbella fere congesta. Spicule usque ad 18 mm. longs, 4-5 mm. late, 20-30-flore, admodum compress, lateribus fere parallelis ; rhachilla subexalata. Glume spisse, ovate, obtusiuscule, breviter mucronate, naviculares, per totam fere latitudinem multistriate, nervis cinnamomeis. Sta- mina 3; filamenta ligulata, persistentia ; anthere lineares, mutica. Stylus nuce subbrevior; rami 3, e gluma plane exserti. N ux j parte glume longior, obtusa, in facie interiore vix concava, in dorso angulata nec carinata, fusco-castanea; cellule extime quadrate, minute.—C. niveo affinis, sed C. niveum umbella radiata nondum visi. India boreali-occidentalis (Royle, h. Kew); Punjab (Clark n. 142, h. Calcutta); Kashmir, Bhimbur (Atkinson n. 24196, h. Kew, speciei typus); Scinde, Kurrachee (Stocks, h. Kew ; Dalzell, h. Calcutta); Bombay (Dalzell, h. Kew). 67. C. OBTUSIFLORUS (Vahl, Enum. ii. p. 308); culmis robustis ; umbella capitata, c. 6-spiculosa; spiculis longis 16 mm., latis 8 mm., admodum compressis, spissis, albidis; glumis ellipticis, obtusis, striatis.— Kunth, Enum. ii. p. 45; Boeck. in Linnea, XXXy. p. 528. C. spherocephalus, 6. leucocephalus, Vahl, Enum. ii. p. 308 ; Kunth, Enum. ii. p. 45. Natal (Grant, h. Calcutta); alt. 0-3000 metr. (Sutherland, h. Calcutta). Var. B. flavissima, Boeck. ; spiculis aureo-flavis. C. flavissimus (sp.), Schrad. Anal. Fl. Cap., Cyp. p. 5, t. 2. fig. 2. C. sphzerocephalus, Vahl, Enum. ii. p. 313; Kunth, Enum. ii. p. 45. Africa australis (Burke, h. Calcutta). 68. C. MARGARITACEUS (Vahl, Enum. ii. p. 307); culmo tenui- ore; umbella capitata, c. 5-spiculosa; spiculis maximis, 20-floris, admodum compressis, lucidis, stramineis; glumis dense imbri- catis, acute carinatis, 19-nerviis.— Kunth, Enum. ii. p.46; Boeck. in Linnea, Xxxv. p. 529. . Niger flumen ; Nupe (Barter, h. Calcutta). Sect. D. Bobartia. Bobartia, Linn. Fl. Zeyl. p. 17 (genus), non Lam.—Galilea, Parl. Fl. Palerm. i. p. 297 (genus).—Sect. Repentes, Boeck. in Linnea, xxxv. p. 536 (C. arenario excl.; C. proteinolepide incl.). MR. C. B, CLARKE ON INDIAN SPECIES OF CYPERUS. 111 Radices lanosi. Folia rigida. Umbella congesta aut capitata. Spieule spissz, stramines aut pallide brunnes. Glume ovate, obtuse aut breviter mucronate. Stamina 3; filamenta late ligu- lata; antheræ majuscule, lineari-oblonge, mutice. Nux cum 4-3 parte glume equilonga, obovoidea truncata, rarius ellipsoidea, inequaliter trigona, in facie interiore plana vel concava, in dorso curvata, compressa; stylus 3-fidus. The species of this small group are so closely allied that Cosson and Boissier unite several specifically with C. conglomeratus. The spikelets are less compressed than in the Platystachyi. The nut and style of C. stoloniferus, Retz., agree closely with this section, and it should perhaps be placed here; but the rhizome is nearer that of C. rotundus : it is, indeed, so continually mis- taken, even by experienced botanists, for a form of C. rotundus, that it is more convenient to arrange it there whether that be its true affinity or not. 69. C. PACHYRRHIZUS (Nees; Boeck. in Linnea, xxxv. p. 545); robusta, radicibus crasse lanosis ; umbella contracta aut capitata ; involucri bracteis 2-3, divaricatis; glumis ovatis, obtusis, minute mucronatis; nuce parva, contra rhachin compressa; stylo quam nux breviore, ramis 3 e gluma vix exsertis. C. conglomeratus, Thwaites, Enum. Pl. Zeyl. p. 343. C. leucocephalus, Wight in h. propr. n. 2381, non Retz. — Bobartia, Linn. Fl. Zeyl. p. 17, non Lam. Rhizoma longe repens, lignosum, tenuius, eastaneo-brunneum. Culmi 1-3ni, longi 1-5 dm., basi multifoliati, castanel, apice crassi, trigoni, straminei. Folia cum culmo sepe zequilonga, lenta, marginibus in sicco involutis. Involueri bracteæ usque ad 1-2 dm. longs, foliis similes. Spice 3 cm. in diam., globose, dense. Spicule 30-80, longs 15 mm., late 5 mm., compresse, lanceolate, 8-16-flore ; rhachilla subexalata. Glume pallide, concave non carinate, 7—11-nervis, lateribus latiusculis enervus. mE 3; filamenta late ligulata ; anther lineari-oblonge, mutice. Nux cum 7-1 parte glume equilonga. Madras peninsula (Wight n. 2381, ^h. n. 105 4. Mus. Brit.) ; Concan australis Zeylania (Thwaites, C. P. n. 808, hh. Kew, Mus. Ins. Laccadive (Hume, hh. Kew, Calcutta). Var. B. minor ; culmis minoribus. l Culmi approximati, basi subincrassati, qu Kew, Mus. Brit., (Law, h. Kew). Brit.). asicæspitosi, longi 112 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 8 cm., tenuiores. Folia plura, culmo subbreviora. Involucri bractez 3, divaricate, usque ad 25 mm. longs. Capitulum glo- bosum, 12-spiculosum. Spicule omnino C. pachyrrhizi. Stylus alte 3-fidus. Aden (Hunter n. 18, h. Kew). 70. C. coxarouERATUS (Rottb. Descr. et Ic. p. 21, t. 15. fig. 7); radicibus lanosis; culmis crassiusculis, sursum teretibus; um- bella simplice, contracta aut congesta ; spiculis parum compressis, lineari-lanceolatis, viridi-stramineis aut cinnamomeo-brunneis, 8-16-floris ; nuce obovoidea, trigona, a dorso subcompressa.— Boeck. in Linnea, xxxv. p. 543; Cosson, Expl. Algér., Bot. iii. p. 243, pro majore parte; Boiss. Fl. Orient. v. p. 369, pro majore parte. C. effusus, Rottb. Descr. et Ic. p. 22, t. 12. fig. 3; Kunth, Enum. ii. . 47. C. faleatus, Boeck. in Linnea, xxxv. p. 545. C. curvulus, Boeck. in Linnea, xxxv. p. 541, partim. Asia australi-occidentalis ; Africa borealis et tropica. Beloochistan (Pierce, h. Kew); Bolan Pass (Griffith nn. 48, 185, h. Kew n. 6146). Sinus Persicus (Kotschy n. 20, hh. Kew, Mus. Brit.). Aden (Ralph n. 666, h. Mus. Brit.; T. Anderson, h. Caleutta ; J. D. Hooker, h. Calqutta; King, h. Calcutta; Brandis n. 2468, h. Calcutta ; Rod. Wiesner, h. Calcutta). Jeddah (Schimper n. 810, h. Mus. Brit., n. 810 partim, forma nana, h. Kew; Schweinfurth n. 122, ‘forma nana, h. Mus. Brit.). Dongola (Ehrenberg, h. Kew). Berber (Schweinfurth n. 645, h. Kew). Nubian desert (Petherick, h. Kew). Var. B. major (Boeck. in Linnea, xxxv. p. 544); spiculis multo majoribus, longis 2 cm., latis 5 mm., 30-floris, glumis densissime stipatis. C. densus, R. Br. ! in Salt. Abyss. Append. p. 62, nomen nudum. C. proteinolepidi fortasse affinior. In exemplo Thomsoni, Spi- cule longe 4 cm., 60-flore. Arabia (Ehrenberg, h. Kew; Schimper n.'801, h. Mus. Brit.); Aden (T. Thomson, hh. Kew, Calcutta). Palestina (Hayne, h. Kew). Egypt: Kosser (Schweinfurth n. 2015, h. Mus. Brit.). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 118 Var. Py. socotrensis ; humillima ; spiculis multo angustioribus, anguste lanceolatis ; glumis castaneo-luteis. Culmi dense cespitosi, 1-2 cm. longi, trigoni. Involucri bracteæ 2—4, vix 1 cm. longe. Umbella e 2-5 spiculis capitata. Spicule longe 1 cm., late 1-2 mm., 8-14-flore. Socotra (Balfour n. 27, h. propr.). 71. C. PRorEINOLEPIS (Steud. Cyp. p. 15) ; radicibus lanosis ; culmis crassiusculis, sursum teretibus; umbella simpliee, con- tracta aut capitata; spiculis compressis, lineari-oblongis, lateri- bus parallelis, griseis aut pallidis, 24—40-floris ; nuce obovoidea contra rhachillam compressa.— Boeck. in Linnea, xxxv. p. 922. C. involutus, R. Br.! in Salt Abyss., Append. p. 62, nomen nudum. C. curvulus, Boeck. in Linnea, xxxv. p. 541 pro majore parte, non Boeck. in Flora, 1879, p. 549. A C. conglomerato glumis obscure mucronatis, densius stipatis, vix sat diversa. Arabia (Schimper nn. 301, 733, k. Kew). Aden (T. Thomson, h. Kew; Sir R. Schomburgk n. 8, h. Cal- cutta). Mare Rubrum (Nimmo, h. Kew). Jeddah (Schimper n. 810 partim, h. Kew; Fischer, n. 55, h. Kew; Zohrab n. 18 partim, h. Kew). Abyssinia (Salt, h. Mus. Brit.). Socotra (Balfour n. 105, h. propr.). 72. C. punaENs (Boeck. in Linnea, XXXV. p. 537); radicibus lanosis ; culmis crassis, sursum teretibus ; umbella simplice, con- tracta aut capitata; spiculis compressis, lineari-lanceolatis, lucide stramincis, 10-18-floris; nuce magna, obovoidea, contra rha- chillam multum compressa. C. conglomeratus, Decaisne, Fl. Sinai, p. 13 ; Cosson, Exp iii. p. 243; Boiss. Fl. Orient. v. p. 369 partim. m A C. conglomerato videtur sat diversa; a C. pachyrrhizo differt capitulis magis lucidis minus densis, nuce duplo longiore. s794 Africa: Egypt (Schweinfurth n. 426, h. Kew; Aucher ns ° h. Kew); Algeria (Balansa n. 944, hh. Kew, Mus. rit. ; Bourgeau n. 22, h. Kew); Sinai (Bové n. 28, ^. Kew). l. Algér., Bot. 78. C. Avcuxar (Jaub. et Spach, 11. Pl. Or. t. 101); radicibus lanosis ; culmis crassiusculis, sursum teretibus ; umbella simplice pauciradiata aut capitata; spicu is magnis, compressis, late LINN. JOURN.—BOTANY, VOL. XXI I 114 MR. €, B. CLARKE ON INDIAN SPECIES OF CYPERUS. linearibus, lucide stramineis aut albidis, 20—40-floris; nuce majuscula, obovoidea, a dorso compressa. C. conglomeratus, Cosson, Expl. Algér., Bot. ii. p. 243; Boiss. Fl. Orient. v. p. 369. Spicule longe 20-45 mm., late 6-8 mm.—Exempla optima Vogelii (quam illa a Jaub. et Spach depicta) paullo majora. Persia australis (Aucher n. 5483, h. Kew). Egypt (Aucher-Eloy, h. Calcutta). Soudan, Agadem (Vogel, h. Kew). Deserta Belbeys (Schubert, h. Kew). 74. C. maypriacus (Gloxin. Obs. p. 20, t. 8); radicibus lanosis ; umbella capitata vel congesta; glumis maximis, aristatis, per totam fere latitudinem ll-nervosis; nuce ellipsoidea sub- obovoidea, a dorso compressa.—Kunth, Enum. ii. p. 48 ; Boeck. in Linnea, xxxv. p. 541. Schoenus mucronatus, Lian. Sp. Pl. p. 63. Mariscus mucronatus, Presl; Sibth. Fl. Grec. t. 43. Galilea mucronata, Parlat. Giorn. Bot. ii. p. 134. Ins. Canaries (Lemann, h. Calcutta). Lisbon (Daveau, h. Calcutta). Montpellier (Bentham, h. Calcutta). Athens (Heldreich, h. Calcutta). Ins. Rhodes (Aucher- Eloy, h. Calcutta). Sect. E. Viscosi. Sulcati sew Glutinosi, Boeck. in Linnea, xxxv. p. 547. Radix fibrosa. Culmi obscurius viseoso-puberuli. Folia an- gusta, involuta, rigidiora. Umbella composita. Glume ovate, navieulares, mucrone breviter excurrente. Stamina 3; anthere lineari-oblonge, mutice. Nux trigona, ellipsoidea aut sub- lanceolata. This group, of three species only in Boeckeler, is held together by trivial characters, while the two species below enumerated are unlike in habit. C. viscosus itself seems near C. elegans, where Kunth placed it, differing in the narrow leaves. C. oæylepis 18 like no other species known to me. 75. C. viscosus (Aiton, Hort. Kew. i. p. 59); umbella laxa, irregulari, spieulis fasciculatis; nuce cum 4 parte glume sequi- longa, obovoidea, triquetra.— Kunth, Enum. ii. p. 28 ; Boeck. in Linnea, xxxv. p. 547. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 115 Exemplum in h. Calcutta (in Galapagos Archip., Ins. Charles a Goodridge lectum) e Kew sub nomine C. rubiginoso, Hook. f., emissum videtur C. viscosus. India occidentalis ; Mexico. 76. C. oxvrzPrs (Nees; Steud. Oyp. p. 25); viscosulus; um- bella fere simpliee, densiuscula ; spiculis dense fasciculatis, luteo- brunneis; glumis laxe imbricatis, plicato-striatis ; nuce oblonga, rostrata, eum 3 parte glume equilonga.— Boeck. in Linnea, xxxv. p. 549. Demerara (A. Calcutta, n. 324). Sect. F. Luzulotdet. Luzuloidei, Kunth, Enum. ii. p. 39; Boeck. in Linnea, xxxv. p. 549 (C. silletensi exci.). Radix fibrosa (in C. reflexo rhizoma lignosum). Folia longius- cula. Umbella composita (aut decomposita); umbellule haud raro capitatim congeste. Spicule multiflore, multum compresse ; rhachilla obscurius alata. Glume concave, obtuse. Stamen 1 (in C. Lechleri 3); anthers oblongz, mutice. Nux trigona, obovoidea aut oblonga, apice breviter vel longius acutata.— Ameri- cani. This is a very natural small group, of which only some characteristic species are taken up here. The Indian C. sille- tensis, included among Luzuloidei by Boeckeler, appears to me to differ considerably, and is exceedingly near C. pulcherrimus, Willd. Kunth included here C. platystylis, R. Br. (under the name C. pallidus), probably because he had no better place to put it, and objected to making a section of a solitary species out of it, as I have done as the lesser evil. 77. C. vecerus (Willd. Sp. Pl. i. p. 283); culmo sursum trigono, levi; umbella 5-6-radiata ; um bellulis globosis, capitatis 10-50-spiculosis ; glumis lutescentibus aut virescentibus ; pace cum 1—: parte glume æquilonga, triquetra, obovoidea, breviter rostrata.— Boeck. in Linnea, xxxv. p. 990. l a Kunth’s statement (Enum. ii. p. 40) that the nut 15 only i-i the length of the glume is misleading. Valdivia (Philippi n. 544, h. Calcutta). Juan Fernandez (Scanlin, h. Calcutta). > I 116 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 78. C. Deum{oxnnii (Hook. et Torrey in Ann. Lyceum New York, iii. p. 437) ; culmo sursum triquetro, scabro ; umbella 5-6- radiata; umbellulis globosis, 80-40-spiculosis; glumis luteo- brunneis aut fuscescentibus ; nuce cum 4-3 parte glume equi- longa, trigona, apice breviter acutata, basi conspicue stipitata.— Boeck. in Linnea, xxxv. p. 592. C. distinctus, Steud. Cyp. p. 24. Texas (Drummond n. 449, h. Calcutta). 79. C. vrnENS (Mich. Fl. Amer.-Bor. i. p.29) ; umbella specie composita, spicis globosis, multispiculosis; spiculis 30—40-floris, lutescentibus; nuce cum 1-$ parte glume equilonga, angusta, apice acuminata subrostrata.— Kunth, Enum. ii. p. 40; Boeck. in Linnea, xxxv. p. 553. Florida (5. Calcutta). 80. C. suntNAMENSIS (Rottb. Descr. et Ic. p. 35, t. 6. fig. 5); umbella composita aut decomposita, densiuscula; spicis globosis, multispiculosis ; spiculis 30—40-floris, pallide luteis; nuce cum l parte glume equilonga, oblonga aut anguste ellipsoidea, utrinque subobtusa.— Kunth, Enum. ii. p. 43; Boeck. in Linnea, XXXV. p. 554. C. virenti simillima. Ins. Trinidad (5. Calcutta). Florida (Chapman, h. Calcutta). 81. C. cxgTOLEPIS (Torrey et Hook. in Ann. Lyceum New York, ih. p. 436); umbella decomposita (specie composita) ; umbellu- larum radiolis pluribus, divaricatis aut deflexis; spicis 1—4 in capitula ovoidea pyramidatis, pallidis; nuce cum } parte glume æquilonga, oblonga, subovoidea apice breviter angustata.— Boeck. in Linnea, xxxv. p. 557. C. Luzule affinis. Florida: Augusta (Wray, h. Calcutta; Chapman, h. Calcutta). Texas: San Felipe (Drummond n. 450, h. Calcutta). 82. C. REFLEXUS (Vahl, Enum. ii. p. 299) ; umbella composita, contracta aut capitata ; spiculis 14-floris, fusco-sanguineis ; glumis ovatis, in utroque latere remote l-nervatis; nuce cum 3-4 parte glume æquilonga late ovoidea, utrinque angustata.— Kunth, Enum. ii. p. 42; Boeck. in Linnea, xxxv. p. 558. C. hematostachys, Steud. Cyp. p. 41 Valdivia (Lechler n. 288, A. Calcutta). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 117 88. C. Luzuræ (Rottb. Descr. et Ic. p. 23, t. 13. fig. 2); umbella composita, specie simplice ; spicis in capitula pyrami- datis ; spiculis lanceolatis, pallidis ; nuce cum 2 parte glume æqui- longa, oblonga, utrinque angustata.—Kunth, Enum. ii. p. 43; Boeck. in Linnea, xxxv. p. 561. Brazil: Rio (Macrae, h. Calcutta); Prov. Rio Negro, Barra (Spruce, h. Calcutta). Neo-Granada: La Paila (Holton, nn. 117, 118, k. Calcutta). 84. C. LECHLERI (Steud. Cyp. p. 27); robusta; umbella com- posita aut decomposita, umbellulis primariis condensatis; spiculis ovatis, brunneis.— Boeck. in Linnea, xxxv. p. 563. Stolones longi, crassi, nigri. Culmus 4-8 dm. longus, basi 15 mm. crassus, nigro-castaneus. Folia plurima, eulmum super- antia, lata 14 mm., spongiosa transversim septato-punctata. Umbella 8-12 cm. in diam., densa, multispiculosa. Spicule longe 7 mm., late 4 mm., circiter 8-flore. Stamina 3 vel 2. Valdivia (Lechler n. 452, h. Calcutta). Sect. G. Pseudanosporum. Species 1; a Kunth (Enum, ii. p. 40) inter Luzuloides, a Boeckeler (Linnza, xxxv. p. 412) in Anosporum collocata ; a Bentham (Fl. Austral. vii. p. 264) et Steudel (Cyp. p. 315), C. alternifolio &c. affinior habita. (Cf. ff. 7, 8, 9.) l This is a very aberrant species; even assuming, as I have endeavoured to put forward, that the peculiar corkiness of the cells of the nut is merely adaptative, there yet remains much to connect it with Anosporwm, i. e. to break down Anosporum as a genus. On the inner face of the nut (fig. 8) two obscure chinks can be made out which appear analogous with the correspondingly situate deep grooves in the nut of C. cephalotes, which Boeckeler interprets as indicating an absorbed perianth. The style is more unlike that of Anosporwm than of Cyperus proper ; it is deciduous papillose-floccose nearly to the base, and more like that of some Finbristylis or Scirpus, whence R. Brown's excellent specific name. The inflorescence will only do for this part of the genus Cyperus, and has doubtless influenced Bentham in his view ; M the spikelets themselves, contrà, have a hardness reminding muc of C. cephalotes. 85. C. prarysryzis (R. Br.! Prod. p. 214); robusta; umbella decomposita, densa, spiculis per 3-6 digitaus, pallidis aut aureo- 118 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. brunneis; styli papilloso-flocculosi, ramis 3 breviusculis linearibus; nuce sessili, cellulis angulorum suberosis stramineis.—Kunth, Enum. ii. p. 111; Benth.! Fl. Austral. vii. p. 264. C. pallidus, Nees! in Wight Contrib. p. 79 ; Kunth, Enum. ii. p. 40. C. caducus, Steud. Cyp. p. 315. C. fluitans, Buch.-Ham.! MS. in h. Wallich. Anosporum pallidum, Boeck. in Linnea, xxxvi. p. 412. Cyperus, Wall.! List n. 3337. Cyperus, Wall.! List n. 3359, D partim. Glabra. In lacubus sepissime natans; radices crasse fibrose inter Pistiam Salviniam &c. intricate, nec repentes nec stolo- nifere. Culmi solitarii, 5-10 dm., sursum triquetri parum scabri. Folia 2-5, cum culmo fere equilonga, plana, lata 1 em., in mar- ginibus sursum antice serrato-scabra, manus incautas secantia. Bractez 3-6, usque ad 2-5 dm. longa, divaricatz, subcorymbose nec plane umbellate, foliis similes. Umbella 3-25 cm. in diam., 5-1000-spiculose ; radii aliquando 10-20, usque ad 1 dm. longi, sepius multo pauciores, minores ; ochrez long 1-2 cm., truncate, minute bidentate; umbellularum partialium bractez breves; radioli ultimi undique divaricati; spice 1-5-spiculose ebracteo- late. Spicule longe 15 mm., late 8 mm., 20-40-flore, anguste lanceolate, compresse, pallide, rubentes aut aureo-castanee ; rhachilla compressa, fere exalata, quadratim excavata. Glume spisse, imbricate, turgidule, ovate, breviter acute, in dorso viridi-3-nervie, in lateribus enervose. Stamina 3, angulis nucis juxtaposita; filamenta lata, levia, persistentia; anthere anguste oblongz, flavæ, a connectivo rubro brevissime excurrente scabro subcristate. Stylus nuce multo brevior, deciduus; rami 3, cum stylo æquilongi, e gluma vix exserti. Nux ellipsoidea in rostrum brevissimum angustata, inequaliter trigona contra rhacheolam adpressa, leviter eurvata, in facie interiore concava, in dorso obscurius carinata, fusca margines versus albo-straminea ; cellule extime quadrate, parve, emarcide, hyaline, persistentes. Cl. Boeckeler transtulit C. caducum, Steud., ad C. Martianum, Schrader; hie autem in Brasilia, ille in Ind. Or. indigena.. A Bengalia (qua vulgaris) usque ad New South Wales; etiamque in Zeylania. (In Coromandelia verisimiliter rara neque in h. Roxburgh neque in h. Wight vidi.) Zeylania (Thwaites, C. P. n. 3559, hh. Kew, Mus. Brit., Calcutta. Assam (Simons nn. 664, 645, h. Calcutta ; Masters, h. Calcutta). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. — 119 Bengalia: Sahebgunje (Kurz, h. Calcutta) ; Mymensingh, Jumalpore (Grifith, h. Calcutta); Bengalia orientalis (Griffith nn. 6158, 6158 bis 2. Calcutta) ; Chittagong (J. D. Hooker n. 139, hh. Kew, Mus. Brit. ; O. B. Clarke n. 19919, h. Calcutta). Burma: Pegu (Kurz, nn. 685, 686, 2685, h. Calcutta); Mergui (Griffith, h. Kew n. 6168). Penang (Wallich n. 3359 D, h. Calcutta). Borneo: Banjermassing (Motley n. 995, h. Kew). Australia: Portus Jackson (R. Brown n. 5907, hh. Kew, Mus. Brit.). Sect. H. Haspani. Haspani, Kunth, Enum. ii. p. 84 (C. pulcherrimo ezcl.); Boeck. in Linnea, xxxv. p. 573 (C. pulcherrimo, albostriato ezcl.). Rhizoma sepius perenne. Folia brevia aut longa, angusta, interdum ad vaginas fere reducta. Umbella composita aut decomposita, spicis ultimis digitatis paucispiculosis. Spicule multiflorze, compresse ; rhachilla obscurius alata. Glume spisse, in speciebus typicis minute mucronate, in ceteris obtuse. Stamina 3-1; anthere sepius cristate, crista modo elongata, modo fere suppressa, albide glochidiato-spinulosa, raro omnino nulla, Nux parva, in speciebus typicis minuta. The next Section (Diffusi) differ by their wider leaves and bracts and larger nut. I think therefore C. albostriatus ought . certainly to be placed among the Elegantes, though the amount of real difference between the two sections is small. As to C. pulcherrimus it seems (with C. silletensis) to me to stand very naturally next C. difformis and C. fuscus; indeed I find C. pulcherrimus, C. silletensis, and C. difformis are much confounded in herbaria. As to C. vaginatus, seeing that it can stand in the Haspani without introducing an exception to their character, it makes one section less to arrange it here. _ 86. C. Haspan (Linn. Sp. Pl. p. 66, partim, nec Linn. h. propr.; matura viridis pallida aut atrosanguinea nec flavida ; radice perenni, repente ; foliis brevibus, longis aut nullis; invo- lucri foliis sepius 2 brevibus, interdum longis ; spiculis parvis, lineari-oblongis; staminibus 3-2; nuce parva, obovoidea obtu- siuscula, acute trigona, pallida aut brunnescente, scabrida vel fere levi.—Zozb. Fl. Ind. i. p. 210; Nees n Wight Contrib. p. 80; 120 ME. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Kunth, Enum. ii. p. 84; Dalz. and Gibs. Bomb. Fl. p. 282; Nees in Mart. Brasil, Cyp. p. 25; Boeck. in Linnea, xxxv. p. 574, var. a excl.; Benth. Fl. Austral. vii. p. 270. C. umbellatus, Burm. Fl. Ind. p. 21, t. 9. fig. 1; Vahl, Enum. ii. p. 362, nec Roxb., nec Benth. C. autumnalis, Vahl, Enum. ii. p. 318 (syn. Rottb. excl.). C. platyculmis, £. Br. Prod. p. 214. C. laticulmis, Spreng. Syst. i. p. 228. , juncoides, Lam. Ill. i. p. 147, fide Vahl. . complanatus, Willd. Sp. Pl. i. p. 270. leptos, Roem. et Sch. Syst. Mant. ii. p. 105; Torrey in Ann. Lyceum New York, iii. p. 273. . gracilis, Muehl. Gram. p. 18. . tunicatus, Roem. et Sch. Syst. Mant. ii. p. 115. . stellatus et C. cayennensis, Willd. ; Link, Jahrb. ii. p. 84. . junceus, Willd.; Link, Jahrb. iii. p. 85. . vaginatus, Willd. ; Link, Jahrb. iii. p. 85, non R. Br. . aphyllus, Rich. in Vahl, Eclog. ii. p. 6. . nudus, H. B. K. Nov. Gen. et Sp. i. p. 203; Presl, Rel. Haenk. p. 166. C. riparius, Schrad.; Nees in Mart. Brasil., Cyp. p. 28. C. hyemalis, Pursh MS. in h. propr. Cyperus, Linn. Fl. Zeyl. p. 15, n. 37. Cyperus, Wall.! List n. 3368. Cyperus, Wall. | List n. 3372. Cyperus, Wall. ! List n. 3369, litt. D, E, F, G n. 1. Cyperus, Wight n. 1822, b. Cyperacea, Griff. Itin. Notes, p. 33, n. 520. Cyperacea Indeterm., Wall. h. propr. n. 4. Rhizoma perenne, repens, culmis solitariis, distantibus (forma typica fig. 23) aut cespitosis (fig. 24); aut in primo anno sspe florigerum, radicibus fibrosis. Culmi 1-5 dm., apice triquetri, inferne sspe inequaliter triquetri aut fere ancipites. Folia plura, in forma typica culmo multum breviora, aut szpe subnulla, vaginis inflatis, apice lanceolatis, acutis. Involucri bractez 2, raro 8, suberectz anguste xiphoides, in forma typica breves, umbellam raro superantes. | Umbelle radii 4-12, usque ad 4-12 em. longi, interdum plane corymbosi; vagine longe 5-10 mm., apice lanceolate ; umbellule 3-8-radiate, interdum iterum diviss. Spicule 3-10, subdigitate, ebracteate, longs 10-15 mm., late 1-2 mm., multum compresse, 10-40-flore; rhachilla subexalata. Glume spissm, fusco-virides, pallide aut atrosanguines, ovate, obtuse, obscure mucronate, compresse, in dorso 8-1-nervis. aaa AQAAQAAQAQA MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 121 Stamina 3 aut 2; filamenta persistentia, granulosa aut subsca- brida; anthere lineari-oblong:z, flavide, apice albide setosæ, (rarius leves), erista obsoleta. Stylus brevis, nuce vix longior; rami 3, e gluma vix exserti. Nux cum j-l parte glume æqui- longa, non (aut sat obscure) compressa, basi umbonato-dilatata, in forma indica pallide brunnea aut flava, cellule extime quad- rate, neque incrassate neque opace.—Rottb. Ic. t. 6, fig. 2, ex exemplo Africano depicta, est C. flavidus, var. africana.—Rottb. Ic. t. 17. fig. 3, involucro bracteis pluribus setaceis additis, est fortasse revera Scirpus, ut a Rottboellio nominatus.—Species parum variabilis; exempla juniora, in primo anno florentia, a C. flavido ope coloris viridis, nucis trigone, spicularum majorum distinguenda sunt. Forme sequentes pro varietatibus vix habendæ :—forma malasica; folis laxe herbaceis quam culmi multo longioribus, bracteis elongatis (in Wall. List n. 3369 usque 3 dm. longis): forma in Malaya, China precipue obvia. Forma americana; rhizomate brevi; spiculis haud raro subcasta- neis; nuce sspe scabrida: in America et in Africa tropicali frequens. Asia australi-orientalis; Australia; America calidior; Africa occidentalis tropica. India orientalis, ubique vulgaris: Sikkim, alt. 600 metr. (G. King n. 4812, ^. Calcutta) ; India centralis (G. King, À. Calcutta) ; Assam (Jenkins n. 740, h. Calcutta) ; Khasia (Oldham, h. Caleutta; Griffith n. 520, h. Kew); Sylhet (Wallich nn. 3368, 3372, h. Calcutta) ; Bengal (Griffith n. 6216, kk. Kew, Calcutta); Madras Peninsula (Hohenacker n. 691, hh. Kew, Mus. Brit.). l Zeylania (Thwaites n. 794, h. Kew, n. 799 h. Mus. Brit., n. 805 partim A. Calcutta). Burma et Malaya, communis: Pegu (Kurz, nn. 677, 679, 681, À. Caleutta); Tavoy (Wallich n. 3369 H, h. Calcutta) ; Tenasserim (Helfer, h. Kew, n. 6216) ; Ins. Bangka (Amand, h. Kew); Java (Zollinger n. 269, hh. Kew, Mus. Brit.) ; Cochinchina (Lebeuf, h. Kew). [Forma nehina C :J oth orgftad, h. Mus. Brit.) ; Ins. Nicobar (Kurz n. 25979, h. Kew); Borneo (Motley n. 98, h. Kew; Barber n. 194, h. Kew) ; Hongkong (C. Wright n. 557, h. Kew; Wilford n. 48, h. Kew); Canton (Sampson n. 260, h. Kew). Australia (Cunningham n. 331, ^. Mus. Brit.) ; sinus Carpen- 122 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. taria (R. Brown n. 5911, hh. Kew, Mus. Brit.); Sinus Moreton (Macgillivray, h. Kew) ; Portus Darwin (F. Schultz n. 310, k. Kew). America australis, vulgaris: Paraguay (Balansa nn. 406, 409, À. Kew); Brasil (Glaziou nn. 5449, 6768, h. Kew; Burchell nn. 1169, 1663, 2640, 2754, 3451, 4192, 4464, 4481-2, 5819, 9907, h. Kew); Organ montes (Gardner n. 718, hh. Kew, Mus. Brit.); Surinam (Hostmann n. 711, h. Kew); Guiana Francorum (Sagot nn. 645, 1178, hh. Kew, Mus. Brit.) ; Venezuela (Funcke n. 230, h. Kew; Moritz n. 766, h. Mus. Brit. ; Funcke n. 705, h. Mus. Brit.). India occidentalis, cum terris adjacentibus: Panama (Hayes n. 316, h. Kew); Mexico (Liebmann, h. Kew); Cuba (C. Wright n. 3359, hh. Kew, Mus. Brit.; C. Wright n. 3358, h. Mus. Brit.); New Orleans (Drummond, h. Calcutta) ; Florida (Curtiss n. 3041, h. Kew; Chapman, h. Calcutta) ; Alabama (Gates, h. Calcutta); Georgia (Beyrich n. 535, h. Mus. Brit.). Africa occidentalis tropica; Niger flumen (Barter n. 1572, h. Kew); Nun flumen (Vogel n. 13, h. Kew); Congo flumen (Christian Smith nn. 19, 47, h. Mus. Brit.). Africa orientalis tropica: Mombassa (Hildebrandt n. 2045, h. Kew). Madagascar (Hilsenberg et Bojer, h. Mus. Brit.) |. 87. C. FLavipus (Retz. Obs. v. p. 18); matura flavida aut nigrescens ; radice fibrosa; folis cum culmo sspe sequilongis; involucri bracteis 2-3, sepe elongatis; umbella composita aut decomposita ; spiculis minimis, lineari-oblongis ; stamine 1, raro 2; nuce minima, obtuse obovoidea, eontra rhachillam compressa, in dorso convexa aut vix carinata, albida marmorata, scabra vel punetata.— Rozb. Fl. Ind. i. p. 200; Dalz. and Gibs. Bomb. Fl. p. 284; Nees in Wight Contrib. p. 80 in nota; Hassk. Pl. Jav. Rar. p. 71 in nota. C. tenuispicus, Steud.! Cyp. p. 11, in Pl. Hohen. n. 1670. C. Fieldingii, Steud. Cyp. p. 11, fide Wallich n. cit. et descr. C. leptostachyus, Nees; Steud. Cyp. p. 33. C. Haspan, Rottb. Descr. et Ic. p.36, t. 6. fig. 2; Thwaites! Enum. Pl. Zeyl. p. 343 partim ; Boeck. in Linnea, xxxv. p. 574, var. a; Oliver! in Trans. Linn. Soc. xxix. p.165; Rottler, MS. in h. propr.; Wight! n. 1822 ah. propr. ; non Linn. C. microcarpus, Boeck. in Rel. Rutenb. i. p. 37. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 123 Cyperus, Wall. List n. 3365. Cyperus, Vail. List n. 3369, A, B, C, Gn. 2. Cyperus, Wall. List n. 3313, plagula tertia. Cyperus, Wight n. 2874, h. propr. Cyperus, Griff. Itin. Notes, p. 14, n. 199. — Pluk. Alm. t. 192. fig. 2. Annua. Culmi cæspitosi, a 2 em. usque ad 3 dm. longi, trigoni. Folia graminea, ssepius elongata, rarius brevia interdum in vaginas reducta. Umbella multiradiata, 5-20 cm. in diam. Spi- cule longz 5-12 mm., late 1 mm., compress, 8-30-flore. Glume quam C. Haspani minores, magis truncate, proventu laxe (vel vix) imbricate, Anthere quam C. Haspani breviores, apice non setose. Nux obtusissima ; cellule extime magne, quadrate. India orientalis, alt. 0-1200 metr., ab Himalaya usque ad Zeylaniam vulgaris, in oryzetis pestis: Kashmir (Jacque- mont n. 1140, h. Kew); Nepaul (Wallich n. 3365 B, h. Cal- cutta); Assam (Jenkins, Simons, h. Calcutta); Khasia colles (Griffith n. 199, h. Kew n. 6209/5); Sylhet (Wallich n. 3365 C, h. Calcutta); Bengal, Burisal (C. B. Clarke n. 16947, k. Calcutta); Chota Nagpore, alt. 600 metr. (C. B. Clarke nn. 20845, 25192, h. Calcutta); Canara (Hohenacker n. 1670, k. Kew) ; Madras Peninsula (Wight n. 2874, h. Cal- cutta); Arracan (Kurz n. 677, h. Calcutta). l Zeylania (Thwaites, C. P. n. 805, hh. Kew, Mus. Brit.). Australia, Victoria flumen (F. Mueller, h. Kew). Ins. Seychelles (Horne n. 634, h. Kew). Africa: Zanzibar (Kirk, h. Kew; Blackburn, h. Kew) ; Nilus flumen superior (Grant, h. Kew); Djur (Schweinfurth nn. 194, 2054, h. Kew) ; Niger flumen (Barter n. 1566, h. Kew); Senegal (Roger n. 38, h. Kew; Heudelot n. 325, h. Mus. Brit. ; Adanson n. 164, h. Mus. Brit.). 88. C. xquauis (Vahl, Enum. ii. p. 320) ; subaphyllus » radis umbellæ primariis, innumerosis, qualibus, tenuibus unt, Enum. ii. p. 37; Boeck. in Linnea, xxxv. p. 941. Umbelle radii primarii usque ad 100, longi 1 dm. Anthere flavidæ a crista minuta rubra albide setose coronatæ. ax minn tissima, cum 2 parte glume vix æquilonga, trigona, P i a Species primo adspectu, causa inflorescentie, Cyperum Papyrus referens, Ins. Mauritius (Bojer, A. Calcutta). Portus Natal (Sutherland, h. Calcutta). 124 MR. C. B. CLARKE ON INDIAN SPECTES OF CYPERUS. 89. C. DENUDATUS (Vahl, Enum. ii. p. 324) ; rhizomate repente; foliis cum 2 parte culmi subzquilongis; umbella composita ; spi- culis 3-6nis digitatis, lanceolatis, 8-10-floris, brunneis.— Kunth, Enum. ii. p. 36; Boeck. in Linnea, xxxv. p. 576. C. spherospermus, Schrad. Anal. Fl. Cap., Cyp. p. 8. Anthere lineari-oblonge, rubescentes ; crista minima, quadrata, albida, non setosa. Caput Bon: Spei: Uitenhage (n. 175, k. Calcutta). 90. C. pentatus (Torrey, Fl. United States, i. p. 61); foliis eum culmo subequilongis ; umbella composita aut decomposita ; spiculis 3—6nis digitatis, 8-16-floris, lucide brunneo-viridibus.— Kunth, Enum. ii. p. 94; Boeck. in Linnea, xxxv. p. 578. Stoloneslongi.. Anthere lineari-oblong:, rubr:e, apice a crista hyalina oblonga serrato-scabra terminate. Nux cum 4 parte glume equilonga, late ellipsoidea, trigona. Massachusetts (Jesup, h. Calcutta). 91. C. erinus (F. Muell. MS., nec Hochst., nec Nees); foliis bracteisque longis, duris; umbella composita, radiis radiolisque divaricatis; spiculis 2—4nis digitatis, ellipsoideis, admodum com- pressis, eastaneo-brunneis; glumis spissis, obtusis. Culmi 15 em. ; basi inerassati, a vaginis corneis striatis brunneis intecti. Spicule long: 5 mm., late 2 mm., 10-14-flore (spiculas Brize quodammodo referentes). Stamina 2 (vel 3); anthers lineari-oblongzs, rubre, mutice. Nux nimis juvenis. Australia: Snowy flumen (F. Mueller, h. Calcutta). 92. C. vaerNATUS (R. Br. Prod. p. 69, non Willd.); sub- aphylla; culmis rigidis, teretibus; involucri bracteis pluribus, 1-3 em. longis; umbella capitata.—Boeck. in Linnea, xxxv. p. 472. Rhizoma lignescens, horizontale. Culmi approximati, 5 dm. longi, usque ad 1 dm. a vaginis intecti; vaginorum pars libera rigida, lanceolata, striata, non foliiformis. Involucri bractes xiphoidez, rigide, scabre. Antherarum criste cum 4-4 parte loculorum equilonge, lineari-lanceolate, rubre, scabride. Australia: Government flumen (Drummond. h. Calcutta). Sect. I. Diffusi. i Diffusi, Kunth, Enum. ii. p. 25, partim ; Boeck. in Linnea, xxxv. p. 931. Radix perennis, valide fibrosa, sæpe nigra, interdum repens. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 125 Folia plura, longa et lata, plana, viridia, conspieue striata; invo- lucri bractez foliis similes. Umbella composita aut decomposita, non congesta, spicis ultimis digitatis, paucispiculosis. Spicule multiflore, compresse, virides pallide» aut fusce; rhachilla an- guste alata. Glum: breviter mucronate aut acute. Stamina sepissime 3; antherw lineari-oblonge, apice sspe cristate aut setose. Nux robusta, ellipsoidea aut obovoidea, acute trigona. This is a small group of very closely allied species, easily recognized by the numerous broad leaves and bracts. It is a striking proof how little real difference in structure there is between the whole of Eucyperus, that Bentham thinks C. multi- spicatus, from its small nut, should, perhaps, rather be placed among the Haspani; while, on the other hand, C. macer, nov. sp., below among the Corymbosi, seems to me to belong really to the Diffusi, but has such short and narrow leaves and bracts that it could only be placed here as a sp. anomala. 93. C. ELEGANS (Linn. Sp. Pl. p. 68); culmo solitario; involucri bracteis pluribus, longis, latis, viridibus, in margine scabris ; umbella composita aut decomposita; glumis late ovatis, subito acutato-mucronatis, multinerviis, subdistantibus, fructigeris laxe imbricatis; nuce cum 3 parte glume zquilonga, late ellipsoidea, utrinque angustata.—Vahl, Eclog. ii. p. 7, Enum. ii. P 328 ; Meyer, Fl. Esseg. p. 28; Kunth, Enum. ii. p. 28; Nees in Mart. Brasil., Cyp. p. 83; Boeck. in Linnea, xxxv. p. 582. C. laxus, Lam. Ill. i. p. 146; Griseb. Fl. Brit. West Ind. p. 563. C. parciflorus, Willd. ; Link, Jahrb. iii. p. 85. . C. toluccensis, H. B. K. Nov. Gen. et Sp. i. p. 206; Kunth. Enum. ii. p. 28. . umbrosus, Lindl. et Nees in Mart. Brasil., Cyp. p. 31. . meestus, Kunth, Enum. ii. p. 31. . racemosus, Steud.! Cyp. p. 52, non Retz. . nigroviridis, Thwaites, Enum. Pl. Zeyl. p. 344. . digitatus, Wall. MS. in h. propr., non Rozb. . microstachyos, Rottler in h. propr., non Vahl. Cyperus, Wall.! List n. 3398. Cyperus, Wall.! List n. 3362. Cyperus, Wall.! List n. 3370, A. Cyperacea indeterm., h. Wallich n. 5. Hypelytrum, Griff. Itin. Notes, p. 17, n. 243. — Sloane, Jamaica, i. p. 117, t. 75. fig. 1. Radix perennis, lignescens, a fibris nigris validis defixa. Cul- Oooooono 126 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. mus 4-10 dm., trigonus, in americanis interdum gracilis, in gerontogeis sepius robustus, triqueter. Folia basin versus culmi plura, cum culmo equilonga, viridia, lata (in gerontogeis sepe 10-15 mm. lata), in marginibus scabra, in gerontogeis sspe serrulata. Involucri bracteæ 4-10, cum foliis consimiles, longs 3-6 dm., late 5-12 mm. Umbelle radii 8-10, in americanis siepe 1-2 dm. longi, in gerontogeis multo breviores; umbellule in americanis pluriradiolatz, sspe bracteolate, radiolis 2-5 cm. longis, in gerontogeis breviuscule divaricatim radiate; umbella in americanis sepius composita aut decomposita, in gerontogeis sepe decomposita aut supradecomposita, fasciculis numerosis densa. Spicule 3-9nz, digitate, in americanis usque ad 12-20 mm. longs, 3-4 mm. late, 10-24-flore, in gerontogeis sæpius minores angustiores, interdum 4-6-flore. Glume laxius (fructi- gere vix) imbricate, naviculares, in dorso late virides 5-nervie, in lateribus latiusculis scariose, pluristriate, mature glauco- virides aut sepius in americanis pallide scariosæ, in gerontogeis lutee aut fere castanes; rhachille ale anguste aut latiores nonsolubiles. Stamina 3-2; antheræ subinclusz, lineari-oblonge, flavide, a crista parva lineari-lanceolata scabra alba sspe coro- nate. Stylus vix ullus; rami 3, e gluma breviter exserti. Nux triquetra, apice obtuse conica, aut (in gerontogeis) magis acutata ; matura nigro-fusca, immatura specie granulosa ; cellule extime quadrate, persistentes, hyaline, nec porose nec reticulate.— Exempla raro prolifera.—Species difficillima: exempla americana formam unicam constituunt; exempla gerontogea polymorpha sunt, neque a C. diffuso bene digerere potui. America calidior; India orientalis, China, Malaya; presertim in sylvis udis inter saxa radicans, neque in oryzetis. America, sat communis: Paraguay (Balansa n. 412, h. Kew) ; Brasilia (Burchell nn. 1044, 1218, 1461, 1496, 1627, 6459, 8438, 8474, 9808, h. Kew; Martius n. 545, hh. Kew, Mus. Brit.; Gardner n. 2838, h. Kew); Bahia (Saltzmann, C. um- brosus, A. Calcutta); Para, Santarem (Spruce, h. Calcutta); Amazon flumen (Traill n. 1167, h. Kew). Venezuela (Funcke n. 711, h. Mus. Brit.) Guiana (Schomburgk n. 461, hh. Kew, Mus. Brit.). Surinam (Berthoud-Coulon nn. 54, 56, h. Mus. Brit.). Ins. Trinitas (Sieber n. 9, A. Kew; Fendler n. 898, h. Mus. Brit.). Jamaica (Swartz, h. Mus. Brit.). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 127 India orientalis, alt. 0-1500 metr., in Bengal et peninsula Malayana: Sikkim, alt. 1500 metr. (C. B. Clarke n. 9836, h. Kew); Sikkim Terai (Gamble, T. Anderson, Kurz, h. Cal- cutta); Assam (Griffith n. 6162, hh. Kew, Calcutta, n. 1605, h. Mus. Brit.; Jenkins n. 574, h. Kew; Simons, Masters, h. Calcutta); Khasia colles (J. D. Hooker n. 1902, h. Kew; Oldham n. 8, h. Calcutta), alt. 1200 metr. (J. D. Hooker et T. Thomson, h. Calcutta) ; Sylhet (Griffith, h. Kew n. 6218); Yunan (J. Anderson, h. Calcutta) ; Chittagong (J. D. Hooker n. 386, h. Kew); Mergui (Griffith, h. Kew n. 6354); Ran- goon ( Wallich n. 3362 A, h. Calcutta); Tenasserim (Helfer, h. Kew n. 6161); Ins. Andaman (Kurz, k. Kew). Madras Peninsula, montes Anamallay (Beddome, h. propr.). Zeylania (Thwaites, C. P. n. 2879, hh. Kew, Mus. Brit., Cal- cutta). Java (Zollinger n. 720, h. Mus. Brit.); Buitenzorg (Kurz n. 1847, h. Calcutta). Borneo (Motley n. 65, h. Kew; Barber n. 327, h. Kew). Whampoa et Canton (Hance n. 2149, h. Kew). 94. C. prrrusus (Vahl, Enum. ii. p. 921); culmo solitario ; involucri bracteis pluribus, longis, latis, viridibus J umbella com- posita aut supradecomposita; glumis spissis, ovatis, subito acu- tato-mucronatis, multinerviis ; nuce cum $-3 parte glume æqul- longa, late ellipsoidea, acute trigona.— Kunth, Enum. u. p. 30; Boeck. in Linnea, xxxv. p. 534, non Roxb. C. iridifolius, Dietr. Sp. Pl. ii. p. 315. C. scirpoides, Presi, Rel. Haenk. i. p. 178. C. pubisquama, Steud. Cyp. p. 20. C. lagorensis, Steud. Cyp. p. 36. Cyperus, Wall.! List n. 3370, B. l Differt a C. elegante glumis arctius imbricatis; an satis ? Folia (ut bracteæ) quam in C. elegante minus scabra, vix serru- lata. Spicule haud raro in radiolis 5-10 mm. longis pedicellate, solitaris, i. e. umbellule ultime 1-2-spiculose. Glume in forma typica luzonensi vix quam in C. elegante densius stipate ; . . i , in unica in forma indica spicule majores, teretes, densiflore, m u umbella usque ad 800. Glume in margine sepe sublaceræ (nec i. e. spiculæ . . in foli i pubescentes), superiores interdum in folia mutate, een prolifere. In forma indica, stylus subnullus, rami autem . anino ut C. giusculi, spiculeque fere comosz evadunt. Nux omn 128 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. elegantis; cellule extime interdum albescentes, opace, i. e. nux granulata videtur. India orientalis, Zeylania, Malaya, ins. Philippine. Sikkim ( Kurz, T. Anderson, h. Calcutta). Assam (Jenkins, Masters, h. Calcutta). Khasia colles (Hooker f. et T. Thomson, h. Kew); alt. 300 metr. (C. B. Clarke n. 5468, h. Kew). Cachar (Keenan, h. Kew). Chittagong (C. B. Clarke, n. 19689 h. Kew, n. 8292 h. Cal- cutta). Pegu (Kurz nn. 665, 666, h. Calcutta). Moulmein (Griffith nn. 320, 324, h. Kew n. 6162). Tenasserim (Helfer, h. Kew n. 6164/1). Bhamo (J. Anderson, h. Calcutta). Rangoon (Wallich n. 3474, h. Calcutta). Ins. Penang et Amherst ( Wallich n. 3370, h. Calcutta). Ins. Andaman et Nicobar (Kurz, h. Calcutta). Zeylania (Thwaites, C. P. n. 3931, hh. Kew, Mus. Brit., Cal- cutta). Borneo (Barber n. 327, h. Kew). Timor (Kunstler n. 331, h. Calcutta). Ins. Philippine (Cuming nn. 445, 533, 534, hh. Kew, Mus. Brit.). Forma pedicellata; spiculis fere omnibus pedicellatis, soli- tariis. Arracan (Kurz, h. Calcutta). Tenasserim v. Andamans (Helfer n. 6161, h. Calcutta). 95. C. Herren (Boeck. in Linnea, xxxviii. p. 360); culmis cæspitosis; involucri bracteis 3-6, longis; umbella composita, laxiuscula; glumis ovato-lanceolatis, submucronatis; nuce cum 2 parte glume equilonga, obovoidea, triquetra. Fimbristylis, Wall. ! List n. 3528. Radix fibrosa, valida, nigra. Culmi 3-5 dm. longi, trigoni, graciliores. Folia plura, cum $ parte culmi equilonga, graminea, lata 5 mm. Involucri bracteæ usque ad 1-2 dm. longe. Um- belle radii erecto-patuli, inzequales, longi 3-12 em. ; umbellularum radioli pauci. Spicule 3-7næ fasciculate, longs 1-2 cm., late 8 mm., glauco-virides, compresse, 10-30-flore ; rhachilia sub- exalata. Glume distantes, navieulares, in dorso virides nervis obsoletis scil. coalitis, in lateribus scariose subenervie. Stylus brevissimus; rami 3, e gluma breviter exserti. Nux obtusa, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 129 pallide brunnea ; cellulæ extimæ quadratæ, hyalinæ, ideoque nux quasi reticulata. Peninsula Malayana: Barma (Griffith, h. Kew); Pegu (Kurz nn. 670, 2680, h. Calcutta); Chappedong (Wallich n. 3528, Ah. Linn. Noc., Kew, Calcutta) ; Tenasserim (Helfer n. 6140, hh. Kew, Calcutta); Mergui (Griffith n. 821, h. Kew n. 6140). 96. C. wuLTISPICATUS (Boeck. in Linnea, xxxvii. p. 962); involucri bracteis 5-6 longis, viridibus; umbella decomposita aut supra decomposita, magna, laxa; spiculis pro majore parte solitariis, multifloris, admodum compressis; nuce cum 4-3 parte glume zquilonga, obovoidea, obtusa, trigona. Radix valide fibrosa. Culmus 3-8 dm. longus, robustior, apice triqueter, deorsum subcompressus. Folia 2-3, cum 3 parte culmi equilonga, in margine scabra vix serrulata. Involucri bractew usque ad 3-5 dm. longæ, sepe 1 cm. late. Umbella 3 dm. in diam. ; radii 8, usque ad 12 em. longi; umbellule bracteate, radi- oli 3-12, suberecti, sepe 3-6 em. longi. Spicule numerose, in exemplo typico fere omnes solitarie, longe 1 cm., late vix 2 mm., 24-flore, pallide brunnes; rhachilla subexalata. Glumæ sub- distantes, leviter imbricatæ, naviculares, ovatæ, obtusæ, minute vel obscure mucronatæ, in dorso sub-3-nerviæ, in lateribus fer- rugineo-maculatæ. Stamina 2; antheræ vix exserte, lineari- oblongæ, rubræ, fere muticæ. Stylus nuce vix longior ; rami 3, e gluma breviter exserti. Nux luteo-brunnea, quasi-puncticulata, —Species a Bentham (causa nucis parve) inter Haspanos or- dinata. Tenasserim vel Andamans (Helfer n. 6168, hh. Kew, Calcutta). Cachar (Keenan, h. Kew). Pun 97. C. Kunzir; foliosa; umbella decomposita, radiolis re rieatis; spiculis l—4nis digitatis, straminco-brumses; 8 s Spissis arctissime stipatis, ovatis mucrouatis, basi late breviter decurrentibus ; staminibus 2. C. multispicatus, Kurz MS. . C. multispicatus, Boeck.; var. Kurzii, C. B. Clarke, MS. in h. Kew. Rhizoma validum, breve, lignosum. Culmi longi P dm., apiso triquetri. Folia 5-8, cum culmo equilonga, lata 7 mm., . - imi sque ad . 8-nervia. Involucri bractez 5-6, foliis consimiles, usq 2-4 dm. longe. Umbella 1-2 dm. in diam., densiuscula ; radii 8-106; radioli tenues, divaricati ; umbellularum braeteolze minute, Spicule (sæpe pedicellate solitariz) long? 12 mm., uM 1-2 mm., LINN. JOURN.—BOTANY, VOL. XXI. 130 ME. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. usque ad 40-flore, multum compress»; rhachille ale late, breves, persistentes. Glume carinate, in dorso 3-5-nervis, in lateribus subenervie, maculate. Anthers juniores lineari-lan- ceolate cum 3-3? parte glume equilonge, rubre; in plurimis spiculis monstrose in phyllodia imperfecte transmutate. Ins. Andaman, Pheacia (Kurz in hh. Kew, Calcutta). 98. C. TURGIDULUS ; culmo subtrialato; involucri bracteis plu- ribus, longis, latis, viridibus; umbella composita aut decomposita, densa, radiolis divaricatis; spiculis ovoideis, teretibus, paucifloris, in capitula parva subglobosa fasciculatis; glumis nucibusque ut C. elegantis. Scirpus trialatus, Boeck. in Linnea, xxxvi. p. 721. Species C. eleganti admodum affinis, cum hac (nisi quoad notas supra datas) arcte congruens. Spicule c. 6-floræ, longæ 4 mm., late 3 mm., inflate, fere teretes. Glume latiores quam longs. Peninsula Malayana, a Burma usque ad Malacca et Cambodia. Pegu (Kurz n. 655, h. Calcutta). ' Moulmein (R. Scott, h. Calcutta). Tavoy (Wallich, h. Kew). Martaban (Kurz, h. Calcutta). Amherst ( Wallich, h. Kew). Mergui (Griffith nn. 88, 98, h. Kew). Tenasserim (Helfer n. 6164, hh. Kew, Calcutta). Malacca (Griffith n. 6164, h. Kew). Siam (Lebeuf n. 457, h. Kew). Cochinchina (Lebeuf n. 827, h. Kew). 99. C. LONGIFOLIUS (Poir. Encye. vii. p. 270) ; foliis longis: simis, latis; umbella decomposita ; spiculis 1—5nis digitatis, 4-8-floris ; glumis ovatis, mucronatis, per totam fere latitudinem 9-13-nerviis; nuce eum à parte glume equilonga, ellipsoidea, trigona, nigra.— Kunth, Enum. ii. p. 30, syn. Decaisne excl. ; Boeck. in Linnea, xxxv. p. 535. Mauritius (Byer, h. Calcutta). 100. C. ALTERNIFOLIUS (Linn. Mant. p. 28); bracteis plurimis, latis; umbella imperfecta, plane subcorymbosa, composita aut decomposita, radiis radiolisque numerosis, tenuibus; spiculis 2-5nis digitatis, brunneis, 14-floris ; glumis ellipticis, elongatis, muticis, tenuiter nervosis ; nuce cum 3-4 parte glume equilonga, anguste ellipsoidea, utrinque angustata.— Kunth, Enum. ii. p. 333 Boeck, in Linnea, xxxv. p. 568. Madagascar.—(Exempla culta plurima visa.) MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 181 101. C. AnBosTRIATUS (Schrad. Anal. Fl. Cap. Cyp. p. 7); bracteis 6-10, longis 1 dm., lineari-lanceolatis, 3-nerviis ; umbella laxa, 2 dm. in diam., radiis filiformibus, umbellulis depauperatis ; spiculis pro magna parte solitariis.— Kunth, Enum. ii. p. 34; Boeck. in Linnea, xxxv. p. 581. Rhizoma repens, a squamis lanceolatis, nigro-brunneis intectum. Culmi solitarii, 8 dm. longi. Spicule long; 8 mm., late vix 2 mm., 8-16-flore, pallide virides; rhachille ale conspicus, hyaline, sanguineo-maculats. Nux gluma parum brevior, ellipsoidea, utrinque acutata, acute trigona. _ Natal (Dr. Sutherland, h. Calcutta). 102. C. pepuncuLosus (F. Mueller, MS. in h. Calcutta) ; foliis bracteisque latis; umbella simplice, radiis longissimis ; spiculis 4—7nis, digitatis, 40-floris. Culmi 3 dm. Folia cum culmo fere equilonga, lata 12 mm., l-nervia, multistriata. Bracteæ 8 usque ad 15 cm. longe, 10 mm. late. Umbelle radii 7-12, usque ad 2 dm. longi. Spicule 2-3 em. longz, compresse ; rhachille ale hyaline, anguste. Glume laxe imbricatz, elliptic, mutice, in dorso virides, in lateribus pallide brunnes.—C. albostriato similis, sed major. Australia, sinus Rockingham (Dallachy, h. F. Mueller). Sect. K. Fusci. Fusci, Kunth, Enum. ii. p. 87; Boeck. in Linnea, xxxv. p. 585 (C. Iria adjecto). Annus, foliate. Umbella simplex, composita aut decomposita, spicis ultimis szepius densis, multispiculosis, in C. Iria laxe spicatis. Spicule multiflorze, compresse ; rhachilla anguste alata. Glume parve, laxe imbricatz, late ovate, obtuse aut truncate. Stamina 1-3; anther» lineari-oblongex, ecristate. N ux cum 2-1 parte glume equilonga aut gluma parum brevior, late ellipsoidea, trigona. l The first 5 species here included are so alike that they require care to sort: C. Iria, on the other hand, is placed by both Kunt and Boeckeler with C. eleusinoides, which appears to me to differ widely. Though the difference in inflorescence between c. [ria and C. difformis distinguishes average examples readily, jay met with large numbers of these two species cross-named iu herbaria ; and there are extreme forms of each species between whieh the difference is merely one of very small degres. At all 132 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. events, if C. Iria is not placed here, I would not place it with C. eleusinoides. * Spicule fasciculate. 103. C. SILLETENSIS (Nees in Wight Contrib. p. 79) ; culmis cespitosis ; umbella composita, contracta aut capitata; spiculis viridi-fuscis ; nuce cum 4 parte glume equilonga, ovoidea, acuta, triquetra— Kunth, Enum. i. p. 98; Boeck. in Linnea, xxxv. p. 555. Cyperus, Wail.! List n. 3363, F. Cyperus, Wall.! List n. 3536. Glauco- vel fusco-virens. Radix fibrosa. Culmi plures 1-4 dm. longi, trigoni. Folia 2-3, anguste graminea, cum culmo sepe squilonga. Involucri bractes 3-5, usque ad 10-18 em. longs, divaricate. Umbelle radii 3-10, longi 1-5 em. ; umbellule 1-5-radiate, congeste, subebracteatz. Spicule in unoquoque radio 5-15, fasciculate, longe 5-8 mm., late 1 mm. et ultra, com- presse, 10-22-flore. Glume imbricate, ovate, obtuse, vix cari- nate, 3—-5-nervie. Stamen 1 (aut 2) inclusum ; anthera oblonga, flava, mutica. Stylus nuce brevior; rami 3, e gluma vix exserti. Nux brunnea.—Species formis majoribus C. pumili, Linn., similis, et cum hac in herbariis admixta. Bengalia orientalis, ab Assam usque ad Martaban, alt. 0-500 metr., sat communis. Assam: Suddiya (Griffith nu. 1029, 1474, h. Kew n. 6212, n. 1474, h. Mus. Brit.); Seebsagur et Gowhatty (Masters in h. Calcutta); colles Naga (Masters n. 1888, h. Calcutta). Bengalia orientalis (Griffith n. 1456, h. Kew n. 6166, n. 6166 hh. Calcutta, Mus. Brit.) ; Cachar (Keenan, h. Kew) ; Dacca (C. B. Clarke nn. 6747, 6885, 17076, 17106, &e.). Bengalia borealis: Rungpore et Siligori (C. B. Clarke nn. 12068, 26510, 26793, &c.). Burma: Pegu,in convalle Sittang (Kurz n. 2679, h. Calcutta) ; Martaban et Tonkhyeghat (Kurz n. 653, k. Calcutta) ; Ran- goon (Cleghorn n. 215, h. Calcutta). 104. C. PULCHERRIMUS (Willd.; Kunth, Enum. ii. p. 35); eulmis cespitosis; umbella composita aut decomposita, densa aut sublaxa; spiculis parvis, viridi-fuscis; glumis ovato-oblongis, apicibus in sicco incurvis; nuce cum’ 2 parte glume squilonga, ellipsoidea, utrinque conico-acutata, triquetra.—.Boeck. in Linnea, Xxxv, p. 573. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS, 133 C. eumorphus, Steud. ! Cyp. p. 22. C. silletensis, Thwaites! Enum, Pl. Zeyl, p. 343. C. Haspan, Benth. Fl. Austral. vii. p. 270 partim, non Linn, Cyperus, Wall.! List n. 3357. C. silletensis forsan varietas.—Spicule quam hujus angustiores, vix 1 mm. late. Spicularum fasciculi (in exemplis bene evolutis) in radiolis longiusculis sustenti.—In h. Kew in ipsissima plagula eum C. Haspan agglutinata commixta, ideoque ad Cyperum Haspan & Bentham relata. Assam: Suddiya (Griffith nn. 1035, 1480, k. Kew n. 6178; Jenkins n. 565, h. Kew). Zeylania (Thwaites, C. P. n. 3558, hh. Kew, Mus. Brit., Oal- cutta). Java (Zollinger n. 638, hh, Kew, Mus. Brit.; Horsfield, h. Mus. Brit.). Borneo: Banjarmassing (Motley n. 992, h. Kew). 105. C. DIFFORMIS (Linn. Amen. Acad. iv. p. 802, Sp. Pl. p. 67); cæspitosa; umbella simplice composita aut capitata ; spiculis minimis, dense fasciculatis; glumis ovatis, truncatis, castaneis aut pallescentibus; nuce cum gluma fere æquilonga, equaliter triquetra, subovoidea, pallida.— Hottb. Descr. et Ie. p- 24, t. 9. fig. 2; Sibth. Fl. Grac. i. p. 32, t. 46; R. Br. Prod. p.215; Roxb. Fl. Ind. i. p. 195; Nees in Wight Contrib. p. 88; Decaisne in Nouv. Ann. Mus. d' Hist. Nat. ii. p. 358; Kunth, Enum. ii. p. 88; Reich. Ic. Fl. Germ. t. 284. fig. 674; Parl. Fl. Ital. i. p. 31; Willk. et Lange, Fl. Hisp. i. p. 188; Dalz. & Gibs. Bomb. FI. p. 282; Boeck. in Linnaa, xxxv. p. 586; Baker, Fl. Maurit. p. 411; Benth. FI. Austral. vii. p. 268; Boiss. Fl. Orient. v. p. 370. C. protractus, Link, Hort. Berol. i. p. 305, non C. fuscus, var. protracta, Delile. Cyperus, Wall.! List n. 3363 (litt. F excl.). Cyperacea indeterm., Wallich h. n. 3. l Radix fibrosa. Culmi sæpe plures, 1-5 dm., triquetri. ‘Folia 2-3, graminea, culmo sæpius breviora, flaccida, in marginibus minute scabrida. Involucri bracteæ 2—4, usque ad 5-25 cm. longs, divaricatea. Umbella sepius contracta; radii 3-8, oeque ad 4 em. longi; umbellule congeste, ebracteate. Spicule g °- boso-fasciculate, longe 5-10 mm., late 1 mm., compresse, m - turgidze, 10-40-flore ; rhachilla subexalata. Glumz vix imbri- 134 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. cate, concave vix carinatz, ovate aut subovate, in dorso virides, 3-nervix. Stamen 1, raro 2; anthera parva, subinclusa, oblonga, mutica, flavida. Stylus nuce multo brevior; rami 3, breves. Nux sursum latior, apice obtuse conica, non (aut obscure) a dorso compressa, neque rostrata, nitide straminea; cellule extime quadrats, emarcide, tenuiter hyaline, vix laxe. In oryzetis Orbis Veteris, ab Hispania usque ad Japoniam, Australiam, Africam australem. Hispania: Valentia fide Willkomm et Lange. Italia: forsan cum oryza introducta e Parlatore; Parma (J. Gay, h. Kew); Piedmont (h. Mus. Brit.) ; Sicilia (Todaro n. 927, hh. Mus. Brit., Calcutta). Egypt: Rosetta (Sieber n. 5, h. Kew; Letourneux n. 148, k. Kew); Fossa Egyptiaca (Sarguet n. 723, k. Mus. Brit.). Russia meridionalis : Lenkoran (Hohenacker, h. Mus. Brit.); Armenia Rossica (Besser, h. Kew); Ussuri (Mazximowiez, h. Kew); Japonia (Oldham n. 901, hh. Kew, Calcutta). India orientalis: alt. 0-2600 metr., ab Himalaya usque ad Zeylaniam et Malaccam vulgaris; Cabul (G7;ffth nn. 191, 892 partim, 4. Kew n. 6165); Kashmir, alt. 2600 metr. (Jacquemont nn. 722, 827, h. Kew); Kumaon (Wallich n. 3363 H, A. Calcutta); Assam (Masters n. 206, k. Kew; Simons &c., h. Calcutta); Oudh (R. Thompson n. 360, h. Calcutta); mons Aboo (G. King, h. Calcutta); India cen- tralis (G. King n. 49, h. Calcutta); Poonah (Jacquemont n. 275, h. Kew); Canara (Hohenacker n. 821, h. Kew); Madras Peninsula (Wight n. 2884, h. Calcutta); Zeylania (Thwaites C. P. n. 3042, hh. Mus. Brit., Calcutta). Burma superior et Tenasserim (Griffith n. 6165, hh. Kew, Calcutta) ; Ava ( Wallich n. 3363 B, h. Calcutta); Pegu( Kurz nn. 657, 2678, ^. Calcutta); Rangoon (R. Scott, h. Calcutta). China borealis: Shantung (Maingay n. 87, h. Kew, n. 158, h. Calcutta). Formosa (Oldham n. 582, hh. Kew, Mus. Brit.). Canton (Sampson n. 464, h. Kew); Shanghai (Maingay n. 546, h. Calcutta); Hongkong (C. Wright n. 558, h. Kew). Ins. Philippine (Cuming n. 549, hh. Kew, Mus. Brit.). Isle of Pines (MacGillivray n. 771, h. Kew). New Caledonia (Vieillard n. 1437, h. Kew; Pancher, h. Mus. Brit.). Timor (Le Guillon, h. Kew). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 135 Java (Zollinger n. 433, h. Kew; Horsfield nn. 1014, 1081, 7. Mus. Brit.); Buitenzorg (Kurz n. 1851, 5. Calcutta). Australia: Victoria (F. Mueller, h. Kew); flumen Brisbane CF. Mueller, h. Calcutta); Australia subtropicalis ( Bidwill n. 141, k. Kew); sinus Moreton (F. Mueller, hh. Kew, Mus. Brit.); sinus Rockingham ( Dallachy, h. Kew); sinus Keppel (R. Brown n. 5906, hh. Kew, Mus. Brit.) ; Baloon flumen (Mitchell, h. Mus. Brit.); Murray flumen (F. Mueller, h. Mus. Brit.). Africa australis: Grahamstown (Macowen n. 1347, hh. Kew, Mus. Brit.) ; Cap. Bone Spei (Drége, h. Mus. Brit.); Dam- mara-land (Hen, k. Mus. Brit.); Natal (Gerrard n. 702, h. Kew; Burchell n. 3350, h. Kew). Congo (Christian Smith n. 13, h. Mus. Brit.). Ins. S. Thomas (Don, h. Mus. Brit.). Ins. S. Helena (Burchell n. 6, h. Kew). Niger flumen (Baikie, h. Kew); Djur (Schweinfurth n. 2478, À. Kew); Nilus superior ( Grant, h. Kew). Socotra (Balfour n. 472, h. propr.). Mauritius (Sieber n. 17, hh. Kew, Mus. Brit.; Bouton, nn. 4, 5, h. Kew). Madagascar: forma maxima (Hildebrandt n. 3425 hh. Kew, Mus. Brit.; Pervillé nn. 468, 482, h. Kew; Hilsenberg et Bojer, h. Mus. Brit.). . 106. C. ruscus (Linn. Sp. Pl. p. 69); ceespitosa ; umbella sim- plice composita aut capitata; spiculis minimis, multifloris, oo fasciculatis ; glumis ovatis, obtusis, nigro-castaneis aut ne notatis ; nuce cum 3 parte glume wquilonga, pallida, ellipsoi los, a dorso compressa, angulata aut carinata, breviter rostrata.— ; Dan. t. 179 ; Host, Gram. iii. p. 49, t. 78; Poit. et Turpin, M Paris, t. 75 (non visa); Desf. Fl. Atlant. 1. p. 47; Sibth. 5 . Grac. i. p. 84, t. 48; Engl. Bot. t. 2626; Sturm, FI. De i xiii. t. 52; Curt. Fl. Lond. iv. t. 85; Koch, Syn. Fl. Germ. p. 849; Kunth, Enum. ii. p. 37; Reichb. Ic. Fl. Germ. t. 280. fig. ad ; Gren. et Godr. Fl. France, iii. p. 360; Boiss. Voy. Eepagn Pe > Fl. Orient. v. p. 370; Ledeb. Fl. Ross. iv. p. 241; Part » 7 M ii. p. 27; Anders. Oyp. Skand. p. 1, t. 1. fig.2; Boeck. in Linnea, xxxv. p. 585. Poi C. ferrugineus, Forsk. Fl. Zig.-Arab. p. 14, non Porr. C. Forskalei, Dietr. Sp. Pl. ii. p. 251. —, Morison, Hist. iii. p. 239, sect. 8, t. 11. fig. 38. 196 MR. C. B. CLARKE ON INDIAN SPECTES OF CYPERUS. Cyperacea, Griff. Itin. Notes, p. 281, n. /52. Cyperacea, Griff. Itin. Notes, p. 293, n. 892. Radix fibrosa. Culmi numerosi, 5-25 cm. longi, trigoni. Folia 2-3, graminea, culmo sepius longiora. Involucri bractesm 3-6, usque ad 5-15 cm. longa, divaricatz:. Umbella 3-6-radiata, interdum contracta aut subcapitata; radii usque ad 1-3 em. longi; umbellule congestz, ebracteate. Spicule 3-20n:, longe 8 mm., late 1-2 mm., lineari-oblonge, compressse, subturgide, 12-24- flore; rhachilla subexalata. Glume imbricatz, obsolete mucro- nate vel omnino obtuse, compress, vix carinate, in dorso virides l- (obscurius 3—5-) nervis, in lateribus nigrescentes castaneo- rubre aut scariose. Stamina 2; anther vix exserte, oblongs, mutice. Stylus nuce brevior; rami 3, e gluma breviter exserti. Nux ellipsoidea, inzqualiter trigona, in dorso convexa aut cari- nata, matura flavide brunnea aut straminea; cellule extime quadrate, emarcide, hyaline, laxiuscule. Ab Algiers, Hispania, Anglia, Gothland, per totam Europam, Africam borealem, Asiam occidentalem, usque ad Kashmir vulgata. Scania: Widstkofte (Ahlberg, h. Mus. Brit.). Podolia (Rehmann n. 169, h. Mus. Brit.). Dresden (Reichenbach, Fl. Germ. Exs. n. 152, hh. Kew, Mus. Brit.). Tours (Delaunay n. 1804a, h. Kew). Belgium (Marchal n. 346, h. Mus. Brit.). Montpellier (Bentham, h. Calcutta). Treves (Billot n. 85, h. Mus. Brit.). Vienna (Breidler, h. Calcutta). Tyrol (Churchill, h. Mus. Brit.). Switzerland (Shuttleworth, h. Calcutta). Portugal (Welwitsch nn. 337, 338, 403, h. Kew, nn. 150, 338, h. Mus. Brit.). Corsica (Jordan, h. Mus. Brit.). Rome (E. Forster, h. Mus. Brit.). Sicilia (Todaro, hh. Mus. Brit., Calcutta). Mauritania (Gowan n. 5, h. Kew); Tangier (Durand n. 86, h. Mus. Brit.); Algeria (Lefranc n. 486, h. Mus. Brit.). Alexandria (Camaritani n. 3597 , h. Kew). Attica (Orphanides n. 398, h. Kew). Damascus ( Gaillardot n. 2297, h. Kew). Tauria (Compin, h. Calcutta). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 187 Narym flumen (Karelin et Kirilow n. 1069, hh. Kew, Mus. Brit.). Soongaria (Schrenk, h. Kew). Ispahan (Haussknecht, hh. Kew, Mus. Brit.). Cabul (Griffith nn. 6174, 6175, 6176, hh. Kew, Calcutta). Kashmir, alt. 1800 metr. (T. Thomson, hh. Kew, Mus. Brit., Calcutta); Kishtwar, alt. 1800 metr. (C. B. Clarke n. 31383, h. Kew). Var. B. virescens (Vahl, Enum. ii. p. 336); spiculis majoribus pallidioribus, albo-virescentibus ; glumis apice sublanceolatis.— Reichb. Fl. Germ. i. p. 72 (sp.), Ic. Fl. Germ. t. 280. fig. 668. C. fuscus, var. protracta, Delile, Fl. Egypt. p. 8, t. 5. fig. 3. C. viridis, Sieb.; Spreng. Syst. i. p. 216. Regio Mediterranea orientalis: ins. Creta (Sieber, h. Kew); Tunis (Kralik n. 308, hh. Kew, Mus. Brit.); Sidon (Foz, h. Kew); Nazareth (Bové n. 385, h. Kew). 107. C. mrcroxeptis (Boeck. in Flora, 1879, p. 551, non Baker); cespitosa, parva, albo-viridis ; involucri bracteis 3-5, longis ; umbella composita; spiculis parvis, fasciculatis ; glumis obovatis, albidis; nuce eum gluma :equilonga, obovoidea, triquetra, apice conica, stramineo-lutea. C. Afzelii, Boeck. in Linnea, 18/9, p. 547 partim. Culmi 10-15 cm., trigoni Folia 2-3, angusta, cum culmo spe equilonga. Involucri bractee usque ad 10-18 cm. longs. Spicule 8-20-florse, quam C. difformis ssepius breviores, latiores. Glum:» concave, rotundats.—O. fusco var. virescenti affinis, differt glumis admodum obtusis. Africa centralis: Djur (Schweinfurth n. 2328, h. Kew); Bongo (Sehweinfurth n. 2195, h. Kew); Teba, in flumine Quorra (Barter, h. Calcutta). ** Spiculis lace spicatis. 108. C. Irta (Linn. Sp. Pl. p. 67, tab. Rheede cit. exel., et h. propr.); umbella simplice aut composita ; spiculis multifloris, laxe spicatis; glumis distantibus, obovatis; nuce quam gluma parum breviore, obovoidea, triquetra.—Rottb. Descr. et Ic. p. 41 (Ira); Roxb, Fl. Ind. i. p. 201; Nees in Wight Contrib. p. 87; Kunth, Enum. ii. p. 88; Hassk. Pl. Jav. Rar. p. 83; Dalz. § Gibs. Bomb. Fl. p. 282; Boeck. in Linnea, xxxv. p. 595 ; Benth. Fl. Austral. vii. p. 276 (var. flavescente inclusa); Boiss. Fl. Orient. v. p. 370. 138 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. Santonici, Rottb. Descr. et Ic. p. 41, t. 9. fig. 1. C. panicoides, Lam. Ill. i. p. 145. C. chrysomelinus, Link, Hort. Berol. i. p. 305. C. parviflorus, Nees! in Wight h. propr. n. 1840, in Wight, Contrib. p. 87, nec Vahl, nec C. umbellatus, Roxb, C. seminudus, Moritz in Zoll. Verz. 1842-44, p. 96, nec Roxb. C. Diaphaniria, Steud. Cyp. p. 23. C. Microiria, Steud. Cyp. p. 23. C. resinosus, Steud.! Cyp. p. 23. C. microlepis, Baker! Fl. Maurit. p. 410, non Boeck. Cyperus, Wall.! List n. 3360 (C. piloso in distrib. commixto). Cyperus, Wall.! List n. 3361. Cyperacea, Griff. Itin. Notes, p. 14, n. 198. —, Pluk. Almag. t. 191. fig. 7. Radix fibrosa. Culmi cespitosi, 1-5 dm., trigoni. Folia 2-3, graminea, subflaccida, cum culmo interdum equilonga. Involucri bractee 3-5, usque ad 1-2 dm. longs, divaricate. Umbella a 5 cm. usque ad 5 dm. in diam. ; radii plures, patuli, apice corym- bosi aut iterum imperfecte umbellati; umbellule torymbose ; radioli 1-spicati, sæpius brevissimi, ab ochreolo caudati suffulti (ideoque umbella specie simplex spicis subdigitatis) aut elongati 1-3 cm. longi a bracteis usque ad 5 cm. longis suffulti. Spice laxissime, longe 1—4 cm., 5-20-spiculose ; spicule antice fascicu- late, gluma ima inconspicua. Spicule longe 1 cm., late 2 mm., compresse, 6—20-flore, flavescentes ; rhachilla subexalata. Glume compress, in dorso congestim 3-5-nervis, nervo viridi obsolete (raro plane minute) excurrente; latera estriata, scarioso-flavida, sursum latiora. Stamina 2 vel 8; anthere vix exsertæ, oblongæ, mutice, flavide ; filamenta basi subconnexa. Stylus nuce multo brevior; rami 3, e gluma breviter exserti. Nux apice obtuse conica, nigra; cellule extime quadratse; emarcids, albide, reti- formes.—Rheede, Hort. Mal. a Linnso multisque auctoribus citata, est verisimiliter C. procerus, Rottb., certe non C. Iria, Linn.— C. Iria, var. flavescens, Benth. Fl. Austral. vii. p. 276, est omnino C. Iria, Linn., neque ullo modo C. flavescentem, Linn., versus tendit.— C. Iria, var., Hance h. n. 1147, est forma expli- cata amabilis, glumis submucronatis. In Asia orientali cum Malaya, Australia, in oryzetis pestis ; in Asia australi-occidentali et Africa tropicali frequens. Per Indiam ubique, a mari usque ad 2200 metr. alt.: Almora, alt. 1200 metr. (Strachey et Winterbottom n. 10, k. Mus. Brit.); Mussoorie (G. King, h. Calcutta); Nepaul ( Wallich MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 189 n. 3360 G, h. Calcutta); Sikkim (T. Anderson n. 1343, k. Cal- cutta); Sonada, alt. 2200 metr. (Kurz, h. Calcutta); Assam (Masters nn. 200, 570, h. Kew, nn. 852, 417, 643, h. Calcutta ; Jenkins, Simons, &c., h. Calcutta); Khasia colles (Griffith n. 198, k. Calcutta); Lucknow (Bonavia nn. 229, 237, ^. Cal- cutta); Parasnath, alt. 450 metr. (C. B. Clarke n. 24914); Chota Nagpore (Wood n. 14, h. Calcutta); India centralis et mons Aboo (GŒ. King, h. Calcutta); Poona (Jacquemont n. 944, h. Kew); Mangalore (Hohenacker n. 188, h. Kew); Quilon (Wight nn. 2869, hh. Kew, Calcutta); Zeylania (Thwaites, O. P. n. 811, hh. Kew, Mus. Brit., Calcutta). Peninsula Malayana: Burma et Mergui (Griffith n. 6178, hh. Kew, Calcutta); Yunan, Hotha (J. Anderson, h. Calcutta); Segain ( Wallich n. 3360 H, h. Calcutta); Arracan (Kurz, h. Calcutta); Pegu (R. Scott, h. Calcutta; Kurz nn. 675, 2677, h. Calcutta); Singapore (Kurz n. 3003, h. Calcutta); ins. Andaman et Nicobar (Kurz, h. Calcutta). Japonia (Oldham nn. 193, 908, h. Kew; Buerger, h. Calcutta) ; Yokohama (Bisset n. 818, ^. Mus. Brit.). l China: Pekin (Hance n. 1147, hh. Kew, Calcutta) ; Shanghai (Maingay nn. 635, 636, hh. Kew, Calcutta, n. 759, h. Cal- cutta); Hongkong (C. Wright n. 559, h. Kew); Canton (Sampson n.259, h. Kew); Formosa (Oldham n. 903, h. Kew, n. 585, k. Mus. Brit.). , Ins. Philippine (Cuming n. 563, hh. Kew, Mus. Brit.) Java (Horsfield nn. 1013, 1057, h. Mus. Brit.; Zollinger nn. 74, 77, h. Mus. Brif.); Buitenzorg (Kurz n. 1852, h. Calcutta). Borneo (Barber n. 52, h. Kew; Motley n. 620, h. Kew). Australia borealis: Sturt’s Creek (F. Mueller, h. Kew); Portus Darwin (Fitzalan, h. Kew). Australia centralis (Gosse n. 11, 4. Kew). . Cabul (Griffith nn. 42, 96, h. Kew n. 6177, h. Calcutta); Persia (Haussknecht, hh. Kew, Mus. Brit.). PD Ins. Mauritius (Bouton, k. Kew; J. Grey, h. Kew; Bojer, ^. Mus. Brit.). Mus. Brit) Ins. Mascarene (Macgregor, h. Mus. bru.). l Africa: Nubia (Kotschy n. 267, hh. Kew, Mus. Brit.) ; Darfur (Pfund n. 624, h. Kew); Cordofan (Pfund n. 335, h. Kew); Djur (Schweinfurth n. 2281, h. Kew). 140 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Sect. L. Glomerati. Majuseuli, foliati. Radix fibrosa, valida, rarius lignosa. Cul- mus sepissime solitarius, trigonus, apice triqueter. Umbella com- posita aut decomposita, spicis numerosis, approximatis, multi- spieulosis. Spicule multiflore, compresse; rhachille ale con- spieus, persistentes. Glume laxiuscule imbricats, ovate aut ellipsoideg, obtuse. Stamina 3 ; anther lineari-oblonge, mutica, rarius apice obsolete cristatz, sanguinem. Nux cum 4-% parte glume æquilonga, ellipsoidea aut oblonga, apice sspe angustata, trigona, haud raro ineequalis, interdum curvata. From this point till we come to the Exaltati there is a long series of species which I cannot divide into any satisfactory sections; but, to avoid one inconveniently long string of species, I have arranged them pretty much as Boeckeler has left them. Boeckeler has placed C. distans as an anomalous species, but I cannot sort it satisfactorily even as a species from C. nutans ; and Boeckeler calls Steber n. 18, C. distans, whereas I think if there is a typical C. nutans, that is it. Boeckeler places C. spectabilis very near C. eleusinoides, but arranges C. glomeratus itself with C. Sorostachys (leucocephalus, Retz.), which 1 think is certainly not the true affinity ; indeed C. spectabilis has been issued from Kew as a mere form of C. glomeratus, and I think it is excessively near it. Benth. et Hook. f. Gen. Pl. iii. p. 1045, have followed Kunth in making their large sections (subgenera) depend upon the development and persistence of the wing of the rhachilla: this, as stated in the introduction, I think is impossible to work ; and, so far as it can be worked, leads to most unnatural combinations. In species like C. rotundus, of which we have abundant material, it is seen that the wing is generally finally soluble: the point of age at which it is soluble appears to be very variable, and in her- barium specimens depends largely on the method by which the specimen was dried. As regards the breadth of the wing, I doubt its value as a subgeneric character: C. cephalanthus is arranged next C. spectabilis by Hooker f., and I incline to think that its true affinity, but the wings of the rhachilla in C. cephalanthus are orbieular, perhaps broader than in any other species of Eucyperus ; 80 that Boeckeler has removed it to quite another place. Finally we see, in common species like C. /ongus and C. tegetum, very great difference in the breadth of the wing in one species.— The MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 141 arrangement here followed is, I hope, tolerably natural as to the East-Indian species. 109. C. arowERATUS (Linn. Amen. Acad. iv. p. 801; Sp. Pl. p. 68); umbella composita ; spiculis in spicas densissimas breviter cylindricas congestis ; glumis elliptico-oblongis obtusis, ferrugi- neis; nuce cum 2 parte glume equilonga, angusta, trigona.— Host, Gram. ii. p. 48, t. 71; Sturm, Fl. Deutschl. xii. t. 52; Reichb. Ic. Fl. Germ. t. 284. fig. 675 ; Kunth, Enum. ii. p. 77 ; Koch, Syn. Fl. Germ. p. 850; Ledeb. Fl. Ross. iv. p. 242; Parl. Fil. Ital. ii. p. 81; Boeck. in Linnea, xxxv. p. 592. C. cinnamomeus, Retz. Obs. iv. p. 10. C. australis, Schrad, Fl. Germ. i. p. 116; Mertens et Koch, Deutsch. Fi. i. p. 424. —, Seguier, Verona, iii. p. 68, t. 2. fig. 2. —, Montt, Gram. p. 14, t. 1. fig. 1. Radix fibrosa. Culmus solitarius (vel 2-3 fasciculati), robus- tior, 3-10 dm. longus, obtuse (apice subacute) trigonus. Folia 2-5, cum culmo ssepe equilonga, 5-8 mm. lata, plana, in margini- bus serrulato-scabra. Involucri bractew 3-5, usque ad 2-3 dm. longs. Umbelle radii 3-8, usque ad 6-12 cm. longi (umbella interdum congesta vel subcapitata) ; umbellularum radioli pauci, brevissimi ; bracteole 2-3, usque ad 4 cm. longs, foliacee. . Spice longs 2 cm., late 15 mm., spiculis usque ad 40-80, densissime. Spieule longs 12'mm., late 1 mm., mature undique divaricate, compress, 12-20-flore. Glume laxe imbricate, vix carinate, fere concolores, in dorso tenuiter 3-5-nervi® ; rhachillee alee lanceolatw, hyaline, conspicue, insolubiles. Stamina 2 vix exserta; antheræ parve, oblongæ, ferrugineæ, muticæ. Stylus nuce multo brevior; rami 3, breviter exserti. Nux utrinque breviter angustata, fusco-atra; cellule extime quadrate, emar- cide, nitide, hyaline, persistentes. Ab Italia usque ad Kashmir et Amurland: Verona (Reichen- bach n. 2310, h. Kew). Caloutée) Adriatic: Aquileia (n. 6, 4. Vateutia). Croatia (Harkas-Vukotinovis n. 859, h. Calcutta). Iberia (Wilhelms, h. Calcutta). Mare Caspium (Karelin n. 311, h. Kew; M MM Kashmir: prope Sonamurg, alt. 2000 meir. (T. Thomson, hh. Kew, Mus. Brit., Calcutta). | China borealis: Shantung (Maingay n. 155, h. Calcutta). 142 MR. €. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Ussuri (Maack, h. Kew). Amurland (Bunge, h. Kew). 110. C. SPECTABILIS (Schreb. ; Kunth, Enum. ii .p.73) ; umbella specie simplice, revera composita; umbellulis e spicis densissime congestis; spiculis compressis, laxiuscule 12-floris; glumis late ellipticis, submucronatis; nuce angustius obovoidea, trigona, apice acutata, subrostrata, cum 4 parte glume æquilonga.— Boeck. in Linnea, xxxv. p. 605. Texas: Drummond (h. Calcutta). 111. C. SIeBERI (Kunth, Enum. ii. p. 96); umbella parce com- posita; spicis in radiis longiusculis sustentis, stellatis, subglo- bosis ; spiculis oblongis, compressis, 6—10-floris, lete cinnamomeo- brunneis; glumis late ovatis, obtusis, mucrone brevissime excur- rente.—.Boeck. in Linnea, xxxv. p. 508. [This is placed hereabout by Boeckeler, nor do I know that it technically may not belong to this section; but it is remote in habit both from the species that go before and those that go after.] Victoria (Robertson, h. Calcutta). 112. C. ELEUSINOIDES (Kunth, Enum. ii. p. 89) ; umbella com- posita, radiis elongatis; spicis multispiculosis, laxis, ssepe sub- fasciculatis; glumis vix imbricatis, nervosis, submucronatis; nuce cum $ parte glume equilonga, oblonga, curvata, trigona, in facie interiore concava.—Boeck. in Linnea, xxxv. p. 596; Benth. Fl. Austral. vii. p. 277 ; Boiss. Fl. Orient. v. p. 871. C. xanthopus, Steud. in Flora, 1842, p. 595; A. Rich. Fl. Abyss. ii. p. 485; Thwaites, Enum. Pl. Zeyl. p. 344. C. infra-apicalis, Nees in h. Wight n. 2382; Aitch. Cat. Punjab Pl. p. 155. C. Santonici, Rottler h. propr., non Retz. C. racemosus, Dalz. h. propr., non Retz. Cyperus, Wall.! List n. 3346. Cyperacea indeterm., Wallich h. n. 2. Cyperacea, Griff. Itin. Notes, p. 322, n. 1967. Radix fibrosa. Culmus 5-10 dm. longus, solitarius (basi autem sepe progemmans ideoque culmi 1-3ni), trigonus, sursum tri- queter. Folia plura, robusta, sepe cum culmo fere squilonga, 5-8 mm. lata, plana, in marginibus minute scabra. Involucri bractew 3-6, usque ad 2-5 dm. longs, divaricatz:. Umbelle radii 6-8, longi 1-2 (rarius 3) dm.; ochres 1-2 cm. longs; umbellularum bracteole longs; radioli breves (sspe subnulli), MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 143 paucispieati. Spice 1—4 em. longs, 5-25-spiculose, spiculis fusco- viridibus, maturis brunneis, adscendentibus neque ullo tempore divaricatis. Spicule longe 10-14 mm., late 3 mm., compresse, 14-22-fore ; rhachille ale late lanceolate, hyaline, persis- tentes; glume 2 ime breves, quadrate. Glume paullo remote, compress, vix carinate, obtuse, 7—-1l-nerviw, in dorso vires- centes, nervo dorsali interdum breviter excurrente. Stamina 3, subinclusa; anthere lineari-oblonge, non aut minute albo-cris- tate, fusco-rubre. Stylus 4 parte nucis brevior; rami 3, exserti. Nux utrinque breviter angustata, nigro-brunnea; cellule extime quadrate, emarcide, hyaline, persistentes, minute porosæ. India Orientalis; China; Australia; Africa tropicalis. India Orientalis, alt. 0-1500 metr., in regionibus occidentalibus a Kashmir usque ad Zeylaniam sat vulgaris: Cabul (Griffith nn. 28, 1267, h. Kew, n. 6167, h. Calcutta) ; montes Suleiman (Dr. Sanders, h. Calcutta); Punjab (T. Thomson, hh. Kew, Calcutta; Aitchison n. 244, h. Kew; Stewart nn. 294, 384, h. Calcutta); Dehra Doon (Jacquemont n. 410, h. Kew); Moradabad (T. Thomson n. 399, hh. Kew, Mus. Brit.) ; India centralis (G. King n. 20, h. Calcutta) ; Sangur (Vicary, h. Calcutta) ; Chota Nagpore (C. B. Clarke nn. 20423, 21220, &c.); Bombay (Dalzell, h. Kew) ; Coimbatore (C. B. Clarke n. 11525) ; montes Nilghiri (G. Thomson, h. Calcutta) ; Zey- lania (Z'waites, C. P. n. 3044, hh. Calcutta, Mus. Brit.). China: Canton (Hance n. 19311, h. Kew). Australia: Portus Darwin (Fitzalan, h. Kew). Africa tropicalis: Abyssinia (Schimper nn. 1021, 1155, Ah. Kew, Mus. Brit.) ; Ethiopia (Kotschy n. 528, h. Kew) ; Gal- labat (Schweinfurth n. 2005, hh. Kew, Mus. Brit.). 113. C. nurans (Vahl, Enum. ii. p. 363); umbella composita aut decomposita; spicis multispiculosis, elongatis, laxis; glumis remotis, ellipticis, obtusis; nuce cum } parte glume sequilonga, vix curvata, trigona, in facie interiore non (aut obscure) concava. — Kunth, Enum. ii. p. 94; Boeck. in Linnea, xxxv. p. 597. C. Jacquini, Schrad. in Ind. Sem. Goettingen, 1835, p. 5, in Linnea, 1837, Litter.-Bericht. p. 87 ; Fenzl in Denksch. Akad. Wissen. 1854, p. 54, t. 1. C. exaltatus, Strachey, Kumaon Pl. p. 74, non Retz. C. distans, Zaker, Fl. Maurit. p. 411 partim ; Hook. Niger Fl. p. 551; Boeck. in Linnea, xxxv. p. 612 partim (quoad Sieber n. 18). Cyperus sp. n. 54, Herb. Ind. Or., Hook. f. et T. Thoms. 144 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Radix fibrosa, valida sepe progemmans (culmis 1—3nis), haud raro lignosa, horizontalis, breve crassa, interdum elongata, de- scendens. Umbella sepe magna, usque ad 1 metr. in diam.; radii elongati cum spicis nutantes. Glume parum compresse, concave, obscurius nervose.—Species a C. eleusinoide ope nota- rum supra datarum difficillime distinguenda; in herbariis etiam cum C. distante confusa que differt spiculis angustioribus, maturis divaricatis, vix aliter. India orientalis, alt. 0-1500 metr., late sparsa; China; insula Mascarensia. India: Sutledge flumen, Rampore (7. Thomson, h. Kew); Gurwhal (G. King, h. Calcutta); Kumaon, Almora (Strachey et Winterbottom n. 8, hh. Kew, Mus. Brit.); Sikkim (J. D. Hooker, h. Kew; C. B. Clarke n. 8899; Kurz, h. Calcutta) ; Assam (Simons, Jenkins, h. Calcutta); Khasia colles ( Grif- Jith, h. Kew n. 6156); Cachar (Keenan, h. Kew); Monghyr (Wallich n. 3347 B, h. Calcutta); Rajmehal (Kurz, h. Cal- cutta); Bombay (.Dalzell, h. Kew); Mysore vel Carnatic (G. Thomson, hh. Kew, Calcutta) ; Coorg (Hohenacker n. 2399, h. Mus. Brit.); montes Anamallay et Pulney (Beddome, h. propr.); Zeylania, Dambool (Thwaites, C. P. n. 3844, hh. Kew, Mus. Brit.). China: Canton (Hance n. 19311, h. Calcutta). Madagascar (Boivin, h. Kew). Mauritius (Sieber n. 18, hh. Kew, Mus. Brit.; Bojer, h. Kew). Rodriguez (Balfour, hh. Kew, Mus. Brit.). Seychelles (Horne nn. 216, 646, h. Kew; Bouton G. 22, h. Kew). 114. C. pistans (Linn. f. Suppl. p. 103) ; umbella composita ; spiculis spicatis, anguste linearibus, multifloris, maturis divari- catis; glumis distantibus, oblongis, obtusis; nuce oblonga, cum 4—5 parte glume xquilonga.—Jacq. Ic. Pl. Rar. ii. p. 8, t. 299; Roxb. Fl. Ind. i. p. 207; Beauv. Fl. d'Owar. i. p. 85, t. 20; Nees in. Wight Contrib. p. 88; Kunth, Enum. ii. p. 93; Dalz. 4 Gibs. Bomb. Fl. p. 288; Griseb. Fl. Brit. West Ind. p. 565 ; Nees in Mart. Brasil. Cyp. p. 40; Baker, Fl. Maurit. p. 411 partim; Boeck. in Linnea, xxxv. p. 012; Benth. Fl. Austral. vii. p. 277. C. elatus, Rottb. Descr. et Ic. p. 37, t. 10; Steud. Cyp. p. 49; Miq.! FI. Ind. Bat. ii. p. 284 (nec Linn., nec Papyrus elatus, Nees). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 145 C. laxus, Vahl, Enum. ii. p. 362. C. squamulatus, Steud.! Cyp. p. 49. C. graminicolus, Steud. ! Cyp. p. 49. Cyperus, Wall.! List n. 3350. Cyperus, Wail.! List n. 3366, litt. A, B, C, D, E, F, G. Cyperacea indeterm., Wallich h. n. 8. Radix fibrosa, interdum brevissime lignescens. Culmus soli- tarius (aut basi progemmans culmis 2-3nis), 3-8 dm. longus, trigonus, apice triqueter, basi longius a vaginis intectus. Folia 3-6, culmum sspe superantia, in marginibus scabra. Involucri bractez 3-5, usque ad 2-4 dm. longi, divaricate. | Umbella sspe 2—4 dm. in diam. ; radii 4-12, usque ad 8-16 em. longi, divaricato- patuli; ochreæ 10-15 mm. longe, apice lanceolato-setacee, folix- formes; umbellule simplices, corymbose, nude aut iterum um- bellate a bracteolis longiuseulis suffulte. Umbellule ultime fere nude, divaricatim 8-4-spicate. Spice 1—4 cm. longæ, 5—30-spi- culose; rhachis glabra. Spicule mature rectangulatim patula, longe 12-24 mm., late 1 mm., 10-20-florz, lineares parum com- presse; rhachille ale anguste, hyaline, persistentes. Glume admodum remote, remote imbricatz, convexs vix carinats, in dorso leviter viridi-7-9-nerviz, in lateribus fusco-rubre, mature sepe pallide. Stamina 3, subinclusa; anthere lineari-oblonge, mutice. Stylus brevissimus ; rami 3, e gluma breviter exserti. Nux obtusiuscule trigona, apice obtuse conica, nigro-fusca ; cel- lule extime quadrate, emarcide, albidx, subnitide, persistentes. —C. nutans, Vahl, differt spiculis sepe pedicellatis (vel si mavis spicis ultimis l-spiculosis), maturis fasciculatim congestis.— C. distans, Sieber, A grost. n. 18, Baker, Fl. Maurit. est C. nutans.— C. distans, Hook. Niger Fl. p. 551, est C. dissolutus, H. B. K. Regio tropica, in fere tota orbe terrarum vulgata. India Orientalis, ab Himalaya usque ad Zeylaniam, communis: Kumaon (Strachey et Winterbottom n. 12, h. Kew); Mora- dabad (T. Thomson n. 1419, h. Kew); Assam (Jenkins nn. 205, 566, k. Kew; Griffith n. 1469, h. Mus. Brit.; Masters, h. Calcutta) ; Khasia colles, alt. 900 metr. (Hook. f. et T. Thomson, h. Calcutta); alt. 1200 metr. (C. B. Clarke n. 5346, h. Calcutta) ; Hindoostan (T. Thomson, h. Calcutta); Ben- galia orientalis (Griffith n. 6156, h. Calcutta); Madras Peninsula (Wight n. 1848, h. Kew ; Wallich n. 3366, B, C, h. Calcutta); Zeylania (Lhwaites, C. P. n. 810, hh. Kew, Mus. Brit., Calcutta). LINN. JOURN. — BOTANY, VOL. XXI. L 146 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Burma et Malay Peninsula, vulgaris: Hotha et Bhamo (J. Anderson, h. Calcutta); Pegu (Kurz nn. 651, 652, 683, h. Calcutta ; Brandis, h. Calcutta); Rangoon ( Wallich n. 3366 G, h. Calcutta); Moulmein (Parish n. 268, h. Calcutta); Mergui (Griffith n. 151, k. Kew); Attran (Wallich n. 3350, h. Cal- cutta); Thyat-Myo (Eug. Oates, h. Calcutta) ; ins. Andaman (Kurz, h. Kew); Malacca (Griffith n. 6197, hh. Kew, Cal- cutta); Singapore (Kurz n. 3001, h. Calcutta; Kunstler n. 44, h. Calcutta). Malaya: Bangka (Amand, h. Kew); Java (Zollinger nn. 459, 460, h. Kew); Buitenzorg (Kurz n. 2740, h. Calcutta); Borneo (Motley n. 91, h. Kew; Beccart n. 868, h. Kew); Ternate (Christian Smith, h. Mus. Brit.). China: Hongkong (C. Wright n. 564, h. Kew ; Forbes, h. Mus. Brit.). Ins. Formosa (Campbell, h. Mus. Brit.). Ins. Philippine (Cuming n. 444, hh. Kew, Mus. Brit.). Australia: Rockingham sinus (Dallachy, h. Kew); Herbert flumen (Bowman, h. Kew). Madagascar (Pervillé n. 473, h. Kew) ; Mauritius (Sieber n. 18, hh. Kew, Mus. Brit.; Bojer, h. Calcutta) ; ins. Joanna (Bojer, h. Kew); ins. Bourbon (Balfour, h. Kew); ins. Comoro, (Hildebrandt n. 1737, h. Kew). Africa: Natal (Buchanan nn. 19, 20, h. Kew) ; Zanzibar (Kirk, h. Kew) ; Abyssinia (Schimper n. 176, hh. Kew, Mus. Brit.) ; Ukamba (Hildebrandt n. 2656, h. Kew) ; Djur (Schweinfurth n. 2314, h. Kew). Ins. S. Helena (Cuming n. 2464, hh. Kew, Mus. Brit.; Melliss nn. 132, 161, h. Kew). America: Brazil (Gardner n. 859, h. Kew; Martius n. 250, h. Kew; Burchell nn. 1070, 1506, 1513, 2745, 4319, A. Kew) ; Guayaquil (Jameson n. 551, hh. Kew, Mus. Brit.). Ins. S. Thomas (Zoepffer n. 145, h. Kew). Jamaica (Swartz, h. Mus. Brit.). Sect. M. Marginati. Marginati, Boeck.in Linnea, xxxv. p. 598, pro magna parte.— Proceri, Kunth, Enum. ii. p. 72 pro majore parte. Majusculi, foliati. Radix fibrosa, valida, aut rhizoma repens vel stoloniferum. Culmus sæpissime solitarius, deorsum com- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 147 planatus, sursum acute triqueter, apice (exsiccatus) faciebus 3 concavis. Umbella composita aut decomposita (in pluribus spe- ciebus interdum depauperata vel congesta) Spicule spicate, compresse, multiflore ; rhachille ale angustissime. Glume concave, elliptice, obtuse (nisi in C. benghalensi), Stamina 8 ; anthere lineari-oblonge, mutice. Nux obovoidea aut oblonga, trigona aut triquetra, apice sspe acuta—Species Asiatice, cum unica Madagascarica, This section is made up of four species evidently closely allied, and C. benghalensis (nov. sp.) which perhaps should not be placed here, and differs a good deal. The two common plants C. pilosus and C. procerus (here treated as conglomerate species) have each a very great range in habit and development, and yet are very difficult to separate from each other. 115. C. xaraCCENSIS (Lam. I.i. p. 146); foliis brevibus ; um- bella composita aut decomposita; glumis remotiusculis, concavis, marginibus apiceque (in sicco) incurvatis; nuce cum 3j parte glume equilonga, anguste oblonga.— Vahl, Enum. i. p. 353; Kunth, Enum. ii. p. 74; Boeck. in Linnea, xxxv. p. 603; Boiss. Fl. Orient. v. p. 872 (syn. C. Enodi excl.). C. monophyllus, Vahl, Enum. ii. p. 352. C. Pangorei, Roxb. Hort. Beng. p. 6, FI. Ind. i. p. 202 (non Rottb. Descr. et Ic. p. 31, t. 7. fig. 3). C. incurvatus, Roxb. ! Fl. Ind. i. p. 196, Ic. ined. t. 2016 in h. Kew. C. spaniophyllus, Steud.! Syn. p. 21; F. Muell. Fragm. Phyt. Austral. viii. p. 260. C. tegetiformis, Benth. Fl. Austral. vii. p. 278, non Roxb. C. gangeticus, Roxb. Ic. ined. n. 1111, h. Kew. Cyperus, Wall.! List n. 3329, litt. M, N. Cyperus, Wall.! List n. 3332, litt. C partim. Cyperus, Wall. ! List n. 3342, litt. C partim. Cyperus, Wall.! List n. 3351 A (=C. gangeticus, h. Roxd.). Cyperus sp. n. 36, Herb. Ind. Or., Hook. f. et T. Thoms. l Rhizoma repens, crassiusculum, a vaginis brevibus nigro-fuscis ovatis acutis intectum ; stolones flaccidi, tenues, elongati interdum adduntur. Culmi parum distantes, 5-10 dm. longi. Folia 3-4; vagine longs, arcte, culmum usque ad 15-20 em. sepe inte- gentes; foliorum lamine lineari-lanceolate, vaginis sæpe bre- viores, 4 partem culmi rarius superantes, interdum brevissimæ, Involucri bracteæ 2-4, divaricatæ, cum Umbellæ radii 3-12, 2-4 cm. L2 in margine fere leves. umbella sæpius subæquilongæ. 148 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. longi aut usque ad 10-13 cm. elongati, apice subcorymbosim umbellulati ; ochree 10-15 mm. longe; umbelle contracte aut subcapitate (=C. incurvatus, Roxb.) aliquando reperiuntur. Umbellule simplices, vel interdum umbellati; bracteole parve. Spice 1-2 em. longe, 4-16-spiculose, rhachis glabra. Spicule patule, lineares, subteretes, cinerem, longe 15-22 mm., late 1-2 mm., 20-40-flore; rhachille ale angusts, persistentes. Glume non carinate, obscurius 5-7-nervie. Stamina 3, subin- clusa ; filamenta subpersistentia; anther lineari-oblonge, mu- tice. Stylus brevissimus ; rami 3, e gluma fere omnino extrusi. Nux -4plo longior quam lata, apice acuta, trigona, nigro-fusca ; cellule extimæ quadrate, emarcide, hyaline, obscuræ.— C. enodis, Boeck., hue a Boiss. adductus, multum differt culmo obtuse trigono, rhachille alis latis, nuce obovoidea &e. Asia ausirali-orientaiis cum Malaya, in estuariis frequens. Persia: Basorali (Haussknecht, h. Kew); ins. Karakin (Noe n. 898, h. Kew). Bengal: Noakhali (J. D. Hooker, h. Kew); Calcutta (Kurz, h. Calcutta) ; Soonderbun (C. B. Clarke) ; Bengal orientalis (Griffith, h. Kew n. 6206; Wallich n. 3342 C, h. Calcutta) ; Dacca (C. B. Clarke n. 16958, hh. Kew, Calcutta; Keenan, h. Kew). Peninsula Malayana: Pegu (Kurz, h. Calcatta) ; Arracan in ripis fl. Coladyne (Kurz, h. Calcutta); Mergui (Griffith n. 317, h. Kew n. 6147) ; Singapore (Kunstler n. 106, h. Cal- cutta). Japonia (Dickens, h. Kew). China (Fortune n. 7, hh. Kew, Mus. Brit.). Cochinehina (Lebæuf, h. Kew). Ins. Lombok ( Wallace, h. Kew). Borneo (Motley n. 1294, h. Kew). Australia: Victoria flum. (F. Mueller, h. Kew). 116. C. prtosus (Vahl, Enum. i. p. 354); foliata; umbella simplice aut composita, umbellulis congestim corymbosis ; spicis elongatis, rhachi pilosa; spiculis undique divaricatis; glumis ovatis, concavis, obtusis, plurinerviis, in margine scariosis; nuce cum 2-3 parte glume :zequilonga, obovoidea, apice acuta, acute triquetra.— Kunth, Enum. ii. p. 80; Thwaites, Enum. Pl. Zeyl. p. 944; Boeck. in Linnea, xxxv. p. 598 (syn. C. procero excl.); Benth. Fl. Austral. vii. p. 275. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 149 C. fimbriatus, Nees! in Wight Contrib. p. 86; Kunth, Enum. ii. p. 100; Miq. in Ann. Mus. Lugd.-Bat. ii. p. 142. C. marginellus, Nees! in Wight Contrib. p.86; Kunth, Enum. ii. p.75; Hook. et Arn. Bot. Beechey Voy. p. 222; Miq. in Ann. Mus. Lugd.- Bat. ii. p. 141. C. honestus, Kunth! Enum. ii. p. 74. C. pauciflorus, Steud. ! Cyp. p. 34. C. hebes, Steud. Cyp. p. 315. C. piptolepis, Steud.! Cyp. p. 40; Miq. Fl. Ind. Bat. iii. p. 279. C. subalatus, Steud. Cyp. p. 31. C. Wallichii ?, Wight in h. propr. n. 2390; Wallich in h. propr. partim, non Nees. Cyperus, Griff. Itin. Notes, p. 27, n. 422. Cyperus, Wall. ! List n. 3334. Cyperus, Wall. ! List n. 3342, litt. C partim. Cyperus, Wall.! List n. 3336, litt. E. Cyperus, Wall. ! List n. 3348. Cyperus, Wali. ! List n. 3355, litt. D, E, F, G, H, I. Cyperacea indeterm., Wallich h. n. 9. Cyperacea, Griff. Itin. Notes, p. 89, n. 1313. Radix fibrosa. Culmus solitarius (e basi haud raro progem- mans), 3-8 dm. longus, apice acute triqueter. Folia 3-5, valida, cum culmo sspe squilonga, in marginibus serrulato-scabra. Involueri bracteze 3-5, usque ad 2-4 dm. longa, divaricate, in marginibus acriter scabre etiamque ssepe plus minus (in C. Jimbriato conspicue) pilose. Umbelle radii 3-8, usque ad 10-15 em. longi (sepius 1-5 cm.); ochres usque ad 3-5 em. longe; umbellule seepius abbreviatim corymbose, 3~10-spicate ; radioli ssepe 1-spicigeri, raro elongati, divisi. Spice 1-6 em. longs, 10—40-spiculoss ; rhachis (in formis typicis vulgatisque) presertim secus angulos fulvo-pilosa, in forma maxime australi (=C. marginello, sp. Nees) fere glabrate; spicule inter ge 0-6 mm. distantes, mature rectangulatim divaricatz, basi a seta subuliformi 2-7 mm. longa sspissime suffulte. Spicule longs 10-15 mm., late 2 mm., compresse, 10-24-flore, pallide rubes- centes aut castaneo-brunne:, in eadem umbella quoad colorem variabiles ; rhachilla vix alata. Glume fructigerze vix imbricate, ovate, obtuse, concave nec carinate, in dorso viridi-7-13-nervie ; latera enervia modo late rubro-scariosa, modo sanguinea, in mar- gine ipso nitide hyalina. Stamina 3, subinclusa ; anther lineari- oblongæ, muticæ. Stylus brevissimus ; rami 3, e gluma breviter exserti. Nux nigra ; cellule extimæ quadratæ, emarcidæ, hyaline, 150 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. non eonspieus.—Species polymorpha a rhachi spicarum fusco- pilosa seepius distinguenda: in C. procero autem rhachis interdum subpilosa videtur; contra C. marginellus, Nees (mihi forma C. pilosi) in h. Kew cum C. procero conjunctus erat. Asia australi-orientalis, eum Malaya, China, Japonia, et Aus- tralia tropica. India Orientalis, alt. 0-1500 metr., ab Himalaya usque ad Zey- laniam, vulgaris: Kumaon (Strachey et Winterbottom n. 11, hh. Kew, Mus. Brit.); Dehra Dhoon (G. King, h. Calcutta); India boreali-oecidentalis (Royle n. 29, h. Kew); Nepaul (Wallich n. 3355 H, A. Calcutta); Sikkim (J. D. Hooker, h. Calcutta; Kurz, h. Calcutta); Assam (Jenkins n. 197, h. Kew); Suddiya (Griffith, h. Calcutta); Khasia colles (Griffith nn. 422, 1313, h. Kew n. 6162/1; J. D. Hooker n. 1517, h. Kew; Griffith n. 422, h. Calcutta); alt. 1300 metr. (C. B. Clarke n. 16154, h. Calcutta); Sylhet (Wallich n. 3355 E, h. Calcutta; C. B. Clarke n. 6961, h. Calcutta); Bengalia orientalis (Griffith n. 6154, h. Kew, n. 6195 hh. Kew, Calcutta); Bhotan, alt. 1200 metr. (Gamble n. 9599, h. Calcutta); Bombay (Burn, h. Mus. Brit); Carnatic (G. Thomson, h. Calcutta); Madras Peninsula (Heyne in Wallich n. 3334 A, h. Calcutta); montes Nil- ghiri (G. Thomson nn. 4, 20, h. Kew); Zeylania (Col. Walker, h. Kew; Thwaites, C. P. n. 797, hh. Kew, Mus. Brit., Cal- cutta). Burma et Malay Peninsula: Arracan (Kurz, h. Calcutta); Pegu (Kurz nn. 648, 649, 2674, h. Calcutta); Irrawaddy flumen (Wallich n. 190 h. propr., h. Calcutta); Mergui (Griffith n. 181, h. Kew n. 6196, n. 6196 h. Calcutta); Te- nasserim (Helfer n. 796, h. Kew n. 6208/1, n. 6208/1 h. Calcutta); Malacca (Griffith, h. Kew nn. 6152, 6208, n. 6152 A. Calcutta); ins. Nicobar (Kurz n. 25977 bis, h. Kew). Malaya: Bangka (Amand, h. Kew); Singapore (Kurz, h. Cal- cutta; Kunstler n. 111, h. Calcutta); Java (Zollinger nn. 267, 457 partim, 458, h. Kew); Buitenzorg (Kurz n. 182, h. Calcutta; Forbes n. 952, h. Calcutta); Borneo (Motley n. 99, h. Kew). Japonia: Nagasaki (Oldham n. 900, h. Kew). China: Hongkong (C. Wright n. 562, h. Kew); Macao (Vachell n. 681, h. Kew). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 151 Ins. Philippine (Cuming n. 535, hh. Kew, Mus. Brit.). Australia: Brisbane flumen (Bailey, h. Kew). Var. B. obliqua ; spiculis paucifloris, interdum 5-6-floris sepe pallidioribus.— Nees! in Wight Contrib. p. 86 (sp.); Kunth Enum. ii. p. 60; Boeck. in Linnea, xxxv. p. 611. C. quinqueflorus, Hochst! in Pl. Hohenack. n. 944; Steud. Cyp. p. 37. C. pennatus, Steud. in Zoll. Verz. Ind. Archip. p. 62, non Nees. Cyperus, Wall. ! List. n. 3334, A partim, B partim, C partim, D partim. Varietas in C. pilosum typicum sensim transeuns. Nux eum 4 parte glume vix equilonga. i Nepaul (Wallich, h. Calcutta). Kumaon ( Wallich n. 3334 C, h. Calcutta). Sikkim (7. Anderson n. 1342, h. Calcutta). Bengal et Nepal (Wallich n. 3334, B, D, h. Kew). Pegu (Kurz n. 649, h. Calcutta). Java (Zollinger n. 457 partim, A. Kew). Var. y. polyantha; spiculis usque ad 20-26 mm. longis, 40-45- floris, angustioribus autem quam C. pilosi typici; glumis minori- bus, fere equilongis ac latis. Bengalia: Mymensingh (C. B. Clarke n. 7763, h. Kew). Var. à. babakensis, Steud.! in Zoll. Verz. Ind. Archip. p. 62 (sp.); spiculis in spicas densissimas breviter cylindricas congestis. —JMiq. Fl. Ind. Bat. ii. p. 257 ; Boeck. in Linnea, xxxv. p. 251. C. Babakan, Steud. Cyp. p. 6. Java (Zollinger n. 693, h. Kew). 117. C. BENGHALENSIS ; umbella simplice; spiculis dense fas- ciculatim spicatis, rhachi glabra; glumis ovatis, obtusis, navi- eularibus, in carina sursum hispido-scabris ; nuce obovoidea, triquetra, brevissime rostrata, cum 3 parte glume vix equilonga. Radix fibrosa. Culmus solitarius, 5 dm. longus, deorsum compressus, sursum triqueter levis. Folia 3-4, cum culmo fere equilonga, lata 5-8 mm., acute carinata, in marginibus scabra. Involucri bractese 3-4, usque ad 3-4 dm. longe, divaricate. Um- belle radii 4-5, usque ad 4-6 em. longi, trigoni. Spice sub- ebracteate, longe 25 mm., late 2 cm., 20-30-spiculosx ; rhachis vix scabra. Spicule undique dense patule, long» 12 mm., lat» 3 mm., eompresse, 12-18-flore ; rhachille ale anguste», sca- riose, rubro-maculate, persistentes; gluma ima spe caudata, Glumæ spissæ, fuscw, in dorso arcte ll-nervie, margine hya- 152 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. lino nullo aut angustissimo. Stamina 3; filamenta angusta, levia; anthere lineari-oblonge, mutice, rubro-notate. Stylus nuce brevior; rami 3 longi, rubro-maculati. Nux nigra; cellule extime quadrate ; emarcide, albide, persistentes, conspicue reti- culate. [I believe this is the same plant as Buchanan-Hamilton’s in Wallich h. n. 3336 E, collected at Nathpur in North Bengal; but at Kew I did not separate this species from C. babakensis.] Bengalia (Griffith, h. Kew n. 6207); Mudhopoor (C. B. Clarke n. 7773, h. Kew; n. 7773, 7787, h. Calcutta). 118. C. naTrFOLIUS (Poir. Encyc. vii. p. 268); foliis bracteisque longis, latis; umbella composita, umbellulis corymbosim dense multispicatis; spiculis pallidis; glumis rotundis.— Kunth, Enum. ii. p. 75; Boeck. in Linnea, xxxv. p. 602. Mauritius ( Prescott, h. Calcutta). 119. C. procerus (Rottb. Descr. et Ic. p. 29, t. 5. fig. 3); robusta, foliata; umbella specie simplice, radiis apice corymbosim 3—5-spicatis ; spiculis admodum remotis, majuseulis; glumis late ellipticis, obtusis, ecarinatis ; nuce cum 3 parte glume equilonga, obovoidea, triquetra.—Roth, Catalecta Bot. p. 5; Roxb. Fl. Ind. i p. 203 partim (nec Roxb. Ic. ined. $. 724) ; Kunth, Enum. ii. p. 72; Nees! in Wight Contrib. p. 883; Thwaites, Enum. Pl. Zeyl. p. 343. C. torosus, Rottb. Descr. et Ic. p. 40, t. 11. fig. 3; Vahl, Enum. ii. p. 339. C. ornatus, R. Br.! Prod. p. 217 ; Kunth, Enum. ii. p. 112; Benth. FI. Austral. vii. p. 276. C. carnosus, Heyne; Nees! in Wight Contrib. p. 83. C. Heynei, Boeck. in Linnea, xxxv. p. 600. C. spadiceus, Heyne! non Lam. (fide Rottler in h. propr.). Cyperus, Wall.! List n. 3355 A. —, Rheede, Hort. Mal. xii. t. 50. Radix fibrosa, non lignescens; stolones elongati, tenuiores, herbacei, e basi culmi erumpentes. Culmus solitarius, 5-8 dm. longus, apice triqueter. Folia 3-5, cum culmo sspe fere æqui- longa, complicata, valida, in margine fere levia. Involucri bracteæ 3-5, usque ad 2-4 dm. longs, divaricate, Umbelle radii 3-7, usque ad 4-15 em. longi, apice subebracteati, 1-8-spicati ; ochreæ 1-2 em. longe; umbellarum radioli 1-spicati, interdum longi- usculi, basi ochreati, apice ebracteolati. Spice 2-8 em. longa, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 153 15-30-spiculosz ; rhachis angulosa, flexuosa, glabra vel in angulis minute pilosa. Spicule inter se admodum remote, mature diva- ricatim patule, basi nudæ aut raro a seta conspicua suffulte, longe 2 cm., late 4 mm., compresss, subturgidz, 30-46-flore ; thachilla subexalata. Glume adpresso-imbrieatz, marginibus involutis, einnamomeo-rubescentes vel pallidiores, in dorso plano- concavo concoloriter 9—11-nervie; latera latiuscule estriata; margines hyalini, crispato-undulati, specie (exsiccati) serratuli. Stamina 3, subinclusa; anthere oblonge, truncatz, obsolete cristate, rubescentes. Stylus cum 4 parte nucis squilongus; rami 3, longiuseuli, rubro-maculati, exserti. Nux obtusiuscula, nigra ; cellule extime quadrate, emarcide, tenues, inconspicuz.— In exemplis C. proceri, ante explicationem florum, spicule vi- dentur subteretes; hee sunt C. torosus lectorum veterum.—In C. ornato, R. Br., glume intensius castaneo-rubre.—C. procerus (in herbariis veterum) eum C. puncticulato maxime confusus est. Asia australi-orientalis cum Australia; in sstuariis maritimis etiamque in montanis. India, alt. 0-1000 metr., ab Himalaya usque ad Zeylaniam (Wallich nn. 3355 B, 1203, h. Mus. Brit.); Chumba (C. B. Clarke n. 23683, h. Kew); Cooch Behar (C. B. Clarke n. 23683, A. Kew); Chota Nagpore (C. B. Clarke n. 25079, k. Kew); Madras Peninsula (Heyne, h. Kew; Koenig, h. Kew; G. Thomson, h. Calcutta; Wight nn. 1832, 2309, R. Kew); Zeylania (Col. Walker, h. Kew; Thwaites C. P. n. 3752, hh. Kew, Mus. Brit., Calcutta) ; Soonderbun, com- munis (C. B. Clarke n. 25079 &c.). Arracan; in estuariis submersis (Kurz, h. Calcutta). Java ( Labillardiére, h. Kew). China: Amoy (Hance n. 1392, h. Kew); Shanghai (Maingay n. 639, k. Kew). Australia (R. Brown n. 5897, hh. Kew, Mus. Brit.) ; Queens- land (Amalia Dietrich n. 609, h. Mus. Brit.). [There is a fragment in h. Caleutta, collected in Maunbhoom (Chota Nagpore) by V. Ball which may be a species closely allied to C. procerus ; the spikelets are large, comose from the long stigmas, the rhachis of spike fuscous-pilose. ] Sect. N. Corymbosi. Corymbosi, Kunth, Enum. ii. p. 53; Boeck. in Linnea, xxxvi. p. 271. 154 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Radix (in speciebus gerontogeis typicis) perennis, rhizomate repente, stolonibus sspe additis; in speciebus americanis (hoc loco dubie locatis) annua aut fibrosa. Culmi trigoni vel teretes. Folia evoluta, aut in paucis abbreviata vel subnulla. Umbella sepius laxa aut aperta, radiis longis aut conspicuis; spicule in pluribus numerosz, spicate (in paucis americanis globoso-fasci- culate), compresse, multiflore, sepius acute. Rhachille ale sepius conspicue, persistentes aut solubiles. Glume naviculares sepius mutice, rarius breviter mucronate. Stamina 3. Styli rami 3, sepius elongati. Nux sepius ellipsoidea, trigona, cum 3-3 parte glume equilonga, nec minima nec maxima. This group here contains all the species of Eucyperus which I have not been able to get into the preceding sections. The Old- World species form a tolerably natural group, except C. stolont- Jerus and C. macer, which perhaps should each have a section to itself. The American species, though some of them brought here by all authors, appear to me to differ considerably ; C. sphacelatus with its annual root and narrowly-winged rhachilla is very near C. compressus. I have brought here C. strigosus and C. congestus, two common species placed by Boeckeler in a section called Fas- eieulati, which from its character differs from the Corymbosi in " spiculæ sspissime disco impositz;"' but neither C. strigosus nor C. congestus is said (in its own description) to have such discs. I hold that two very different things are included as discs by Boeckeler. In many species of Cyperus where the spikelet is at all oblique in its axis, in the dried state the rhachilla contracts very near the base below the lowest glume: here it seems to sit on a small cushion, but it is very firmly attached, and does not disarticulate at this point. This appearance (often only a result of drying) is not uncommon throughout the genus, asin C. poly- stachyus, where it is often prominent; and I estimate it as of no classificatory importance. But in a number of species of Cyperus (Mariscus for me) above the two lowest barren glumes, the rhachilla of the spikelet dis- articulates by a clean cut, leaving a small dise or cushion within the two barren glumes: this character may, I think, be made of the greatest use in a natural classification of the species. I have divided this large section into three “ series " for conve- nience; but the series pass into one another at various points. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 155 . * . . . Series * Folia culmo breviora, sepe abbreviata vel subnulla. 120. C. arricunarus (Linn. Sp. Pl. p. 66); subaphyllus ; culmo terete, apice vix trigono, in sicco nodoso ; umbella com- posita ; spiculis breviter spicatis, pallidis vel subferrugineis ; glumis elliptieis, arctius imbricatis; nuce late oblonga, cum 1 parte glume squilonga.—Z. B. K. Nov. Gen. et Sp. i. p. 202; Torrey in Ann. Lyceum New York, iii. p. 256; Kunth, Enum. ii. p.53; Nees in Wight Contrib. p. 80, in Mart. Brasil. Cyp. p. 35; Griseb. Fl. Brit. West Ind. p. 564; Baker, Fl. Maurit. p. 412; Boeck, in Linnea, xxxvi. p. 274; Benth. Fl. Austral. vii. p. 278; Boiss. Fl. Orient. v. p. 874. C. niloticus, Forsk. Fl. ZEg.-Arab. p. 13; Beauv. Fl. d'Owar. ii. p. 63, t. 97. fig. 23; Vahl, Enum. ii. p. 302. C. nudus, Roxb. Fl. Ind. i. p. 187 (neque homonymus, Roxb. Fl. Ind. i. p. 209), neque H. B. K. C. gymnos, Roem. et Sch. Syst. Mant. ii. p. 97. Cyperus, Wall. ! List n. 3364. —, Sloane, Jamaica, i. p. 121, t. 81. fig. 1. Stolones crassi, a squamis ovato-lanceolatis laxis atro-fuscis intecti. Culmi usque ad 1-2 metr., basi crassi, vivi leves nec nodosi, in sieco nodosi neque articulati. Vagine 3-4, culmum usque ad 2-3 dm. integentes, rarius breviter foliigere. Involucri bracte; longs 1-5 cm., lanceolate, basi erect, raro usque ad 1 dm. elongatz, in marginibus fere leves. Umbelle radii 4-12, usque ad 8-12 em. longi; ochrese 1 em. truncate; umbellule congestm, subcorymboss; spice numerose, ebracteate, breves, multispiculose. Spicule fasciculatim confluentes, longs 1-4 em., late 2 mm., 12-50-flore ; rhachille ale oblonge, solubiles, rubre aut albo-hyaline; gluma ima brevis. Glums concave, non carinate, in dorso obscurius 3-5-nervis, viridescentes, in lateribus rubre aut pallide. Stamina 3, inclusa; anthers lineari- oblonge, mutice. Stylus 4 parte nucis brevior; rami 3, breviter exserti. Nux utrinque angustata, trigona, nigra; cellule extime quadratz, emarcidz, hyaline, inconspicus, persistentes. In regionibus calidioribus orbis fere totius. Asia: India Orientalis, Calcutta (Roxburgh, h. Kew); Bengal (Griffith, h. Kew n. 6214); Soonderbun, vulgaris (C. B. Clarke) ; Noakhali (C. B. Clarke n. 8204) ; Madras Penin- sula (Wight n. 1823, hh. Kew, Mus. Brit.); Zeylania (Thwaites C. P. n. 3561, hh. Kew, Mus. Brit., Calcutta ; 156 MR. €. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Ward n. 74, h. Mus. Brit.) ; Bangkok (Koenig, h. Mus. Brit.). Australia: Portus Essington (Armstrong n. 568, h. Kew). Madagascaria (Gerrard, h. Kew) ; Mauritius (h. Calcutta) ; ins. Seychelles (Horne nn. 237, 637, h. Kew) ; ins. Bourbon (Balfour, h. Kew). Africa australis (Drége, h. Kew); Zambesia; Shire flumen (Kirk, h. Kew); Zanzibar (Hildebrandt n. 1058, h. Mus. Brit.). Arabia (Ehrenberg, h. Kew) ; Oriens (forskael, h. Mus. Brit.); Egypt, Ramleh (Parquet n. 44, h. Mus. Brit.); Damietta (Sieber, h. Mus. Brit.). Nilus Albus (Petherick, h. Kew); Terra Monbuttu (Schwein- Surth n. 3446, h. Kew). Africa occidentalis (Mann n. 894, h. Kew); Angola ( Mon- teiro, h. Kew) ; Congo (Christian Smith nu. 14, 26, h. Mus. Brit.) ; Senegambia (Brunner, h. Kew); Senegal (Sieber n. 101, h. Mus. Brit.). Madeira (Lowe, h. Mus. Brit.). America: Florida (Curtiss n. 8024, h. Kew; Rugel n. 156, h. Mus. Brit.); New Orleans (Drummond n. 383, hh. Kew, Mus. Brit.) ; Texas (Berlandier n. 3222, h. Kew); Mexico (F. Mueller n. 2151, h. Kew); Nicaragua (Tate n. 87, hh. Kew, Mus. Brit.). Jamaica (Purdie, h. Kew). Martinique (Hahn n. 1258, h. Kew). America equatoria: Guiana, Cayenne (Rothery, h. Mus. Brit.) ; Columbia (Karsten, h. Mus. Brit.) ; New Granada, La Paila (Holton n. 119, hh. Kew, Calcutta) ; Venezuela (Funcken. 599, h. Mus. Brit.; Sagot n. 644, h. Mus. Brit.) ; Surinam (Hostmann n. 432, hh. Kew, Mus. Brit.) ; Para (Spruce n. 87, h. Kew) ; Pernambuco (Gardner n. 1210, hh. Kew, Mus. Brit.); Amazon superior (Traill n. 1163, h. Kew); Bahia (Blanchet nn. 576, 2929, h. Kew). 121. C. noposus (Willd. Enum. Hort. Berol. i. p. 72) ; culmo terete apice trigono, in sicco nodoso ; foliis 1-2, culmo multum brevioribus; involucri bracteis umbella brevioribus; spiculis breviter spicatis, vix divaricatis, castaneo-brunneis, paullo com- pressis; glumis ellipticis in dorso viridi-nerviis; nuce cum 3 parte glume æquilonga, trigona.— Kunth, Enum. ii. p. 54; Boeck. in Linnea, xxxvi. p. 275. C. articulatus, Benth. Pl. Hartw. p. 124. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 157 Stolones tenuiores. Culmi 5 dm. longi, tenuiores. Folia 18 em. longa. Peruvia borealis (Spruce n. 6413, h. Mus. Brit.). Guayaquil (Hartweg n. 702, h. Mus. Brit.). 122. C. rEGETIFORMIS (Roxb. Hort. Beng. p. 6, Ic. ined. t. 1821 in h. Kew); subaphyllus; culmo trigono, apice ipso sepe triquetro; involucri bracteis brevissimis, rarius cum 1 parte umbellz composite zquilongis; spieulis spicatis, subtereti- bus, subulatis, pallidis, floris tempore rectangulatim patentibus ; styli ramis e gluma floris tempore arctius imbricata longius exsertis.— Arnott in Wight Contrib. p. 19 in nota; Kunth, Enum. ii. p. 56. C. nudus, Roxb. Fl. Ind. i. p. 209 neque homonymus, Roxb. Fl. Ind. i. p. 187. C. bengalensis, Spreng. Neue Entdeck. iii. p. 101. C. Gula-Metthi, Roem. et Sch. Syst. Mant. ii. p. 125. C. enodis, Boeck. in Linnea, xxxvi. p. 271. Cyperus, Wall.! List n. 3351, A n. 2, B, C partim, H nn. 1, 2. Stolones crassi, longiusculi, teretes; squame 3-4 cm., ovato- acute, nigro-cinerez. Culmi 6-16 dm., in sicco sspe obscurius transversim nodosi, trigoni, basi bulboso-incrassati, apice in tabula Roxburghii n. 1321 triquetri, in exemplo autem Roxburghii (in Wall. List 3351 A n. 2 conservato) fere teres, in exemplis numerosis vivis visis triquetri aut saltem trigoni. Folia subnulla aut usque ad 2-8 cm. elongata; vagine plures, usque ad 15 cm. eulmum integentes. Involucri bracteæ 2-3, 2-5 cm. longe, lan- ceolate, rigide, erectz, basi auriculate, in marginibus scabra, rarius paullo elongate sed in exemplis ab umbella contracta coro- natis cum umbella fere equilonge. Umbella 3-35 em. in diam., composita aut decomposita ; radii usque ad 15, usque ad 13 cm. longs, tenuiores ; ochrez usque ad 15 mm. longs, acumine lan- ceolato. Umbellule corymboss ; radioli usque ad 8, usque ad 6 em. longi, conspicue ochreolati, aut umbellule sepius congestz. Spice 4-16-spiculose. Spicule 1-3 em. longe, 10-30-florz, floris tempore divaricato-patule, vix compresse; rhachille ale ob- longs, solubiles. Glume paullo remote, longe 2 mm., elliptiez, obtuse aut emarginate, concave non carinate, in dorso 5-9- nervis, pallidz fusco- vel rubro-notate. Stamina 3, subinclusa ; antherz lineares, mutice, rubre. Stylus nuce brevior; rami 3, longi. Nux cum 4 parte glume equilonga, oblonga, trigona, sub-plauo-convexa, nigra; cellule extime quadratz, emarcide, 158 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. persistentes, inconspicue.—Species a C. corymboso vix satis diversa. Bengalia; China; Japan. Bengalia (Roxburgh, h. Kew; Griffith, h. Kew n. 6204) ; Cal- cutta (Kurz, h. Calcutta; T. Thomson, h. Calcutta) ; Chit- tagong (Hooker f. et T. Thomson n. 401, h. Kew) ; Noakhali (J. D. Hooker n. 12, h. Kew; C. B. Clarke nn. 8194, 8208, hh. Kew, Calcutta); Burisal (C. B. Clarke n. 20036, h. Kew) ; Mymensingho (C. B. Clarke n. 17308, hh. Kew, Cal- cutta) ; Pundua (J. D. Hooker n. 356, h. Kew). Assam (Griffith n. 1460, h. Mus. Brit.; Jenkins, h. Cal- cutta). China: Whampoa (Hance nn. 1392, 6193, k. Kew). Japan (Bissett, h. Kew). 123. C. CORYMBOSUS (Rottb. Descr. et Ic. p. 42, t. 7. fig. 4); subaphyllus aut breviter foliatus; culmo subtrigono, apice ipso terete aut obscurius trigono ; involucri bracteis brevibus aut um- bella decomposita sublongioribus ; spiculis spicatis, paullo com- pressis, subferrugineis aut rubescentibus, florigeris fastigiatim suberectis; glumis florigeris imbricatis, styli ramis modice ex- sertis.— Boeck. in Linnea, xxxvi. p. 277 partim, nec Kunth. C. diphyllus, Retz.! Obs. v. p. 11; Kunth, Enum. ii. p. 54; Boeck. in Linnea, xxxvi. p. 2/2; Benth. Fl. Austral. vii. p. 279 partim ; Boiss. Fl. Orient. v. p. 374. C. Koenigii, Vahl, Enum. ii. p. 302; Kunth, Enum. ii. p. 54. C. monophyllus, Vahl, Enum. ii. p. 352. C. seminudus, Rozb.! Fl. Ind. i. p. 187 ; Nees in Wight Contrib. p. 80; Kunth, Enum. ii. p. 55. Papyrus Pangorei, Nees! in Wight Contrib. p. 88 partim. Cyperus, Wall.! List n. 3351, C, E partim, F, G, H n. 3, A. C. tegetiformi vix differt spiculis paullo latioribus (in eadem spica quoad longitudinem spe conspicue variabilibus), florigeris erecto-confluentibus (nee divarieato-patulis), culmo apice non triquetro, bracteis brevibus sed ssepius cum 4-3 parte umbelle equilongis (ut in Wall. List n. 3351, G, H n. 3, &c., vidende sunt). In exemplo typieo C. diphylli, Retz., in bh. Mus. Brit. conservato, folia 2 usque ad 14 em. longa, bractee umbellam plane superant.— C. fegetum, Roxb., omnino differt glumis magis distantibus, fruetigeris vix imbricatis. India Orientalis: Assam (Griffith n. 1459, kh. Mus. Brit.; Masters, Simons, h. Calcutta) ; Kumaon (Wallich n. 3351 E, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 159 h. Calcutta); Khasia (Hooker f. et T. Thomson, h. Cal- cutta); Sylhet (Wallich n. 8351 F, h. Calcutta) ; Bengal (C. B. Clarke nn. 8523, 17951) ; Madras Peninsula (Rottler, h. Kew; Koenig, h. Mus. Brit.; Wight in Wallich n. 3351 C, h. Calcutta) ; Zeylania (Thwaites C. P. n. 809, hh. Kew, Mus. Brit., Calcutta). Arracan (Kurz n. 667, h. Calcutta). Pegu (Kurz nn. 668, 669, 2683, h. Calcutta). Australia borealis: Portus Essington (Armstrong, h. Kew). Var. P. Pangorei, Rottb. Descr. et Ic. p. 31, t. 7. fig. 3 (sp.) ; spiculis usque ad 3 cm. longis, 18-36-floris—Kunth, Enum. ii. p. 57, nec Roxb., nec Papyrus Pangorei, Nees. C. corymbosus, forma macrostachya, Boeck. in Linnea, xxxvi. p. 277. Madras Peninsula (Wallich n. 3351 C, k. Kew); Tinnevelly (Bidie, h. Kew, tegetum materies fide Bidie). Ins. Nossibé prope Madagascar (Boivin, h. Kew). 124. C. scan1osus (E. Br.! Prod. p. 216) ; culmo pertenui; folis 2-3, 4 parte culmi brevioribus ; involucri bracteis 3, cum umbella composita contracta vix æquilongis ; spiculis fere C. tege- tiformis sed suberectis.— Kunth, Enum. ii. p. 112. C. pertenuis, Roxb. Fl. Ind. i. p. 198, Ic. ined. t. 1320 in hh. Kew, Calcutta; Nees in Wight Contrib. p. 83; Kunth, Enum. ii. p. 99. Stolones graciliores. Culmi solitarii aut approximati 5-7 dm., inter omnes vicinos tenuissimi, apice trigoni vix triquetri. Folia 3, summum ad 6-12 cm. longa. Involucri bractes 3, usque ad 6 cm. longz, tenuiores. Umbella composita aut subsimplex, 5-8 em. in diam., radii usque ad 7, graciles ; umbella 30—50-spiculosa. Spice 3-7-spiculoss. Spicule longs 1 cm., late 1 mm., teretes, pallide fusco-roses ; rhachille ale late, hyaline, solubiles. Glume arcte imbricate, subacute, apice sspe eroso-denticulate. — Inflorescentia fere C. corymbosi; spicule autem quam C. tegeti- formis tenuiores; culmi tenues ab utraque specie distant. Spe- cies dubie bona (an C. tenuiflori, Rottb., forma depauperata ?), pluribus aliis simillima, cum nulla alia omnino congruens. Bengal: Calcutta (fide Roxburgh); Jessore (C. B. Clarke n. 8465, hh. Kew, Calcutta). Pegu (Kurz nn. 683, 684, h. Calcutta). Australia: in ora boreali (R. Br. nn. 5898, 5894, h. Mus. Brit., n. 5993 partim h. Kew, n. 5994 partim h. Kew) ; in ora orientali (R. Brown n. 5895, h. Mus. Brit.). 160 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 125. C. MACER; culmis cespitosis, gracilibus ; foliis brevibus, angustis ; umbella tenui, laxa, subcomposita; spieulis 2—4nis, digitatis, multifloris ; glumis ovatis, obtusis. Rhizoma brevissime repens, gracile, radicibus filiformibus nigris. Culmi numerosi, angustissimi, longi 5-8 dm., trigoni. Folia pauca, vix vagina longiora, cum 3 parte culmi equilonga. Involueri bractee 2-3, anguste, umbelle breviores, 2-4 cm. longe. Umbelle simplices vel parce composite, 5-8 em. in diam.; radii tenues ; umbellule laxe corymbosz pauci-spiculose ; umbellularum braeteole inconspicue. Spicule 8-16-flors, compresse ; rhachille ale anguste, non solubiles. Glume remotiuscule, in dorso 5-7-nervie, virides, in lateribus pallide brunnee, enervie. Stamina 3 ; anthere lineari-oblonge, mutice. Stylus longus, glumam superans, trifidus. Nux cum $ parte glume equilonga, obovato-oblonga, obtuse trigona, nigra; cel- lule extime quadrate, proventu albide, decidus.— Exempla Kurzii copiosa, optima. Spicule iis C. tegetwm similes; habitus autem cum nuce speciem cum C. Helferi (inter Elegantes) con- jungunt. Pegu: in ripis fluminis Khaboung (Kurz n. 671, h. Cal- cutta). Chittagong: in saxis fluminis Kornophullee (C. B. Clarke n. 19779, hh. Kew, Calcutta). 126. C. TEGETUM (Roxb.! Fl. Ind. i. p. 208) ; robustior ; foliis culmo multo brevioribus aut subnullis; involucri bracteis 3-5, umbellam sspe multo superantibus; umbella composita aut sepius decomposita; spiculis breviter spicatis, florigeris erecto- patulis, multifloris, compressis; rhachille alis angustis, pro- ventu solubilibus; glumis ellipticis, in exemplis exsiccatis jam ante florum expansionem distantibus, marginibus arcte in- curvis; nuce oblonga, trigona, cum 3 parte glume equilonga. C. corymbosus, Koenig in h. propr. partim; Kunth, Enum. ii. p. 55; Strachey, Cat. Kumaon Pl. p.74; Boeck. in Linnea, xxxvi. p- 277 pro majore parte; non Roxb. C. Schimperianus, Steud.! Cyp. p. 34; Boeck. in Linnea, xxxvi. p. 277. C. dehiscens, Steud. Cyp. p. 31. C. longus, Hochst. in Schimper n. 57 ; Oliver in Trans. Linn. Soc. xxix. p. 165, non Linn. C. Pangorei, Thwaites, Enum. Pl. Zeyl. p. 344, non Rottb. C. elongatus, Hochst. in Schimp. n. 1994, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 161 C. nudiculmis, Sieber, MS. in h. propr. Papyrus dehiscens, Nees ! in Wight Contrib. p. 89. C. Pangorei, Nees! in Wight n. 1844, h. propr., in Wight Contrib. p. 88 pro majore parte. C. corymbosus, Nees in Wight Contrib. p- 89 (syn. Rottb. excl.). Cyperus, Wall.! List n. 3332 A (=C. tegetum, A. Roxb. !). Cyperus, Wall.! List n. 3329, I partim. Cyperus, Wall.! List n. 3352, A, B, C (eum C. malaccensi mixta), D. Cyperus, Wall.! List n. 3330, exemplum alterum. Tegetum materies, a Doctore G. King sub nomine C. tegetum a Calcutta communicata. — Rhizoma lignosum, breviter repens. Culmi solitarii aut approximati, 5-8 dm. longi, trigoni. Folia in exemplo Roxburghii 3, summum usque ad 14 em. longum; vagine sæpius nude, a folio 2-8 em. longo rarius terminate. Involucri bracteæ sepe 3 dm. longer, erecto-patentes proventu patule, in marginibus serrulato-scabre, umbellam fere semper conspicue superantes. Umbella 1-3 dm. in diam.; radii 4-12, usque ad 8-16 em. longi; ochree 15 mm. longs, laxiuscule. Umbellu- larum radioli 3-8, usque ad 4-9 em. longi, ssepe iterum umbel- lulati. Spice longe 5-10 mm., 4-10-spiculose ; spicule longe 2 cm., late 2-3 mm. (in exemplo Roxburghiano majores), 10-20- flore, in dorso virides, in lateribus pallide aut (mature) sepius brunnes aut rubre, in Africanis interdum castanee. Glume longs 2-3 mm., obtusæ, fere truncate, vix carinate, 3-7-nervie. "Stamina 8, subinclusa; anthere lineari-oblonge, mutice vel obsolete cristatz, fusco-sanguinee. Stylus nuce multo brevior; rami 3, e gluma breviter exserti. Nux breviter angustata, nigro- brunnea; cellule extime quadrate, emarcide, hyaline, porosz. [This plant, abundant in India, is the authentic C. fegetum, Roxburgh; it differs decisively from C. corymbosus in the much more distant glumes, which in the dried specimens have the margins incurved not overlapping. The spikelets are more com- pressed than those of C. corymbosus. The colour in India varies from pale to a high red-brown: with the more highly coloured Indian examples many African are absolutely identical; but there are other African specimens chestnut or almost black. It is far more difficult to distinguish C. tegetum in Africa from C. longus and its various forms called C. badius; the only absolute distinction appears to lie in the much longer leaves of C. longus. The rhizome of C. longus differs a good deal from that of C. tege- tum, as is evident enough when you have the whole of the rhi- zomes to compare, which may be once in a hundred specimens. LINN. JOURN.— BOTANY, VOL. XXI. M 162 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. The narrow wing of the rhachilla is in C. Schimperianus, as in C. longus, less soluble than in the Indian C. tegetum, but I doubt the value of this character. | India Orientalis, Abyssinia et Egypt. India, ab Himalaya usque ad Zeylaniam, vulgaris: Almora, alt. 1200 metr. (Strachey et Winterbottom n. 9, hh. Kew, Mus. Brit.); Chumba, alt. 900 metr. (C. B. Clarke n. 23688, h. Kew); Kumaon (Wallich n. 3352 D, h. Calcutta); Nepaul (Wallich, h. Kew n. 1018, h. Mus. Brit.); Sikkim, alt. 200 metr. (C. B. Clarke n. 10127, h. Calcutta) ; Khasia colles (Hooker f. et T. Thomson, h. Kew); Soane flumen (J. D. Hooker n. 421, h. Kew); Bengal (Wallich nn. 3351, 3352 A, h. Calcutta); Chota Nagpore (Wood n. 161, A. Calcutta; C. B. Clarke n. 21184, h. Calcutta); alt. 600 metr. (Kurz, h. Calcutta); Sarunda (Gamble n. 9108, h. Calcutta); Car- natie v. Mysore (G. Thomson, h. Calcutta); Zeylania (Thwaites C. P. n. 818, hh. Kew, Mus. Brit., Calcutta). Burma: Karen colles (Kurz n. 673, h. Calcutta). [Mauritius: sub nomine “ Cyp. textili " introduite de Pondi- cherry, k. Kew.) Abyssinia (Schimper nn. 57, 854, 1994, h. Kew); Nilus supe- rior ( Grant, h. Kew). Egypt: Elephanta (Kralik, h. Kew). Var. B. ambigua ; spiculis longis 45 mm. ; rhachille alis angustis non solutis; glumis longis 4 mm. et ultra. Cyperus, Wall.! List n. 3330, exemplum alterum. Cyperus, Wall.! List n. 3329 F (h. Kew, non h. Linn. Soc.). Madras Peninsula? (Heyne in h. Wall. propr.; Wallich n. 3330, h. Calcutta). Var. ? y. protracta; protracta; culmo tenui; folio summo cum culmo xquilongo; umbella subcapitata, paucispiculosa; spiculis longis 45 mm., 50-floris. Culmus 3 dm. Folium angustissimum. Umbelle radii 1-3, 0-3 cm. longi, tenues. Spice 4-spiculose lax. Socotra (Balfour, h. propr.). Series ** Foliate, rhizomate (nisi in C. amauropus) perenni, spiculis spicatis; omnes gerontogei quedam etiam in Ame- rica reperte. 127. C. Gracttinux; umbella laxiuscula, simplice aut com- posita; spiculis laxiuscule spicatis, divaricatis, acutis, pallidis ; MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 163 glumis subdistantibus; rhachille alis conspicuis, solubilibus ; nuce angusta, cum 3 parte glume equilonga. C. tenuiculmis, Boeck. in Flora, 1879, p. 554, non Boeck. in Linnea, xxxvi. p. 286. Stolones plures, graciliores, a vaginis numerosis laxis lanceo- latis brunneis intectx. Culmi solitarii, 5 dm. longi, mediocres vix tenues, trigoni, apice triquetri. Folia 2-5, cum 2 parte culmi equilonga, angustiora, in marginibus minute scabra. Involucri bractez 3-5, usque ad 2-3 dm. longe. Umbelle radii 4-8, usque ad 14 cm. longi; ochres longs 2 cm., a folio setiformi sspe ter- minate. Umbellule radioli 3-5, usque ad 4 cm. longi; bracteolæ sepe 2-4 em. longer, setacez. Spice 1-2 em. longs, 5-12-spi- culose ; rhachis glabra; gluma ima sepe setiformis. Spicule longe 2 cm., late 3 mm., compresse, 8-14-flore. Glume elliptice, obtuse, compresse vix carinatz, in dorso 5-7-nervie pallide virides, in lateribus luteo-stramines. Stamina 3, vix exserta; anthere lineari-oblonge, obtuse, rubescentes. Stylus nuce multo brevior; rami 3, plane exserti. Nux lanceolata, paullo curvata, triquetra, contra rhachillam compressa, ibique con- cava. Africa centralis: Djur (Schweinfurth nn. 2016, 2029, h. Kew). 128. C. tonaus (Linn. Sp. Pl. p. 67) ; culmo elongato, foliato, basi deeumbente, repente, crassiusculo nec bulboso; umbelle radiis elongatis; spicis sepius pedunculatis; spiculis breviter spicatis, paullum compressis; glumis imbricatis, florigeris non distantibus ; nuce cum 3 parte glume equilonga, ellipsoidea aut oblongo-obovoidea.—Jacg. Ic. Pl. Rar. ii. p. 8, t. 297; Vahl, Enum. ii. p. 846; Host, Gram. ii. p. 51, t. 76; Engl. Bot. t. 1309; Kunth, Enum. ii. p. 60; Reich. Ic. Fl. Germ. t. 282. fig. 672; Sturm, Fl. Deutsch. xiii. t. 52; Koch, Syn. Fl. Germ. p. 850; Bert. Fl. Ital. i.p. 269; Gren. et Godr. Fl. Franç. m. p.358; Parl. Fl. Ital. i. p. 12; Cosson, Espl. Algér. Glum. p. 248; Boeck. in Linnea, xxxvi. p. 279 (var. e, ¢,n excl.); Boiss. Fl. Orient. v. p. 375. C. depauperatus, Steud. Cyp. p. 34, e descr. C. Heldreichianus, Boiss, Diagn. i. fase. 13, p. 39. — Cyperus, Griff. Itin. Notes, p. 277, n. 697. — Cyperus, Griff. Itin. Notes, p. 280, n. 747. Culmus ima basi non bulboso-incrassatus, curvatus, horizon- M2 164 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. talis, interdum repens, crassiusculus, 8—7 dm. longus, apice acute triqueter, robustior. Folia 2-4, cum 2-3 parte culmi equilonga, robustiora, in marginibus serrulato-scabra. Involucri bractee 3-5, usque ad 2-6 dm. longs, umbellam longe superantes. Um- belle radii 3-10, erecto-patuli, usque ad 3-12 em. longi; ochreæ longs 1-3 cm., apice a seta unica sepe terminate. Umbellule iterum umbellate vel ad unicam spiculam reducte ; radioli ultimi (in forma typica) elongati, basi ochreolati, apice unicam spicam proferentes. Spica 5-15 mm. longa, 5-15-spiculosa; rhachis glabra. Spieule erecto-patule, rarius divaricate, ferrugineo- rubescentes, rarius atro-castanee, longe 15 mm., late 2 mm., 10-18-flore ; rhachille ale conspieuz, vix solubiles. Glume Spisse, compresse vix carinate, ovate, obtuse, in dorso virides 4—7-nervie, in lateribus coloratz vix striate. Stamina 3, sub- inclusa; anthere lineari-oblonge, mutice aut submucronate. Stylus nuce brevior; rami 3, breviter exserti. Nux subobtusa, apiculata, trigona, nigra; cellule extime quadrate, emarcile, hyaline, porose. Regio Mediterranea; ab Anglia usque ad Cabul. Anglia: Somersetshire (Thwaites, h. Kew; Scott, h. Calcutta ; Dyer, h. Calcutta) ; Wilts, Boyton (Henslow, h. Calcutta). France: Indre et Loire (Vercier n. 224, h. Kew); ins. Jersey (Stevens, h. Kew; Haslar, h. Calcutta). Africa: Algiers (Bové n. 316, h. Kew); Khartoum (Schwein- Surth n. 616, h. Kew). Ins. Rhodes (Bourgeau n. 150, h. Kew). Syria borealis (Kotschy n. 74, h. Kew ; Lowne, h. Calcutta). Byzantium (Postian n. 654, h. Kew). Armenia Ressica (Besser, h. Kew); Kurdistan (Kotschy nn. 292, 893, h. Kew); Tauria (Compin, h. Calcutta). Aberbeidschan (Fischer, h. Calcutta). Cabul (Griffith nn. 29, 30, h. Kew n. 6145; Griffith nn. 6145 6180, 4. Calcutta) ; Quetta (Hamilton, h. Calcutta). India: mons Aboo (G. King, h. Calcutta). Var. B. pallescens, Boiss. Fl. Orient. v. p. 375; spicis laxis; spiculis distantibus, paullo majoribus, pallidis, fere stramineis.— Desf. Fl. Atlant. i. p. 45, t.9; Willk. et Lange, Fl. Hisp. i. p. 139. C. Lamarckianus ?, Hochst. ! in Kotschy n. 222. C. ochreoides, Steud. Cyp. p. 34. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 165 C.longus, var. pallida (saltem pro parte), Boeck. in Linnea, xxxvi: p. 280. C. mitis, Steud. Cyp. p. 315, fide Boeck. Jn exemplis authenticis visis radix deest; hæc var. forsan sub C. rotundo melius militabit.—OC. mitis a Griffith in Bengalia lectus est, species mihi omnino dubia. Egypt: prope Canalem Alexandrie (Letourneux n. 152, k. Kew). Cordofan (Kotschy n. 222, h. Kew). Hispania: Beetica et prope Malaga, fide Willk. et Lange. Var. y. cyprica; spiculis acutis, lucide rubris. C. rotundus, Sintems et Rigo, MS. in h., non Linn. Rhizoma repens, crassiusculum. Culmi distantes, longi 1-5 dm., tenues. Umbella simplex; radii 2-4 cm. longi. Spicule in unoquoque radio 3-5, subfasciculate.—Planta mihi difficilis ; e rhizomate cum C. rotundo neutiquam jungenda. Ins. Cyprus (Kotschy n. 771, h. Kew; Sintems et Rigo n. 872, a, b, h. Kew). Var. à. badia, Desf. Fl. Atlant. i. p. 45, t. 7. fig. 2 (sp.) ; spicis ultimis sepe approximatis, subfasciculatis; spiculis patulis aut divaricatis, sepe fuscis aut atro-castaneis.—Reichb. Ic. Fl. Germ. t. 283. fig. 673; Vahl, Enum. ii. p. 346; Parl. Fl. Ital. ii. p. 38; Gren. et Godr. Fl. Franç. ii. p. 358; Koch, Syn. Fl. Germ. p. 849; Willk. et Lange, Fl. Hisp. i. p. 139; Boiss. Pl. Orient. v. p. 376. C. brachystachyus, Presl! Gram. et Cyp. Sicul. p. 15, non Steud. C. thermalis, Dumort. Fl. Belg. p. 145. C. denudatus, Heyne in Wall. List n. 3349. C. myriostachyus, Tenore, Fl. Nap. iii. p. 47; Bert. FI. Ital. i. p. 271; Parl. Fl. Ital. ii. p. 41. C. amoenus, Drege, litt. e (nec c, d). In C. myriostachyo, inflorescentia fere C. badii, spicularum color C. longi videtur. Ab Europa australi usque ad Caput Bone Spei. Portugal (Welwitsch n. 341, h. Kew). Hispania (Bourgeaw n. 509, h. Kew); Navarre (Willkomm n. 220, h. Kew). Toulon (Bourgeau n. 407, h. Kew). Palermo (Todaro n. 899, h. Calcutta). Attica (Heldreich n. 2548, h. Kew). 166 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Algiers (Munby, h. Kew); Tangier (Sulzmann, h. Calcutta) ; Marocco (J. D. Hooker, h. Kew; Rein. et Fritsch n. 394, h. Kew). Ins. Madeira (Findlay, h. Kew; Lowe n. 892, h. Kew). Ins. Teneriffe (Findlay n. 15, h. Kew). Ins. Azores (Godman, h. Kew; H. C. Watson n. 262, h. Kew). Caput Bone Spei (Drége litt. e, h. Kew; Bolus nn. 3926, 5813, h. Kew); Vaal flumen (Nelson n. 86, h. Kew; Burchell n. 385, h. Kew). [India Orientalis: Madras? (Wight in Wallich h. propr. n. 3349); an indigena ? | Var. e. elongata, Sieber; spiculis minus compressis ; glumis laxiuscule imbricatis, sæpe aureo-brunneis aut rufis. C. tenuiflorus, Jacq. Ic. Pl. Rar. ii. p. 8, t. 296; Presi, Gram. et Cyp. Sicul. p. 15; Ten. Fl. Nap. ii. p. 47; Guss. Fl. Sicul. Suppl. i. p. 780; Webb et Berth. Iles Canar. Phyt. iii. p. 364; Boeck. in Lin- nea, xxxvi. p. 281, non Rottb. C. Preslii, Parl. Fl. Ital. ii. p. 40. C. intermedius, Guss. Fl. Sicul. Suppl. i. p. 780, non Steud. C. amoenus, Kunth, Enum. ii. p. 58, non Roxb. C. emarginatus, Schrad. Anal. Fl. Cap. Cyp. p. 5; Kunth, Enum. ii. p. 60. C. lateriflorus, Steud. in Flora, 1829, p. 152. In C. ameno, Kunth, stolones lignosi, tenuiores ; culmi plures fasciculati; spicule pallidiores. Egypt inferior (Kotschy n. 945, h. Kew; Schweinfurth n. 2029, h. Kew); Tentyra (Sieber, hh. Kew, Calcutta). Sennaar (Kotschy n. 171, h. Kew). Africa australis (Drége, h. Kew; Burchell nn. 1937, 2651, h. Kew). [As regards many of the specimens witbout, rhizome, as Sieber's Tentyra plant with black-chestnut glumes, Schweinfurth's n. 2029 with short leaves, they perhaps would be better placed under C. tegetum. Var. e. adoensis, Boeck. in Linnea, xxxvi. p. 282=Schimper h. Abyss. n. 186, looks a distinct species. Var. Z. maculatus, Boeck. in Linnea, xxxvi. p. 282, as to Barter n. 1213, from Nupe, also looks distinct. Var. s. gracilis, Boeck. in Linnea, xxxvi. p. 282, is Wall. List n. 3329 F; this in Wallich h. propr. is C, rotundus, but the Cal- cutta specimen of that number is C. fegetum, var. ambigua. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 167 From Boeckeler’s description I guess his plant to have been C. tegetum, Roxb., fere typ.] 129. C. noruNDvus (Linn. Sp. Pl. p. 67, non Linn. h. propr.) ; culmo foliato, basi bulboso-incrassato; stolonibus tenuibus, ligno- sis, sepe tuberiferis; umbella simplice aut composita; spiculis breviter spicatis, suberectis, multifloris, compressis, nunquam lute- scentibus; glumis florigeris arctius imbricatis, ovatis, obtusis, per 2-3 latitudinem nervosis; stylorum ramis longius exsertis ; nuce cum 3-3 parte glume equilonga, late ellipsoidea, acuta aut sub- obtusa, triquetra.— A. Br. Prod. p. 216; Roxb. Fl. Ind. i. p. 197 ; Nees, Pl. Medic. Abbild. t. 25. fig. A; Presl, Rel. Henk. i. p. 175; DC. et Lam. Fl. Franç. ài p. 146; Kunth, Enum. ii. p.98; Webb et Berth. Iles Canar. Phyt. ii. p. 363; Parl. FI. Ital. i. p. 86; Hassk. Pl. Jav. Rar. p. 80; Benth. Fl. Hongk. p. 887, FI. Austral. vii. p. 279; Griseb. Fl. Brit. West Ind. p. 564; Boeck. in Linnea, xxxvl. p. 283; Baker, Fl. Maurit. p. 410; Cosson, Expl. Algér. Glum. p. 247; Boiss. Fl. Orient. v. p. 377. C. hexastachyus, Rottb. Descr. et Ic. p. 28, t. 14. fig. 2; R. Br. Prod. p. 216; Nees in Wight Contrib. p. 81 ; Decaisne in Nouv. Ann. Mus. d' Hist. Nat. iii. p. 358. . C. Hydra, Mich. Fl. Amér. Bor. i. p. 27; Vahl, Enum. ii. p. 344; G. W. F. Meyer, Esseq. p. 31; Presl, Rel. Henk. i. p. 175; Elliot, Bot. Carolina, p. 68,t. 2.fig. 25; Torrey in Ann. Lyceum New York, iii. p- 265. , | C. tetrastachyus, Desf. Fl. Atlant, i. p. 44, t. 8; Vahl, Enum. ii. p. 343. C. bicolor, Vahl, Enum. ii. p. 340. C. olivaris, Targ.- Tozz. in Mem. Soc. Ital. Scienz, xiii. pars 2, p. 338; Bert. Fl. Ital. i. p. 267; Reichb. Ic. Fl. Germ. t. 281. fig. 671; Gren. et Godr. Fl. Franç. iii. p. 359; Willk. et Lange, Fl. Hisp. i. p. 138. C. officinalis, Nees, Pl. Medic. Abbild. t. 25. fig. D. C. radicosus, Sibth. Fl. Grec. i. p. 32, t. 45; Griseb. Spic. Fl. Rumel. p. 421. C. comosus, Sibth. Fl. Grec. i. p. 31, t. 44; Benth. Pl. Hartw. p. 157; non Kunth. C. leptostachyus, Griff. Itin. Notes, p. 321, non Nees. C. esculentus, Gouan, Fl. Monsp. p. 389, non Linn. C. maritimus, Bojer, Hort. Maurit. p. 3/8, non Poir. C. stoloniferus, var. pallida, Boeck. ! in Linnea, xxxv. p. 490, non Retz. C. proteinolepis, Boeck.! in Flora, 1879, p. 549, non Steud. C.longus, Boeck.! in Aitch. Kuram Pl. n. 684, in Journ. Linn. Soc. xviii. p. 104, non Linn. 168 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. — Cyperus, Linn. Fl. Zeyl. p. 15, n. 15. — Cyperus, Wall.! List n. 3317, A partim. — Cyperus, Wall.! List n. 3322, A, B, C, D, E. — Cyperus, Wall.! List n. 3329, B partim, C, D, E, F, G, H, I. — Oyperus, Wall.! List n. 3332, C. — Cyperus, Wall.! List n. 3353, A, B. — Cyperus, Griff. Itin. Notes, p. 276, n. 689 (Kew n. 6179). — Cyperus, Griff. Itin. Notes, p. 10, n. 142 (Kew n. 6193). — Cyperacea, Griff. Itin. Notes, p. 277, n. /02 (Kew n. 6144). — Cyperacea, Griff. Itin. Notes, p. 319, n. 1225 (Kew n. 6180). — Cyperacea, Griff. Itin. Notes, p. 321, n. 1253 (Kew n. 6180). — Agrostotheca, Sieber n. 109 (sed non C. littoralis, R. Br.). — Rumph. Herb. Amb. vi.t. 1. fig. 1, 2. Rhizoma repens, lentum, tubera 1-3 em. longa, ovoidea aut oblonga, sæpe proferens. Culmi solitarii (aut approximati vix fasciculati) 1-6 dm. longi, trigoni, apice triquetri, basi incrassati aut bulbosi. Folia 3-5 aut plura, validiora, cum 2 parte culmi sepius zequilonga, in marginibus scabrida. Involucri bractex 3-5, patule, umbellam ssepius multo superantes, interdum abbreviate. Umbella simplex aut composita, radiis ssepius pluribus, elongatis, modo autem congesta aut depauperata, modo effusior magna; ochree 1 cm. longs, subtruncate. Spice 5-20 mm. longs, ebracteatz, 3-20-spiculoss; rhachis glabra. Spicule florifere stepius erecto-patulz, longs 3 cm., late 3 mm. (vel multo minores majoresve), 32-florz (vel 8-70 florz), rubro-virides, pallide virides, intensius sanguineæ aut castanes, nunquam (ut in C. esculento aliisque vicinis) aureo-virides aut lutew; rhachille ale hya- line, rarius sanguineo-maculate, interdum solubiles. ^ Glum:e fructigerz imbricate, laxiores, vix turgid, ovate, obtuse, com- presse vix carinate, in lateribus (quam in C. esculento) latiuscule enervie, estriate. Stamina 3; anthere exsertz, lineari-oblonge, rubescentes aut pallide, mutice. Stylus cum nuce fere æqui- longus vel multo brevior ; rami 3, longi, sepe sanguineo-maculati. Nux quoad magnitudinem variabilis, utrinque angustata, latius- cula, subequaliter acute trigona, nigro-fusca; cellule extime quadrat, emarcide, tenuiter hyaline, porose, aut albide, opace ; ideoque nux obscurius (raro lucide) reticulatim albo-velata.— Species inter omnes Cyperos vulgatissima, in tropicis utriusque orbis pestis. Varietates infra enumerate melius pro formis meris habende. Species plures C. rotundo similes, ab auctoribus optimis difficillime discrete, caute cognoscende sunt, nempe :— C. stoloniferus, nuce inequaliter trigona, spiculis teretioribus. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 169 C. longus, culmo basi non bulboso, rhizomate quasi culmi pro- longatione, nuce minore. C. esculentis, glumis per 2-1 partem latitudinis striatis, viridi- luteis, aureis aut aureo-brunneis. C. tenuiflorus, Rottb., umbella late composita, spiculis divari- catis, nuce trigona neque acute triquetra. C. jeminicus, culmi basi bulbosa nigro-squamata, umbelle radio imo subdistante. C. gracilinux, nuce anguste lineari-oblonga. C. pertenuis, culmo tenui, foliis cum 4 parte culmi equilongis. C. tegetum, glumis florigeris (in sieco) vix imbricatis. C. corymbosus, foliis ssepius brevibus, culmo basi non bulboso. Per regiones calidiores totius orbis terrarum vulgaris; quam in oryzetis aliisque cultus tam in graminosis pestis. EvRoPA.—Portugal (Broteiro n. 401 [C. bulbillosus], 4. Kew; Welwitsch n. 339 [C. badius], A. Kew; Lisbon (Daveau [C. esculentus], h. Calcutta). Hispania (Bourgeau n. 1539, h. Kew; Willkomm n. 875 [C. lon- gus], h. Kew) ; Barcelona (Bentham, h. Calcutta) ; Gibraltar (Salzmann [C. humilis], 4. Kew). Corsica (Mabille n. 284 [C. olivaris], h. Kew). Candia (Sieber (C. comosus], hh. Kew, Calcutta). Athens (Orphanides n. 259 [C. radicosus], k. Kew; Heldreich n. 2547 [C. comosus], 4. Kew). Arrica.—Algiers (Munby [C. tetrastachys], k. Kew). Egypt (Sieber n. 55 [C. elongatus], h. Kew; Schweinfurth n. 2023, h. Kew); Lacus Mareoticus (Aucher-Eloy, h. Cal- cutta). Nubia (Kotschy n. 358, h. Kew; Kotschy n. 28 [C. elongatus], h. Kew). Libyian Desert (Ascherson n. 542 (C. esculentus], k. Kew). Ethiopia (Kotschy n. 170, h. Kew). Niger flumen (Barter n. 853, h. Kew; Vogel n. 62 [C. pal- lescens], h. Mus. Brit.). Madeira: Funchal (A. Calcutta). Ins. S. Helena (Burchell n. 11 [C. tuberosus], 2. Kew). Gallabat (Schweinfurth n. 2026 [C. longus, var. elongata, Boeck.], h. Kew). Djur (Schweinfurth n. 1822, ^. Kew). Berber (Schweinfurth nn. 188, 405, 470, 500, 525, 530, 754, h. Kew). 170 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Nilus Albus (Schweinfurth n. 945, h. Kew). Cordofan (Pfund n. 628 [C. longus], 4. Kew). Abyssinia (Schimper nn. 370, 546, 1291, 1379, 1556, h. Kew, =var. spadicea, Boeck. in Linnea, xxxvi. p. 281). Comoro (Bojer (C. platystaehys], k. Kew; Hildebrandt n. 1736, h. Kew). Zambesi flumen (Kirk, h. Kew). Caput Bone Spei: Albert District (Cooper n. 1865, h. Kew). Natal (Buchanan n. 312, h. Kew). Ins. Mascarene: Madagascar (Hildebrandt nn. 3320, 3355, h. Kew); Bourbon (Balfour, h. Kew); Mauritius (Szeber n. 6, k. Kew) ; Seychelles (Bouton n. 23, h. Kew). Asta.—Syria (Lowne, h. Calcutta). Iberia et Tauria (Wilhelms, h. Calcutta). Bagdad (Reichenbach f. | C. Tenorii], k. Kew). Cabul (Griffith n. 6179, h. Calcutta). India: Gilgit (Winterbottom n. 928, h. Kew) ; Kashmir (Jac- quemont n. 663, h. Kew) ; Simla (Jacquemont n. 2372, h. Kew); Sutledge flumen (Jacquemont n. 1102, h. Kew); India boreali-occidentalis (Royle [C.tenuiflorus], h. Kew); Gur- whal (G. King, h. Calcutta) ; Lahore (Brandis n. 2615, h. Calcutta); Kuram convallis (Aitchison n. 684 [C. lon- gus], ^. Calcutta); mons Aboo et India centralis (G. King, h. Calcutta) ; Khasia colles (Griffith n. 142, h. Calcutta) ; Bengalia (Griffith n. 6193, k. Calcutta); Poona (Jacque- mont nn. 340, 382, h. Kew); Madras Peninsula (Wight nn. 1826, 1827 [C. hexastachyus], k. Kew; Wallich n. 3310, h. Calcutta). Peninsula Malayana: Pegu (Kurz n. 682, h. Calcutta) ; Mergui (Griffith, h. Kew, n. 6194); Moulmein (Parish n. 264, h. Calcutta); ins. Andaman (Kurz, h. Calcutta); Tenas- serim (Helfer n. 6194, h. Calcutta) ; ins. Penang (Stoliczka, h. Calcutta). Malaya: Singapore (Kurz n. 3007, k. Calcutta; Kunstler n. 112, k. Calcutta) ; Java (T. Anderson n. 3738, h. Calcutta ; Zollinger n. 282, h. Kew); Buitenzorg (Kurz n. 197, h. Cal- cutta) ; Banca (Teysmann [C. bulbuso-incrassatus], 4. Kew) ; Borneo (Barber n. 323, h. Calcutta) ; Labuan (Motley n. 323, h. Kew). China (Sir G. Staunton [C. odoratus, Osbeck non Linn.], h. Kew): Hongkong (C. Wright n. 560, h. Kew); Canton e MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 171 (Hance n. 1247, h. Kew); Shanghai (Maingay nn. 640, 763, h. Calcutta); ins. Formosa (Oldham n. 583, h. Kew; Swin- hoe, h. Calcutta). Japonia (k. Calcutta). Ins. Loochoo (Beechey, h. Kew). Ins. Samoa (Biker n. 1227, h. Kew). Ins. Philippine (Cuming nn. 715, 557, h. Kew). AUSTRALIA.— Australia (R. Brown n. 5891, h. Kew). Australia tropicalis (Bidwill n. 197 partim, h. Kew). Queensland (F. Mueller, h. Kew). America.—Florida (Curtiss n. 8055, h. Kew). Texas (Elihu Hall n. 684, h. Kew). Ins. Martinique (Sieber n. 14, h. Kew; Hahn n. 888, h. Kew). Ins. New Providence (Brace n. 397, h. Kew). Guiana: Cayenne (Sagot n. 1361, h. Kew). Bahia (Salzmann [C. sphacelatus], k. Kew; Glocker n. 216 [ C. helodes], 5. Kew). Rio Janeiro (Glaziow nn. 523, 11654, h. Kew). Brazil (Martius n. 860 [C. hexastachyus], k. Kew). Respublica Argentina: Cordoba (Hieronymus n. 740, h. Kew). Var. a. Salsola ; culmis 1-2 dm. longis; foliis cum $-1 parte culmi equilongis; umbella simplice pauciradiata ; spicis sspe 3-4-spiculosis (imo culmis 1—4-spiculigeris) ; involucri bracteis umbella multo brevioribus, interdum 1-3 em. longis. India Orientalis, presertim in maritimis ; vulgaris. Australia: forma frequens. Neo-Zeelandia (Kirk, h. Kew). Var. (3. 100-ffora ; spiculis elongatis, szepe curvatis vel tortis. — Oyperus, Wall.! List n. 3373. Spicule, in G. Thomson n. 258, usque ad 9 cm. longe, 118- florz. Africa tropicalis: frequens. India Orientalis: Madras Peninsula (G. Thomson n. 258, À. Kew). Australia tropicalis (Bidwill n. 187 partim, À. Kew). Var. y. Amalia; rhizomate squamato; foliis smpe brevibus ; umbella simplice contracta, sublaterali ; involucri bracteis brevi- usculis, umbellam superantibus; spiculis sublucide brunneo- rubris. Australia : Queensland, Rockhampton (Amalia Dietrich n. 712, 172 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. h. Kew); Victoria, Murray et Darling flumina (F, Mueller, h. Kew) ; Portus Molle (Macgillivray, h. Kew). Var. à. procerula ; culmo elongato robustiore; umbella expla- nata, modo subsimplice modo composita; spiculis laxe spicatis, majusculis, acutis, sub-lineari-lanceolatis, rubescentibus aut pal- lidis nec lucidis ; glumis laxiuscule (fructigeris vix) imbricatis ; nuce maxima, lata, acute triquetra.—Nees! in Wight Contrib. p. 82 (sp.); Kunth, Enum. ii. p. 51. C. Retzii, Nees! in Wight Contrib. p. 82; Kunth, Enum. ii. p. 733 Thwaites, Enum. Pl. Zeyl. p. 343; Roxb. Ic. ined. t. 720, in h. Kew. C. tuberosus, Roxb. Fl. Ind. i. p. 199; Boeck. in Linnea, xxxvi. p. 285, syn. Kunth et Rottb. excl. C. Pangorei, Retz, Obs. iv. p. 10, nec Rottb. nec Roxb. C. viridis, Roxb. Ic. ined. t. 714, in h. Kew. C. rotundus, var. acuta, Boeck. in Linnea, xxxvi. p. 284. — Cyperus, Wall.! List n. 3356 [=C. albidus, Heyne]. Cyperi proceruli, Nees, exemplum typicum (Wight n. 1829, h. propr.), a manu Neesii nomine inscriptum, nucibus maturis gaudens, mihi videtur omnino idem ac Wight n. 1830 h. propr., exemplum junius florigerum, a manu Neesii nomine C. Retziz inseriptum. Inter hee et C. rotundum typicum exempla plurima sensim transeuntia videnda sunt. In exemplo typico C. procerulo, nuces mature magne, tri- quetre, ideoque auctores C. procerulum cum C. compresso infauste contulerant. In C. rotundo typico ut in C. Hydra (Anglice * Nut-grass ") rhizomata pestifere reptante, nux rarius maturat. Madras Peninsula (Wight nn. 1829, 1830, h. Kew). Zeylania (Thwaites C. P. n. 3750, h. Kew). Var. e. lavata ; spiculis longis 18 mm. latis 3 mm., 10-14-floris, compressis, rubris; glumis majusculis, elongatis, sursum angus- tatis, subdistantibus, florigeris patulis laxe vel vix imbricatis ; rhachillz alis angustissimis, junioribus saltem non solubilibus. Exemplum (nuce ignota) mihi sat dubium; an C. glabro affinior ?—color autem spicularum distat. Naples (.Heldreich [C. olivaris], k. Kew). 130. C. STOLONIFERUS (Retz, Obs. iv. p. 10) ; rhizomate longe repente, lignoso; culmis basi bulbosis, apice trigonis ; umbella simplice; spiculis breviter spicatis, subteretibus; glumis spissis, MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 173 ovatis, obtusis, pallidis, purpureo-notatis ; stylo alte 3-fido, bre- viuseulo; nuce obovoidea, trigona, a dorso compressa, nigro- castanea.— Nees in Wight Contrib. p. 81; Boeck. in Linnea, xxxv. p. 489 (syn. C. tuberoso et var. B exclusis). C. littoralis, R. Br.! Prod. p. 216. C. spadiceus, Lam. Ill. i. p. 147, fide Nees; non h. Koenig. C. bulboso-stoloniferus, Steud.! in Zoll. Verz. Ind. Archip. p. 62. C. tuberosus, Baker, Fl. Maurit. p. 410, nec Rottb. C. rotundus, Koenig in h. propr.; Benth. Fl. Austral. vii. p. 279 partim, non Linn. C. arenarius, Hance in h. propr., nec Retz. C. lamprocarpus, Boeck. in Linnea, xxxv. p. 490. — Cyperus, Wall.! List n. 3315. — Cyperus, Wall.! List n. 3309 partim (Fimbristylidis specie admixta). Stolones elongati, validi. Culmi distantes, 1-2 dm. longi. Folia cum culmo equilonga aut breviora, sursum minute dentata, in sicco plana vel complicata. Involucri bractes sepius 3, quarum una, junior suberecta, usque ad 5-8 cm. longa, aut brevior. Umbelle radii 1-4, longi 1—4 em. Spicule longs 15 mm., late 2-3 mm., parum compresss, sepius suberectz in- terdum curvate, in unoquoque radio 3-8, ebracteate, parum distantes, 12-22-flore; rhachille ale anguste, hyaline, per- sistentes; gluma ima parva, obtusa. Glume dense imbricata, concave nec carinate, obscurius 5—9-nervie, leves. Stamina 3; filamenta levia; anthere lineares, mutice, sanguineo-notate. Stylus brevissimus ; rami 3, plane exserti, erassiusculi, sanguineo- notati. Nux eum 3-3 parte glume equilonga, obtusa, in facie interiore plana subconcava, in dorso convexa tenuiter angulata, nigra; cellule extime quadrate, emarcide, lucide albidæ, reti- culate, plus minus detergibiles.—Nux matura non minus quam in C. lamprocarpo, nitida.— Cyp. stoloniferus (3. pallida, Boeck., in exemplo nimis juveni stabilita (— C. complanatus, Presl; Steud. Cyp. p. 3) mihi videtur (e spieulis compressis) C. rotundus.— Boeck. (in Linnea, xxxv. p. 490) scripsit “ stylo elongato (quan- doque abbreviato) profunde bi- vel tri-fido, aut indiviso." In exemplis sat numerosis a me visis, stylus videtur 3-fidus; stylum indivisum aut bifidum nunquam vidi. Asia australi-orientalis, cum Malaya, Mauritio et Australia septentrionali in maritimis, frequens. Ins. Mascarene (Macgregor, h. Mus. Brit.); Mauritius (Sieber nn. 11, 28, 4. Kew, n. 11 h. Mus. Brit.; Ayres, h. Kew). 174 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. India Orientalis : Goojerat (4. Mus. Brit.) ; Zeylania (Koenig, h. Mus. Brit.); Bombay (Jacquemont n. 430, h. Kew); Madras Peninsula (Wight n. 1825, h. Kew, n. 2861 A. Cal- cutta); Tranquebar (Klein, h. Kew); Nilghiri montes (Foulkes, h. Kew) ; Carnatic (G. Thomson, h. Calcutta). Ins. Penang (G. King, h. Calcutta). Singapore vel Borneo (Burbidge, h. Kew). Singapore (Kurz n. 2991, h. Calcutta). Ins. Nicobar (h. Mus. Brit.). Sumatra (Heyne, h. Mus. Brit., Cuming n. 2437, hh. Kew, Mus. Brit.). Java (Zollinger n. 1321, h. Mus. Brit.; Horsficld, h. Mus. Brit.). China: Amoy et Macao (Hance n. 1886, h. Kew). Australia: in ora septentrionali (R. Brown n. 5892, h. Kew). 131. C. TENUIFLORUS (Rottb. Desc. et Ic. p. 30, t. 14. fig. 1, non Jacq.); stolonifer; culmis elongatis, foliaceis, basi incras- satis; umbella composita, tenui, laxa; spiculis angustis, florigeris sspe rectangulatim patulis; glumis subremotis, imbricatis.— Vahl, Enum. ii. p. 847; Roxb. Fl. Ind. i. p. 199, Ic. ined. t. 1109 in h. Kew; Dalz. and Gibs. Bomb. Fl. p. 284; Nees in Wight Contrib. p. 82 partim; Kunth, Enum. i. p. 57 partim; (non C. longus var. tenuiflora, Boeck. in Linnea, xxxvi. p. 281). C. torsus, Vahl, Enum. ii. p. 339. C. Roestelii, Kunth, Enum. ii. p. 58. C. pertenuis, Bojer, Hort. Maurit. p. 379, non Roxb. C. longus, Baker, FI. Maurit. p. 411, non Linn. — Cyperus, Wall. ! List n. 3329, A, I partim. Culmi basi lignosim bulbosi, 6-9 dm. longi, tenuiores, plane trigoni. Folia 2-4, usque ad 5 dm. longa, angusta. Involucri bractee 3-5, usque ad 3-4 dm. longe, divaricate, anguste. Umbella tenuiter composita; radii 3-7, usque ad 5 cm. longi, admodum tenues ; ochrez 1 em. longs, angustissime ; umbellu- larum radioli elongati, tenues. Spice 10-15 mm. longs, 6-10- spiculose, laxe. Spicule longe 12 mm., late 2-3 mm., ferru- ginee aut pallide, fere ut C. rotundi sed tenuiores.—Species fortasse in C. rotundum var. procerulam transeuns, a C. longo remota. Mauritius (Horne, h. Kew; Bouton, h. Kew; §c.). Socotra (Schweinfurth n. 691, h. Balfour). | | MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 175 Zeylania (Thwaites C. P. n. 8966, k. Kew). Bengal (Griffith, h. Kew, n. 6141) ; Caleutta (Kurz, h. Cal- cutta; C. B. Clarke n. 8515, h. Kew). Madras (Dr. Shuter, h. Kew). Forma australica; umbella composita aut decomposita, ramulis patule corymbosis ; spiculis pallidis aut lete brunneis; nuce parva, anguste obovoidea, obtusissime trigona. C. rotundus partim, Benth. Fl. Austral. vii. p. 279. Culmus basi bulboso-incrassatus ; rhizoma lignosum, tenue. Nux cinereo-fusca, matura nitide albo-reticulata, quam C. rotundi multo minor, minus acute triquetra. l [This plant appears well separable from C. rotundus by the much smaller obtuser nut, and it seems to match well with the Indian C. tenuiflorus; but the nut of the latter T have never found (itis probably rare as in C. rotundus), and this casts a shadowy doubt over the whole business.] Australia: in regione interiore (McDougal, h. Kew); in ora australi-oceidentali (Cunningham, h. Kew, n. 345 h. Mus. Brit); Arnheim's Land (F. Mueller, h. Kew); in ora boreali-orientali (Bidwill, h. Kew); interiore boreali-occi- dentali (Mitchell, h. Kew). 132. C. JEMINICUS (Rottb. Descr. et Ic. p. 25, t. 8. fig. 1); culmo basi bulboso, a squamis crassis nigris ovatis acutis detergibilibus intecto ; umbellæ contractæ, radio imo subdistante l-bracteato ; glumis ovatis, subacutis, rubescentibus aut sanguineis, per totam fere latitudinem nervosis.— Willd. Sp. PL i. p. 281; Roxb. FI. Ind. 1. p. 191. C. bulbosus, Vahl, Enum. ii. p. 342; Nees in Wight, Contrib. p. 80; Dalz. and Gibs. Bomb. Fl. p. 284; A. Rich. Fl. Abyss. ii. p. 483; Boeck. in Linnea, xxxvi. p. 300; Bois. Fl. Orient. v. p. 377. C. laxus, R. Br. in Salt Abyss. Append. p. 62, nomen nudum. C. bulbiferus, Dietr. Sp. Pl. ii. p. 324. C. oleraceus, Roxb. Ic. ined. t. 733, in h. Kew. C. geminatus, hh. Koenig et Rottler, non Schrader. C. rotundus, Kunth, Enum. ii. p. 58 partim, auctorumque aliorum, non Linn. C. rotundus 8 pendulus, Nees in Wight, Contrib. p. 12 pro minore parte (habitat plurimis exclusis). Hemichlzna bulbosa, A. Rich. FI. Abyss. ii. p. 509. — Cyperus, Wali. ! List n. 3317 A partim, B, C. Culmi S-24 em. tenuiores, basi bulbosi; stolones tenues, 176 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. bulbillos tuberiformes remote proferentes. Squame nigre longe 1 em. basin culmi normaliter integunt ; his detersis, exempla ad C. rotundum sspe relata sunt. Folia plura, sæpe plurima, cul- mum longe superantes vel hoc multo breviores, latiusculi vel interdum angusti. Umbella sæpius contracta pauciradiata, rarius 1 dm. in diam., laxior. Umbella fere semper imperfecta ; rami inferiores inter se distantes, unibracteati; bractez 4-8 em. longs, apice setacew; etiamque e medio umbelle 1-2 bracteæ setaceæ sspe exstant. Umbelle radii brevissimi, rarius 2-4 cm. longi, simplices, rarius divisi. Spice 3-12-spiculose, dense vel laxiores ; rhachis glabra. Spicule longe 15 mm., late 3-4 mm., compress, 8-14-flore ; rhachille ale solubiles. Glumx com- presse, subdistantes, imbricate, in dorso anguste virides, 10 lateribus lucide rubre aut castaneo-ferrugines, concoloriter multinervie. Stamina 3, breviter exserta; antherw lineares, muticæ. Stylus brevis; rami 3, e gluma breviter exserti. Nux oblongo-obovoidea, cum 3 parte glume subequilonga.—Si squama nigre culmi basin integentes desint, species a C. rotundo glumis per totam fere latitudinem striatis, umbella imperfecta, coguo- scenda est; a C. esculento glumis rubescentibus nec lutescentibus distingui potest. A Yemen usque ad peninsulam Madras et Senegambia. Jedda (Fischer n. 54, h. Kew ; Schimper n. 809, h. Kew). Beloochisthan (Frere, h. Kew). Scinde (Stocks, h. Kew; Dalzell, h. Calcutta). Zeylania (Trimen, h. Kew). Madras Peninsula (Heyne in Wallich n. 8317, h. Calcutta; Wight nn. 1824, 1828, h. Kew, n. 2868 h. Calcutta; G. Thomson n. 102, h. Kew, n. 58 partim h. Calcutta). Abyssinia (Schimper n. 2238, h. Kew). Africa centralis: Uganda (Wilson n. 751, h. Kew); Berber (Schweinfurth n. 406, h. Kew). Flumen Senegal (J. Gay, h. Kew). Caput Verde (Bolle, h. Kew). 133. C. usitatus (Burchell; Roem. et Sch. Syst. ii, Mant. p. 477) ; bulbo squamato a basi culmi vix incrassati, ope rhizo- matis tenuis lignosi paullo distante; umbella contracta, imper- fecte umbellata; glumis ovatis, rubescentibus aut sanguineis, per totam fere latitudinem striatis.— Nees in Linnea, x. p. 196; Kunth, Enum. ii. p. 107; Boeck. in Linnea, xxxv. p. 511. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 177 C. semitrifidus, Schrad. Anal. Fl. Cap. Cyp. p. 6; Kunth, Enum. ii. p. 107; Boeck. in Linnea, xxxv. p. 513. C. herbivagus, Kunth, Enum. ii. p. 53. C. solidus, Kunth, Enum. ii. p. 76. A C. jeminico vix nisi rhizomate distinguenda. In C. usitato bulbus a basi culmi 1—4 em. distat. Quoad rhachillam spicularum discrimen mihi videtur nullum.—C. vestitus, Hochst. in Flora, 1845, p. 755 ; Steud. Cyp. p. 41, e descriptione ad C. jeminicum affinior est. Africa australis (Burchell nn. 1999, 2082, 2684, à. Kew; Zeyher n. 1743, h. Kew). Africa australi-orientalis (Welson n. 97, h. Kew). 194. C. Tnowsoxr (Boeck. in Linnea, xxxvi. p. 204) ; culmo robustiore, foliato; spiculis numerosis, spicatis, undique patenti- bus, majusculis, acutis, stramineo-brunneis ; glumis laxe imbri- catis, ellipticis, subaeutis; nuce cum 4-4 parte glume squilonga, angustius oblonga. Cyperus sp. n. 55, Herb. Ind. Or., Hook. f. et T. Thoms. Rhizoma breve, lignosum, nigrosquamatum, culmo 1, vel paucis approximatis. Culmus 8-6 dm. longus, basi incrassatus, trigonus, apice triqueter. Folia 3-5, longius vaginata, valida, culmum sepe superantia, marginibus sursum scabris. Involucri bracteæ 3-4, usque ad 4 dm. longe, patule. Umbella sepius densa, interdum radiis longioribus 2 dm. in diam., in uno exemplo ad unieam spicam reducta; radii robusti, interdum ad 8 cm. longi, sæpius breves; ochree 10-15 mm. longe, subinflate, apice lanceolate; umbellule corymbose ; radioli sepius brevissimi 1-spiculosi, interdum 1-2 em. longi, plurispieulosi. Spice longe 2-3 cm., 20-40-spiculose ; rhachis glabra. Spicule longs 22 mm., late 4-5 mm., modice compresse, 16-24-flore; rhachille ale admodum anguste, persistentes ; gluma ima bracteiformis, longius setuligera. Glume vix carinate, in dorso arcte 9-11-nervie, in lateribus latis scariose, brunneo-maculate. Stamina 3 ; filamenta angusta; anthere lineari-oblongs, rubescentes, mutice (apice sanguine obsolete cristatæ). Stylus nuce multo brevior ; rami 3, e gluma breviter exserti. Nux trigona, apice subacuta, nigrescens; cellule extime quadratze, emarcide, nitide hyaline (7. e. nux albo- reticulata). Bengalia: in paludibus Jheel dictis (J. D. Hooker n. 263, h. Kew); Cachar (Keenan, h. Kew); Sylhet (C. B. Clarke LINN. JOURN.—BOTANY, VOL. XXI. N 178 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. n. 7235, hh. Kew, Calcutta) ; Siligori (C. B. Clarke n. 11688, h. Kew). Assam (Jenkins, Masters, h. Calcutta) ; Seebsagur (Kurz, h. Calcutta). 135. C. ESCULENTUS (Linn. Sp. Pl. p. 67) ; culmo basi incras- sato, foliato; stolonibus tenuibus, sspe tuberiferis; umbella simplice aut composita ; spiculis spicatis, patulis, viridi-lutescen- tibus aut brunnescentibus ; glumis ovatis, obtusis, per latitudinem fere totam striatis.— Vahl, Enum. i. p. 945; Host, Gram. ii. p. 90, t. 75; Kunth, Enum. i. p. 61; Bert. Fl. Ital. i. p. 256 ; Boeck. in Linnea, xxxvi. p. 287; Coss. Expl. Algér. Glum. ni. p. 246; Boiss. Fl. Orient. v. p. 377. C. Sieberianus, Link, Hort. Berol. i. p. 313. C. tuberosus, Rottb. Descr. et Ic. p. 28, t. .fig. 1; Vahl, Enum. i. p. 340; Pursh. Fl. Am. Sept. i. p. 52; Kunth partim, nec Boeck. C. damiettensis, Dietr. Sp. Pl. ii. p. 269. C. melanorrhizus, Delile; Parl. Fl. Ital. i. p. 33; Moggr. Contrib. Fl. Mentone, t. 97. C. aureus, Tenore, Fl. Nap. ii. p. 45; Reichb. Ic. Pl. Crit. ii. p. 10, t. 212; Gren. et Godr. Fl. Franc. iii. p. 360, non H. B. K. C. Tenorii, Presl, Fl. Sicul. p. xliii; Bert. Fl. Ital. i.p. 264; Reich. Ic. Fl. Germ. t. 281. fig. 670. C. Tenorianus, Roem. et Sch. Syst. ii., Mant. p. 544. C. Hydra, H. B. K. Nov. Gen, et Sp. i. p. 205, non Mich. C. rotundus, var. pallida, Benth. Fl. Austral. vii. p. 280. C. rotundus, Benth. Pl. Hartw. p. 28. C. gracilis, Humb. et Bonpl.; Link, Jahrb. ii. p. 84. C. gracilescens, Roem. et Sch. Syst. ii., Mant. p. 100 ; Nees in Mart. Brasil. Cyp. p. 36. C. nervosus, Bert. ; Roem. et Sch. Syst. ii., Mant. p. 113. C. retusus, Nees in Linnea, ix. p. 285; Kunth, Enum. ii. p. 115, non Hochst. C. phymatodes, Muehl. Gram. p. 23; Link, Jahrb. ii. p. 87 ; Torrey, Fl. United States, i. p. 62; Kunth, Enum. ii. p. 62; Sereno Watson, Bot. California, i. p. 215. C. repens, Elliott, Bot. South Carolina, i. p. 69 ; Torrey in Ann. Lyceum New York, ii. p. 264; Chapman, Fl. South. States, i. p. 508. C. Hermanni, Zuckley in Proc. Philad. Acad, 1862, p. 10. — Cyperus sp. n. 32, Herb. Ind. Or., Hook. f. et T. Thomson. — Cyperus, Wall.! List n. 3354, A. Stolones flexuosi, 1-9 mm. crassi, a squamis longis 5-8 mm. ovatis acutis luteis aut brunnels ornati; tubera 1 cm. in diam., MA l 3 is i^ Mitt [ m i - m M pa aki o on Mit ia hii t ie RA MEM Tea HÀ — - € ai a e eroe LAENA MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 179 ovoidea, nigrescentia aut pallida, transversim zonata aut tomen- tosa vel sublevia. Culmi 1-5 dm. trigoni, sursum triquetri. Folia 2-5 aut sæpe plara, culmo longiora aut breviora, rigidiora, latiuscula, in margine scabra. Involueri bracteæ 3-5, usque ad 8-18 cm. longe, umbellam sepius plane superantes, patule. Umbella simplex aut composita, sepius 5-12 cm. in diam, interdum (presertim in Americanis) evoluta, haud raro contracta, in C. Hermanni, Buckley, fere capitata; radii 8-10, longi 1-8 cm.; ochreæ 5-15 mm. longs; radii apice 1-spicati aut subcorymbosim paucispicati, a bractea angustissima sepe suffulti. Spies 5-30 mm. longs, laxe 4~14-spiculose; rhachis glabra. Spicule florifere rectangulatim patule, non congeste, aures, brunnesm aut pallide, nec rubr: nee sanguineo-castanew, longe 5-30 mm., late 2-4 mm., quoad formam magnitudinem florum nu merum sat variabiles, compresss aut subinflatse; rhachille ale solubiles. Glumæ remotiuscule aut arctius stipate, florigere spisse aut rarius causa marginum ineurvorum vix imbricate, ovate, obtuse, truncate aut obscurius apicatz, compresse vix carinate, in dorso virides 3~5-nervie, in lateribus lucidz concoloriter plicato-striate. Stamina 3, subinclusa aut longe exserta; anther lineari-oblonge, mutice, lutescentes, rarissime rubre. Stylus nuce brevior ; rami 3, e gluma vix exserti vel longissime exserti (quoad longitudinem in diversis formis admodum diverse). Nux cum 1-5 parte glume equilonga, latissime vel angustius oblonga, obtusa, obscurius apicata, subæqualiter trigona, nigra; cellule extime quadrata, emarcidz, albide, lucide porosz (ideoque nux “ albo-reticulata ' ). — Inter tot exempla, tot synonima, hzc mihi pro formis melius habenda :— l Forma princeps ; umbella 3-10 cm. in diam., simplex aut parce composita; spicule longs 15 mm., late 2 mm., 16-flore, aurez aut aureo-brunnes ; glume remotiuscule, florigerz plane imbri- cate ; styli e gluma breviter exserti; nux oblonga.—Forma in Europa, Asia, Africa, America vulgaris ; nec differt C. phymatodes, Muehl. l Forma melanorrhiza ; umbella quam in forma principe paullo magis evoluta; spiculæ longs 8-10 mm., late 3 mm,, subinflate, 14-florz, pallide aut brunnescentes; glume florigeræ laxe imbri- cate ; styli e gluma longissime exserti; nux late oblonga.—In regione Mediterranea cum ins. Mascarensibus. Forma tuberosa (sp. Rottb.) ; spicule lineari-oblonge, com- presse, subtruncate, 8-12-flore.—In Madras indigena ; a forma N é 180 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. melanorrhiza paullo recedens spiculis magis compressis, stylis parum exsertis. A Rottler cum C. hyalino, Vahl, cujus spicule similes, admixta. Forma hindustanica (Cyperus sp. n. 32, Hook. f. et T. Thoms.); culmi fasciculati, umbella divaricatim composita; spicule longs 1 cm, late 2-3 mm., 14-flore, subinflate ; glume in lateribus fere estriate, florigere vix imbricate; nux fere ut in forma melanor- rhiza.—In India boreali-occidentali. Forma evoluta; umbelle radii 6-9 cm. longi, divaricati, apice corymbosi; spiculs aureo-lutex, usque ad 35 mm. longe, 40-flor@; glume distantes, florigere non imbricate; styli rami e gluma breviter exsertii—A Florida usque ad Monte-Video (in formam principem sensim transeuns); etiamque in Africa australi. Ab Europa australiore usque ad Indiam, Australiam, Africam australem ; in omni America calidiore. Europa: Mentone (Moggridge, h. Kew); Naples (Tenore, h. Kew) ; ins. Crete (Sieber, h. Kew). India: Punjab, Chenab flumen (T. Thomson, h. Kew); Mora- dabad (T. Thomson, h. Kew); Delhi (C. B. Clarke n. 23361, h. Kew); Madras Peninsula (G. Thomson n. 69, hh. Kew, Calcutta); montes Nilghiri (Hohenacker n. 1294, h. Kew) ; Poona (Jacquemont n. 277, h. Kew). Australia: Queensland (Wuth, h. Kew). Madagascar (Hilsenberg et Bojer, h. Mus. Brit.); Mauritius (Bojer [C. maritimus], k. Kew). Africa: Caput Bone Spei (Buchanan n. 212, h. Kew; Wood n. 1581, h. Kew; Natal (Buchanan nn. 88, 316, k. Kew; Krauss n.97,h. Kew) ; Abyssinia (Schimper n. 1240, h. Kew); Djur (Schweinfurth n. 1731, h. Kew); Sierra Leone (Vogel n. 31, k. Kew); Niger flumen (Baikie, h. Kew); Caput Verde (Bolle, h. Kew) ; ins. Verde (Moseley, h. Kew). Ins. Madeira (Burchell nn. 632, 633, h. Kew; Lowe n. 370, h. Kew; Mandon n. 255, h. Kew). Ins. Azores (Godman, h. Kew; H. C. Watson n. 263, h. Kew; Hunt, h. Calcutta). mE America borealis: Canada (Goldie, h. Calcutta) ; Canada occidentalis (Macoun, k. Kew); New Jersey (Darlington, h. Kew); Kentucky, Lexington (Short, h. Kew); Carolina borealis (Curtiss, h. Kew) ; New Orleans (Drummond n. 387, h. Kew) ; Mexico ( Bourgeau nn. 526, 2386, h. Kew; Hartweg 254, h. Kew; C. Wright n. 1948, h. Kew). - MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 181 America australis: Amazon flumen (Traill n. 1176, h. Kew) ; Brasil (Swainson, h. Kew); Peru, Lima (Matthews n. 427, h. Kew); Uruguay (Gilbert n. 523, h. Kew); Pampas (Purdie, h. Kew) ; Montevideo (Gilbert nn. 13, 188 [spi- culis 40-floris], k. Kew). Ins. Galapagos (Anderson n. 53, h. Kew). Var. B. helodes (sp. Schrader) ; spiculis 16-30-floris, angustis, pallidis aut subfuscis.—JVees in Mart. Brasil. Cyp. p. 34. Stolones tenues, tuberiferi. Spicule florigere rectangulatim patule. Glume per totam fere latitudinem striatze.—A cl. Boeck- elero C. rotundo appendiculato; spicule autem divaricate, glume striate. Amazon flumen (Spruce n. 1127, h. Kew): Rio Negro, Barra (Spruce, h. Calcutta). Brasil (Burchell n. 1656, h. Kew; Swainson, h. Calcutta). y. Sprucei; spicis densioribus; spiculis majusculis, florigeris aureis viridi-nerviis, fructiferis fuscis; glumis laxiuscule imbri- catis; styli ramis longiuscule exsertis. Spicule usque ad 28 mm. longer, 3-4 mm. late, 32-48-flors, sat compress. [These large American forms of C. esculentus very possibly contain among them a separable species; for Spruce n. 6114, which may be considered the type of this var., appears to have a strictly fibrous annual root, though that is a point on which it is unsafe to rely on dried specimens that may have been seedlings of the first year and yet in flower. I wrote up this var. at Kew as esculentus, var. y. leta; but there is another species very near this (but not it) already named C. [efus by Presl; so that the name lætus must be altered. | America calidior: Chanduy, in litore maris Pacifici (Spruce n. 6414, hh. Kew, Calcutta); Texas (Drummond nn. 259, 315, 359, k. Kew) ; Mexico, Potosi, alt. 2000 metr. (Parry et Palmer n. 908, h. Kew). 136. C. nurEsCENS (Zorrey and Hook.! in Ann. Lyceum New York, ii. p. 433); umbella subcomposita; spieulis spicatis, florigeris erecto-patulis, majusculis, densifloris, compressis, fusco- brunneis; glumis admodum compressis, ovatis, apicatis per totam fere latitudinem plane striatis.— Chapman, Fl. South. States, p. 508. C. rotundus, J. D. Hook. ! in Trans. Linn. Soc. xx. p. 177. C. esculentus, var. y, Boeck. in Linnea, xxxvi. p. 291. 182 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Stolones tenues: tubera in exemplis non visa. Culmi 3-4 dm., robustiores. Spicule longs 3 cm., late 3 mm., 30-40-flore. America: Texas (Drummond n. 452, h. Kew) ; Florida (Chap- man, h. Kew); ins. Galapagos (Macrae, h. Kew). Var. B. fulvescens; umbella magna; spiculis magnis; glumis magnis ellipticis, remotis, fructigeris non imbricatis; styli ramis longe exsertis ; nuce parva angusta. C. fulvescens (sp.), Liebm. Mex. Halfgr. p. 22, e descr. C. lutescens, Torrey, Bot. U.S. and Mex. Bound. Surv. p. 227. Culmi 5-6 dm. Spicule usque ad 35 mm. longe, 30-36-flore. Mexico: Tortuga (Berlandier nn. 980, 2410, h. Kew) ; Mexico Nova, el Paso (C. Wright n. 705, h. Kew). 137. C. coxazsrus (Vahl, Enum. ii. p. 850) ; umbella simplice (vel fide Kunth composita); spicis breviter cylindricis, densis ; spiculis undique rectangulatim patulis, lineari-lanceolatis, com- pressis, 8-16-floris; glumis ovatis vix aeutis.— Kwnh, Enum. ii. p. 87; Boeck. in Linnea, xxxvi. p. 347. Rhizoma lignosum, breve. Culmi 2-5 dm., trigoni. Folia cum culmo subequilonga. Umbelle radii 2-10, usque ad 5 em. longs. Spice 15-25 mm. in diam., 20-40-spiculoss ; rhachis 1 cm. longa, glabra. Spicule longe 10-15 mm., late 2 mm.; rhachille ale anguste oblonge, hyaline, insolubiles. Nux ellipsoidea, utrinque angustata, acutius trigona, cum 3 parte glum: fere sequilonga. Caput Bone Spei (Greville, Mrs. Miller, h. Calcutta) ; Natal, alt. 1800 metr. (.Dr. Sutherland, h. Calcutta). Series *** Foliate, a series ** (Corymbosis typicis) paullo rece- dentes vel radice fibrosa vel spiculis fasciculatis.—Plerseque Americans. 138. C. amavroPus (Steud. Cyp. p. 33) ; culmo a vaginis in- flatis intecto, quasi-incrassato, basi ipsa tenui radicibus fibrosis ; folus longis, admodum angustis; umbella simplice; spiculis spicatis, compressis, castaneo-brunneis ; glumis ovatis, obtusis, per latitudinem fere totam multinervosis. C. leptophyllus, Hochst. MS.; Boeck. in Linnea, xxxv. p. 641. C. bulbosus var., Schimper, MS. in h. Iter Abyss. n. 1391. Culmi 2-4 dm., tenues. Umbelle radii 4, usque ad 2 em. longi. Spice longe 10-15 mm., 8-16-spiculose. Spicule long: 12 mm., MR. C. B. CLARKE ON INDIAN SPECIES OF OYPERUS. 183 late 3 mm., 10-14-flore; rhachille ale late, purpureo- maculate, insolubiles. Abyssinia: in lateribus montium Schoata (Schimper n. 1391, h. Calcutta). 139. C. seracELATUS (Rottb. Descr. et Ic. p. 26); viridis, radice fibrosa; umbella composita aut fere simplice; spiculis spicatis, linearibus; rhachille alis admodum angustis, insolu- bilibus.—Kunth, Enum. ii. p. 63; Boeck. in Linnea, xxxvi. p. 292. Brasilia: prov. Rio Negro, Barra (Spruce, h. Calcutta). 140. C. sETIGERUS (Torr. and Hook. in Ann. Lyceum New York, iii. p. 434) ; umbella magna, composita; umbellulis corym- bosis, radiolis longius bracteatis; spiculis spicatis, acute lineari- lanceolatis.— Boeck. in Linnea, xxxvi. p. 298. [This has been issued from Kew marked C. longus, Linn. ; and it is so exceedingly like some of the large forms of C. longus, Linn., that, in the absence of the rhizome, it is difficult to express an opinion whether it should be considered specifically distinct. | Texas (Drummond, h. Calcutta). 141. C. srricosus (Linn. Mant. p. 62, excl. syn. Sloane); foliis braeteisque angustis, longis; umbella subsimplice; spicis eylindricis, densiusculis; spieulis rectangulatim patentibus, lineari-lanceolatis, compressis; rhachille alis angustis, insolu- bilibus.— Kunth, Enum. ii. p. 87; Boeck. in Linnea, xxxvi. p. 346. Radix fibrosa. Culmus basi ipsa inerassatus, aliquando con- spicue bulbosus (bulbo parvo globoso). ^ Umbelle radii 5-6, usque ad 4 em. longs, 12-80-spiculose. Spicule longe 16 mm., late 2-3 mm., 14-20-flore, straminese aut luteæ. Nux $j parte glume brevior, oblonga, trigona. [This plant appears to me allied to O. Thomasi, and still more closely to C. lucidulus. The glumes fall one by one from the spikelets, beginning with the lowest glume and proceeding regularly upwards till the rhacbilla of the spikelet is left entire still attached to the spike with only a few glumes at the apex. This is altogether the character of Eucyperus. It is true that subsequent to this the spikelets (very late) are broken from the spikes, as occurs also in all other species of Eucyperus at last. Also, in C. strigosus, the disk-like papille on the old rhachises 184 ME. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. give the suggestion of a quasi-articulation ; but this is also so in C. polystachyus and other altogether remote species. | l America borealis: Canada (Goldie, h. Calcutta); Pennsylvania (Poeppig, h. Calcutta) ; Chieago (Babcock, h. Calcutta). 142. C. CEPHADANTHUS (Torrey and Hook. in Ann. Lyceum New York, ii. p. 181); foliis bracteisque longis; umbella com- posita, densa; umbellulis sæpe condensatis; spicis cylindricis, densis; spiculis undique rectangulatim patentibus; rhachillæ alis magnis, rotundis, subsolubilibus. C. lætus, var. y, Boeck. in Linnea, xxxvi. p. 295. Radix fibrosa. Culmus 4 dm., triqueter. Umbella 12 em. in diam., densa. Spiculæ longs 12 mm., late 2 mm., laxe 14-floræ. [This species, arranged by Boeckeler in the centre of the typical Corymbosi, has been actually issued from Kew as C. glomeratus, Linn. ; and it is so like C. glomeratus that, but for the extra- ordinary wings of the rhachilla, it might possibly be united with it. Boeckeler has, on account of these wings, removed the species to the Corymbosi; but there is nothing in the Corymbosi like the wings, to match which we must go into Mariscus. This is one (among many) species that leads me to think too much systematic importance has been given these wings; I doubt whether, on the whole, C. cephalanthus is not really nearer C. glo- meratus than it is to anything else; the root, stem, habit, in- florescence all point that way.] Texas (Drummond, h. Calcutta). Sect. O. Papyri. Papyri et Exaltati, Kunth, Enum. ii. pp. 64, 70; Boeck. in Linnea, xxxvi. pp. 302, 317. Radix fibrosa. Culmus solitarius, robustus, foliatus. Umbella magna, composita aut decomposita, spiculis spicatis, innumerosis. Spieule compress, multi- vel pauciflore; rhachille ale magne, solubiles, aut minus conspicus aut insolubiles. Glume parve, ovato-rotundse, sæpius breviter mucronate. Stamina 3; antherx longe aut breviter cristate vel mutice. Nux ellipsoidea aut obovoidea, trigona aut triquetra, cum 3-3 parte glume exqui- longa. This is a very natural group, easily recognized by the innu- merable spikelets and small glumes: in some of the species small examples with a depauperated congested umbel occur, but such MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 185 are rare. The wings of the rhachilla are in many species very soluble and tawny yellow ; similar (though narrow) soluble tawny wings are present in some forms of C. exaltatus itself; and it is, I think, inconvenient to make two sections upon such a character. Our sections of Eucyperus are so unmanageably numerous that our object should be to throw two into one wherever possible. * Spicule lineari-lanceolate, rhachille ale insolubiles vel angustissime. 143. C. napraTUs (Vahl, Enum. ii. p. 869); robusta, foliata, umbella majuscula, composita specie simplice; spicis longis, a spiculis dense antice fastigiatis, sessilibus, umbellulatis ; glumis approximatis; rhachille alis angustis, insolubilibus; antheris muticis; nuce quam 4 pars glumæ breviore.— Kunth, Enum. ii. p. 71; Boeck. in Linnea, xxxvi. p. 817. .. C. verticillatus, Roxb. FI. Ind. i. p. 206. Radix fibrosa. Culmus solitarius (interdum cespitosi), robus- tior, 3-8 dm. longus, apice (ssepe obtuse) trigonus. Folia plura, eum culmo sspe equilonga, 3-nervia, multistriata, in marginibus minute scabra. Involucri bractez 3-6, usque ad 3-5 dm. longe, divaricate. Umbelle radii 3-12, usque ad 1 dm. longi; ochreæ usque ad 2 em. longs, bidentate, dente altero usque ad 1 cm. longo, lineari. Umbellule bracteolatz ; radioli sepius brevissimi, interdum 0-15 mm., sepissime 1-spicati ; spice ipse semper ses- siles, nud: (causa autem radiolorum suppressorum spice specie 3-7 digitate bracteate), longe 2-6 cm., late 1 cm., undique a spiculis sessilibus, antice fastigiatis, densissime obsite. Spiculee longe 5-8 mm., late 1 mm., 16-30-flore ; gluma ima parva, non setaceo-caudata. Glum: approximate, dense imbricate, margi- nibus in sicco non incurvis, late ovate, minute apiculate, com- presse, in dorso viridi-3-nervix, in lateribus sanguineo-striate, in margine lutescentes. Anthers parve, anguste oblonge. Stylus vix cum nuce zquilongus ; rami 3, e gluma breviterexserti. Nux ellipsoidea, utrinque angustata, subacuta, trigona, parva, lutea aut brunnea, levis. Species fide Boeckeler Z. e. in tropicis Asie, Africe, Americæ late sparsa; hoc loco habitat. Indicas paucas notare potui e. g. Bengalia orientalis (Grifith nn. 6148, 6218 | 1, ^. Calcutta). Bengalia: Sahebgunj (Kurz, h. Calcutta) ; Dacca (C. B. Clarke n. 17124, h. Calcutta). Assam: Gowhatty &e. (Jenkins, Simons, h. Calcutta). 186 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Chittagong, alt. 0-300 metr. (J. D. Hooker et T. Thomson). Pegu (Kurz n. 650, h. Calcutta). 144. C. EXALTATUS (Retz, Obs. v. p. 11) ; exaltata, foliata, um- bella magna, composita; umbellularum radiolis sepius elongatis, apice ebracteatis, 1—4-spicigeris ; glumis approximatis; rhachille alis angustis, insolubilibus ; antheris muticis ; nuce cum 4 parte glum:» subzquilonga, in tribus faciebus subconcava.—JVees in Wight, Contrib. p. S4; Kunth, Enum. ii. p. 71; Boeck. in Linnea, xxxvi. p. 821; Benth. Fl. Austral. vii. p. 285. C. canaliculatus, Retz, Obs. vi. p. 20, fide Vahl, sed? C. festivus, Willd.; Link, Jahrb. iii. p. 88. C. umbellatus, Roxb.! Fl. Ind. i. p. 205, Ic. ined. t. 729, in h. Kew; Dalz. and Gibs. Bomb. Fl. p. 284, nec Benth. C. venustus, R. Br. Prod. p. 217; Thwaites, Enum. Pl. Zeyl. p. 432; nec Nees, nec Kunth. C. altus, Nees in Wight, Contrib. p. 84. — Cyperus, Wall.! List n. 3328, A, B, C. — Cyperus, Wall. ! List n. 3343, A, B, C, D E, F, G, H, I, J, K. — Cyperus, Wall.! List n. 3327, A, B, C (forma juvenis). Radix fibrosa. Culmus solitarius, usque ad 1 dm. longus, basi tumidus, apice triqueter. Folia plura, cum $ parte culmi sspe squilonga, in marginibus scabra. Involucri bracteæ 3-6, usque ad 2-5 dm.longw. Umbella 2-4 dm. in diam.; radii 5-10, usque ad 18 em. longi; ochree 2-4 cm. longs, bicuspidate. Spice longe 2-5 cm., digitate, 20-40-spiculose. Spieule longe 6-10 mm., late 1-2 mm., distantes, rectangulatim patentes, lanceolate, castaneæ, compressz, 8-16-flore; gluma ima rarius setaceo-cau- data. Glume arcte stipate, ovate, minute apicate, in dorso viridescentes, 3-nervie ; in lateribus castaneo-rubre aut pallidiores lutescentes, obscure nervose. Stamina 3, vix exserta; antheræ lineari-oblonge. Stylus nuce brevior; rami 3, e gluma brevius- cule exserti. Nux cum 3 parte glume subequilonga, ellipsoidea, utrinque angustata, apice vix acuta, plane trigona, pallide brunnea aut lutescens ; cellule extime parve, quadrate, persistentes. Zeylania (Thwaites C. P. n. 3040, h. Calcutta). Madras Peninsula (Heyne in Wallich n. 3343 G- [C. extensus], h. Calcutta). Mysore v. Carnatic (G. Thomson [C. altus, Nees], h. Calcutta). Saugor (Vicary, h. Calcutta). Chota Nagpore, alt. 200 metr. (C. B. Clarke n. 20871, h. Cal- cutta). MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 187 Rajmahl (Wallich n. 3328 A [C. canescens, Ham.], h. Calcutta). Oudh (R. Thompson n. 302, h. Calcutta). Bengal (Wallich n. 3343 K, h. Calcutta). India centralis et mons Aboo (G. King, h. Calcutta). Canton (Hance n. 9308 [C.? racemosus, Retz), h. Calcutta). Var. B. amæna; spiculis multifloris, in spicas nutantes densas antice fastigiatis. C. ameenus, Koenig ; Roxb. Ic. ined. t. 731 in hh. Kew, Calcutta ; non Kunth. C. alopecuroides, Rozb. Fl. Ind. i. p. 208, Ic. ined. in h. Calcutta ; non Rottb. Spieule longe 10-14 mm., late 1-2 mm., usque ad 20-50-flore. Calcutta (Hort. Bot. Caleutta [C. seminudus], k. Calcutta ; Kurz [C. rotundus ?], k. Calcutta; C. B. Clarke n. 19532). Madras Peninsula (Wallich n. 3329 B [C. alopecuroides, Wight), h. Calcutta; Wight n. 2863; Heyne in Wallich n. 3343 A. [C. maximus ,C. distachyos?, Vahl], hh. Calcutta, Madras [C. fastigiatus], k. Calcutta; Heyne in Wallich n. 3343 D [C. spicatus], ^. Calcutta; Gingee, Wight in Wal- lich n. 3343 H [C. verticillatus, Roxb.], h. Calcutta). Zeylania (Thwaites C. P. n. 9788 [C. venustus, R. Br.], h. Cal- cutta). [Forma assamica; spiculis minus numerosis, majoribus, lati- oribus. Assam (Wallich, Griffith et Jenkins, h. Calcutta).] Var. y. dives, Delile, Egypt. v. t. 4. fig. 3 (sp.) ; spiculis dense spicatis, undique rectangulatim patentibus, lucide brunneis aut aureis ; glumis approximatis, breviter mucronatis ; rhachille tenuis alis angustis, insolubilibus; stylo trifido; nuce ellipsoidea aut subovoidea, trigona, cum 4 parte glume vix equilonga.— Kunth, Enum. ii. p. 71. C. alopecuroides, Boeck. in Linnea, xxxvi. p. 322, var. a, nec Rottb. Umbella quam in C. exaltato typico rigidius, divaricato-corym- bosa, lutescens; spicule longiores, crassiores, subturgide, lucide flavido-brunnez, 30-40-flore ; gluma ima sepius longe setaceo- acuminata. . The type specimens of this at Kew agree exactly with my Caleutta example, so that if C. dives is a distinct species, it is an Indian one. The specimens differ from those of Retz's exaltatus by the much more numerous glumes to the spikelet, while they differ from those of Koenig's amenus in the rigid, not tasselled, 188 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. nodding spikes. C. dives may therefore yet be held a distinct species; but whatever it is considered, it must be remote from C. alopecuroides, Rottb., which has a 2-fid style, a compressed nut, and a very thick rhachilla. Egypt: Rosetta (^. Calcutta). Caleutta: Mutlah (C. B. Clarke n. 24707, h. Calcutta). Var. ? 6. Oatesii ; inflorescentia ut in C. exaltato typico; spi- eulis nitidis, pallidis, 14-floris; rhachille alis parvis, linearibus, flavidis, mox solutis; nuce magna, cum $ parte glume equilonga, trigona, pallida. Spicule angustze. Nux cum ? parte glume interdum equi- longa.—Species fortasse bona, sectionem Papyrum cum Exaltatis arcte conjungens; cum C. exaltato, Retz, admodum affinis. Burma: Thyat Myo (ug. Oates, h. Calcutta). ** Rhachille ale conspicue, solubiles ; spicule anguste lineares; antherce sepe cristate. 145. C. auricomus (Sieber; Spreng, Syst. i. p. 2830) ; exaltata, foliata; umbella magna, composita aut decomposita ; spicis longis, digitatis ; spieulis densis, multifloris, undique rectangulatim pa- tentibus, linearibus, basi bracteolatis; glumis remotiusculis ; rhachille alis lanceolato-rhomboideis, flavidis, solutis; antheris muticis; nuce cum 3 parte glume equilonga, oblongo-ellipsoidea, trigona, apice breviter cylindraceo-umbonata. C. digitatus, Roxb. Fl. Ind. i. p. 205. C. venustus, Kunth, Enum. ii. p. 68; Boeck. in Linnea, xxxvi. p. 316 (non R. Br.?). C. Neesii, Kunth, Enum. ii. p. 101, syn. Linn. excl. ; Boeck. in Linnea, xxxvi. p. 315. C. racemosus, Kunth, Enum. ii. p. 100, quoad syn. citat., nec quoad descript., nec Retz. Radix valide fibrosa. Culmus solitarius, 3-8 dm. longus, sur- sum trigonus vel triqueter. Folia plura, cum $ parte culmi sepe eequilonga, in marginibus scabra. Involucri bractex 5-8, usque ad 5 dm. longa, divaricate. Umbella 1-6 dm. in diam., compo- sita aut decomposita; radii 4-10, sepe 1 dm. longi; ochres 1-3 em. longs, apice 2-cuspidate. Umbellularum bracteole 3-5, foliacez, cum umbellulis szpe equilonge. Spies 2-4 em. longs; rhachis glabra a glumis imis spicularum persistentibus undique dense ornata. Spicule lineares, compress:, 10-24-flore, quoad magnitudinem variabiles ; in C. auricoma Sieb. typico anguste sub- MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 189 teretes, aciculares, glumis densius imbricatis; in C. Weesii lati- uscule, laxiuscule 10-flore ; in forma majuscula (in Assam &c. vulgari) longe 16 mm., 20-flore, multum compress, glumis lon- gioribus, remotiusculis; rhachille alis magnis, caducis. Glume ovate aut elliptieze, compresse, dorsi nervi 3, virides, in acumen excurrentes, latera rubra vel saturate colorata, margines scariosi. Stamina 3; anthere lineari-oblonge. Stylus nuce brevior; rami 3, e gluma breviter exserti. Nux paullo unsymmetrica, curvata, obtuse trigona, apicem versus in brevem cylindrum subito angus- tata; stylobasis ab apice cylindri obtusi facile secedens. Nux nigra; cellule extim: quadrat, crasse, albide. This plant is the C. auricomus of h. Kew; but the synonymy I did not work at Kew, and cannot discover at Calcutta. The localities following are all from the Calcutta Herbarium :— India Orientalis: Moradabad (Z. Thomson [C. auricomus, Sieber]); Sikkim Terai (G. King) ; Assam (Jenkins, Simons, Masters); Khasia; Nongram, alt. 1400 metr. (C. B. Clarke) ; Jengal; Rajapore Jheel (Kurz [C. digitatus, Rorb.]); Ben- gal (Hooker f. et T. Thomson [C. auricomus, Sieber]) ; Bengal orientalis (Griffith n. 6149) ; Dacca (C. B. Clarke nn. 7417, 17131 B); Nilgherries, alt. 1900 metr. (C. B. Clarke n.11368 ; G. King n.1318); Zeylania (Thwaites C. P. n. 3043). Burma: Meaday (R. Scott); Pegu (Kurz nn. 647, 672, 2672) ; Attran et Pagamew (Wallich n. 3438 [C. digitatus, Roa. |). Egypt (Sieber n. 42). Nupe, in flumine Niger (Barter) Brasil (Swainson). Var. B. khasiana ; nuce apice acutata in stylobasin sensim transeunte. C. Hookeri, Boeck. in Linnea, xxxvi. p. 308. Papyrus elatus, Hook. f. et T. Thoms. MS., non Nees. Bengalia orientalis (Griffith n. 6151). Khasia v. Assam (Wallich et Griffith, Assam Deputation). Khasia: regio tropicalis (Hooker f. et T. Thomson n. 23 [Papyrus elatus, Nees]). . Khasia: Sohra Reen, alt. 1600 metr. (C. B. Clarke n. 19173). 146. C. ELATUS (Linn. Amen. Acad. iv. p. 301, Sp. Pl. p. 67); robusta, foliata; umbella magna, decomposita; spicis elongatis, densis; spiculis parvis, angustis, antice fastigiatis; antherarum arista cum 4 parte loculorum subequilonga. 190 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. C. racemosus, Retz, Obs. vi. p. 20 (?) ; Kunth, Enum. ii. p. 101; Boeck. in Linnea, xxxvi. p. 311. Cyperus, Wall.! List n. 3341 A. Radix valide fibrosa. Culmus solitarius, 3-8 dm. longus, apice triqueter fere levis. Folia cum $ parte culmi equilonga, in marginibus scabra. Involucri bractesz 4-8, divaricatz, usque ad 5 dm. longe. Umbella 3-4 dm. in diam.; radii 1-2 dm. longi, robusti: ochree 3-5 cm. longs, apice in altero latere foliaceo- lanceolate ; umbellularum bracteole 3-6, cum umbellulis sepe equilonge. Spice in umbellula secundaria numerose, usque ad 8-16, longz 5-7 cm., late 5-7 mm., 20-50-spiculose ; rhachis a spiculis tecta. Spicule longe 5-7 mm., late 1 mm., 12-flore, luteo-fusce, parum compress, basi non (aut obscure) bracteo- late; rhachille ale magne, late lanceolate, flavide, mox solute. Glume ovate, compress, vix mucronate, in dorso virides 3-nervie, in marginibus scariose. Stamina 3; anthere parve, oblongz, flav, a crista lineari-lanceolata scabra rubra coronate. Stylus brevis; rami 3, e gluma breviter exserti. Nux cum 3-3 parte glumæ subequilonga, ellipsoidea, sursum angustata, trigona, nigra; cellule extime quadrate, albidz, crasse. Penang (Wallich n. 3341 A, h. Calcutta). Java (Kurz n. 2751, h. Calcutta). Var. B. macronux; umbella specie composita; umbellularum bracteolis umbellulas longe superantibus; spiculis minus nume- rosis, majoribus, subpatentibus ; nuce vix quam sit gluma breviore. Bengal: Comilla (C. B. Clarke n. 14188, h. Calcutta). [I believe I have written this variety up in h. Kew as a distinct species. ] 147. C. PLATYPHYLLUS (Roem. et Sch. Syst. ii. p. 876) ; maxima, foliis latis; umbella composita; spicis elongatis, a spiculis dense fastigiatis ; antherarum crista lineari-lanceolata cum 3 parte loculorum zquilonga. C. elatus, Roxb. Fl. Ind. i. p. 204, non Linn. C. maximus, Roxb. MS. in. h. propr. C. eminens, Klein ; Kunth, Enum. ii. p. 70; Boeck. in Linnea, xxxvi. p. 312. C. Roxburghii, Nees in Wight, Contrib. p. 84. Cyperus, Wall.! List n. 3341 B, h. propr. Radix valide fibrosa. Culmus solitarius, 7-14 dm. longus, crassus, basi teres, sursum acute triqueter, in angulis serrato- scaber, manus incautas secans. Folia 5-10 dm. longa, 2-3 cm. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 191 lata, in marginibus scabra. Involucri bractes 4-6, usque ad 9 dm. longe. Umbella 3-4 dm. in diam.; radii 10-15 cm., ro- busti; ochree 3 cm., bicuspidate. Umbellularum bracteole parve, 1-3 cm. longs; radioli 3-8, breves. Spicw longe 5 cm., late 8 mm., 1-4-digitate. Spicule longe 8 mm., late 2 mm., 10-14-flore, pallide; rhachille ale magne, late lanceolate, flavide, solubiles. Glume arctius imbricatz, compresse, margi- nibus in sicco non incurvatis, in dorso viridescentes, 3-nervie, in lateribus stramineo-fusce, obscurius multistriate. Stamina 3; antherarum loculi flavi; crista rubra, scabra. Stylus brevis ; rami 3,e gluma breviter exserti. Nux ellipsoidea, sursum angus- tata, trigona, cum 4—§ parte glums squilonga; cellule extime albs, crasse, quadratw#.—Species C. elato affinis, ab hoc bene distincta; culmus major, apice serrato-scaberrimus ; umbellu- larum bracteole multo minores; spicule majores; anthere majores, longius appendiculate. Zeylania (Thwaites C. P. n. 8041, h. Calcutta). 148. C. Papyrus (Linn. Sp. Pl. p. 70) ; umbelle radiis usque ad 100, 1-2 dm. longis; involucri bracteis umbella multo brevi- oribus.— Kunth, Enum. ii. p. 64; Boeck. in Linnea, xxxvi. p. 304. Rhachille ale lanceolate, caduce. Anthera oblonga, apice acutata vix cristata. Nux cum § parte glume equilonga, ellip- soidea, utrinque angustata, trigona. In Calcutta Hort. Bot. culta. Subgenus 5. Drcrrprux. Diclidium, Nees in Martius, Fl. Brasil. Cyp. p. 51, t. 1, 2 (genus Schrader); Boeck. in Linnea, xxxvi. p. 392; Benth. et Hook. f. Gen. Pl. iii. p. 1045. Spicule rhachilla per articulos disjuncta; gluma persistens cum articulo decidua; rhachille ale conspicue, hyaline, nucem amplectentes, insolubiles. Umbella composita vel decomposita, sepius laxa. Spicule spicate, subulate, subteretes. Glume remotiuscule. Anthere mutice. Stylus 3-fidus. Nux trigona, equalis vel curvata. 149. C. renax (A. Rich. in Act. Soc. Hist. Nat. Paris, i. p. 106); robusta, foliata; umbella composita, specie subsimplice; spicis longis undique patentim spiculosis; spiculis pauci- vel multifloris, linearibus, teretibus, sepe curvatis.—C. B. Clarke in Journ. Linn. Soc. xx. p. 295, cum syn. 192 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. Radix fibrosa. Culmi solitarii, aut ope progemmationis basalis 2-3ni, longi 3-5 dm., apice trigoni. Folia plura cum $ parte culmi subequilonga, in marginibus scabra. Involucri bractes (in umbellis majoribus) usque ad 3-4 dm. longs, valide, in mar- ginibus scabre ; in umbellis interdum contractis vix 1 dm. longs, debiliores. Umbella 1 dm. in diam. contracta, spicis densa, aut usque ad 5-6 dm. in diam., radiis longis 2 dm., validis; ochree usque ad 5 em. longs, in altero làtere foliaceo-appendiculate. Umbellule divaricato-corymbose ; bracteole cum $ parte umbel- lule sæpæ equilonge. Umbellula specie simplex sed revera sepius iterum divisa, 7. e. radioli plus minus elongati, conspicue ochreati, apice l-paucispicigeri. Spicule longe 5-25 mm., in diam. 1-2 mm., 5-15-flore ; gluma ima lanceolata, setaceo-caudata. Glum:e distantes, parum imbricatz, arcte adpresse; rhachilla suberosa alternatim excavata; articulus unusquisque e nuce in rhachilla abscondita cum gluma floris superioris exstructus. Glume late ovate, obtuse, minute (aut non) apiculate, in dorso rotundatz», rubro-striate. Stamina 3; anthere mutice. Stylus brevissimus ; rami 3, e gluma breviter exserti. Nux anguste oblonga, cum $ parte glume equilonga, utrinque breviter angus- tata, nigra; cellule extime quadrate, tenues, hyaline. Mr. Bentham in herb. Kew has treated this as a composite species, widely spread in the tropics of both hemispheres, and I followed him in Journ. Linn. Soc. xx. p. 295. I made no notes on this species at Kew, and therefore here at Calcutta can only refer to that place for the synonymy and geography. The description above is from the following :— Bengal: Furidpore (C. B. Clarke n. 7511, h. Calcutta). Pegu (Kurz n. 2686, h. Calcutta). Borneo (Barber, h. Calcutta). 150. C. MrcnAUXIANUS (Torrey in Ann. Lyceum New York, iii. p. 259) ; robusta, foliata; umbella composita ; spicis cylindricis, densis; spieulis rectangulatim patentibus, linearibus, subtereti- bus, 6-floris; glumis remotis, ovatis, obtusis.— Boeck. in Linnea, xxxvi. p. 400. Radix fibrosa. Culmus 4 dm. Folia lata 5 mm. cum culmo fere equilonga. Umbella 7-radiata, 8 dm. in diam. Nux cum $ parte glume equilonga, oblonga, subtrigona, curvata, luteo- brunnea. ° Florida (Chapman, h. Calcutta), MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 193 151. C. wEMAcAULOS (Steud. Cyp. p. 43); foliis bracteisque longis, tenuibus; umbella simplice densiuseula ; spiculis spicatis, linearibus, longis, patentibus; glumis remotis, adpressis; nuce longa, angusta. C. filiculmis, Schrader ; Nees in Mart. Brasil. Cyp. p. 40, non Vahl. Rhizoma perenne, tenue. Culmi 4 dm., tenues. Spice longs 10-25 mm., 10-22-spiculoss. Spicule longs 22 mm., late 1-2 mm., 12-flore. Glume oblongs, obtuse. Rhachille ale angustiores, insolubiles. Para: Santarem (Spruce, h. Calcutta). Subgenus 6. Marıscus (char. ampliato). Mariscus, Vahl, Enum. ii. p. 372 (genus); Kunth, Enum. ii. p. 115 (genus) ; Boeck. in Linnea (Leptostachyis inclusis), xxxvi. p.372; Benth. et Hook. f. Gen. Pl. iii. p. 1045. Stylus 3-fidus ; glume persistentes; spicule caduce.—Foliate. Spicule sspius pauciflore, interdum 1-flore (i. e. 4-5-glume, glumis 2 imis vacuis, tertia hermaphrodita, nucigera, quarta mascula vel vacua depauperata) rarius multiflore. ^ Rhachille ale latz vel anguste, insolubiles, sepius conspicue. Rhachilla continua, szpe crassiuscula aut alternatim excavata, cum glumis supra glumam secundam articulatim caduca. Glume 2 ime vacus (ima szpe caudata, secunda sepius parva truncata), sepe post casum spicule persistentes. Stamina 3; anthere lineari- oblong, mutice. Nux trigona, haud raro inequalis, sepius majuscula haud raro augusta. [I have not before me a sufficiently large number of species to propose a complete arrangement: the following “sections” are merely a list of the Indian species, with notes on some non-Indian that appear allied to them.] * Spicule aciculares pluriflore, umbella sepius composita (Leptostachyi, Boeck.). 152. C. prnurus (Vahl, Enum. ii. p. 857); robusta, foliata ; umbella simplice aut composita; spicis stellato-globosis ; spiculis 3-9-floris, linearibus ; nuce anguste oblonga, acuminata, rostrata, —Kunth, Enum. ii. p.92; Boeck. in Linnea, xxxvi. p. 354. C. spinulosus, Roxb. Fl. Ind. i. p. 20. Radix robuste fibrosa. Culmi solitarii, aut sspe cespitosi, 3-5 dm. longi, sursum teretes, vix trigoni leves. Folia plura, LINN. JOURN.— BOTAN Y , VOL. XXI. o 194 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. lata, basi longe vaginata, spongioso-incrassata, sursum in mar- ginibus spinuloso-scabra. Involucri bractez 4-8, usque ad 2-4 dm. longs. Umbella spherica, 1-2 dm. in diam. ; ochreæ incon- spicuæ. Spice ebracteatæ, usque ad 3 cm. in diam. (sæpius multo minores), rubro-brunneæ aut cinnamomeæ, 15-40-spiculose. Spiculæ dense digitatæ (brevissime spicatæ) lineares, compresse, usque ad 15 mm. longe, 12-14-florie (swpius multo minores 3-6-flore) ; rhachilla tenuis, continua, supra glumam secundam articulatim decidua; ale oblonge, hyaline, insolubiles. Glume elliptic, obtuse, multinervose. Stamina 3; anthere oblong, mutice. Stylus brevis; rami 8, e gluma breviter exserti. Nux (rostro incluso) cum $ parte glume tequilonga, anguste oblonga, trigona, nigro-brunnea; cellule extime quadratie, tenues, incon- Spicuz. [India Orient. Malacca. Java. Ins. Philippinenses, fide Boec- keler ; the subjoined localities are all from the Calcutta Herbarium.] Bengalia orientalis (Griffith n. 6245). Bengal: Sahebgung (Kurz) ; Sylhet (C. B. Clarke n. 6960). Planities Gangetica superior (T. Thomson; Falconer n. 1143). Assam (Jenkins, Simons) ; Suddiya (Griffith). Khasia colles (J. D. Hooker et T. Thomson n. 28). Madras Peninsula (Wight in h. Wallich n. 3439 ; G. Thomson). Zeylania (Thwaites C. P. n. 815). Pegu (Kurz nn. 660, 661, 2671); Sittang (R. Scott n. 897). Ins. Penang (Wallich n. 9439). Ins. Andaman et Nicobar (Kurz). Malaya: Goping (k. Caleutta n. 952). [Allied to C. dilutus are C. lucidus, R. Br., C. multifolius, Kunth, C. incompletus, Link, C. nitidulus, Boeck., C. dissitiflorus, Torrey, &c., but they all have the spikes much more elongate.1 ** Umbellule divaricato-corymbose ; spicule lanceolate, subteretes, Pluriflore. ` 153. C. PENNATUS (Lam. ; foliis bracteisque longis, spo culis divaricatis, arcte imbricatis. vii. p. 284. C. canescens, Vahl, Enum. ii. p.32 Radix valide fibrosa. Poir. Encycl. vii. p.240) ; robusta; ngiosis; umbella decomposita; spi- 5-12-floris, pallidis subrubescentibus ; glumis —Kunth, Enum. ii. p. 80; Benth. Fl. Austral. 55; Boeck. in Linnea, xxxvi. p. 340, Culmus solitarius, 3-7 dm. longus, MR. €. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 195 crassiusculus, sursum trigonus. Folia plura, cum culmo sspe equilonga, basi longius vaginata. Involueri bracteæ 3-5, usque ad 5 dm. longe, divaricatz, in marginibus scabre. Umbella 1-3 dm. in diam., rigida, divaricata, densa; radii 6-12, usque ad 5-8 em. longi; ochreæ 1-2 em. longs, apice unidentate. Um- bellularum radioli 1-2, corymbosi, divaricati (umbella sepe specie simplex); bracteole parve, rarius 1 em. longs. Spice 1-2 em. longz, 10-24-spiculose ; rhachis glabra; spicule undique rectan- gulatim (et subretrorsim) patentes, lanceolate, subteretes, supra glumam secundam articulatim caducw, 5-14-florz, pallidæ, sub- turgid: ; rhachilla alternatim excavata; ale hyaline, latiuscule, insolubiles. Glume semper imbricatæ, concave, ovate, obtuse, 9-11-striatz, stramineo-grisez, pallide purpureo-maculate. Sta- mina 3; anthere lineari-oblonge, mutice. Stylus nuce brevior; rami 3, longiusculi. Nux obovoidea, acute trigona, cum 4 parte glume equilonga, nigra; cellule extims lucide albs, minute poross. [I know no species to compare with this. The subjoined loca- lities are from the Calcutta Herbarium: the plant also grows in Australia and the Seychelles.] Bengalia orientalis ( Griffith n. 6159). Caleutta (C. B. Clarke n. 24705). Madras Peninsula (Wight n. 2866). Zeylania (Thwaites C. P. n. 678). Moulmein (Wallich n. 3359 E). Arracan et ins. Andaman (Kurz). Burma v. Malay Peninsula (Griffith n. 6159). Tenasserim v. Andamans (Helfer n. 6159). Borneo ( Barber). *** Snicule oblonge 3-6-flore, plus minus compresse ; umbella composita spicis cylindricis, vel subsimplex spicis densissimis. 154. C. SEEMANNIANUS (Boeck. in Linnea, xxxvi. p. 390); umbellz specie simplicis radiis elongatis, digitatim 1-4-stachyis ; spicis cylindrieis usque ad 60-spiculosis ; spiculis rectangulatim patentibus, 4-7-glumis, 1-3-nucigeris ; nuce obovoidea, triquetra, apicata. i i Mariscus lævigatus, Hook. et Arn. Bot. Beechey Voy. ii. p. 72, non Roem. et Sch. , An differt C. monostachys, Boeck. in Linnea, xxxvi. p. 389 ? Ins. Feejee (Seemann n. 669, ^. Calcutta ; Milne, h. Calcutta). 196 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 155. C. ANDERSSONII (Boeck. in Linnea, xxxvi. p. 988); um- bella simplice; spicis digitatis, densis, eylindricis, terminali sepius multo longiore ; spiculis anguste oblon gis, 2-4-floris ; nuce eum ? parte glume equilonga, ellipsoidea, utrinque angustata, triquetra. (A Kew sub nomine C. brachystachyo, Hook. f., distributa.) Ins. Galapagos (Douglas, h. Calcutta). [Allied closely to this last appear to be Mariscus AMutisii et M. Jacquini, H. B. K., to one of which I suppose Fendler n. 1601, from Tovar in Venezuela, to belong. | 156. C. LIGULARIS (Linn. Amoen. Acad. v. p. 31, Sp. Pl. p. 70); robustior, foliacea; umbella specie simplice; spicis cylin- dricis, densissimis, digitatim 1—3nis; spieulis 3—5-floris, rufes- centibus ; rhachille (in articulos insolubilis) alis latis insolubili- bus; nuce acute triquetra, acuta, inequali—Kunth, Enum. ii. p. 69; Boeck. in Linnea, xxxvi. p. 332; C. B. Clarke in Journ. Linn. Soc. xx. p. 295. Mariscus brachystachyus, Hook. in Trans. Linn. Soc. xx. p. 179. Africa; Australia; America tropicalis. Para: Santarem (Spruce, h. Calcutta). Galapagos archip.: James et Charles ins. (Douglas, h. Cal- cutta). 157. C. FLAVUS (C. B. Clarke, nec Presl, nec Boeck.); involucri bracteis numerosis, longis; umbella simplice, congesta, radiis subnullis ; spicis cylindricis, densis ; spiculis 3-6-floris, 2—4-nuci- geris; nuce cum 3 parte gluma equilonga, late ellipsoidea, utrinque subacuta, acute trigona, paullo inequali, castanea. Mariscus flavus, Vahl, Enum. ii. p. 374; Kunth, Enum. ii. p. 118. Rhizoma perenne. Culmi 3 dm., robustiores, apice triquetri. Folia plura 5-6 mm. lata, cum culmo exquilonga. Involucri bractez 4—7, usque ad 18 cm. longs, divaricate. Para: Santarem (Spruce, h. Calcutta). This species is exceedingly like C. ligularis, but has the rays of the umbel obsolete. I am very possibly wrong (in the absence of authentic type specimens) in taking it as the original Mariscus flavus ; but it certainly isnot Boeckeler’s Cyperus flavus (Linnea, xxxvi. p. 384), which has the nut “late obovata basi leviter an- gustata apice obtusissima, subretusa, apicata," and is a plant exceedingly like in general aspect to what I call C. Jlavus, but MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 197 has a totally different nut, and is = Holton n. 116, from La Paila in New Granada. **** Umbella simplex vel subcapitata, spicule pluriflore, digitato-fasciculate ; glume subdistantes. 158. C. nurus (Rottb. Descr. et Ic. p. 20, t. 4. fig. 5) ; culmis cespitosis, foliatis; spicis 2-5 in capitulum unicum congestis ; spiculis lineari- -oblongis, turgidis, 3-5-floris.— Roxb. Fl. Ind. i. p.189; Boeck. in Linnea, xxxvi. p. 336; C. B. Clarke in Journ. Linn. Soe. ii. p. 285. C. kyllingizoides, Vahl, Enum. ii. p. 312; Kunth, Enum. ii. p. 94, non Pursh. Radix fibrosa. Culmi plures, 2-3 dm. longi, basi paullo tumidi, à vaginis pluribus laxis rubescentibus intecti, apice triquetri. Folia plura, angusta, debilia, viridia, cum culmo spe equilonga. Involucri bractes plures, longs 8-16 cm., flaccide. Capitulum 8-16 mm. in diam., albo-viridis. Spicule undique divaricatz, dense, 5-8 mm. longe ; rhachille ale lanceolate, hyaline, in- solubiles. Glume imbricatz, longe 2 mm., late ovate, obtuse, compress, per totam fere latitudinem 15-19-nervie. Stamina3; filamenta linearia, levia; anther lineari-oblonge, mutice, rubre. Stylus nuce brevior; rami 3, e gluma breviter exserti. Nux cum $ parte glume equilonga, ellipsoidea utrinque angustata, trigona, nigra; cellule extime quadrate, hyaline. . In endeavouring too hastily to compress the number of sections of Cyperus (in Journ. Linn. Soc. xx. p. 285) I have put this Species with C. niveus and the capitate set of Eucyperus: this I now call a great error. This species has been issued from Kew as Mariscus Kraussii, Hochst.; but Boeckeler says M. Krauss?? is altogether different. . Malay Peninsula: Wellesley (G. King, h. Calcutta); ins. Penang (A. Calcutta n. 1724); in oris Burme australis (Kurz, h. Calcutta). Madras Peninsula (Wight et Heyne in Wallich n. 3326, h. Calcutta; G. Thomson, h. Calcutta). Ins. Zeylania (Thwaites C. P. n. 2942, h. Calcutta). Ins. Mascarene: Madagascar, Seychelles, Rodriguez, Comoro. Africa australis: Portus Natal (Grant, h. Calcutta). 159. C. 1scunos (Schlecht. Bot. Zeit. 1849, p. 99) ; tenuiter 198 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. stolonifera; capitulo congesto; nuce late obovoidea vel exacte oblonga. I have no specimen. Isaw at Kew Hooker f. et T. Thomson n. 25, and it appeared conspecific with the American specimens. Here I am only copying Boeck. in Linnea, xxxvi. p. 334. India Orientalis (Madras Peninsula). Mexico. Columbia. 160. C. Barpwrnit (Torrey in Ann. Lyceum New York, in. p. 270); umbella simplice; spiculis dense breviter spicatis aut fere faseiculatis, substellatis, linearibus, acutis, parum compressis, 7-floris; glumis ovatis, obtusis, multinerviis ; rhachille alis latis, insolubilibus.— Boeck. in Linnea, xxxvi. p. 857. Rhizoma lignescens, horizontalis; culmi parum distantes, basi globoso-bulbosi, 8 dm. longi. Folia angusta, viridia. Involucri bracteæ 6, usque ad 15 em. longs, divaricate. Umbelle radii 4-9, usque ad 6 cm. longi, tenues. Spice 10-18 mm. in diam., 5-20-spiculose. Spieule longe 7-8 mm., lat: 1 mm. Glume subremote. Nux cum § parte glume equilonga, late oblonga vel subellipsoidea, obtuse trigona, fere symmetrica, utrinque bre- vissime angustata, nigra; cellule extime quadrate, hyaline. Florida (Chapman, h. Calcutta). New Orleans (Drummond, h. Calcutta). New Jersey (Buckley, h. Caleutta). 161. C. sETIFOLIUS (Torrey, non D. Don); folis angustis cum culmo æquilongis; umbella simplice; spiculis in apice radiorum stellato-globosis, fusco-viridibus, 6-12-floris; glumis ovatis trun- eatis per totam fere latitudinem multinerviis ; nuce cum gluma subzquilonga, late oblonga aut subellipsoidea, trigona. C. Grayii, Torrey in Ann. Lyceum New York, iii. p. 260. C. filiculmis, var. Grayii, Boeck. in Linnea, xxxv. p. 609. New Jersey: Pines (A. Gray, h. Calcutta). In these excellent specimens, communicated and named by A. Gray, the wing of the rhachilla is very broad; and the plant appears to me exceedingly near C. Baldwinii, and very far re- moved from C. filiculmis, in which the wing of the rhachilla is exceedingly narrow. It is possible, indeed, that A. Gray has sent the wrong plant under the name C. setifolius ; but I think not. ltis not the business of the present article to review cri- tically American species; but I wish to point out that in this part of the series also the wing of the rhachilla has been probably ME. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 199 made of too much importance as a sectional character, and that C. filiculmis perhaps should be removed to this neighbourhood. = Umbella simplex, spicule pauciflore 1-2-(rarissime 3-4-) nucigerc. 162. C. ovvramrms (Torrey im Ann. Lyceum New York, iii. p. 278); foliis bracteisque longis; umbella simplice, radiis longis apice a spicis globosis densis terminatis; spiculis 3-4-floris 1-2- nucigeris.—-Boeck. in Linnea, xxxvi. p. 376, var. a excl. Mariscus ovularis, Vahl, Enum. ii. p. 374. Alabama; New Jersey; Florida (h. Calcutta). 163. C. RETROFRACTUS (Asa Gray, Man. ed. ii. p. 494, non Engelmann); undique pubescens; umbella simplice ; spiculis in apice radiorum longorum dense stellatis (inferioribus sspe retro- fractis), linearibus, acutis, 1-nucigeris.— Boeck. in Linnea, xxxvi. p. 365. Mariscus retrofractus, Vahl, Enum. ii. p. 375. Louisiana (Drummond, h. Calcutta). Sierra Leone (Vogel n. 41, h. Calcutta). 164. C. prGruMis; robustior; umbella simplice; spicis ovoideis aut breviter cylindricis, densis, breviter pedunculatis aut con- gestis; spiculis majuseulis, anguste lanceolatis, 3-6-floris, 2-3- nucigeris, flore infimo quam sit spieula multo breviore. Mariscus biglumis, Gertn. Fruct. i. p. 12, t. 2. fig. 8. M. paniceus, Kunth, Enum. ii. p. 119 pro parte (syn. Rottb. eacl.). Radix perennis, valida, fibrosa; stolones laterales breves. Culmi 3-5 dm. longi, acute trigoni. Folia plura, culmum sæpe superantia, lata 5-8 mm. Involueri bractes 3-8, usque ad 2-3 dm. longs. Spice 3-9, usque ad 100-florz, 1-3 cm. Iongse, densissimæ ; pedunculi 0-2 em. Spicule usque ad 15-18 mm. long, 6-flore, sepius 8-10 mm. longs, 3—4-florz. Glume ellip- tice, obtuse, parum compress, in dorso viridi-nervose, in late- ribus late enervie, pallid, rubro-maculate aut viridi-notate, marginibus ipsis in sicco incurvatis. Nux cum 3 parte glume equilonga, late oblonga, trigona, in una facie concava.—Sp. se- quentis fortasse pro var. insigni melius habenda. Zeylania (Thwaites C. P. n. 816, h. Calcutta). Assam (Jenkins, h. Calcutta). Ins. Andaman (Kurz, h. Calcutta). Sumatra: Lamponga (Forbes n. 1620, A. Calcutta). 200 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 165. C. UMBELLATUS (Benth. Fl. Hongk. p. 386, non Roxb.) ; viridis ; umbella simplice aut congesta ; spicis solitariis (rarissime 1-3 digitatis ebracteatis) elongato-cylindrieis aut brevioribus aut globosis, sspissime multispieulosis; spiculis l—4-floris, nucem unieam maturantibus. C. ovularis, Boeck. in Linnea, xxxvi. p. 337, var. a, non Torrey. Kyllinga umbellata, Linn. f. Suppl. p. 105; Rottb. Descr. et Ic. p. 15, t. 4. fig. 2; Roxb. Fl. Ind. i. p. 182. Mariscus umbellatus, Vahl, Enum. ii. p. 3/6 ; Kunth, Enum. ii. p. 118. M. polyphyllus, Steud. ! in Flora, 1842, p. 596, in Schimper n. 1124. Species polymorpha vel composita, per regiones calidiores Orbis Veteris late vulgata. a. typica (Rottb. Descr. et Ic. t. 4. fig. 2). Culmi cæspitosi, 2-4 dm. longi, sursum trigoni Radix perennis, valida fibrosa nigrescens, haud raro breviter rhizomatosa, lignescens, interdum (Heyne in Wallich h. n. 3433) longe repens, tenuis, lignosa, stolonibus longis nigris tenuibus squamosis additis. Folia plura longa, plana, in marginibus scabra, culmum sepe superantia. Involucri bractez 3-12, umbellam longe superautia. Umbelle radii 5-14, usque ad 5 em. longi. Spice longs 10-15 mm., late 7-8 mm., cylindrice, densissime, 40-60-spiculose. Spicule longe 4 mm., late 1 mm., oblique lanceolate, 4-glume; glume 2 ime parve, vacue; tertia hermaphrodita, quarta vacua (vel rarius mascula) tertiam non (aut vix) superans. Gluma fertilis ovata, subobtusa, 3-nervia, multistriata, viridis, fusco-spadiceo- maculata. Stamina 3; filamenta angusta,levia; anthere lineari- oblongz, mutiez vix apicatz. Stylus nuce multo brevior ; rami 3, e gluma breviter (ccnspieue) exserti. Nux cum $-3 parte glume squilonga, late oblonga vel subellipsoidea, utrinque brevissime angustata, trigona; cellule extime quadratze, persistentes. Per Indiam orientalem vulgaris: Dehra Doon (G. King, h. Calcutta); Madras (Wallich n. 3434, h. Calcutta) ; &c. In these typical forms the 3rd glume is ovate hermaphrodite, the 4th ligulate barren shorter than the 3rd, cutting away the distinction given by Bentham to separate Kyllinga from Cyperus (sect. Mariscus) But from this typical form there are grada- tions (by me indefinable) which carry us into the forms cylindro- stachys, cyperina ; in short, into all the other varieties. We have, first, examples in which the upper (4th) glume is a little longer than the 3rd, then usually male; after these come examples with the 4th glume § the length of the 8rd, often her- maphrodite (rarely perfecting a nut), and a fifth glume is present. MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. 201 The spieula becomes more slender, or finally acicular, and we arrive at forms named by some cyperologists C. cylindrostachys, by others C. cyperinus. Among the great diversity of forms I do not find one that seems to me to be conspecific with C. ovularis, Torrey ; and that is Mr. Bentham's opinion. Var. f. panicea; spicis parvis subsessilibus; spiculis parvis, anguste oblique lanceolatis, gluma mascula hermaphroditam parum superans. C. paniceus, Boeck. in Linnea, xxxvi. p. 361. Kyllinga panicea, Rottb. Descr. et Ic. p. 15, t. 4. fig. 1, non Mariscus paniceus, Kunth. Zeylania (Thwaites n. 2878, h. Calcutta). Madras Peninsula ( Heyne in Wallich nn. 3435, 3436, h. Cal- cutta); Palaverum Hill (Wight in Wallich n. 3435, h. Cal- cutta). Var. y. picta, Wallich MS. ; spicis fere sessilibus subglobosis, albidis aut glauco-purpurascentibus ; spiculis late oblongis, gluma mascula quam hermaphrodita vix longiore. Mariscus Wallichianus ? Kunth, Enum. ii. p. 117. Tranquebar (Rottler n. 648, h. Calcutta). Ins. Nicobar (Kurz, h. Calcutta). Var. à. laxata; rhizomate elongato, ramoso, tenui ; culmis elongatis, tenuibus, flaecidis ; spicis breviter pedunculatis, laxis, ovoideis, 1—5-spiculosis ; spiculis anguste lanceolatis, gluma her- maphrodita quam spicula multo breviore. Zeylania (Thwaites C. P. n. 817, h. Calcutta). Madras Peninsula (Heyne, h. Calcutta). . Var. e. cylindrostachys; spicis elongatis cy lindraeeis, siepe longius peduneulatis, undique patentim vel subdefloxim dense spiculigeris; spieulis anguste lineari-lanceolatis ; gluma erma- phrodita tertia quam superiores vix breviore aut multo breviore. Bengalia orientalis ( Griffith n. 6242, A. Kew). Nepaul (Wallich n. 3437 a, h. Culeutta). Tenasserim v. Andamans (Helfer n. 6243 | 1, h. Calcutta). Borneo (Barber, h. Calcutta). This vanity is defined only by the inflorescence : the structure of the spikelets varies from thatin C. umbellatus type very el to that in C. biglumis: the spikelets vary 1n length from 2 M millim. Those examples in which the 4th (and 5th) glumes mu overtop the 3rd are named C. cyperinus &e. P LINN. JOURN.— BOTANY, VOL. XXI. 202 MR. C. B. CLARKE ON INDIAN SPECIES OF CYPERUS. DESCRIPTIO TABULARUM. Tas. I. Fig. 1. Cyperus cephalotes, Vahl. Pistillum juvene. 5 2. Pistillum paullo maturius. 3. —— ——. Nux, cum 3 staminibus, a facie visa. 4, —— ——. Hujus sectio transversalis per basin nucis. 5. —— ——. Hujus sectio transversalis per basin pedicelli. 6. —— Nux a dorso visa. 7. —— platystylis, R. Br. Nux, cum 3 staminibus, a facie visa. 8. — Nux a dorso visa. 9, —— ——. Hujus sectio transversalis per basin nucis. j Tas. II. Fig. 10 & 10a. Cyperus pygmeus, Rottb. Nux, cum sectione transversali. 11 & lla. Isolepis Micheliana, Roem. et Sch. Nux, cum sectione transversali. 12. Cyperus alopecuroides, Rottb. Nux, cum sectione transversali. Spieulz rhachilla ampliata. 13. — dives, Delile. Nux, eum sectione transversali. Spiculs rhachilla ampliata. 14. longus, Linn. Rhizoma (squamis amotis). 15. —— esculentus, Linn. Ditto. 16. rotundus, Linn. Ditto. 17,18. Jjeminicus, Rottb. Ditto. 19. —— tegetum, Roxb. Ditto. Tas. ITI. Fig. 20, 21. Cyperus levigatus, Linn. Rhizoma (squamis amotis). | 22. —— scariosus, R. Br. Ditto. | 23, 24. Haspan, Linn. Ditto. 25. — flavidus, Retz. Ditto. | 26. filicinus, Vahl. Umbelle radius unicus. | 27. polystachyus, Rottb. Ditto. | 28. —— lucidulus, Klein. Spicule rhachilla. | 29. —— Monti, Linn. Ditto. Tas. IV. | Fig. 30. Cyperus puncticulatus, Vahl. Spicule rhachilla. 31. —— auricomus, Sieber. Spicule rhachilla. 32. platyphyllus, Roem. et Sch. Anthera. 33, —— levigatus, Linn. Ditto. 34. Haspan, Linn. Ditto. 35-39. Cyperi sp. Nucis superficies ampliata, 41. Cyperus flavescens, Linn. Nux. 40. —— Hujus superficies magis ampliata. 42, 44. Pyerei sp. Nux, cum sectione transversali. 43. Juncelli sp. Ditto, MR. J. BALL ON THE FLORA OF NORTH PATAGONIA. 203 Contributions to the Flora of North Patagonia and the adjoining Territory. By Jons Br, F.R.S., M.R.I.A., F.L.S., &c. [Read February 21, 1884.] I RETURNED last year from Brazil to England in company with M. Georges Claraz, a Swiss gentleman who had passed several years in the Argentine territory, chiefly at Bahia Blanca, and who during his residence at that place made frequent excursions through Northern Patagonia as far as the river Chubat, and even beyond that river. Having a fairly extensive knowledge of the principles of physies and biology, and being gifted with keen and accurate powers of observation, M. Claraz has accu- mulated a large store of valuable information respecting a region still very imperfectly known, which will, I trust, be socn given to the world. Along with collections in other branches of natural history, M. Claraz preserved specimens of most of the plants observed during his residence in South America. The greater part of these were sent some years ago to Switzer- land, but have unfortunately been lost or mislaid. He retained a smaller collection, chiefly from the neighbourhood of Bahia Blanca and the adjoining territory of North Patagonia, which he was good enough to send to me after his return to Europe. It included about 190 species from the Patagonian region, and a much smaller set of about 60 species from the province of Entrerios. The interest of the collection is much enhanced by the notes added by M. Claraz respecting many of the species, their properties and uses, with the vernacular names of those familiar to the Indian tribes. It is well known that the coast-region of North Patagonia was visited by Charles Darwin in 1833. He collected a good many plants, chiefly near Bahia Blanca, specimens of which are preserved in the Kew Herbarium, and has left an admirable description of the prominent physical characteristics of the Pata- gonian region. But neither Darwin nor any of the botanists who have since been able to visit the country have been able to penetrate far into the interior, or to remain long enough to acquire any but a very partial acquaintance with the flora. This is evidenced by the very meagre account of the vegetation given in Grisebach’s great work on the Vegetation of the Earth, and by the fact that in the very useful volume on the Argentine Republic, edited by Mr. R. Napp, the late Professor Lorentz, LINN. JOURN.— BOTANY, VOL XXI. Q 204 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA in the chapter treating of the Vegetation of the Argentine ter- ritory, was forced, when attempting to give an account of the Patagonian flora, to rely altogether upon information supplied by two settlers, MM. Heusser and Claraz, who were not botanists, and were therefore unable to identify with species known to science most of the plants referred to by them under local names. I have therefore thought it desirable to prepare a list of the species received from M. Claraz, as a contribution to the existing scanty materials towards a knowledge of the Patagonian flora, adding a few preliminary remarks on its general characteristics. The political boundaries of Patagonia have been altered at various times; but we may safely assign as its natural physical limit to the north the valley of the Rio Cclorado, which reaches the Atlantic about fifty miles south of Bahia Blanca. In his phyto-geographical map of the Argentine territory, Prof. Lorentz included a tract north of that river extending to the Sauce, whose estuary forms the port of Bahia Blanca; while for political purposes a considerable part of the country lying between the rivers Colorado and Negro has been annexed to the province of Buenos Ayres. Three considerable rivers carry the drainage of the Cordillera to the Atlantic across Patagonia, but receive only few and inconsiderable affluents—the Rio Colorado, forming the northern boundary, flowing S.E. and having its mouth about 40° S. lat.; the Rio Negro, uniting two considerable branches which between them drain the eastern face of the Araucanian Cordillera for a distance of fully 200 miles, and flowing E.S.E. into the Atlantic about 41° S.; and the Chubat, flowing a little S. of E. to its mouth about 433° S. The Chubat may be looked upon as the boundary between North and South Patagonia, dividing it into two territories of nearly equal extent. Of these the northern has been partially, though imperfectly, explored ; and annually receives a gradually increasing number of European colonists ; while Southern Patagonia, in spite of the remarkable journey of Lieutenant Musters, continues to be one of the least- known portions of the earth, and, excluding the northern shores of the Magellan Strait which are subject to quite different physical conditions, contains in a territory as large as Spain but one petty trading-port—that of Santa Cruz, about lat. 50? S. ; while the indigenous Indian population, estimated by Musters at 3000, is believed to be now reduced to less than half that number. Speaking in general terms, Patagonia may be said to consist of AND THE ADJOINING TERRITORY. 205 a plateau region extending from the base of the Cordillera to the Atlantic coast, the whole of which has been upraised within a very recent geological period, and in which the rivers and minor streams have excavated valleys varying much in depth and breadth. The most remarkable feature in the flora of this region is its extreme poverty. Its components include the plants growing on the dry stony plateaux and those of the comparatively moist and sheltered valleys; but, putting together all that has been collected and published in Europe, I doubt whether more than 300 indigenous species can be said to be certainly known to grow south of the Rio Colorado. No doubt this number will be largely increased whenever naturalists are able to reach the country at the eastern base of the Cordillera, where Lt. Musters observed many plants not seen by him elsewhere. With regard, however, to the region now comparatively known, it is certain that the extreme poverty of the flora of such an extensive con- tinental area is a fact quite exceptional. If this be true as to North Patagonia, the case is much stronger as to the southern part of the territory. In the neighbourhood of Santa Cruz Dr. C. Berg was able to collect only 60 species, including in that number a few Cryptogams; and he notes the rarity of grasses in that district, while in M. Claraz’s collection I find 24 species of indigenous grasses, besides 6 others, probably introduced by man. The causes of the poverty of the Patagonian flora do not seem to me to have been adequately explained. Prof. Lorentz * is disposed to attach most importance to the uniformity and steri- lity of the soil and the rudeness of the climate. The soil of the plateau is no doubt both`very uniform and very sterile, but similar tracts in other parts of the world support a very varied vegetation, and Engler has assigned good reasons for believing that dry soils are in general favourable to the development of new vegetable forms. Further, it may be remarked that the soil of the valleys must exhibit a sufficient degree of variety of moisture, of constituents, and of exposition to favour the develop- ment of many species not yet established there. Still less can I admit the severity of the climate as an expla- nation of the poverty of the flora. So far as I know, we have no continuous observations from any place in Patagonia proper ; but we cannot suppose the climate of the northern districts to * See ‘The Argentine Republic,’ by Richard Napp (Buenos Ayres, 1876), Chapter vii. by the late Prof. Lorentz. 2 Q 206 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA differ much from that of Bahia Blanca, where we find the fol- lowing temperatures as the result of 20 years' observations :— o Three summer months ............ 75:6 Fahr. Three winter months ............ 462 , Mean of the year ................-- 594 , This at once shows that the climate is eminently of a tem- perate character; and this conclusion is fortified by the fact that in 20 years the mean maximum temperature of the year was 10097 F.* and the mean minimum 26*8 F. A very similar climate is to be found in Europe in the corresponding latitude about 100 miles east from the coast of Portugal towards the Spanish frontier. If further proof were requisite, it is to be found in the fact that wheat is very productive, and the vine thrives in the valley of the Rio Negro. As already remarked, the dryness of the climate does not aecount for the poverty of the flora. As is well known, the opposite coasts of extratropical South America present exactly opposite conditions as regards the distribution of moisture. In travelling southwards on the west coast you pass from the rainless zone of Peru and the extreme dryness of Northern Chili to the climate of Central Chili, where at Valparaiso the annual rainfall is only 13:6 inches, yet supports a varied flora, including a good many indigenous trees whose structure has been adapted to the climatic conditions. The rainfall increases very rapidly along the southern coast of Chili till it reaches a maximum, probably about 42? S. lat., of from 130 to 140 inches a year. On the Atlantie coast you find in South Brazil, about 26? S. lat., an annual rainfall of about 90 inches, which at Monte Video is already reduced to about one half, and at Buenos Ayres, less exposed to the direct influence of the Atlantic, is not more than 35 inches. But this, it will be remarked, is nearly three times the fall at Valparaiso, in the same latitude, on the west coast. At the hilly station of Tandil, not far from the sea, though further south by three degrees, the rainfall is the same as at Buenos Ayres; but at Bahia Blanca we find only a fraction less than 20 inches, and Dr. Hann gives a return for one year from the banks of the Chubat showing a rainfall of 17 inches. There * The comparatively high maximum temperature given for Bahia Blanca may depend on local causes or some defect in the protection of the thermo- meter from radiation. At Buenos Ayres, more than 4° nearer the equator, the mean yearly maximum is 94° F. AND THE ADJOINING TERRITORY. 207 is a further point to be noted—that on the west coast the rainfall is almost exclusively confined to winter, whereas in Patagonia it is spread pretty uniformly over the seven warmer months of the year from October to April, not more than five inches falling in the five colder months, from May to September, a condition evidently favourable to tree-vegetation. The true explanation, in my opinion, of the exceptional poverty of the Patagonian flora is to be sought in the direction long ago indicated by Charles Darwin, when, in discussing the absence of tree-vegetation from the pampas, he remarks that in that region, recently raised from the sea, trees are absent, not because they cannot grow and thrive, but because the only country from which they could have been derived—tropical and subtropical South America—could not supply species organized to suit the soil aud climate. So it happened in Patagonia— raised from the sea during the latest geological period, and bounded to the west by a great mountain-range mainly clothed with an Alpine flora requiring the protection of snow in winter, and to the north by a warm temperate region whose flora is mainly of modified subtropical origin—the only plants that could occupy the newly formed region were the comparatively few species which, though developed under very different conditions, were sufficiently tolerant of change to adapt themselves to the new environment. ‘The flora is poor, not because the land cannot support a richer one, but because the only regions from which a large population could be derived are inhabited by races unfit for emigration. The rapidity with which many introduced species have spread in this part of South America is perhaps to be accounted for less by any special fitness of the immigrant species, than by the fact that the ground is to a great extent unoccupied. Doubtless, if no such interference had taken place, and the operation were left to the slow action of natural causes, a gradual increase in the vegetable population would eome about. l Fresh species of Andean plants would gradually become modified to suit the climate of the plain (perhaps one such recent instance is supplied in Boopis laciniata of the following list) ; sian di slowly new varieties would have been developed among the indi- genous plants, from which, by natural selection, new species would have been formed, No doubt these causes have been in aetion during the short time that has elapsed since Patagonia has existed as part of the continent; but the time has been far 208 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA too short to allow of the development of a rich and varied flora. We are apt, I think, to underrate the extreme slowness of the operation of the agencies that modify the forms of vegetation and the fact that change in arboreal vegetation must, other things being the same, proceed much more slowly than with herbaceous, especially annual, plants. How many of the plants found in fossil Miocene deposits, enormously more ancient than the commencement of the Patagonian flora, are more than slightly modified forms of existing species ? Although the collection at my disposal gives a very incomplete view of the flora of North Patagonia, there may be a little interest in comparing the proportion borne by the chief natural orders to the whole known flora in this as compared with the flora of the Argentine region as made known by Grisebach's * Symbolæ.’ Table showing the Proportional Number of Species belonging to the chief Natural Orders in the North-Patagonian and Argentine Floras. North Patagonia. Argentine. Natural Orders. No. of |Percent.on| No. of |Percent. on species. [whole flora.| species. [whole flora. Cruciferz ............... 4 2:0 21 0:9 Caryophyllem ......... 3 l5 30 1:35 Malvacez ............... 0 0 51 2:25 Geraniacesm ............ 5 2:6 31 1-4 Leguminosee ........ ... 8 41 174 T Rosaeem ..............- 7 36 14 06 Lythrariem ............ 1 0:5 25 l1 Loasacem ............... 4 2:0 11 0:5 Umbelliferze ............ 4 2:0 98 1:7 Rubiaceæ ............... 1 6:5 39 17 Composite |. ............ 36 18°65 377 166 Asclepiader ............ 1 0:5 38 17 Convolvulacez ......... 0 0 44 1:9 Solanaces. ............. 9 4:6 95 42 Serophularinem ...... 5 2:6 33 15 Labiatg.................. 1 0:5 33 15 Verbenacem ............ 7 3:6 43 19 Amarantacez ......... 7 9:6 37 1:65 Chenopodiaceze......... 8 41 19 0:84 Polygonaces............ 5 2:6 19 0-84 Euphorbiacee ......... 0 0 76 34 Urticacez ............... 1 0:5 22 i0 Irideæ .................. 2 1:0 22 1:0 Liliaeeg.................. 1 0:5 26 115 Cyperacege............... 5 2:6 66 2:9 Graminez FEN 30 155 187 8:25 Filices ........ ......... 6 31 66 2:9 AND THE ADJOINING TERRITORY. 209 The latter naturally includes a large proportion of species belonging to subtropical types. Both lists include the intro- duced as well as the indigenous species. Seven species enumerated in the following list, but which pro- bably do not extend to Patagonia, are excluded from this table. The most noticeable features in the comparison are the much larger proportion of Rosacee and Graminew in the Patagonian flora, with relatively few Leguminose, and a complete absence of Malvacew, Convolvulacee, and Euphorbiaces, of which orders the Argentine flora contains 171 species, or more than 73 per cent. Besides the above, the Argentine flora, as might be expected, includes representatives of numerous natural orders (with about 400 species) not yet known to be represented in Patagonia. The small collection of plants from Entrerios sent to me by M. Claraz contains only about 60 species, several of which are imperfect fragments, and of course cannot be considered as representative of the comparatively rich and moderately well- known flora of that region. Nevertheless, I find the following eight species not enumerated by Grisebach from that province, and only two of which he appears to have received from any part of the Argentine territory :— Pavonia glechomoides, A. St. Hil., =P. cymbalaria, DC. Con- cepcion del Uruguay (No. 276, G. C.). Cassia cespitosa, Lam. Entrerios (No. 263, G. C). Cuphea glutinosa, Cham. & Schlecht. Near S. José (No. 260, G. C.). ` . Tabernemontana affinis, var.? Banks of the Uruguay (No. 251, G. C.). Grisebach does not include any species of this genus. Convolvulus dissectus, Cav., var. angustifolia. Forests and islands of the Uruguay (No. 287, G. C.). Nicotiana acutifalia, A. St. Hil. Entrerios (No. 262, G. C.). Received by Grisebach from the province of Cordoba. » Nectandra amara, var. australis, Meisn. Entrerios (N o. 2: , G.C.). Received by Grisebach from Oran (the subtropical region in the extreme north of Argentine territory). Andropogon virginicus, L. Entrerios (No. 130, G. C.). eh I may here remark that many of the statements made by travellers which have been adopted in works of authority, M specting the vegetation of South America, are open to muc 210 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA question, and illustrate the tendency (from which even men of science are not free) to draw large inferences from a slender foundation of observed facts. What has been stated in general terms respecting the flora of Patagonia, if supposed to apply to the whole region, is much as if one should discuss and describe the flora of France who had landed at Dieppe and Nantes and Bordeaux, and travelled at each place a few miles into the interior. Thus Grisebach has borrowed from Mr. Page the statement that the plains of North Patagonia are so devoid of tree-vegetation that a single stunted tree of Acacia* near the Rio Negro is worshipped by the natives as a sacred object. But I learn from M. Claraz that groups of small trees or shrubs tall enough to conceal a man on horseback, especially those of Jodina rhombifolia, are dotted at rather wide intervals ; while along the valleys Salix Humboldtiana and other small trees are not unfre- quent, and in the valley of the Upper Limay the apple-tree, intro- duced from Europe, forms considerable groves. Again I may remark, with reference to the valley of the Uru- guay, that the assertion, also accepted by Grisebach, that the trees on the banks are all low, not exceeding 10 metres, or 33 teet, may be true at some points, but certainly not as a general state- ment. At many places, both along the shores and on the nume- rous islands, the trees range from 50 to 60 feet in height ; oneof these is Luhea divaricata, and several appear to be Leguminose, which, however, I was not able to approach. Again, with regard to the asserted poverty of the flora of Uruguay and Entrerios, I venture to entertain much douUt. Much of the territory of both provinces is flat and uniform in physieal conditions, but about 1000 species are known from Entrerios, to which, out of a small set of 60 species, as above mentioned, 8 additions have to be made, showing how incomplete is our present knowledge. The sur- face and the soil of Uruguay, on the opposite bank of the same river, are much more varied, besides which there is a long stretch of sea-coast with moderately high land. If M. de St. Hilaire, while travelling in the best season, was unable to collect more than 500 species, I conclude that his itinerary was not well chosen for botanical purposes. To judge from two short excur- sions made in the depth of winter (end of June) from Monte- video, and from Paisandu on the Uruguay, I should have formed a very different conclusion. Near the latter place I found the * Perhaps the same spoken of by Darwin. AND THE ADJOINING TERRITORY. : 211 open country interspersed with a considerable variety of shrubs, in great part evergreen, reminding one of the aspect of many parts of the Mediterranean region. Lastly, with regard to the asserted paucity of subtropical forms in the regions bordering on the lower course of the Paraná and the Uruguay, it seems to me that, as compared with districts in ihe same latitude north of the equator, whether in North America or North Africa, the statement is quite opposed to the facts. Of the. tree-vegetation nearly all are subtropical forms. Of the bushes the majority seem to belong to types characteristic of temperate South America, most of them common to the east and west sides of the Andes, while the herbaceous vegetation exhibits about an equal proportion of elements characteristic of the sub- tropical and the South-temperate American floras. With refer- ence to this subject much information is to be found in a paper by Sir Charles F. Bunbury, in the 21st volume of the * Trans- actions of the Linnean Society,’ lately reprinted, with additions, for private circulation, along with other valuable papers by the same author. It may be well to state that all the information contained in the following list respecting the habitats, the local names, and the uses of the plants enumerated is derived from M. Claraz. For the botanical matter alone I am responsible. List of Plants collected in North Patagonia and South Part of the Province of Buenos Ayres by M. Georges Claraz. The species marked with an asterisk probably do not extend to Patagonia. BERBERIDES. BERBERIS HETEROPHYLLA, Juss.? A bush 4 to 5 feet high, common in Patagonia. Its northern limit is about halfway between the Rio Negro and the Chubat (No. 147, G. C.). The Struthious bird, Rhea Darwinii, Gould, is very fond of the berries, but these give a disagreeable taste to the flesh. l The Patagonian Indian name for the plant 18 Gayaukhia, and for the berries Khalgo. The specimen is a small incomplete fragment, and the determination uncertain. The plant may be undescribed. PAPAVERACES. FUMARIA cAPREOLATA, L.,var. Common about Buenos Ayres and further south ; doubtless introduced from Europe (No. 157, G. C). 212 MR. J. BALL ON THE FLORA OF NORTH. PATAGONIA CRUCIFERE. VxsicARIA ANDICOLA, Gill., =V. montevidensis, Hichl. in FI. Bras. Onthe plateaux near Bahia Blanca and in North Pata- gonia (No. 132, G. C.). The name given by Gillies was changed by Eichler because the plant is mainly an inhabitant of the low country in the Argentine and Uruguay territories. But it also extends to the eastern slopes of the Andes, and mere inappro- priateness, apart from positive error, is not a sufficient ground for changing an established name. Brassica Rapa, L., var. CAMPESTRIS, = B. campestris, L. Intro- duced from Europe, and now widely spread from the warm low regions of Entrerios to the barren banks of the Chubat in Mid- Patagonia (No. 186, G. C.). It sometimes produces tough and stringy turnips, and is very troublesome to farmers. From the Spanish name Nabo the Araucanian Indians have formed for the plant a native name, Napur. LEPIDIUM BONARIENSE, L.,—L. pubescens, Desv. Very com- mon throughout the province of Buenos Ayres (No. 143, G. C.). Grisebach (Pl. Lorentz. p. 25) identifies the L. pubescens of Desvaux with the plant generally taken by authors as L. bona- riense of Linnsus, but impliedly holds that the plant of Linnwus is a different species. I do not know the grounds for this con- clusion, and am disposed to believe that there is one somewhat variable species with a wide range from Mexico to temperate South America, which includes L. bonariense, L., L. pubescens, Desv., L. bipinnatifidum, Desv., and L. Chichicara, Desv. Lerum? Sierra de la Ventana, near Bahia Blanca (No. 64, G. C.) Specimen too immature. Doubtless a form of ZL. bonariense, L. POLYGALER. PoLYGALA SPINESCENS, Gill, var.? ASPALATHOIDES, nobis. Grows about Bahia Blanca, where it is used for making brooms (No. 184, G. C. As this is a well-marked variety, if not à distinct species, I subjoin a brief description. Fruticulus ramosissimus, ramis erectis, novellis herbaceis foliaceis, annotinis rigidis sublignosis aphyllis; foliis parvis, lineari-oblongis apice rotundatis, subglaberrimis ; floribus in racemum laxum foliaceum dispositis, parvis lilacinis ; petalis lateralibus ope tubi staminei cum carina coherentibus. Capsulain AND THE ADJOINING TERRITORY. 213 non vidi.— P. spinescens, Gill. in Hook. Bot. Misc. iii. 146, differt statura minore, ramulis flexuosis, et (an semper P) aphyllis. Species est parum cognita, cujus exemplar unicum incompletum exstat in Herbario Kewensi, a Gillies ad Salto de las Aguas prope Mendoza lectum. I have not seen Acanthocladus microphyllus, Griseb., to which that author doubtfully refers P. spinescens, Gill. This appears to be a true Polygala. CARYOPHYLLESX. SILENE ANTIRRHINA, L., var. PTERONEURA, nob. Frequent in the valleys near Bahia Blanca (No. 116, G. C.). Silene antirrhina, L., is widely spread throughout both North and South America. A large series of specimens in Kew Herbarium, including one gathered at Bahia Blanea by Darwin, exhibits but trifling varie- ties. The plant sent by M. Claraz differs remarkably in having the veins of the calyx expanded into prominent herbaceous corru- gated ridges. The calyx-teeth are also longer, and the solitary capsule ovoid and not globose. Is this an instance in which a new variety has been developed within the last half-century ? CERASTIUM SEMIDECANDRUM, L. Bahia Blanea,by watercourses (No. 108, G. C.). CERASTIUM | CoMMERSONIANUM, Ser., = C. chilense, Baill. Bahia Blanca, by watercourses (No. 111, G. C.). Distinguished from O. arvense, L., mainly by the foliaceous bracts, with the edges not scarious. HYPERICINEEX. Hypericum connatum, Lam. Sierra de la Ventana, and North Patagonia (No. 71, G. C.). ZYGOPHYLLEX. LARREA NITIDA, Cav. Valley of the Limay and elsewhere in North Patagonia (No. 87, G. C.). GERANIACEE. GERANIUM PATAGONICUM, Hook. fil. Common about Bahia Blanca (No. 48, G. C.). The local Spanish name for this, as well as the next very different plant, is Alfilerillo. To judge from a specimen in Kew Herbarium, G. intermedium, Bert., is not distinct from this. 214 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA Eropium CICUTARIUM, L'Hér. Widely spread from Buenos Ayres southward through Patagonia. It is called Alfilerillo, and is considered useful for cattie, because it springs up afresh after each fall of rain (No. 106, G. C.). This species, doubtless intro- duced from Europe, has spread throughout nearly the whole of South America. I have seen it in Peru from the coast up to 3700 metres in the Andes, and equally common in Chili; and Grisebach records it from all parts of the Argentine territory. Like many other introduced plants, it owes its diffusion much more to auimals than to the direct agency of man. *TnoPXOLUM PENTAPHYLLUM, Lam. Inawood near La Mag- dalena, not far from the shores of La Plata (No. 232, G. C.). This species is, I believe, common in Uruguay and Entrerios, but does not spread far southwards. It is sometimes cultivated for ornament. OXALIS ? Very common at Bahia Blanca and in North Patagonia, springing up after the rains in spring and autumn, and very variable in the colour of the flowers: the form with yellow flowers (No. 57, G. C.) ; with violet flowers (No. 58, G. C.) ; with pale-blue flowers (No. 100, G. C.). This plantis called by the Spaniards Vinagrillo, and Tschilki by the Araucanian Indians, who eat the tuberous root. There isa specimen at Kew collected by Darwin at Bahia Blanca, but I cannot identify it with any of the described species. So many South-American species of this genus have been published of late years, that I think it imprudent to giveaname tothis. If undescribed, it may properly be named O. Darwinii. Oxaris ManrIANA, Zucc. ? From the Isla Verde, near Bahia Blanca, and elsewhere. It is widely spread northwards to Entrerios, and southward through North Patagonia (Nos. 104 & 159, G. C.). This has tomentose leaves, blue flowers, and a tuberous root. Though very different from the last, it has the same name both among the Spaniards and Indians, and the latter also eat the roots. If the name above given is correct, this extends also to Brazil. I should be more disposed to identify it with O. floribunda, Lehm., but M. Claraz’s specimen was named by the late Professor Lorentz. OXALIS AMARA, St. Hil. Like the last species, this is widely spread from Entrerios through Buenos Ayres to North Patagonia AND THE ADJOINING TERRITORY. 215 (No. 103, G. C.) Sometimes improperly called Vinagrillo. This species has not a tuberous root. My specimen is very incomplete. CELASTRINE Æ. Mayrenvs CnHiLENsIS, DC. Valleys of the Upper Limay and Chubat, in Patagonia (No. 59, G. C.). A small tree, rather variable in habit. RHAMNES. DiscanrA Lonaispina, Hook. 4 Arn. Common in the west and south of the province of Buenos Ayres (No. 83, G. C.). The thick roots serve as fuel, and the flowers are sold in Buenos Ayres. Further south, in Patagonia, the flowers of this (perhaps a different?) species give out an odious smell. D. Sebrifuga, Mart., is scarcely a distinct species. ANACARDIACER. Duvava precox, Griseb. A common and characteristic shrub throughout North Patagonia (four specimens, Nos. 150-153, G. C.). This forms a bush 5 or 6 feet in height. The Patagonian Indians call the plant Reréle and the fruit Yssgitz. The Arau- canjan name is Mifschi. There is another species, called Ahnec by the Patagonians, a small tree, not more than 15 feet in height, with broader leaves and larger fruits, from which the Indians pre- pare a fermented drink. D. precos appears to be widely spread and very variable; I do not think that D. fasciculata, Griseb., can be separated from it. l LEGUMINOSE. Mepicago LuPULINA, L.? Very young, but undoubted. Bahia Blanca (No. 105, G. C.). Introduced from Europe with other species of the same genus. Grycrrruiza ASTRAGALINA, Hook. § Arn. Appears to have descended from the Cordillera along the valleys of the Chubat, Rio Negro, and Rio Colorado (No. 162, G. C.). This, which has blue flowers and glutinous leaves, flowers in spring (October to December). The Spanish name is Curuzu ; the Araucanian name Milpi. A decoction of the resinous leaves is administered to women after childbirth. Avesta GnrsEA, Hook. fil. Bahia Blanca and south of the province of Buenos Ayres. Common, especially in sandy ground 216 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA (No. 109, G. C.) A. incana, Vogel, seems to be too near to this species. AprsMIA PENDULA, DC. Bahia Blanca; Cabo S. Antonio; North Patagonia (No. 228, G. C.). This is a yellow-flowered species. The specimen represents a small form. ApEsMIA ——? Bahia Blanca (No. 110, G. C.). The speci- men is too imperfect for determination, but is apparently the same as an unnamed specimen in Kew Herbarium from Mal- donado, collected by Lieut. Carr. ADESMIA P Valley of the Rio Colorado. Collected in March (No. 206, G. C.). The specimen (wanting leaves and fruit) is too imperfect for determination, but appears to be near A. longipes. *Ervem HIRSUTUM, L. In a wood at Ajó (No. 169, G. C.). Doubtless introduced from Europe. Not mentioned by C. Berg in his * Enumeracion de las Plantas Europeas que se hallan como silvestres en la provincia de Buenos Ayres y en Patagonia.’ Ruyncosta Senna, Gill., var. foliolis lanceolatis Sierra de la Ventana, near Bahia Blanca (No. 80, G. C.). Rayncosta SENNA, Gill., var. foliolis late ovatis. Bahia Blanca (No. 189, G. C). This species,originally described from Chilian specimens, appears to be widely spread from Texas and Mexico to the borders of Patagonia. Ido not think that R. texana, Torr. & Gr., can be separated from it. It is an insignificant climber, with small yellowish flowers. Rosace®. Mareyricarpus SETOSUS, Ruiz § Pav. Bahia Blanca (No. 66,G.C.). Called by the Spaniards “ Yerba de la perdiz,” which is merely a translation of the Araucanian name Silio lahuen, Silio meaning partridge, aud lahuen remedy. The Tinamous (Nothura maculosa, Temm.), or so-called partridges, of South America, are used to eat the fruit. This species was collected at Bahia Blanca by Darwin. MARGYRICARPUS -——? Sierra de la Ventana and North Pata- gonia (No. 67, G. C.). This is probably a variety of M. setosus, but is prostrate and much more robust, bearing to it much the same relation that Juniperus nana does to slender forms of Juni- perus communis. AND TITE ADJOINING TERRITORY. 217 MARGYRICARPUS ? Valley of the Upper Limay and else- where in North Patagonia (No. 65, G. C.). This is too imperfect for description ; evidently allied to M. setosus, but the leaves are different. The rhachis is tomentose, rigid, subspinescent, and the leaflets shorter and more equal in size. It is probably M. microphyllus of Niederlein, who has described several new Pata- gonian plants in the * Monatschrift zur Beförderung des Garten- baus in den Preussischen Staaten’ for 1881. Maneyricarrus CLARAZII, n. sp. A native of Middle Pata- gonia, extending from about lat. 41? S. to the Chubat, and south- ward from that river (No. 155, G. C.). This is a small plant allied to M. alatus, Gill. in Hook. Miscell. iii. p. 385, but certainly distinct. I subjoin brief diagnostic characters. A M. alato, Gill, differt foliolis subtus lanatis, stipulis (basi petioli adnatis) fimbriato-ciliatis, calyce florifero sericeo-piloso, fructifero in alas duas acutas expanso tuberculis acutis 2-3 inter- mediis minuto. The fruit in this plant is very singular. In JL alatus we find 4 or 5 membranous wings, recalling the appearance of the fruit of Laserpitium. In this the two rigid, pointed, wing- shaped expansions are as long as twice the diameter of the fruit, and the stiff, pointed, intermediate tubercles are more than half that diameter in length. ACXNA SPLENDENS, Hook. & Arn. Upper Limay, North Pata- gonia (No. 90, G. C.). Apparently brought down by the streams from the Andes. Aca#Na MYRIOPHYLLA, Lindl. Bahia Blanca, especially com- mon about the bizcacheros, or bizcacha* warrens (No. 95, G. C.). Pyrus marus, L., var. Valley of the Upper Limay (No. 54, G.C.). The apple is said to have been introduced into this region by the Spanish missionaries. It has thriven wonderfully, and in the interior, especially between 40° S. and 40°30’, it forms extensive groves and even small forests. It has developed two varieties ; the fruit of one is sweet, of the other somewhat tart, but not uneatable. My specimen has the leaves duplicato. serrate, velvety on the under surface. SAXIFRAGE.E. l RIBES MAGELLANICUM, Poir. Valley of the Upper pinay i jern à (No. 91, G. C.) I have seen no specimens from so norther locality. * The “ Viseacha," Lagostomus trichodactylus, Brcokes, of zoologists. 218 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA CRASSULACES. Titnxa muscosa, L., var, =T. minima, Miers. South of pro- vince of Buenos Ayres and North Patagonia (No. 156, G. C.). This was gathered by Darwin at Bahia Blanca. HALORAGEJE.. MXYRIOPHYLLUM ? Streams at Bahia Blanca (No. 123, G. C.). This has neither flower nor fruit, but is almost certainly M. proserpinacoides. Yt also grows in Entrerios (No. 122, G. C.). LYTHRARIEX. CUPHEA SPICATA, Oav., var. RACEMOSA, Koehne? Marshes of the Naporta Chico, Bahia Blanca (No. 117, G. C.). I do not feel certain of the identity of this with the Uruguay plant of Koehne, but it is certainly very near to C. spicata. ONAGBARIER. CENOTHERA MENDOCINENSIS, Gill. Bahia Blanca and North Patagonia (No. 137, G. C.). It is called Balsamo by the Spaniards, and by the Indians Inuwlei, and is used for application to wounds. This, which was collected by Darwin in Patagonia, differs from Œ. odorata, Jacq., in being altogether more glabrous, and having the leaves more attenuated at the base, but I doubt whether it should be ranked as more than a variety of that species. It is doubtful whether Œ. mollissima, L., should not be included under the same specific type. LOASACER. MENTZELIA ALBESCENS, Griseb., = Bartonia albescens, Gill. North Patagonia (No. 92, G. C.). This species is widely spread throughout temperate South America. Loasa PROSTRATA, Gill.? Sierra de la Ventana, near Bahia Blanca (Nos. 61 & 107, G. C.). Iam somewhat doubtful as to this. The species of Loasa are very variable, and their limits ill- defined. BLUMENBACHIA MULTIFIDA,— Loasa multifida, C. Gay. In woods at Ajó, near Cabo S. Antonio (No. 167, G. C.). Called Amor seco by the Spaniards. GRAMMATOCARPUS VOLUBILIS, Presl. Bahia Blanca and North Patagonia (No. 176, G. C.). Araucanian name Fuéo. Gathered by Darwin at Bahia Blanca. AND THE ADJOINING TERRITORY. 219 FricorpEx. SEsUVIUM PORTULACASTRUM, Z. Salt-marshes of the Rio Colorado (Nv. 182, Œ. C.). Called by the natives Gume or Jume, but that is a general name for sueculent herbaceous plants. UMBELLIFER &. AZORELLA GLEBARIA?, =Bolax glebaria, D'Urv. Mid-Pata- gonia, northern limit about lat. 41? S., and extending southward beyond the river Chubat (No. 149, G. C.). I have but three minute fragments (flowering rosettes) of this plant, each about the size of a pea, but I feel sure that they belong to A. glebaria ora near ally. The Indian names for the plant are Kethdla and Gethenn. They use the resinous root as a masticatory, and also roast and grind it into a sort of flour. EnvNaruM PANICULATUM, Lam. Bahia Blanca, thence extend- ing northward through Buenos Ayres and Entrerios (No. 85, G. C.). This, as well as two other large Eryngia of similar habit, is called Shetd by the Araucanians. It is much less common than it formerly was. It seems as though it were driven out by the European thistles. Enxwsaivw ——? Bahia Blanca (No. 50, G. C.) ; Entrerios (No. 52, G. C.). The specimens are too immature for determina- tion. They agree best with the Chilian E. humifusum. The plant is sometimes used for dyeing wool; and M. Claraz thinks that it may have been introduced. But it has, as he says, an Indian name, Lief huintschu. APIUM RANUNCULIFOLIUM, DO. Bahia Blanca and North Patagonia (No. 236, G. C.). Called by the Spaniards Apio a Maron. This is a wide-spread and variable species, but quite distinct from the A. australe, which is merely a southern form of the cosmopolitan A. graveolens, and which, according to M. C'araz, also grows on the coast of Patagonia. RUBIACER. Gantum PUSILLUM, Endl., non Sm. Bahia Blanca and b orth Patagonia (No. 115, Œ. C.). Called by the Indians Re fen oF Relvun. The roots are used to produce a red dye for wool. This is scarcely distinct from G. corymbosum, Ruiz & Pav. R LINN. JOURN.—BOTANY, VOL. XXI. 220 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA DIPSACER. DIPSACUS SYLVESTRIS, L. Naporta Grande, near Bahia Blanca (No. 39, Q. C.). Introduced from Europe, and has now spread widely. It is found on the banks of streams in the Sierra de la Ventana and Sierra de Tandil, in the valleys of the Rio Negro and Rio Colorado, and as far as Valeheta in Patagonia. Called by the Spaniards Cardo de la Sierra. This is an instance of an introduced plant extending its area very rapidly. It appears to be of very recent introduction, and is not mentioned by Grise- bach or by Berg. CALYCERE. Booris ANTHEMOIDES, Juss. Bahia Blanca and adjoining country (Nos. 114 & 140, G. C.). Boorrs LACINIATA, = Nastanthus laciniatus, Miers. Grows on the gravelly bed of the Rio Negro, and the Chubat in Patagonia (No. 93, G. €.). This belongs toa group of species hitherto seen only in the higher regions of the Andes, and has doubtless been carried down to the plain by the streams from the Southern Andes. It is, perhaps, a variety of B. scapigera, Remy. COMPOSITA. STEVIA SATURELEFOLIA, Sch. Bip., var. ANGUSTIFOLIA, Baker. isla Verde, south of Bahia Blanca (No. 187, G. C.). Flowers rose-coloured. STEVIA SATURELEFOLIA, var. PATAGONICA? Sierra de la Ven- tana, Sierra de Tandil, Bahia Blanca, and throughout North Patagonia (No. 188, G. C.). Fiowers red. I suppose this to be one of the forms of this very variable species. It approaches the variety laxa. EvpaToriuM ERODIIFOLIUM, DC. Sierra de la Ventana and North Patagonia (No. 79, G. C.). This species has been hitherto known only from South Brazil and Uruguay. The Araucaniau Indians call the plant Meñuèké, and use the decoction as a sudorific. MIKANIA P Grows in the water of the Naporta, near Bahia Blanca, climbing up the stems of reeds (No. 73, G. C.). This is a small plant, with leaves sinuato-lobate, the lobes rounded obtuse, the stem thin and feeble, the flower-heads small and con- AND THE ADJOINING TERRITORY. 221 gested. It is either an undescribed species or an extreme form of the very variable M. scandens, L. GUTIERREZIA LINEARIFOLIA, Hook. § Arn. North Patagonia (No. 193, G. C.). HYSTERIONICA JASONIOIDES, Willd. Bahia Blanca (No. 96, G.C.). SOLIDAGO LINEARIFOLIA, DC. Widely spread about Bahia Blanca (No. 136, G. C.). The Indians call the plant Felel, and use it for dyeing wool yellow. This and S. microglossa, DC., seem to be forms of the same species. ERIGERON BoNARIENSE, L. Valley of the Rio Colorado and North Patagonia (No. 191, Œ. C.). The Spanish name given to this and several other Composites of similar appearance is Chileca ; the Araucanians call it Sassi. BACCHARIS AnRTEMISIOIDES, Hook. § Arn. Common in North Patagonia, on the plateaux and the flanks of the valleys (No. 94, G. C.). BACCHARIS CORIDIFOLIA, Pers. Sierra de Tandil, flowering in (March) autumn (No. 121, G. C.). Spanish name Zomerillo. Widely spread, especially in Uruguay, and said to be very poisonous to cattle. BACCHARIS GILLIESII, 4. Gr. Proc. Am. Ae., V. 123; var. foliis longioribus profundius dentatis. Sand-hills near the sea, at Isla Verde, the mouth of the Rio Colorado, and elsewhere in Patagonia (No. 181, G. C.) This is used to make brooms, and called in Spanish Zscoba. BACCHARIS sALICIFOLIA, Pers: Sandy ground in the valleys of the Rio Negro, Rio Colorado, aud the Chubat (No. 203, G. C.). A shrub 7-9 feet high; the leaves somewhat resinous and aro- matie; the wood spongy, used by the Indians to make fire by friction. The limits of the South-American species of Baccharis are not easily defined; and this with B. glutinosa and their allies form a group which is widely spread through temperate and subtropical S. America. Bacowaris TENELLA, Hook. d Arn.? Patagonia (No. 210, G. C.). A very imperfect fragment, and the determination uncertain. Bacemamrs —— ? Bahia Blanca, Sierra de la Ventana, and n2 Plateaux of North 222 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA elsewhere (No. 200, G. C.). A small fragment, too imperfect to be named. AcHYROCLINE SATUREIOIDES, DC. Sierra de Tandil; flower- ing in autumn, end of March (No. 120, G. C.). It is somewhat remarkable to find this species, widely spread throughout tropical South America, extending so far south under very different climatal conditions. Neither from dried specimens nor from the description (DC. Prod. vi. p. 220) can I see any note of dis- tinction between this and A. flaccida, DC. GNAPHALIUM LUTEO-ALBUM, D. Bahia Blanca and North Patagonia (No. 63, G. C.). Itis rather remarkable that this cos- mopolitan weed, which is found almost everywhere else in South America, is not mentioned in Grisebach’s enumeration of the very full collections made by the late Prof. Lorentz and others in the Argentine territory. It is not included by Berg in his list of plants introduced about Buenos Ayres, perhaps because he con- sidered it indigenous. GNAPHALIUM AMERICANUM, Mill., = G. spicatum, Lam. About Ajó (No. 168, G. C.). AMBROSIA ARTEMISIÆFOLIA, L.; var. foliorum segmentis an- gustis lanceolato-linearibus. Common at Ajó, and extends throughout the province of Buenos Ayres (No. 173, G. C.). This plant is said to give a disagreeable taste to the flesh of cattle feeding on it. It has no Indian name. Perhaps an introduced species. XawTHIUM AMBROSIOIDES, Hook. § Arn. South of the pro- vince of Buenos Ayres, very common about the bizeacheros (No. 53, G. C.. Unlike the other Xanthia, this very distinct species is confined to temperate South America; butits area appears to be extending, it being doubtless transported by animals, to whose hair or fur the fruit adheres. M. Claraz informs me that in his district the plant has been seen only of late years, and was not found anywhere in that region by Professor Lorentz in his earlier exeursions to the south of Buenos Ayres. VERBESINA AUSTRALIS, Baker, — Ximenesia microptera, DC. Bahia Blanca. A weed which seems to follow the colonists (No. 43, G. C.). SPILANTHES HELENIODES, Hook. j Arn. From Bahia Blanca AND THE ADJOINING TERRITORY. 223 westward and southward through Patagonia (No. 97, G. C.). The Indians call this plant Nim ñim, and chew the root to allay thirst. A decoction is also used with a compress for headache. M. Claraz's specimen is named Gaillardia Doniana on the authority of Lorentz; but there must be some mistake, as that is a very different plant. * BIDENS CHRYSANTHEMOIDES, Michæ. About Cabo S. Antonio, in inhabited places (No. 195, G. C.). Probably spread by the agency of man. B. helenioides, H. B. K., seems to be a form of the same species. FrAvERIA CowTRAYERVA, Pers. Bahia Blanca (No. 51, G. C.). Perhaps an introduced plant; but its use asa yellow dye is known to the Indians. TAGETES ? Bahia Blanca (No. 220, Œ. C... An imper- fect specimen which I have been unable to make out. Even the genus is uncertain. GAILLARDIA MEGAPOTAMICA, Spreng., var. SCABIOSOIDES, Baker, =Cercostylos scabiosoides, Arn. Bahia Blanca and North Pata- gonia (No. 75, Œ. C.). The Araucanian name is Tschóike catschu, meaning Rhea-grass. The leaves are very aromatic, and the infusion is used as a sudorific. It is an illustration of the justice of the precept of M. Alphonse de Candolle, that you should never attribute to an author a name which he has not used, to observe that this species is cited by Grisebach ( Plantx Lorentziane, P. 140, and ‘Symbolæ ad floram Argentinam,' p. 199), » Gail- lardia scabiosoides, Benth. Hook.” Ifthe authors of the Genera had undertaken to name the species, they would probably have formed the same conclusion as Mr. Baker has expressed in the ‘ Flora Brasiliensis.’ Senecio PINNATUS, Poir. Sierra de la Ventana and North Patagonia (No. 78, G. C.). The Araucanian name is Tschacalia. SENECIO PINNATUS, Poir., var.? GLANDULOSUS. Bahia Blanca, and through North Patagonia (No. 98, G. C.). 1 am somewhat doubtful as to this being referred to a form of 8. pinnatus. A closely resembles S. punctatus, Hook. & Arn., which has been by Mr. Baker united to S. pinnatus; but this specimen differs from all the forms of that species in the rather dense glandular pubes- cence of the upper ramifications aud the involucres. 224 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA SENECIO LARICIFOLIUS, H. B. K.? A small shrub from the plateaux of North Patagonia (No. 209, G.C.). This is a poor specimen, but it agrees with one collected by Darwin near Bahia Blanca, and thus labelled in Kew Herbarium. It agrees pretty well with the description of Kunth. Nevertheless I do not feel sure of the specific identity of this and the Peruvian plant. SENECIO ? Bahia Blanca (No. 205, G. C.). A very im- perfect fragment, looking like S. oligoleucos, Baker. Cyicus LANCEOLATUS, L. Introduced from Europe of late years, now widely spread in the district of Bahia Blanca. Called in Spanish Cardo negro. HYALis ARGENTEA, Don. Bahia Blanca aud North Patagonia : one of the characteristic plants of this region (No. 164, G. C.). Tbe Spaniards call this Oliva and also Maqui blanco. The resi- nous excrescences on the stem caused by insects are gathered by the Indians, and used as a masticatory. CHUQUIRAGA ERINACEA, Don. Widely spread through North Patagonia from the Atlantic coast to the Cordillera, the northern limit being about Bahia Blanca (No. 82, G. C.). The Arauca- nian name is Tschitri kekelu. CHUQUIRAGA HYSTRIX, Don. North Patagonia, northern limit some way south of the Rio Negro, extending thence to the Chubat and beyond that river, very common in some parts (No. 148, G. C.). A shrub 4-5 feet high, with the heads (invo- lueral scales) bright yellow, flowering in summer (December and January) The leaves prick like needles, but easily drop off from the dried specimen. This isa larger and stronger plant than the allied C. erinacea. The leaves, which are quite glabrous, are broader below, terminating in a sharp brown point. There is slight araneous pubescence (cobweb) about the margin of the involucral scales. CnuvqQuinAGA Kinet, n. sp. Extends from the mountains of Trencta southward to the Chubat and beyond that river (No. 146, G. C). I have received but a very incomplete fragment of this undeseribed species, but it is evidently the same as two speci- mens in the Kew Herbarium ; the first, from Port St. Elena, col- lected by Captain King, the second, labelled * Patagonia," from Captain Middleton. I subjoin a brief diagnostie character which may serve to identify the plant. AND THE ADJOINING TERRITORY. 225 Chuquiraga Kingii differt a proxima C. spinosa, Don, spinis axillaribus brevissimis, foliorum nervis lateralibus obsoletis, et presertim involueri squamis latioribus apice cartilagineo sub- muticis, nec in mueronem spinescentem acuminatis. In hac specie squame interiores flave, nee ut in C. spinosa saturate aurantiac:. The Indians distinguish the above-mentioned three species of Chuquiraga, and the Patagonian name for this is Amtrac-trac- tschic. TRICHOCLINE HETEROPHYLLA, Less. Very common about Bahia Blanca (No. 141, G. C.). Called by the Spaniards Chucho, but improperly, as the true Chucho is a Solanaceous plant. Nassauvia ROSULATA, = Ácanthophyllum rosulatum, Hook. § Arn. Comp. Bot. Mag. ii. p. 43 (Nos. 88 and 145, G. C.) A stunted shrub growing in rough stony ground in Patagonia from the Rio Negro southward to the Chubat and beyond that river. It is used to make rude combs for the Indian women ; and as the root is very tough, the Patagonians, who call it Yahenele, drag it up by passing the leather thong which is attached to the saddle of the horse round the base of the stem. This isa curious plaut which, though apparently common in Patagonia, is rare in her- baria. There are but two fragments in the herbarium at Kew; the first collected by Darwin at Port Desire, the other somewhere on the coast of Patagonia by Captain Middleton. It belongs to a small group including two other species from the Chilian Andes, first described by Lagasca, and referred by him to Triptilion, but more correctly placed in the genus Nassauvia by Don in 1882. Soon after, on account of their very peculiar habit aud mode of growth, they were constituted into a separate genus by Hooker and Arnott (Comp. Bot. Mag. i. p. 37) under the name Acantho- Phyllum. That name having been previously adopted for a very different genus by C. A. Meyer, the group received the nane Strongyloma in DC. Prod. vii. p. 51. In the present species re upright stems have very numerous branches, wbich are r uces to rounded glomerules of minute leaves about the size of a pea, with usually 1, sometimes 2, small flowering heads at the apex. My specimens have no primary leaves; butiu Darwin s specimen these much resemble the involueral scales, having a strong medial nerve which is produced to a sharp spiuy point. I have not seon specimens of Nassauvia glomerulosa, Don (Strongyloma, DC. 226 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA Prod.), which is said to grow in the higher region of the Chilian Andes; but from the description it appears to be so similar to the Patagonian plant, that the latter should, perhaps, be referred to it as a variety. SONCHUS ASPER, Fl. Dan. Naturalized, and now widely spread (No. 49, G. C.), This and the nearly allied S. oleraceus are widely spread throughout tropical and temperate South America, and their diffusion has probably been to a great extent inde- pendent of the agency of man. HyrocuærIs ——? Common throughout the province of Buenos Ayres (No. 194, G. C.). The specimen is in bad con- dition, and too imperfect for recognition. PLUMBAGINEX. STATICE BRASILIENSIS, Boiss. Salt-marshes of Tuyu, and in North Patagonia (No. 102, G. C.). The Patagonians call this Guaicuru, and make a deep-red decoction from the roots, used in popular medicine to purify the blood. PRIMULACES. ASTEROLINON SERPYLLIFOLIUM, = Lysimachia serpyllifolia, Poir., =Pelletiera verna, A. St.-Hil. Bahia Blanca (No. 69, G. C.). I do not feel sure that Pelletiera serpyllifolia, Webb, Phyt. Can., is the same as the South-American plant. APOCYNACER. MEsopora ? From the base of the Sierra de la Ventana near Bahia Blanea (No. 218, G. C.). Of this I have a mere frag- ment without leaves; but it does not agree well with any of the described species. From JM. decemfida, to which it approaches, it differs by a larger corolla and the form of the ealyx. ASCLEPIADER. BracuYreris CawpoLLEANUS, Hook. § Arn. N. Patagonia, flowering late in autumn (April, May). Also found on the Sierra de la Ventana and in Entrerios (No. 213, G. C.). * - - . PHILIBERTIA SOLANOIDES, H. B. K., — Sarcostemma incanum, Dene. At Tuyu, near Cabo S. Antonio; also seen in Entrerios (No. 227, G. C.). This climbs up the stems of Gramineæ and other small plants. AND THE ADJOINING TERRITORY. 227 *OXYPETALUM SOLANOIDES, Hook. At Ajó, near Cabo S. Antonio (No. 165, G. C.). HYDROPHYLLACER. PHACELIA GLANDULOSA, Nutt., var. PATAGONICA. South of provinee of Buenos Ayres (No. 211, G. C.). A very imperfect fragment, but evidently the same as a plant in Kew Herbarium gathered by Darwin at Bahia Blanca, and another from the Pata- gonian coast, sent by Captain Middleton. It differs from the variety mneo-mexicana by a smaller corolla and less exserted stamens and style. BonaAGINEX. HELIOTROPIUM ANCHUSEFOLIUM, Poir., var. ANGUSTIFOLIUM. Marshes of the Naporta Chico near Bahia Blanca (No. 119 5, G.C.). This was collected and sent along with Stemodia lan- ceolata, Benth. HELIOTROPIUM CURASSAVICUM, L., var. PARVIFLORUM. Salt- marshes at the mouth of the Rio Negro, and elsewhere in Pata- gonia and Buenos Ayres (No. 221, G. C.). EnrrRICHIUM ? Valley of the Rio Negro on moist ground (No. 199, G. C.). Iam unable to determine this plant, not find- ing it possible to fix the characters of the numerous American species with minute flowers. It has some resemblance to Æ. albi- Jlorum, Griseb. (Myosotis albiflora, Banks. & Sol.), but cannot, I think, be referred to that species. Its true position, I believe, either as a variety or as a distinct species, is between E. califor- nicum, A. DC., and E. tenellum, Gray. SoLANACER. é SoLANUM ELEAGNIFOLIUM, Cav. South of the province of Buenos Ayres (No. 81, Œ. C.). This plant, whose original home is the subtropical zone of South America, appears to have followed the spread of colonization southward. It is very common about inhabited places, and a troublesome weed in cultivated soil. SOLANUM ? About Bahia Blanca, and along the valley of the Rio Negro (Nos. 55 and 99, G.C.). The Araucanian Indians cook the leaves and eat them as a vegetable, calling the plant Liaghé. Without a familiar acquaintance with the plants in their home, it is not easy for a botanist to find his way among 228 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA the many hundred species of Solanum described from South America. This is certainly the same as Argentine specimens (unnamed) preserved in Kew Herbarium from Tweedie and Miers. It is nearest to S. Tweedianum, Hook., but much smaller than that species. Whether it should be ranked as a variety or as a distinct species, I will not attempt to determine. Puysarts viscosa, L. This is widely spread from Entrerios through the province of Buenos Ayres to the Rio Negro, and be- yond itin North Patagonia, but always near inhabited places (No. 175, G. C.), This is commonly called Camanbá, a Guarani name, showing that the plant has spread from the north (7. e. Brazil). The same name is given to the fruit of another Solanaceous plant which has also spread from South Brazil to Patagonia, Salpichroma rhomboideum, Miers*. The Araucanian Indians have no name for the present species, but call the fruit Kilièn, the name given to the Strawberry (Fragaria chilensis). LyciuM FILIFOLIUM, Gill. var.=L. minutifolium, Miers, IU. S. Am. Pl. ii. p. 130. Bahia Blanca and through North Pata- gonia (No. 158, G. C). This is common about Buenos Ayres, but appears to have been introduced there. The Araucanian name is Tschayem. This is a small bush, 2-3 feet in height, with feeble branches, which, when not supported, usually rest on the ground. Lycium ? Bahia Blanca; Sierra dela Ventana; North Patagonia (No. 208, G. C.). Araucanian name Zuingan, called by the Spaniards Mataperro, probably because dogs, attempting to penetrate the bushes, are seriously injured by the sharp tough spines. I cannot identify this with any described species; butas there are several cf which I have not seen specimens, I hesitate to give this a name. It agrees with an unnamed specimen in Kew Herbarium from the Rio Negro, from Wilkes's Explorin g Expedi- tion, and approaches L. elongatum, Miers ; but is, I think, certainly different. In the present plant the calyx is minute, with very short segments, the corolla attenuated, almost entire, scarcely lobed, but rather crenated at the mouth; the pedicels, which are very short in flowering-time, are somewhat elongated when the plant bears its'bright-red berries. The leaves are small, linear- * D may remark that the locality “ Magellania,” given by Dunal in DC. Prod. for this Sa/pichroma, ow the supposed authority of Commerson, is an in- dubitable mistake. Commerson found the plant near Monte Video. AND THE ADJOINING TERRITORY. 229 spathulate. If, as I believe, undescribed, it may properly be called Lycium Witlkesii. Cestrum Pargcr, L’ Hér., var. South of the province of Buenos Ayres (Nos. 174 and 219, G. C.). This species, almost universal in South America, is extremely variable. M. Claraz’s specimen represents a small and slender form. PETUNIA NYCTAGINIELORA, Juss. North Patagonia, Sierra de la Ventana, Sierra de Tandil (No. 192, G. C.). This was known to extend from South Brazil to Buenos Ayres; but M. Claraz has added considerably to its known area. NIEREMBERGIA HIPPOMANICA, Miers. On the dry stony and rocky plateaux near Bahia Blanca, and through North Patagonia (No. 135. G. C.). l NIEREMBERGIA RIGIDA, Miers. Common about Bahia Blanca (No. 144, G. C.). This is very near to, and perhaps only a variety of, N. filicaulis, Lindl. ScROPHULARINEE. LINARIA CANADENSIS, Spr. Banks of the Naporta near Bahia Blanea (No. 72, G. C.). This has possibly been introduced by man; but the species is very widely spread through the American continent. STEMODIA LANCEOLATA, Benth. Marshes of the Naporta Chico near Bahia Blanea (No. 119, G. C.). Flowers of a violet-blue tint. This species appears to extend from the northern provinces to the extreme south of the Argentine territory. Herrestis nADICATA, Bentham. Same locality as the last species (No. 118, G. C.). Scoparta FLAVA, Cham. et Schlecht. Same locality as the two preceding species (No. 1184, G. C.). GxmaRDIA GENISTAEFOLIA, Cham. et Schlecht. Sierra de la Ventana and Sierra de Tandil (No. 161, G. C.). Flowers full rose-colour. VERBENACES. LANTANA Crarazu, n. sp. Plateaux of North Patagonia above the valley of the Rio Negro, flowering in spring (September) (No. 214, G. C.). I subjoin a brief description of this very distinet plant :— 230 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA Fruticulus ramosissimus ; foliis (saltem superioribus) sessilibus, lineari-lanceolatis, utrinque attenuatis, margine subintegerrimo smpius revoluto, nervo medio prominente, lateralibus obsoletis ; floribus capitatis in spicas brevissimas ramorum lateralium ter- minales congestis; bracteis floriferis ovatis pubescentibus calyci subzquilongis; calyce tubuloso brevissime 4-dentato; corolle tubo exserto, paululum recurvato, fauce bilabiato. Ob fructum ignotum inter Lantanam et Lippiam subambigua, priori tamen, ut videtur, referenda. The nearest ally appears to be Lantana Sellowiana, Link, from which, however, it differs both in habit and structural cha- racters. LIPPIA LYCIOIDES, Steud. Bahia Blanca and North Patagonia (No. 133, G. C.). The leaves, which are very aromatic, are put into the Paraguay tea; and they are also used medicinally in infusion. The Spaniards call the plant Poléo; the Araucanian name is Tschoiké mamuel, meaning ostrich-wood. LIPPIA sERIPHIOIDES, A. Gray. Bahia Blanca and through North Patagonia, extending westward to the Cordillera (No. 84, G.C.). This is a characteristic species of the Patagonian flora, not extending far beyond, unless perhaps along the flanks of the Cordillera. It is called in Spanish Tomillo del Campo, and in Araucanian Loom. It is used as thyme for flavouring soup &c., and the infusion, which is aromatic with a bitter taste, is used medicinally. From a note by Tweedie on a label in Kew Her- barium, it appears that in his time, nearly fifty years ago, it was sent from the south to Buenos Ayres, and sold by apothecaries. Though long in herbaria, having been collected near Bahia Blanca by Darwin, and on the Patagonian coast by several suc- ceeding travellers, this appears to have been first published by Asa Gray. The manuscript name * L. rubiginosa, Gill.,” is affixed to specimens collected near Mendoza by Gillies ; but that specific name has been applied to a very different plant by Schauer in DC. Prod. xi. *VERBENA LITORALIS, Kunth. Tuyu, near Cabo S. Antonio (No. 222, G. C.). VERBENA BONARIENSIS, L. Valleys of the Rio Colorado and the Rio Negro (No. 224, G. C.). Grows in damp ground and by water-courses. Flowers rose-colour or violet. AND THE ADJOINING TERRITORY. 281 VERBENA TEUCRIOIDES, Gill. 4 Hook. Common at Bahia Blanca and in North Patagonia; flowers white (No. 188, G. C.). VERBENA ERINOIDES, Lam. Bahia Blanca; North Patagonia (No. 139, G. C.). This has lilac flowers, with a scent of Vanilla, which come out early (November). VERBENA INcISA, Hook.? Bahia Blanca (Nos. 163 and 204, G. C). The specimens are very incomplete fragments, and the name therefore rather uncertain. The flowers are red, and the plant is commonly called Margarita. LABIATÆ. HEDEOMA MULTIFLORA, Benth. Bahia Blanca and North Patagonia (No. 125, G. C.). This plant, called by the Spaniards Menta del Campo, is much used medicinally, the decoction being considered a remedy for diarrhea and various stomach derange- ments. From the fact of having no Indian name, M. Claraz supposes the plant to be of recent introduction into Patagonia. It was, however, long ago found by Gillies, Tweedie, and others at various places in the south and west parts of the Argentine territory, while it extends northward to South Brazil, PLANTAGINEE. PLANTAGO MAJOR, L. Common at Ajó and elsewaere (No. 196, G. C.). Doubtless introduced from Europe. PLANTAGO PATAGONICA, Jacg., var. Bahia Blanca and North Patagonia, very common (No. 127, G. C.). Among a large set of specimens of this very variable species, I find none quite agree- ing with this variety. It comes nearest to the form most common in North America (P. gnaphaloides, Nutt.). The pubescence, especially of the calyx, is abundant, cottony, white, and not, as in many forms of the species, rufous. PrANTAGO Bismarck, Niederl. Sierra de la Ventana (N o. 60, G. C.). This is a curious and very distinct species, described by Niederlein in the * Monatschrift zur Beförderung des Gartenbaus in d. k. Preussischen Staaten ’ for 1881, p. 16. It forms a small bush, 1 to 2 feet high, with a woody stem nearly an inch in dia- meter, and numerous crowded tufts of linear-subulate leaves on the lateral branches. 232 MR. J. BALL ON THE FLORA OF NORTH PATAGONTA ILLECEBRACER. PENTACENA POLYCNEMOIDES, Bartl., var. Bahia Blanca; Sierra de la Ventana: North Patagonia, extending to the foot of the Cordillera (Nos. 68 and 70, G. C.). The Indians call this Ancue, and use the infusion for stomach complaints. AMARANTACER, AMARANTUS cirroRosTACHYS, Willd. Bahia Blanca, where it was found by Darwin; North Patagonia; about Cabo S. Antonio (Nos. 177 and 178, G.C.). The young plant is eaten as a vegetable like spinach. -PUPALIA ? Bahia Blanca and North Patagonia. Flowers yellowish white (No. 184, G. C.). A mere fragment, as to which I am quite uncertain. GUILLEMINEA LANUGINOsA, Mog. (sub Gossypiantho). In the province of Buenos Ayres, ruderal (No. 47, G. C.). Also seen in Entrerios. PrArriA LANATA, Poir. (sub Gomphrena). Bahia Blanca; North Patagonia. Also grows in Entrerios (Nos. 56 and 101, G. C). ALTERNANTHERA ECHINATA, Sm. Ruderal throughout the province of Buenos Ayres (No. 46, G. C.). GoMmPHRENA VILLOSA, Mart. Sierra de Tandil ; flowering late (March) (No. 1244, G. C.). Sent along with the next species, from which it is certainly different. GoMPHRENA ROSEA, Griseb.? Bahia Blanca and North Pata- gonia (No. 124a, G. C.). This is said to flower in February in the south, but a month earlier in Entrerios. The flowers (i. e. coloured bracts) are of a bright rose-colour. The specimen is a mere scrap, as to which I am not quite certain. CirENOPODIACEE. CHENOPODIUM ALBUM, L. Throughout the province of Buenos Ayres (No. 42, G. C.). CHENOPODIUM MURALE, Z.? Bahia Blanca and elsewhere, common (No. 197, G. C.). The specimen has neither flower nor fruit, AND THE ADJOINING TERRITORY. 233 CHENOPODIUM AMBROSIOIDES, L. Everywhere about Bahia Blanca &e. (No. 40, G. C.). Commonly called Paico hembra. CHENOPODIUM AMBROSIOIDES, L., var.? Common at Bahia Blanca and elsewhere (No. 41, G. C.). This is sent as a different plant from the last, and said to be called Paico macho; but it appears to be no more than one of the numerous varieties of that protean species. ATRIPLEX PAMPARUM, G'riseb. Near Cabo S. Antonio (No. 166, G. C.) I suspect that this must be ranked as a variety of A. patagonica. ATRIPLEX Lamera, Gill, var. ANGUSTIFOLIA ? Valley of the Upper Limay (No. 99, Œ. C.). This is very possibly an unde- scribed species of the Section Obione, and certainly allied to A. Lampa, Gill. My specimen represents the male plant only. The whole plant is hoary white, very much branched, the male flowers in little spherical or egg-shaped heads. Hererostacuys RiTTNERIANA, Ung. Sternd.? Salt-marshes at Bahia Blanca (No. 180, G. C.). The Araucanian name is Tschkilpé. The specimen is incomplete; and in this very difficult group I feel no certainty. This may possibly belong to the closely allied Spirostachys patagonica, Hook. fil. SALICORNIA FRUTICOSA, L., var. PERUVIANA. Salt-marshes of the Rio Colorado (No. 183, G. C.). The native name for this and several other succulent plants growing in salt-marshes is Gume. POLYGONACES. RUMEX MAGELLANICUS, Griseb. In the valleys of all the rivers and principal streams draining North Patagonia (No. 128, G. C.). The inflorescence is of a rich purple-red hue. The Indian name is Calcatréo. Though a well-marked form, this seems to be closely allied to R. cuneifolius, Campd. Rumex pratensis, Mert. § Koch. Valley of the Rio Negro, North Patagonia (No. 160, G. C.). Perhaps introduced from Europe? Rumex concromeratvs, Murr. At Tuyu and Ajó (No. 231, G. C.. I suppose that this must be ranked among the forms of the cosmopolitan R. sanguineus, L. POLYGONUM camporum, Meisn., var. AUSTRALE. At Tuyu (No. 223, G. C.). 234 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA MUEHLENBECKIA CHILENSIS, var. INJUCUNDA, Meisn. in DC. Prodr. Valley of the Limay, North Patagonia (No. 86, G. C.). This is commonly called Salsaparilla, and is used medicinally. SANTALACER. ARJONA PATAGONICA, Hombr. et J.,— A. tuberosa, Cav., var. patagoniea, 4. DC. Bahia Blanca; North Patagonia (No. 126, G. C.) This is called Macachina by the Spaniards, and Sakel by the Araucanian Indians. In winter one or more tubers, about an inch long and half as thick, are developed. These are eaten, either raw or cooked. The tubers are detached from the parent stem when the plant flowers in spring. M. Claraz states that, further south in Patagonia, beyond the Chubat, this species is replaced by a different one, which produces longer and thinner tubers. I have not seen several of the species (or forms) de- seribed by Mr. Miers in the Journal of the Linnean Society, xvii. IODINA RHOMBIFOLTA, Hook. § Arn. Widely spread, but not common, extending from the northern frontier of Patagonia to. Entrerios and Uruguay, and to the interior of the Argentine territory (Nos. 142 and 190, G. C.). This shrub occurs here and there in small groves at wide intervals, usually on dry ground. One of M. Claraz's specimens comes from the Salina Chica, 12 leagues west of Bahia Blanca, and another grove is at the foot of the Sierra de la Ventana. The Araucanian name is Trallian. At Bahia Blanca the Spaniards call it Sombretoro ; but in Entre- rios that name is given to a species of Maytenus. UnTICACERF. URTICA SPATHULATA, Sm. Common about Bahia Blanca, in ruderal stations, and especially around the bizcacheros (No. 44, G. C.). This seems to be nearly allied to the Andean U. echinata, Benth. BROoMELIACEX. TILLANDSIA COanCTATA, W.? Valley of the Rio Negro and elsewhere in North Patagonia, on the branches of shrubs (No. 207, G. C.). The specimen is too imperfect to be determined with certainty ; it undoubtedly belongs to the group Diaphoranthema. A good many species of Tillandsia have been recorded from the northern and central parts of the Argentine territory, but the AND THE ADJOINING TERRITORY. 235 extension of the genus so far southward as Patagonia is, I believe, altogether new. Inrprx. SISYRINCHIUM CHILENSE, Hook. Extends southward from Entrerios through the province of Buenos Ayres to the borders of Patagonia. Common at Bahia Blanca (No. 113, G. C.). This is called by the Spaniards Thé pampa and by the Indians Picun pelia. The infusion is used as tea, and is valued as a sudorific. StsyRINCHIUM CLARAZIT, Baker MSS. Bahia Blanca (No. 112, G. C.). On showing this plant to Mr. Baker, whose knowledge of this natural order is so extensive and complete, he at once pro- nounced it to belong to an undescribed species; and he has been good enough to draw up a deseription, with the remarks which 1 subjoin :— “S. Clarazii, Baker; acaule, dense cæspitosum ; foliis basi- laribus 3-4 rigidulis, anguste linearibus, scapo brevioribus ; scapo 8-4-pollieari, conspicue ancipite; florum fasciculis 2-3, singulis 2-3-floris, ad scapi apicem aggregatis ; spathe valvis exterioribus duris, lanceolatis, infima exteriore in apicem foliaceum producta ; pedicellis ex spatha protrusis ; perianthii segmentis obverse lan- ceolato-oblongis, luteis, venis pluribus nigro-brunneis parallelis percursis ; ülamentis liberis; capsula subglobosa glabra. “A dwarf acaulescent densely tufted yellow-flowered species, with a conspicuously two-edged flower-stalk, allied to S. pusillum, S. bogotense, and S. tinctorium, H. B. K. The leaves are rigid in texture, not more than half a line broad, with about seven distinct ribs, The perianth is about } inch in length." From a note of M. Claraz, it appears that this is used for tea in the same manner as S. cAilggse. AMARYLLIDEX. ZEPHYRANTHES ANDERSONI, Herb. North Patagonia; also seen in Entrerios (No. 76, G. C.). Springs up commonly after the rains. LinrAcEX. Bnoprza aurea, Benth. 4 Hook. f.?, — Milla aurea, Baker. Bahia Blanca, common after the rains (No. 35, G. C). The specimen is very imperfect, and the determination somewhat uncertain. LINN. JOURN.—- BOTANY, VOL. XXI. 936 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA JUNCACER. Juncus acutus, L. Bahia Blanca and elsewhere, common (No. 37, G. C.) The succulent rhizomes are sought after by pigs at Bahia Blanca. TYPHACED. Typna ANGUSTIFOLIA, L. Extends from the valleys of North Patagonia, through the provinces of Buenos Ayres and Entrerios (No. 229, G. C.). The Spanish name is Totara, the Araucanian Trapal. This cosmopolitan species probably owes its dissemina- tion to water-birds, and not to the agency of man. Cx PERACE RF. Cyperus vkaETUS, Willd. Valley of the Rio Negro (No. 171, G. C.). SCIRPUS MARITIMUS, L. Common at Tuyu and elsewhere (No. 234, G. C.). In default of better material, this is sometimes used for thatching. SCIRPUS ? Valley of the Rio Negro (No. 170, G. C.). A very imperfect fragment, not determinable. CAREX PHALAROIDES, Kunth. Bahia Blanca, on banks of streams (No. 77, G. C.). CAREX RIPARIA, L. Bahia Blanca, North Patagonia, and throughout the province of Buenos Ayres; common along water- courses (No. 38, G. C.). This is the form described as C. incras- sata, Schlecht. GnRAMINEE. SETARIA CAUDATA, Roem. & Sch. Bahia Blanca and North Patagonia, common (Nos. 11 and ¥79, G. C.). Spartina COARCTATA, Trin. Bahia Blanca, and also from Tuyu (No. 225, G. C.). A tall, wiry grass, used for thatching. ANDROPOGON ARGENTEUS, DC. Bahia Blanca (No. 14, G. C.). CHRYSOPOGON STIPOIDES, Trin. (ex Munro, MSS.), =Sorghum nutans, 4. Gray (L. sub Audropogone), — Andropogon avenaceus, Michx. Valley of the Naporta Grande, district of Bahia Blanca (No. 9, G. C.). This species was collected somewhere near Bahia Blanca by Darwin. It has a wide geographical range in the American continent. According to strict rules of nomenclature, AND THE ADJOINING TERRITORY. 237 this should be named Chrysopogon nutans, the latter being the trivial name given by Linneus; but it seems undesirable to add to the present synonymy. ALOPECURUS GENICULATUS, L., var. ARISTULATUS, =A. aristu- latus, Mich». Valley of the Rio N egro and elsewhere, in moist situations (No. 198, G. C.). Commonly called Gramilla de bañado. ARISTIDA SETIFOLIA, H. B. K. Bahia Blanca (No. 30, G. C.). STIPA INTERMEDIA, Trin. Bahia Blanca (No. 20, G. C). Exactly agrees with a specimen collected by Darwin near Bahia Blanca, and named by Munro. Stipa BICOLOR, Vahl. Bahia Blanca (No. 22, G. C.). Stipa CAUDATA, Trin. Bahia Blanca and elsewhere, common on the plateaux (No. 19, G. C.). A specimen collected by Tweedie from the “ plains of Patagonia," and named by Munro, differs only in the awns being somewhat shorter. Stipa PULCHELLA, Munro MSS. Bahia Blanca and North Patagonia (No.:25, G. C.). This is scarcely different from S. pogonathera, Desv. Stipa Cramazir, n. sp. Bahia Blanca (No. 21, G. C.). This appears to be well distinguished from all the described species, and deserves to bear the name of its discoverer. I subjoina brief description :— i Stipa Clarazii. Perennis, culmis 2-3-pedalibus ; foliis radi- calibus filiformibus, involutis, flexuosis, vix rigidis, caulinis ex vagina ampla subinflata latiusculis, apicem versus involutis, basiu versus margine subscariosis, eximie striatis ; paniculo laxæ ramis 1-2-polliearibus ; glumis pro grege maximis, fere pollicem Jones superiore paullo longiore apice acuminata; glume fertilis pilis albis sericeis, arista elongata 2-3-pollicari, scabro-villosa. PonyPoaow MONSPELIENSIS, L. Very common in the A ^ Rio of Buenos Ayres, about Bahia Blanca, in the valley of the Rio Negro, and elsewhere (Nos. 12, 172, and 230, G. C.). ; ia Blanca PAPPOPHORUM ALOPECUROIDES, Vahl, var. P Bahia Bl i (No. 74, G. C.). In Kew Herbarium several specimens, seeming d ; 7 a, : » same cover by Genera widely different, have been placed in the same cov ^ N ; n I d -< : :h-branched and wel- Munro, Some are tall grasses, with a much brane 238 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA furnished panicle, others dwarfed, with a few nearly sessile spi- cule. There are, however, intermediate forms, and it remains for closer study to decide whether they should be united as varieties of a single species. GYNERIUM ARGENTEUM, ees ab Esenb. Bahia Blanca and North Patagonia (No. 28, G. C.). The Spanish name for this wide-spread species is Corfadera. KaLkRIA crisrara, Pers. Bahia Blanca, common (No. 10, G. C.). Doubtless introduced from Europe, but not included in Berg's * Enumeracion.’ Enaanosris MEGASTACHYA, A@l. Common about Bahia Blanea (Nos. 16 and 26, G. C.). Confined to the neighbourhood of inhabited places and cultivated ground, and apparently intro- dueed. Called Gramilla. Enaanosris DELICATULA, Zrin. Bahia Blanca and North Patagonia (No. 18, G. C.). This species affords good pasture in the valleys. Meuica Macra, Nees ab Esenb. Common about Bahia Blanca (No. 36, G. C.). This is called Pasto bravo, or wicked grass. It is scarcely more than a variety of M. papilionacea, L» Merica vroracka, Cac. Very common at Bahia Blanca and through North Patagonia (No. 15, G. C.). The Araucanians call this Nahuel catschu, probably because the puma is supposed to crouch in the thick of it. Disricuris MARITIMA, Raf., = Brizopyrum spicatum, Hook. d Arn. Sandy and salt soil at Bahia Blanca ; in North Patagonia and at Ajo (No. 233, G. C.). Called Gramilla de Salitral. Briza LaManckKIANA, Nees ab Esenb., = Chascolytrum subaris- tatum, Desv. Bahia Blanca, on the plateaux (No. 13, G. C.). Poa aLoPECURUs, Kunth. Widely spread from Entrerios through Buenos Ayres to North Patagonia (No. 31, G. €). Araueanjian name Fueo catschu. Poa LANIGERA, Nees ab Hsenb. Bahia Blanca (No 32, G. €). Poa pEkNUDATA, Steud. Bahia Blanca (No. 33, G. C.). Poa DENUDATA, Sleud., var. MINor. Bahia Blanca (No. 34, G. C). AND THE ADJOINING TERRITORY. 239 *GLYCERIA FLUITANS, Æ. Br. Tuyu &c., in province of Buenos Ayres; also in Eutrerios (No. 230, G. C.). Introduced, but per- haps not by the agency of man. FESTUCA BROMOIDES, auct. (incluso Munro), non L. Very common at Bahia Blanca (No. 29, G. C.. Doubtless introduced from Europe. Bromus uNiOLOIrDEs, H. B. K. (Ceratochloa, auct.). Com- mon in the province of Buenos Ayres and about Bahia Blanca (Nos. 24 and 27, the latter a small form, G. C.). The Arau- canians call this grass Lancu, and formerly used the pounded seeds as food. M. Claraz supposes this to be an introduced species ; but it is widely spread as au indigenous plant in South America, and appears to have been long known to the Indians, therefore probably indigenous. LOLIUM PERENNE, Z. Bahia Blanca; a common ruderal plant (No. 17, G. C.). Introduced from Europe, but now widely spread in South America. The larger specimens represent the variety brasilianum of Nees, with the rhachis scabrous and the fertile glumes setiferous. HORDEUM pratense, L. Bahia Blanca, very common (No. 28, G. C.). HonpEuM JuBATUM, L. Bahia Blanca and elsewhere (No. 226, G.C.). This has a very wide range, from the Arctic coast of North-west America to the Strait of Magellan. It is said to afford good herbage for cattle. FILICES. The following five species, along with Adiantum cuneatum, T > : -ere collected by M. Claraz in of which I have seen a specimen, were collected by AM. Ul the mountains of Tandil and La Ventana; but no other ferns were observed by him during his very numerous excursions im the interior :— - Ta 5 1 CASSIBEERA TRIPHYLLA, Aaulf. (No. 5, G. C.) "ED * PELLÆA TERNIFOLIA, Fée. (No. 1, G. C.) , 7 3. C.) Also seen in CHEILANTHES MICROPTERIS, Sw. (No. 4, G. C ) Patagonia. e Th t LC Also seeh 1n Burcuntm wasrarum, Willd. (No. 3, G.C.) Alsi Patagonia. 940 MR. J. BALL ON THE FLORA OF NORTH PATAGONIA ETC. AsPIDIUM CAPENSE, Willd. (No.2,G. C.) Also seen in Pata- gonia. ADIANTUM CUNEATUM, Langsd. No specimen iu the col-- lection. P.S.— Since the foregoing pages were in type I have received the part relating to Botany of the official publication* respecting the expedition to the Rio Negro under General Roca iu 1879, prepared by the late Professor Lorentz and Mr. G. Niederlein, who was the botanist of the expedition. The Catalogue includes 324 species or varieties of flowering plants, and 13 vascular Cryptogams. Of these 20 are described as new species, and as many as 66 are merely assigned to genera, but not specifically determined. Of the 258 named species of flowering plants only 53 are certainly identical with as many species in Mr. Claraz's collection ; but I have reason to believe that a consider- able portion of the unnamed species should also be included in that category. To account for so large an amount of difference, I may remark that the plants collected by Mr. Niederlein were all found in the interior, while a majority of Mr. Claraz's plants come from the coast-region. In both collections there are a good many species not yet certainly known to grow in Patagonia proper, as several of Mr. Niederlein's plants were found north of the Rio Colorado. His journey was effected in autumn and winter; and many of the specimens being in bad condition, it is easy to understand the difficulty of accurately determining the species. It is probable that the discrepancy between the two lists is partly due to differences between botanists working under very different conditions. I have not seen authentic specimens of many species named by MM. Lorentz and Nieder- lein, and they have not seen the types on which Hooker and Arnott and A. Gray have founded many species of this region. * Informe Oficial de la Comision Cientifica Agregada al Estado Mayor General de la Expedicion al Rio Negro (Patagonia) bajo las ordenes del General D. Jurio A. Roca. Entrega II. Botánica. 4to, Buenos Aires, 1881. DE. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGXE. 24] Notes on Afghanistan Algæ. By Dr. Jurivs ScuaaRsCHMIDT, Lecturer on Cryptogamie Botany and Vegetable Anatomy, Assistant in the Botanie Garden of the Royal Hungarian University at Kolosvár [Klausenburg]. (Communicated by Prof. Orrvzgn, F.L.S.) [Read March 6th, 1884.] (Prare V.) Tue following list of Afghanistan Alge is the result of the investigation of some dried Phanerogamie Plants collected in Afghanistan by J. E. T. Aitchison, F.L.S., Surgeon-Major in H.M. Bengal Army, in the Afghanistan Expedition of 1880. I am indebted to Professor Dr. A. Kanitz, whose kindness has enabled me to draw up this account, for permitting me to exa- mine many plants from the interesting duplicates distributed by the Kew Herbarium. Ifound the enumerated Alge chiefly adhering to the specimens of Ammannia pentandra, Roxb. (417. Shalizan, in rice-fields, pro- fuse”*], forming fine bluish-green incrustations around the stems and on the leaves. Many interesting forms were found (perhaps Bacillariacee) in the small earthy particles remaining attached to the roots. One species, viz. Hantzschia Amphioxys, was only found on the roots of Anemone tetrasepala, Royle [* 186, 237, 332=463 (1879). Saféd-koh range, on the margins of forests, and in open grassy slopes on the inner hills, at an altitude of from 7000 to 10,000 feet " 1]. CYANOPHYCES. 9 1. Curoococeus (Sxnecnococcus ? GL@oTHECE ?) sp.? (See Anabena (Cylindrospermum) ) 2. GOMPHOSPHÆRIA APONINA, Kiitzing. Size of cells 0004 millim. 3. OSCILLARIA sp.? 4. Mrcrocorevs A1rcursontt, n. sp. (Plate V. fig. " Miere coleus aquaticus, strato ærugineo-chalybeo, tr ichomatıbus æq . " By J. E. T. * « On the Flora of the Kuram Valley &c., Afghanistan. Part 1T na Aitchison, F.L.S., Surgeon-Major H.M. Bengal Army. Journat er te Society, Botany, xix. Nos. 117-119 (1882), p. 162. t L. c. p. 148. 242 DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALG. bus in fasciculos filiformibus, articulis diametro sub:zequalibus vel paullo brevioribus, subtiliter granulatis, apieulo subobtusis vel obtuso-rotundatis; vagina universali crassa, undulato-constrieta et transversim striata [ut in Symphyosiphoni Thelephoroidi (Mon- tagne), Rabenhorst], achroa.— Differt ab aliis Microcoleis vagina distinetissime undulato-constricta, transversim striata. Crassi- tudo fascicul. 0010 millim., vagin: 0:016 millim. ; diameter artic. 0:003 millim. 5. ANABENA (CYLINDROSPERMUM) sp.? (Plate V. figs. 2-10.) From the filaments of this Anabæna a Chroococcus arises in a very interesting manner. We find here a curious case of vegetative polymorphism of the Cyanophyeeous Algae (Schizophycea). What some years before many conjectured as possible is by recent observers assertedas true, not only for the Cyanophyceous *, but for the Chlorophyllophyceoust Algæ, viz., that many, if not all, the unicellular species, and some of the composite species of the Cyanophyceous Algæ, and perhaps of the Chlorophyllo- phyceous, are merely stages in the life-history of higher plants. These states being fixed, the different forms have been defined and distinguished as different species. It will probably be finally found that some of the so-called species of Glwocapsa, Chroococcus, &c., or, on the other hand, of Cylindrocystis, Protococcus, &e., have their own peculiar species of the higher Algæ, from which they spring, and into which alone they cau develop. In the order of Cyanophyceous Algæ species of Glaocapsa are degraded by recent discoveries as subordinate forms of higher Cyanophyceous Algæ, as S’rosiphon, Oscillaria, Ke. Gleeocapsa Itzigsohnii, Bornet, for example, was proved to be a peeuliar state or development of Siroszphon Bornetit, Zopf t. Whilst no proof whatever has as yet been furnished for the vast majority of the plauts of this Order (Cyanophyeesw), with regard to the genus Anabena (Cylindrospermum), we will briefly characterize the interesting relation betweeu a minute Chroococcus and Glæothece and a species of Anabena (Cylindrospermum). The * W. Zopf, ‘Zur Morphologie der Spaltpflanzen (Spaltpilze und Spaltalgen),’ 1882. t Cienkowski, “Ueber Palmellen-Zustand bei Stigeocloniui,” Botanische Zeitung, xxiv. (1879), p. 17. For the further literature and for the Proto- coccus stage of Conferva, see Schaarsehmidt, “ Némely Chlorosporeák vegetativ alakváltozásairól," Magyar Növénytani Lapok, vii. (1883). t Z. e. p. 58, t. vii, figs. 1-9. DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGÆ. 243 filaments of this Anabena (Oylindrospermum) are coraposed of cylindrical or more or less globular cells. The spores are much larger than the ordinary cells, from which on their first appear- ance they are not readily distinguishable (Plate V. fig. 3a); but when the frond has attained a certain age, the spore-cells begin to enlarge, and finally assume a form and size apparently fixed ; the matured spores are cylindrica), rounded at the ends, and sur- rounded with distinct yellowish coats. The relative position of spores and heterocysts is various. The contents of the hetero- cyst is paler, the colour of the cell-wall yellowish. The filaments are seen in division in Plate V. fig. 2. The cell at first seems merely to elongate until it obtains nearly twice its original length, when the division commences. A median constriction divides the cells into two daughter cells, the filaments then assu- ming a moniliform aspect. The cells are closely connected, no nuclei being discernible as in the Noctoc-cells*. By this process the filament rapidly increases in length ; but at the ends the new joints become further separated until they are almost detached (Plate V. fig. 7a). The separated cells remain together and form irregular masses (Plate V. fig. 8) resembling Synechococcus. These changes were observed from dried material, but were traced from uninjured filaments, and remind us of the metamorphosis of Poly- pothrix amphibia, Zopf t. Besides these filaments we have seen many of Conferva-like appearance (Plate V. figs. 3 and 4), which are composed of cells three or four times as long as broad (Plate V. figs. 3 and 4), filaments with swollen yellowish cell-walls (Plate V. fig. 5), and filaments which are composed of thicker-coated cells (Plate V. figs. 6a & b). The cells of these filaments (figs. 5 and 6) sepa- rate in a similar manner as before described. The result of the fragmentation (at the ends of the filaments) is a Chroococeus or Gleothece-form (Plate V. figs. 9 and 10). The cells delineated in figs. 9 and 10 spring from filaments of the shape and zppear- ance of fig. 6. The Glaothece-form springs from the Chroococcus- * Sehaarsehmidt, ‘A chlorophyllés a nóvényi sejtmag morphologSjéhor (Contributions to the Morphology of the Chlorophyll and Vegetable Nucleus) 1881, p. 46, fig. 17. t “Weitere Stützen für meine Theorie von der Inconstanz der Spaltalgen (Phycochromaceen).” Berichte d. deutschen Bot. Gesellsch. i. n. 7 (1883), p. 319, t. ix. LINN. JOURN.— BOTANY, VOL. XXI. T 944 DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGA. form from the rounded cells (fig. 9) by repeated division of these cells. 6. Nosroc sp. ? Size: family 0:185 millim. ; cells 0:008 millim., by one third as long. 7. GrGornicHiA sp.? Insufficient material. Size: cells at the base of the filament 0:007-0:008 millim. by one half as long; resting spore 0:015 millim. x 0:012 millim. ; heterocyst 0:008 millim. BaACILLARIACER. 8. ENcYONEMA VENTRICOSUM (AKüfzing) Grunow; Van Heurck, ‘Synopsis des Diatomées de Belgique, t. ii. f. 15. Size 0:058 x 0'014 millim. Also “forma inter E. cespitosum et Lunula," Van Heurck, l. c. t. ii. f. 18. Size 0:033-0:046 millim. x 0:010-0:014 millim. In conjugation also. 9. STAURONEIS PucaNICENTERON, Ehrenberg ; Van Heurck,l. c. t. iv. f. 2. Size 0'070-0:078 millim. x 0:015-0:020 millim. 10. S. acuta, W. Smith, nov. f. tenuis; lateribus levissime concavis, area lineari, non dilatata. (Plate V. fig. 11.) Long. 0:100 millim., lat. max. 0:014 millim. 11. S. DILATATA, W. Smith, British Diatomacece, p. 60, t. xix. f. 191. Size 0'046 x 0'016 millim. 12. S. ANCEPS, Ehrenberg, nov. f. intermedia. Between the figs. 4 and 5 of Van Heurck, l. c. t. iv.; bridge parallel. Size 0:060 x 0:012 millim. nov. f. tenuicollis; sub polis valde constricta. (Plate V. fig. 12.) Long. 0040 millim., lat. max. 0:012 millim. 13. NavicuLA VIRIDIS, Kützing, var. commutata, Grunow, Van Heurck, Z. c. t. v. f. 6. nov. f. longior. Size 0'060 x 0'010 millim. 14. N. Brepissonu, Aiitzing; Van Heurck, l.c. t. vi. Size 0:060-0:098 millim. x 0:010-0:015 millim. 15. N. TanELLARIA (Ehrenberg, p. p.), Grunow. Size 0'070- 0:078 millim. x 0:011-0:020 millim. 16. N. APPENDICULATA, Aüfzing; Van Heurck, l. c.t. vi. f. 18. Size 0'038 x 0:008 millim. DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGA. 245 17. NAVIOULA OBLONGA, Küfzing; Van Heurck, l.c. t. vii. f. 1. Size 0:064 x 0:014 millim. 18. N. AwBrGvA, Ehrenberg; Van Heurck, l.c. t. xii. f. 5. Size 0°100-0°105 millim. x 0:022 millim. 19. N. timosa, Küfzing; ò. bicuneata, Grunow, in Verhandl. d. zool.-bot. Gesellsch. Wien, x. (1860), t. v. (3) f. 7. Size 0:045 x 0:014: millim. 20. GOMPHONEMA GRACILE, Ehrenberg, f. major, Grunow ; Van Heurck, Z. c. t. xxiv. f. 12. Size 0060 x 0'010 millim. 21. G. pnicmorowvwM, W. Smith (G. diehotomum, Kwtzing, P. p; Van Heurck, l. c. t. xxiv. f. 19, 20. Size 0:030 x 0008 millim. 22. Œ. PARvULUM (Küfzing), Grunow ; Van Heurck, l. c. t. xxv. f.9. Size 0:022 x 0'008 millim. 23. SYNEDRA sp.? 24. FRAGILARIA sp.? Size 0052 x 0004-0005 millim. 25. HawrzscuiA Amputoxys (Ehrenberg), Grunow; Van Heurck, l. c. t. lvi. f. 1l. Sáfed-koh, on the roots of Anemone tetrasepala, Royle. Var. vivax, Grunow; Van Heurck, l. c. t. lvi. f 6. — Size 0:100 x 0:015 millim. 26. NriTZsCHIA VERMICULARIS (Kitzing), Hantzsch; Van Heurck, l. c. t. lxiv. f. 2. Size 0135 x 0012 millim. 27. SuninAYA ANGUSTA, Kiitzing; Van Heurck, l. c. t. lxxii. f. 18. Size 0'038 x 0011 millim. 28. SURIRAYA sp. ? DESMIDIACEX. 29. EvasrnUM SPINULOSUM, Delponte, subsp. inermis, Nord- stedt, De Algis et Characeis, p. 9, t. i. f. 17. Var. nov. OLIVERI; lobis tumoribus et verrucis in circulo dis- positis munitis. (Plate V. fig. 18.) Long. 0:048-0-058 millim., lat. 0*040-0:050 millim., lat. isthmi 0'016 millim. 30. Cosmarium Botrytis (Bory), Meneghini. Var. nov. ar@nantcum. (Plate V. fig. 19.) Membrana tota Verrucis minutissimis in series regulares peripherias et radiantes T 2 946 DB. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGÆ. concentricas ordinatis munita. Long. 0'056 millim., lat. 0'044 millim. 31. COSMARIUM PULCHERRIMUM, Nordstedt, in Warming’s Symbole ad floram Brasilia, p. 175, t. ii. f. 24. Also in division. Size 0'050 x 0:040 millim. 32. C. uNpULATUM, Corda ; Wittrock, Anteckningar om Skan- dinaviens Desmidiacéer, p. 11, f. 3. Var. nov. ogNATUM. (Plate V. fig. 23.) Membrana tota puuctis distinctis in series radiantes ordinatis munita. Long. 0'020 millim., lat. 0:013 millim. 33. C. AITCHISONII, n. sp. (Plate V. fig. 20.) Cosmarium parvum, subhexagonum, tertia parte longius quam latius, pro- fundissime constrictum, sinu lineari; semicellule subtrapezice, sursum angustate, dorso subrotundate, lateribus subconvexis, angustis inferioribus truncato-rotundatis, sub dorso tumore in- structis. Membrana achroa, glabra. Long. 0:032 millim., lat. 0:024: millim. Differt a C. nitidulo, De Notaris (Nordstedt, Desm. Ital. t. xii. f. 10) seimicellulis late rotundato-trapezoideis, angulis inferiori- bus valde rotundatis, dorso tumore magno instructis. 34. C. HookEnr, n. sp. (Plate V. fig. 21.) Cosmarium par- vum, paullo longius quam latius, medio profundissime constrictum, sinu lineari, semicellulæ reniformes, a vertice visæ oblongo- ellipticæ, dorso 2-3 verrucis acute conicis munitæ. Membrana achroa, glabra. Differt a C. Phaseolo, Brébisson (Ralfs's * British Desmidiace:e,' p. 106, t. xxxviii. f. 5), membrana glabra, semicellulis vertice oblongo-ellipticis, medio non tumidis, dorso verrucis conicis 2-3 munitis. 35. C. OurveRt, n. sp. (Plate V. fig. 22.) Cosmarium par- vum, tertia parte longius quam latius, ellipticum, medio profunde constrictum, sinu lineari angustissime ; semicellule a basi recta magis magisque angustate, apice valde rotundate, lateribus leviter convexis, angulis inferioribus subrectis, a vertice vise oblongo- elliptice, medio inflate, breviter cuspidate. Nuclei amylacei singuli. Membrana achroa, glabra. Long. 0'040 millim., lat. 0:031 millim. Differt a C. microsphincto, Nordstedt (Desm. Ital. p. 33, t. xii. DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGAX. 247 f. 9), et a C. microsphincto, Nordstedt, f. parvula, Wille (Fersk- vandsalger fra Novaja Semlja, p. 38, t. xii. f. 22), precipue semi- cellulis vertice oblongo ellipticis medio breviter cuspidatis. 36. CosMARIUM PYRAMIDATUM (Ralfs), Brébisson ; De Notaris, Elementi per lo studio delle Desmidiacee Italiche, p. 40, t. iii. f. 22. Size 0040 x 0030 millim. 37. C. GRANATUM, Brébisson; Ralfs, British Desmidiacee, p. 96, t. xxxii. f. 6. Size 0:040 x 0'030 millim. Var. ELONGATUM, Nordstedt, Desin. Spetsberg. t. vi. f. 6. With zygospores. Size 0:036 x 0'024 millim. ; zygospore 0:028 millim. 38. C. ABRUPTUM, Lundell, De Desmidiaceis que in Suecia invente sunt observationes critice, p. 43, t. ii. f. 22. Forma nov. simplex; semicellulis verrucis centralibus desti- tutis. (Plate V. fig. 24.) Long. 0:019 millim., lat. 0'015 millim. 39. C. Menecurnit (Meneghini), Brébisson; De Bary, Conju- gaten, t. vi. f. 46. Size 0015 x 0'011 millim. 40. C. MINUTUM, Delponte, Specimen Desmid. Subalp. ii. p. 105, t. vii. f. 37. Size 0:028 x 0020 millim. 41. CrosrERIUM Cornu, Ehrenberg; Ralfs, British Desmi- diacec, p. 176, t. xxx. f. 6. Size 0'120 x 0'006 millim. 42. DESMIDIUM QUADRATUM (Delponte), mihi; Delponte, Spe- cimen Desmid. Subalp. i. p. 63, t. iii. f. 20-23. Var. nov. EXCAYATUM ; semicellule vertice vise duobus lateribus excavatis; teniis chlorophyllaceis quattuor. (Plate V. fig. 25.) Long. cell. 0:023 millim., lat. 0'024 millim. ZYGNEMACER. 43. Movarorra sp.? The conjugated filaments of this Mou- geotia were soaked for a few minutes in warm water. By this treatment the cell-walls of those either in conjugation or about to become so, swelled in such a manner as to occupy the cell- cavity, and compress the dry contents of the cells, the primor- dial utricle, the irregular chlorophyll, and imbedded starch- granules (Plate V. fig. 13). The vegetative cylindrieal cells exhibited no alteration (Plate V. fig. 14). u When the filament is stained with iodine (in alcohol), it 1s seen that the swollen cell-wall is distinctly composed of layers ; by the action of alcohol (extracting the water) the,layers of the 248 DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALG. cell-wall will be contracted. We see, then, that the cell-wall is composed of three parts (Plate V. figs. 15 and 16) [exclu- ding the common coat of the cells, the cuticle by which all the tells of the filament are uninterruptedly covered (Plate V. fig. 16 a)], viz., the inner and outer thickened portion (Plate V. figs. 15a, b, and 165, c), and between these the exceedingly translucent and easily seen portion. By this simple method I have found a constant difference between the fertile and sterile cells of this Mougeotia. Size: vegetative cells 0:020 millim., 13-5 times as long ; zygo- spore (Plate V. fig. 17) 0:028 x 0:036 millim. 44. SPIROGYRA MIRABILIS (Hassall), Kützing; Petit, Spirogyra des Environs de Paris, p. 14. t. ii. f. 4. Some zygospores are cylindrical and rounded at the ends as in S. lutetiana, Petit. Size: zygospore 0:084 x 0'032 millim. (1 : 23). 45. S. porticatis (O. F. Müller); Cleve, Svenska arterna af Algfamiljen Zygnemacee, p. 22, t. v. f£. 13; Petit, l. c. p. 21, t. v. 9. Size: zygospore 0:056 x 0°028 millim. (1 : 2). 46. S. PUNCTATA, Cleve, Svenska arterna af Algf. Zygnemacee, p. 23, t. vi. f. 4. Size: zygospore 0°065-0°075 millim. x 0*030— 0:033 millim. (1: 2). PALMELLACEX. 47. PLEURococous MUCOSUS, Kiitzing ; Cooke, British Fresh- water Alge, ii. p. 4,t. ii. f. 9. Size: cells 0:003 millim. 48. DacryLococcts INFUSIONUM, Nägeli in Gattungen einzell. Algen, p. 85, t. iii. f. F. Size: cells 0'009 x 0:004 millim. 49. Oocystis NaorLr, Alex. Braun. —Eight-celled family. Size: family 0:036 millim. ; cells 0*018 x 0:009 millim. 50. GrocvsrIS vESICULOSA, Nägeli; Cooke, British Fresh- water Alge, i.p. 7,t. ii. f. 2. Size: family 0°050-0-070 millim. ; cells 0:007-0:010 millim. PROoTOCOCCACEX. 51. PnoTococcus sp.? Size: cells 0'036 millim. 52. PoryEpRiUM MINIMUM, Alex. Braun, Lagerheim Bidrag till Sveriges Algflora, p. 69, t. i. f.27. Size: cells 0:006 millim. DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGJE. 249 53. SCENEDESMUS QUADRICAUDA (Turpin), Brébisson; Schaar- schmidt, Specimen phycologie ZEquatoriensis, p. 14. Var. a. &oonwis (Ehrenberg), Ralfs, British Desmidiacee, p. 190, t. xxxi. f. 12,7. Size: cells 0010 x 0:002 millim. VoLVOCACEA. 54. Panportna Morum (O. F. Müller), Bory. Two very small eight-celled families, with 1—4 (two-four-eight-celled) daughter families. Size: family 0:035-0-040 millim.; cells 0°007-0°008 millim. CoNFERVACEZ. 55. CoNFERYA sp.? Size: cells 0007—0:008 x 0'004 millim. CEDOGONTACER. 56. CEpogontum LONGICOLLE, Nordstedt, Alge aque dule. et Charac. Sandvic. p. 20, t. ii. f. 11-12. Var. SENEGALENSE, JVordstedt, De Algis nonnullis Mus. Lugd. Bat. p. 13,t. i. f. 23. Forma nov. afghanicum ; oogoniis 2-7 continuis. (Plate V. fig. 27.) Size: vegetative cells 0:007—0*008 millim.,three times aslong; oogonia 0:022 x 0:024 millim. ; oospore 0:021 x 0019 millim. 57. (E. Prinesnermu, Cramer; Cooke, British Freshwater Alga, v. p. 166, t. lxiii. f. 2. Size: female cells 0030 millim. 2-21 times as long; male cells 0:023 millim., 2-23 times as long ; oogonia 0°045-0:065 millim. ; oospore 0038-0049 millim. 58. BULBOCHÆTE PYGMUA, Pringsheim in Jahrbücher f. wiss. Bot. i. 1. (1857), p. 74, t. vi. f£. 10. Size: cells 0°016 millim. by twice as long; oogonia 0:024 x 0'044 millim. CoLEOCH ATACES.. 59. COLEOCHÆTE scurata, Brébisson, Pringsheim Jahrbücher f. wiss. Bot. ii. 1. (1860), p. 35, t. iii. f. 3. With corticated and matured carpogonia. Size: plant 0:050-0160 x 0:060—0:200 millim. ; carpogonia 0:070 x 0'080 millim. ; cells 0'015 millim. by 1-2 times as long. CHARACE E. 60. Cnama sp. Chlorophyll-granules in division. only, not sufficient for identification of the species. For the Fragments 250 DR. JULIUS SCHAARSCHMIDT ON AFGHANISTAN ALGZ. description of the ehlorophyll-granules, cfr. Schaarschmidt's * Re- searches on the Division of the Chlorophyll-granules, and upon the Occurrence of Hypochlorin in the Cyanophycew and Bacilla- riacez," Nature, xxvi. 1882, No. 672, pp. 476-477. DESCRIPTION OF PLATE V. Fig. 1. Microcoleus Aitchisonii, n. sp. Xx 1500. 2, Anabena (Cylindrospermum), filament, cells in division. X 1000. 9 & 4. Conferva-like filaments. x 1000. 5. Filament, with swelled yellowish cell-wall. x 1000. . Thicker-coated filaments. x 1000. - The cells at the end of the filament (of a) separating. x 1000. - The separated cells remaining close together and forming a Synecho- coccus-like family. x 1000. . Chroococeus-like cells, which are derived from the filaments, fig. 6. x 1000. 10. Gleothece-like cells, springing from the former (fig. 9) cells by divi- sion. x 1000. ll. Stauroneis acuta, Ehrenberg, n. f. tenuis. x 400. 12. S. anceps, Ehrenberg, n. f. cenuicollis. x 870. 13. Mougeotia sp.? Oells in conjugation, soaked for a few minutes in warm water. The cell-walls of the conjugated or cells ready for conjugation swollen so as to fill out the inner cavity of the cells. x 600. 14. The vegetative cells remain unaltered. x 600. 15-16. The cells after treatment with warm water stained with iodine (in alcohol); the cell-wall is, as seen, composed of layers. Two of the layers (the outer and the inner) are thicker (fig. 15a, 5; fig. 165, c). The cuticula (fig. 16 a) is the common coat of the cells. x 600. 17. Matured zygospore. x 600. 18. Euastrum spinulosum, Delponte, * inermis, Nordstedt, n. var. Oliveri. x 600. 19. Cosmarium Botrytis (Bory), Meneghini, n. var. afghanicum. x 600. 20. —— Aitchisoni,n. sp. x 760. 21. —— Hookeri, n. sp. x 760. 22, —— Oliveri, n. sp. X600. 23. —— undulatum, Corda, n. var. ornatum. x 760. 24, — abruptum, Lundell, n. f. simplex. x 1000, 25. Desmidium quadratum (Delponte), n. var, excavatum. End view, . . Xx 600. 26. Edogonium longicolle, Nordstedt, var. senegalense, Nordstedt, n. f. afghanicum. Xx 870. ano eo POLLEN FROM FUNEREAL GARLANDS IN AN EGYPTIAN TOMB. 251 On some Pollen from Funereal Garlands found in an Ec yptian Tomb, circa s.c. 1000. By Cuartes FnEDERICK Wu E, F.LS. [Read 3rd April, 1884.] (PLATE VI.) Tues drawings of the pollen of Papaver Rhœas I have been enabled to make by the kindness of Sir Joseph Hooker, who pro- cured for me a few anthers from the funereal garlands from the cofin of the Princess Nzi Khonson, of the XXI. Dynasty, about B.C. 1000, from Egypt, of which several were lately exhibited here by Mr. Dyer. And on the same sheet I have added copies of previous drawings from recent gatherings of the plant. The former are generally slightly larger, and somewhat more tapering towards the ends; the anthers also are larger. These latter are hard and brittle, and of a blackish-red colour, as though charred. Perhaps there is no family in which the size of the pollen-grain varies more than in that of the Papaveracew, nor in which the proportionate number of aborted or malformed grains is greater. And this character appears to belong to the ancient as well as to the modern. Still the grains of normal size and shape are not only larger, but more regular in their markings,— certainly cleaner and brighter in colour than some that I have endeavoured to pre- serve on slides for the microscope, collected only a few years ago. The point, however, to which I desire especially to call atten- tion is the readiness with which these minute objects (presum- ably deprived of all moisture for so long a period) absorb water, and expand into that subspherical shape so usual with pollen of simple form, with the peculiarity that the Egyptian assume the three-lobed shape common to many pollens, the furrows becoming deeper than when dry, instead of, as generally happens, being nearly obliterated when placed in water. No indication of the appearance of the pollen-tubes can be detected, excepting that in several grains I observed at one of the three points at which they would be produced a small bubble of air. Imay mention that the characters of the spiral tissue of the inner lining of the anther can be, when in water, sufficiently made out. DESCRIPTION OF PLATE VI. win of th Fig. 1. Pollen-grains of the Poppy (Papaver Rheas) from A co 7 " e Egyptian Princess Nzi Khonson, XXI. Dynasty, s.c. 1000. . 2. Some of the same, immersed in water. X 500. 3. The recent pollen-grains of P. Blew. x 50D. x 500 4. Some recent pollen-grains immersed in water. 2 5. An anther of the Poppy from the coffin (B.C. 1000) above mentioned. X50. 6. An anther from a Poppy grown in 1883, x 00. 252 MR. C. B. CLARKE ON } / Notes on the Flora of Parasnath, a Mountain /North-western Bengal, in a Letter from C. B. Cranxk&, F.R.S., F.L.S., to, and with an Introductory Note by, Sir J. D. Hooker, K.C.S.I., F.R.S. [Read 19th June, 1884, ] [Tne mountain, to the vegetation of which the following pages are devoted, is one of the most interesting in trans-Gangetic India from its isolation and position. It stands on the north-eastern extreme of the elevated region which forms the high land of the Deccan, and is prolonged northward to the bend of the Ganges , from which the descent is abrupt to the plains of the Gan- getic valley, which again extend northward uninterruptedly to the foot of the Himalaya. The latitude of Parasnath is nearly 24? N.; longitude 86° 6' E.; its distance from Calcutta is 122 miles in a north-eastern direction, and it is 88 miles distant from the Ganges north of it. Its height is about 4500 feet. The vegetation of Parasnath is that of the Deccan, with a slight admixture of Himalayan and Malayan types. Of the former of these there are fewer than might have been anticipated from its elevation and position, but not fewer than may well be accounted for by the heat and dryness of the surrounding country. The first described ascent made of Parasnath was, as far as I am aware, my own in February 1848, as narrated in the Journal of the Asiatic Society of Bengal (vol. xvii. pt. 2, p. 855), and subse- quently in my Himalayan Journals (Ed. i. vol. i. p. 18), when I collected or observed about 300 species of plants. Its botany has since been investigated successively by Drs. Thomson, Anderson, and King, when Superintendents of the Calcutta Botanical Gardens, but no complete account of its flora has been published —J. D. Hooxzn.] Hazaribagh, Oct. 12th, 1883. My pear Sra J. D. HOOKER, I ascended Parasnath on 6th October; and you may be interested to get some modern news of a mountain which you first made known to the European scientific public. The road you made the ascent by is still sound, but covered with tall grasses and disused. There is a railway station now only 20 THE FLORA OF PARASNATH. 258 miles from Mudobun at the northern base of the mountain; and from Mudobun to the summit there is a road (6 miles) which you could drive up in a hansom. Along the north base of the mountain is a tea-plantation; but Parasnath itself is not, I think, likely to be encroached on by planters for many years to come, nor has it yet been injured botanically by visitors. As my visit was a short one, I did not attempt to make a general col- lection of plants: I confined myself mostly to the Monocotyledons, the Glumacee in particular; but there are some Dicotyledons which, though well known, it is hardly possible to keep one’s hands off, such as the Begonia, the Sonerila, the Exacum, all abundant. There is only one plant on Parasnath which I cannot get elsewhere (so far as I know), the pink Knozia brachycarpa, R. Br., which you got here, and which is, I believe, more plentiful in Burma ; I collected a stock of it, and find it produces a very thick woody root, bright yellow when cut across. I collected (which I had not got before on Parasnath) the common orange- tipped ZEschynanthus ; and the leaves only of a very large Hoya. I got up into two trees but could find neither flower nor any remains of fruit: I suppose it flowers in April. On the central summit (which is 200 feet lower than the western, a. e. is 4300 feet about) there is left a cluster of the problematical Phania (acaulis ?): I measured the stem of one, which is 14 feet from the ground to the lowest leaf, and nearly uniformly 34 inches in cir- cumference for its whole length: there are other stems 8-10 feet long. I endeavoured to collect a good set of the large Grasses along the ridge—Oymbopogon, Anthistiria, Pollinia, &c. Living on these I found (not a tree-frog but) a grass-frog ; a very slaty- dirt coloured species. "When I coaxed him downwards he seemed to have a horror of terra firma; he jumped about on the Anthi- stirias, holding the culm with his two fore feet, and swinging on them like a bird on a twig. In this season the summit of Parasnath had not one drop of dew in the morning, while at 3000 feet the dew was plentiful : this difference, as well as the dif- ference in altitude, will account for the number of plants that grow along the crest of Parasnath but will not descend. MP a Carex I never got here before—a very slender culm, wit " remote female grass-green spikes; it 1s very nearly (or quite) the | same as the low-level Carex which grows 1n the Teesta gorge in Sikkim. I kept my eye on the Gingers, as they are numerous 1n species, and abundant in individuals here. The large Amomum 254 MR. C. B. CLARKE ON was in full flower at 4100 feet, with leaves 5 feet long and more (the lamina 82 by 11 inches), the flower-spikes very large, white, the rhizome gamboge-yellow when cut across. It is a very striking plant; but, comparing it with the medium-sized common white Amomum of the 1500-feet level, I could find no specific difference. The Hedychium common here is now (October) very shabby, but is throwing up young stems to flower next April. These are very red, with the young leaves red and seale-like, and so exceedingly like the flowering (leafless) stems of the Zingiber that (the two plants growing mixed) it is difficult, without taking the plant in the hand, to say whether you have got a young stem of Hedychium or a leafless flower-stem of Ginger. Is the leafless flower-stem of Ginger really only an ordinary stem gradually forced abnormally early into flower by a secular change in climate? There was a fine Hlettaria in fruit, which is nearly or quite the same as the one that grows at Khursiong in Sikkim. I saw three ground Orchises in fruit: I have never seen on Parasnath any epiphytie Orchis; the air seems too dry, though there is Vanda in abundance on the Mohwa trees (Bassia) at the base, as you found it. In spite of the dryness, there are a good many Ferns even near the summit. Asplenium (Drepanophyllum is the name) abounds, while the nearly allied common A. (Athyrium) macrocarpum is absent. Parasnath is the north-east limit of the A. (Drepanophyllum) Hohenackerianum, which is, I believe, abundant in Central India. I also collected several specimens of Ophioglossum vulgatum: I say vulgatum, relying on Luerssen, who has (I doubt not correctly) run a number of Ophioglossums into that species. The Indian Ophioglossums are not less in- teresting considered as forms than considered as species: this Parasnath form is, however, much more like the common English Ophioglossum than either the Sikkim or Levinge’s Kashnior one. It is 15 years ago since I first ascended Parasnath, and I see wonderfully little difference in the vegetation: there are the same plants, in the same relative degrees of abundance, and I can hardly pick out one that I can say has either advanced or retreated 100 yards, so nicely are the forces balanced. There is the Tha- lictrum, the Geranium, the Barberry, the Senecio, &., on the ridge exactly where you found them, and, as of old, the giant scandent Bauhinia ascending a few feet higher than any of them. THE FLORA OF PARASNATH. 255 The Himalayas and Khasia rather spoil one for a small thing like Parasnath; but Parasnath is really very rich ; the number of irees that flourish at or above 4000 feet is remarkable, and some of them, as Grewia asiatica, appear to grow higher up at this elevation than elsewhere. I noticed that Heptapleuron com- mences its life here as a scandent epiphyte, but subsequently reaching the ground, it grows to a large size as a tree, and shows no sign of its early history. Other Araliads do this in Sikkim, thus imitating Wightia and several of the Figs. I attempted photographs on Parasnatb, but I had only some gelatine plates that have been through the rains at Caleutta, and I cannot tell how they may develop, possibly not at all. By the north railway Darjeeling is now so quickly and easily reached that all the pressure of Calcutta visitors is taken off Parasnath, which is chiefly visited from the neighbourhood. The coal-mines here giveemployment to a large English population, and the East-Indian Railway has some of its largest works not far off, so that the Bungalow at 4200 feet is still used: indeed I met a lady who went up for the whole hot weather, nearly two months. The air is perfect on Parasnath, and the change sufficient for the Englishman from the plains who, if he goes to 7000 feet at Dar- jeeling, may there get “chill,” or “fever,” or “ hill-dysentery." Parasnath is perfectly healthy ; and the succession of visitors still bless Sir Œ. Campbell, who purchased the bungalow (very cheaply) with the express purpose that Europeans might recruit there. I collected at about 2500 feet alt., on Parasnath, the remark- able Rice we know (perhaps wrongly) as Oryza coarctata, Roxb. It has the hard, horny, almost woody root, and the broad plicate ribbed leaves that so many of the rocky-wood inhabiting grasses possess ; but the flowers and fruit are completely Oryza. Indeed the natives tell me that it is a high-class food-rice, and that though so scanty and troublesome to collect, they sometimes set their children to collect a little, as exceptionally good in quality. Yours very truly, C. B. CLAREE. 256 MR. ER. A. ROLFE ON X NEW GENUS OF On Hyalocalyz, a ngw Genus of Turneracee from Madagascar. By R. A. Rois, Herbarium, Royal Gardens, Kew. (Com- municated by Prof. D. Oxrver, F.R.S., F.L.S.) [Read 17th April, 1884.] (Puate VII.) Tue Order Turneracee has recently formed the subject of an excellent monograph by Dr. Ignatius Urban, of Berlin. In this monograph he has shown that the Order consists of 83 species belonging to five genera. Of these Zurnera has 54 species, all American, and ranging from Mexico to the Argentine republic ; Wormskioldia with 7 species, and Streptopetalum with 2, are con- fined to Africa, where they range from Abyssinia to Mozambique ; 2 species of the former genus being confined to the small island of Zanzibar; the remarkable monotypic genus Mathurina is peculiar to the island of Rodriguez; while Piriqueta has 19 species, of which 16 are American, and ranging from North Carolina to the Argentine Republic, one is confined to a limited region at the Cape of Good Hope, and the remaining two are limited to North Madagascar. An additional genus may now be added from the small island of Nossi-bé, on the north-west of Madagascar. I am indebted to Dr. Vatke, of Berlin, for the opportunity of examining and describing this plant, which was collected by the late Dr. Christian Rutenberg, and is preserved in the Herbarium at Bremen. A careful and leisurely examination of this plant has shown one or two peculiar characters which seem to require a new genus for its reception. I have therefore made a drawing of it, which, with the accompanying description, I have pleasure in offering to the Linnean Society. Following Urban’s arrangement, the position of this genus is between Mathurina and Turnera, with both of which it agrees in the absence of a corona at the base of the petals. In habit it is very similar to Turnera, being a small herb with erect flowers, which are solitary in the axils of the leaves; the styles, seeds, and arillus are also very similar. The only approach to Mathu- rina is in the subhypogynous insertion of the petals and stamens. The flower (as far as I know) is by far the smallest in the Order, as it is only two lines long and very slender; it is also peculiar in having a number of long white hairs on the pedunele. TURNERACEH FROM MADAGASCAR. 257 The calyx is, perhaps, its most remarkable character, and it was only by careful dissection in water that I was able to understand it, The name I propose explains its peculiarity ; instead of the usual herbaceous calyx, we have a membranous one, hyaline, and without chlorophyll; it is tubular, cleft to the middle into five, quincuncially arranged, concave oblong segments, close to the apex of which are two diverging set», quite a fourth of the calyx in length, and bent back from it at an angle of 45°. There are three faint nerves on each segment to the base of the calyx, at which point the petals and stamens are inserted (as shown in the drawing). The ovary contains nine ovules, three on each placenta, and apparently arranged in a single series ; the number may not be constant, and a few of the ovules do not develop. The fruit is altogether anomalous in the Order: after the flower withers the short peduncle elongates considerably ; but the side nearest to the branch grows more rapidly than the opposite one, and the capsule becomes inverted, and closely appressed to the eurved peduncle (fig. 10), the long white bristles giving it a singular appearance. The absence of chlorophyll in the calyx, and its hyaline cha- racter, seem to be very easily accounted for. The flowers occur at the ends of the branches, where the leaves are much congested and sessile; these upper leaves are clothed at the base with a quantity of long whitish hairs, amongst which the minute flowers are developed. The light is evidently prevented from reaching them in this position, so much so that the greer colour is absent in the bases of these upper leaves. It is probable that the elongation of the peduncle after flowering is an attempt on the i he fruit under the full influence part of the plant to bring t of the light, and thus assist its development. It would be interesting to watch this development under cultivation. HyaLooaLyx, gen. nov. Sepala ad medium in tubum subcylindraceum coalita, aliter sime hyalino-membranacea, inconspicue trinervia, quincunch » imbricata, apice bisetifera. Petala ima basi calycis inser A e perigyna, inferne cuneata, nuda. Stamina thie breves calycis inserta ; filamenta linearia, basi dilatata; ant ere E cordato-ovoideæ, apice leviter mucronulatze. Ovarium Nn ” glabrum; styli 3, recti, filiformi, glabri, apice breviter Habe 258 ON A NEW GENUS OF TURNERACE A. multiparti ; placentæ 8, 3-ovulate, ovulis uniseriatis, funiculis longiusculis inserto. Pedunculus fructiferus auctus, apice arcte incurvato. Fructus invertus, levis, pene ad basin dehiscens. Semina oblongo-obovoidea, in hilum subito contracta, curvata ; testa reticulato-striata, striis elevatis; arillus unilateralis semen dimidium squans, tenuiter membranaceus margine integro. H. sETIFERUS, sp. unica.— Herba parva, ramosa; rami tereti, hirsuti, pilis simplicibus adpressis. Folia alterna, inferne sparsa, ad apices ramorum conferta, breviter petiolata, elliptica v. lan- ceolato-elliptica, acutiuscula, crenato-serrulata, basi attenuata, integra, eglandulosa, utrinque strigoso-hirsuta. Flores solitarii in axillis foliorum congestorum ad apices ramorum, erecti, breviter peduneulati, 2 lin. longi; pedunculo dense setoso pilis albidis. Fructus 1j lin. longus, pedunculo fructifero 2} lin. longo. In insula Nossi-bé, Madagascariensis ; Rutenberg! DESCRIPTION OF PLATE VII. Hyalocalyx setiferus, Rolfe, nov. gen. et sp. Fig. 1. Branch, of natural size. 2. Unexpanded flower; showing the short setose flowering peduncle and the peculiar setiform calycine appendages, x4 diam. 3. Bract removed, x 2 diam. 4. Part of expanded flower; showing the position and relative lengths of the parts (two of the stamens in each flower are a little longer than the other three, but a little shorter than the styles*); x4 diam. 5. The delicate. hyaline membranous calyx laid open; showing the apical appendages and the insertion of the petals (a), and stamens (5), x8 diam. 6. Petal, x8 diam. 7. Stamen, x16 diam. 8. Ovary and styles, x 16 diam. 9. Ovule with funieulus, x 16 diam. 0. Fruit, showing the elongated curved setose fruiting peduncle with inverted capsule, x 8 diam, 1l. Capsule after dehiscence, showing the placentas and insertion of the seeds, x 4 diam. 12. Seed, showing the reticulate-striate testa and the unilateral mem- branous arillus, x8 diam. * The species will probably prove to be dimorphous, like many others in the Order : in this case our specimen will represent the long-styled form, REY. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 269. On the Algo-Lichen Hypofhesis. By the Rev. Jamrs M. Crows1, F.L.S. [Read 12th April, 1884.] (Prates VIII. & IX.) I NEED scarcely say that one of the most interesting biological problems of the day, in so far at least as relates to Vegetable Morphology and Physiology, is connected with the theory usually known under this appellation, or, as it might more correctly be called, the Algo-Fungal-Lichen hypothesis. This is sufficiently proved by the now very extensive literature of the subject and by the various discussions to which it has given rise. To give any detailed résumé of that literature, or to enter minutely into these discussions, is as foreign to my present purpose as it is unnecessary for the elucidation of the subject. It will suffice that I bring under your notice, as succinctly as is consistent with a due understanding of the question, the presumed grounds on which the hypothesis rests and the ascertained facts by which it is subverted. Now the problem to be solved is—What is the origin of the lichen-gonidia, and in what relation do they stand to the thallus ? In premicroscopical days, as might be expected from the nature of the case, nothing on these points was either known or written. Indeed, by the earlier lichenists the gonidia were scarcely, if at all, distinguished ; and it was not until 1825 that they were first distinctly recognized by Wallroth (* Naturgeschichte der Flech- ten’), who supposed that they were the asexual reproductive organs of Lichens, and so gave them their distinctive appella- tion. Even in Koerber's Dissertation * De Gonidiis Lichenum j (1839), in which they are treated more fully and accurately than in any previous work, nothing is adduced as to their genesis or their relation (except in the case of soredia) to the other elements of the thallus. The first who gave any explanation of the matter was Bayrhoffer, who, in 1851 (‘Einiges über die Lichenen und deren Befruchtung’) stated that “the threads of the fibrous stratum” swell at the top, which swellings after- wards become “male gonidia." This was subsequently con- firmed by Speerschneider (Bot. Zeit. 1853 &e.), and supported (with a slight modification) by Schwendener (in Nageli, Bei- träge zur Wissen. Bot. ii. 1859, p. 125, t. i. fig. 18, t. v. fig. 6), LINN. JOURN.— BOTANY, VOL. XXI. U 960 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. as also by De Bary (‘Morphologie und Physiologie der Pilze Flechten, &c., 1865). According to this theory the gonidia originate from the hyphz in the manner succinctly described by Th. M. Fries, in Scand. (p. 7), who embraces and defends it as the result of his own observations. “The hyphz," he says, “are not only elongated into filaments, but they also put forth short branchlets. The terminal cell of such branchlet is gradually dilated, becomes subglobose, and is at length filled with chloro- phyll (or a subsimilar coloured matter); in a few that cellule is changed into a gonidium, and then, variously divided, originates other gonidia.” This view of the genesis of the gonidia from the hyphe, which was quite erroneous, was for some time generally accepted with- out further inquiry ; although, as we shall afterwards see, there were certain other and different observations made previously to those of Speerschneider, which were either entirely overlooked or not rightly understood. In 1868, however, Prof. Sehwen- dener, reviewing the original notion on this subject, towards the end of a paper entitled * Untersuchungen über den Flechten- thallus,” and more especially in a subsequent treatise, ‘ Die Algen-typen der Flechten Gonidia' (1869), rightly affirms that the actual development of a gonidium from the terminal cell of à hypha had not with certainty been observed, but only assumed by authors. Accordingly he enunciated an entirely new theory on the subject, which was evidently based upon a suggestion previously made by De Bary. This latter celebrated crypto- gamist, in his paper already referred to (p. 291), observing the resemblance between the gonidia of some Collemacei and certain of the lower Algs, came to the following conclusion :—“ Either the Lichens in question," he says, “are the perfectly developed states of plants whose imperfectly developed forms have hitherto stood amongst Algæ as Nostocaceew and Chroococcacesm, or the Nostocacex and the Chroococcacee are typical Algae which assume the form of Collema, Ephebe, &c., through certain parasitic Asco- mycetes penetrating into them, spreading their mycelia into the continuously growing thallus, and frequently becoming attached to their phycochrome-bearing cells." Taking the latter of these alternatives as his starting-point, Schwendener, commencing with the Collemaeei, made various observations and experiments, which were afterwards extended to other tribes and genera, the result of which led him to the REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 261 conclusion that the lichen-gonidia are unicellular or filamentose Algæ, and the lichen-thallus a parasitic Fungus. The various Algal types whieh he regards as constituting the gonidia he includes under two groups, viz. the * Phycochromacex ” and the “ Chlorophyllacez," so called from the colour of their respective cell-contents. To the former group, that with bluish-green cellules, he assigned five Algal types, viz. :—1. Sirosiphonee ; 2. Rivulariz ; 3. Seytonemes ; 4. Nostochacee ; 5. Chroococcacem: and to the latter group, that with chlorophyll-green cellules, he assigns three Algal types, viz.:—6. Confervacee; 7. Chroolepidex; 8. Palmellaceze. Moreover, failing in his researches to observe any genetic connexion between the hyphe and the gonidia, and unable otherwise to account for the presence of the gonidia in the lichen-thallus, he propounded the theory now so well known as — the “Schwendenerian hypothesis." This, briefly stated in his own words, is as follows :—“ As the result of my researches,” he says, “ all these (lichen) growths are not simple plants—not individuals in the usual sense of the term ; they are rather colo- nies, which consist of hundreds and thousands of individuals, of which, however, only one acts as master, while the others, in per- petual eaptivity, provide nourishment for themselves and their master. This master is a fungus of the order Ascomycetes, a parasite which is accustomed to live upon the work of others ; its slaves are green algals, which it has sought out or indeed caught hold of, and forced into its service. It surrounds them, as a spider does its prey, with a fibrous net of narrow meshes, whieh is gradually converted into an impenetrable covering. While, however, the spider sucks its prey and leaves it lying dead, the fungus incites the Algæ taken in its web to more rapid activity, nay, to more vigorous increase.” In the conclusion of the latter paper of Schwendener, above cited, he argues that the algal nature of the lichen-gonidia, which he maintains has been established in the cases he has reviewed, is extremely probable also in every other case, and that consequently the gonidium is not, as previously supposed, a self. developed organ of the nd But though not thus to be regarded, gonidia, he says, would sti is in a physiological point of view, remain as instruments of assi- milation and of asexual increase. For, although they have not the power in themselves to form a thallus, they are an essential constituent of it, and are undeniably the most important, though not the only, ministers of nutrition to the composite plant called v2 262 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. a “lichen,” inasmuch as this also is partly furnished by means of the substratum. Passing over two other and subsequent papers of Schwendener, viz. * Erórterungen zur Gonidienfrage " (in ‘ Flora, 1872) and “Die Flechten als Parasiten der Algen" (in Verhandl. der Naturf. Gesellschaft in Basel, 1873), in which, after replying to several objections that had been adduced by Krempelhuber * and Th. M. Fries, he maintains and further illustrates the hypothesis, though without adducing any absolutely new arguments, we now come to notice the contributions made to the theory by Dr. E. Bornet, in some respects the ablest and most plausible of its advocates. In a memoir entitled * Recherches sur les Gonidies des Lichens" (in Ann. des Sci. Nat. 5° sér. t. xvii., 1873), he accepts in its full extent Schwendener's theory as the only one capable of explaining satisfactorily many obscure phenomena in the morphology and physiology of Lichens. After pointing out that Schwendener did not sufficiently show that the relations between the hyphx and the gonidia were such as necessarily to involve the idea of parasitism, and arguing that the great simi- larity between the gonidia of Lichens and certain Algaltypes was not an accidental circumstance, he passes in review an extensive series of Lichens belonging to numerous different genera. Adopt- ing the two Algal groups of Schwendener, already referred to, he examines first the Chlorophyllacew, viz. Trentepohlia, Mart., Phyllactidium, Kütz., Protococcus, Ag. Of these he traces the resemblance between Trentepohlia and lichen-gonidia in thirteen genera, viz.:—(1) Roccella, as already shown by Schwendener ; (2) Lecanora, in many species ; (3) Dirina repanda, Fr.; (4) Caeno- gonium Linki, Ehrh., and C. confervoides, Nyl.; (5) Byssocaulon niveum, Mont. ; (6) Lecidea lutea (Dicks.), and L. microsperma, Nyl.; (7) Graphis elegans, Ach.; G. contexta, Pers., and G. hetezospora, Nyl. ; (8) Opegrapha varia, Pers., and O. herbarum, Mont. ; (9) Stigmatidium crassum, Dub.; (10) Arthonia cinnaba- rina, Wailr.; (11) Melaspilea arthonioides (Fée) ; (12) Chiodec- ton myrticola, Fée, and C. nigrocinctum, Mont. ; (13) Verrucaria nitida, Scnrad., in which, as well as in several Graphidei, it had formerly been recognized by De Bary. The genus PAyllacti- dium he regards as furnishing the gonidia to Opegrapha filicina, Mont.; while a species of Protococcus, viz. P. viridis, is similar to * In ‘Geschichte und Litteratur der Lichenologie,’ iii, Bd. (1872), REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 263 the gonidia of Physcia parietina, Similarly he reviews Schwen- dener’s other group, the Phycochromacee, and finds the following genera belonging to it furnishing gonidia to various genera and species of Lichens :—(1) Calothrix, Ag., to Lichina, Ag.; (2) Scyto- nema, Ag., to Ephebella Hegetschweileri, Itz., to the cephalodia of Stereocaulon ramulosum, Sw., and to various Pannarie; (3) Lyngbya, Ag., to Stereocaulon ramulosum, Sw. ; (4) Nostoc, Vauch., to Collema, Arnoldia, Physma, Leptogium, Obryzum, the cepha- lodia of some Stereocaula, Nephromium, Stictina, and certain Pannarie ; (5) Stigonema, Ag., to Lichenospheria Lenormandi, Born., Spilonema paradoxum, Born., Ephebe pubescens (L.), and the cephalodia of Stereocaulon furcatum, Fr. ; (6) Gleocapsa, Kiitz., to Synalissa, Omphalaria, Phylliscum (as formerly shown by Schwendener), Cora, and certain cephalodia of Stereocaulon. Of these Algals, as observed by Bornet, a small number of species furnish the gonidia for a great many different species and even genera of Lichens. As to the relation between these and the hyphe, he affirms that there is not the least evidence that the gonidia originate from the hyphe any more than the hyphe from the gonidia; but that, on the contrary, the two are originally quite independent, and that their union is a subsequent event. This union, he acknowledges, is difficult to be seen in the higher Lichens; but in several of those enumerated above, and reviewed by him, be has been able to detect the manner of ad- hesion. For, while in the great majority of Lichens the hyphe envelop the Algals and form a more or less embracing network around them, yet in some instances, according to his observations, the union is something more than mere contact, being a pene- trating by the hyphe into the interior of the algal-cells. In such instances of penetration, e. g. in Arnoldia minutula, Born., the cell, he says, becomes enlarged, the wall thickens, the con- tents become colourless ; at length the wall shrivels up, and the gonidium finaliy becomes a dead membrane. Various other minor points as to the relation between the hyphæ and the gonidia are touched upon by Bornet, which it is not necessary to enter upon for our present purpose. Suffice it to say that, as the general result of his investigations, he concludes that he is fully warranted in laying down the two following propositions. l. Every gonidium of a lichen may be referred to a species M algal. 2. The connexion of the hyphe with the gonidia is 0 such a nature as to exclude all possibility of the one organ being 264 REV. J. M. CROMBIE ON THE ALGO-LICHEN IIYPOTHESIS. produeed by the other, and the theory of parasitism can alone explain it satisfactorily. It is in this way only, he maintains, that we can aecount for the gonidia of very diverse Lichens being almost identical, as well as for the marked differences between the gonidia of other Lichens, of which the thallus and frue- tification are identical. Thus, also, he thinks, can only be rightly explained the origin of dead gonidia found in the interior of Lichens, as well as the occurrence in the same thallus of dissimilarly coloured gonidia. But not to dwell further, at present, upon the conclusions at which, upon these and similar grounds, the supporters of Schwendenerism have arrived, we proceed to notice what would naturally suggest itself as being a very simple solution of the problem. At first sight it might seem that the connexion between the hyphe and the gonidia (if any such existed) could readily be proved by cultivating lichens from their spores, and tracing the subsequent evolution of the thallus at every stage of growth. If the gonidia were proper organs of the lichen, we should thus be able to observe when and how they first made their appearance ; but if, on the contrary, they were foreign bodies, then evidently they would nowhere present themselves in the thallus thus obtained. Indeed Schwendener himself (in ‘ Die Algentypen,’ &c.) rightly observes that the question cannot be settled by hypotheses or isolated one-sided observations, but, as De Bary had previously advised, by numerous and carefully con- ducted experiments in the culture of lichen-spores, lichen-gonidia, and unicellular Algæ. In this way only, he says, can it be definitely established whether the germinating lichen-spore de- velops gonidia or not, and whether such free gonidia-groups as he takes to be algals can form from themselves a hypha-bearing thallus or not. Accordingly, to say nothing of earlier attempts, numerous others have been made since the promulgation of Schwendenerism, in the way of lichen-cultivation. Most of the experiments thus instituted, as will presently appear, proceeded upon an erroneous principle; while all of them confessedly failed in producing gonidia and much less a perfect thallus. The only legitimate cultures made were those of Treub (* Onderzoek- ingen over de Natuur der Lichenen,’ 1873), in so far at least as these relate to the spores of certain species of Physcia, Ramalina, and Lecanora, which, in the manner long before indicated by Tulasne, were sown upon various moistened substrata. Nay, not REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 265 contented with leaving these to develop in a natural way, Treub, on their germinating further, tried to cultivate them by the aid of various nutrient fluids *. After protracted experiments, however, he acknowledges that the results obtained were in all cases very unsatisfactory, owing to the spores not rightly ger- minating, or to the presence of moulds, or to the hyphal filaments not becoming fully developed. This has also been the experience of myself, and no doubt of many others who have attempted spore-eultivation (cfr. Nylander *On the Germination of the Spores of Varicellaria,” in ‘ Flora, 1868). Indeed, except under the same conditions of exposure to atmospheric and other influences as those under which they grow in nature, it is not possible to cultivate lichens beyond the earliest stages of evolu- tion T. Their vital phenomena being very intermittent, and unlike that of all other plants—now dormant in dry weather, now active in wet,—their subsequent growth, in consequence of this twofold life, is extremely slow, and the formation of the. perfect plaut in most cases a long-protracted process. Hence spore- cultures carried on at home, where the conditions referred to cannot be obtained, can hardly lead to any very definite or final results. And this, owing not to any want of co-operation of the requisite algals, as alleged by Schwendenerians, but to the “extreme impatience of situation and air which is inherent in the different species; whence it follows that these vegetables immediately perish, where the natal situation and necessary access of air is disturbed " (NyL, in Pyr. Or. p. 31). But another method of culture has been adopted by Schwen- denerians, who, by the process called Synthesis, or rather Sym- biosis (as it has been termed by De Bary), have made various attempts to manufacture lichens by sowing their spores upon certain algals, or presumed algals. Thus, amongst other experi- ments of this kind, Rees sowed the spores of Collema glaucescens upon Nostoc lichenoides (vide Monatsb. K. Akad. der Wissensch. zu Berlin, 1871); Bornet sowed the spores of Physcia parietina * These nutrient fluids were such as had been used by Boussingault for mho Mucidinos. No development of the Mycoderms and by Pasteur for that of the Med o wonder * that the results were very small," since the nutriment of Fungi an of Lichens is totally different in its nature! dis directi ave bee re, and f Of all recorded experiments in this direction, none have been more, a . : 7 (QVO r M " A Io 23 > 1 a s 1852 . few as, successful as those of Tulasne (vide ‘Mémoire sur les Lichens, 1892) 206 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. upon Protococcus viridis *, and those of Lecidea muscorum upon another species of Protococcus (vide Recherch. sur les Gonidies, &e.); while Treub sowed the spores of Physcia, Ramalina, and Lecanora upon Cystococcus humicola (vide Memoir already cited, pt. ii.). All these experiments, however, met with but a very limited amount of success, just as in the case of spore-culture by itself without any added “algals.? Even where the spores successfully germinated and produced hyphz, all that could be affirmed was that these formed with the Alge a structure re- sembling in some degree the more or less rudimentary thallus of alichen. But it is to be observed with respect to these syn- thetical cultures, and all other experiments of a similar kind, that even were the results more pronounced than they have been, they would prove absolutely nothing as to the truth of the hypothesis. The lichen-spore must, from its very nature, produce lichen-hyphe, whether with or without the addition of algals or pseudo-algals ; though what the fate of true added Alge in the subsequent evolution of the thallus might be, the experiments in Symbiosis do not show. A very singular synthetical experi- ment was made by Dr. Stahl (vide Beitr. zur Entwick.-Gesch. Flechten, 1877, Heft ii., * Ueber die Bedeutung der Hymenial- gonidien ”), which has been imagined in some quarters to supply “the missing link” in the chain of evidence by which the Schwen- denerian hypothesis can be demonstrated T. He cultivated the spores of a lichen which he called Endocarpon pusillum, Hedw. (but which evidently was Verrucaria Garovaglii, Mont.), along with the hymenial gonidia (more correctly gonidimia) which adhered to or surrounded the spores f. Each spore sent forth from its * That this Profococcus constituted the gonidia of PAyscia parietina had long previously been suggested by Kützing (vide Phyc. Gen. 1843, p. 167) ; but, as I have elsewhere shown, it is a true algal, and has only a superficial resemblance to any lichen-gonidia (wide Pop. Se. Rev. 1874, p. 271, t. 112. ff. 3, 4). t For a detailed and crushing reply to Stahl's observations and conclusions, vid. Richard's valuable treatise, * Etude sur les substratums des Lichens’ (1883), pp. 6-13. i These hymenial gonidimia have an “intraconceptacular origin " (vide N yl. in Flora, 1877, p. 357), evidently originating from the parietal cellules of the pyrenium in which they arise before the thee. Consequently they do not find their way in after the formation of the “tubule” ond by the open apo- thecia, as Schwendener affirms in the conclusion of his paper “ On the Gonidia ques'ior," Like the gonidimia of the thallus, they no doubt act chiefly as vivifying stimulants. REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 267 segments certain hyphal filaments which, as they became more evolute, invested the gonidimia, and these, as he says, Increased in size owing to the influence of the hyphe, which prevented them from undergoing division. From this hyphal involution, under special eireumstances of cultivation, a more or less differentiated thallus at length became apparent, and a lichen was, strange to say, synthetically manufactured out of a lichen-spore and lichen- gonidimia. Here, however, it is but right to notice that in Symbi- osis the spore is “ assumed " to be a fungus * and the gonidia to be algals, which, like other assumptions in Schwendenerism, is clearly a “begging of the whole question." Any logical synthesis which could substantiate the hypothesis would evidently be the forma- tion of a lichen by means of the coalition of the filaments of a well-ascertained fungus and an authentic algal. In regard to this, Dr. Lindsay, in a communication against Schwendenerism, which appeared in ‘ Nature’ for January 27th, 1876, says :—“ If, by artificial cultivation, such a union (2. e. of a fungus and algal) could be made to produce a lichen, the theory might be held as proved" f. Now direct observations made in nature itself have shown me that such contact between the hyphe of a Pyre- nomycetes and a Protococcus, instead of producing anything in the shape of a lichen, is simply destructive to the algal, which the mycelium overruns and involves. Besides, and as is well known to every practical lichenist, in cases where fungal hyphe attack (as they often do) either the epithecium or the hymenium of lichen-apothecia, the texture is at once destroyed and the apothecium is killed by the parasite. Hence, instead of there being any affinity, as presumed, there is a mortal antagonism between a lichen and a fungus, which would at once have become apparent had the Schwendenerians tried to cultivate lichens, * The spores of Lichens, notwithstanding a certain similarity, are of a totally different nature from those of Fungi, differing, amongst other characters, in their consistency, in the nature of the hymenial gelatine, and in their producing licheno-hy phx (vid. infra). Indeed, that the spore ofa lichen cannot bea fungal- spore is evident from the fact that several maritime lichens are intertidal, whereas even a short immersion in salt water is destructive alike to the protohyphe and the apothecia of Fungi. 7 t By some singular misunderstanding of the subject, Dr. Vines (in Quart. Journ. Mier. Se. vol. xviii. n. s. p. 153), after quoting these words of Dr. Lind- say, immediately adds, ‘Such evidence is afforded by Stahl’s paper on the nature of hymenial gonidia ”! 268 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. according to the principle involved in their theory and upon which itis directly founded *. This is fully corroborated by the observations of Mr. Marshall Ward on Strigula complanata, Fée (vide Trans. Linn. Soc., Bot. vol. ii. pt. 6, 1884), though the author strangely fancies these to be entirely confirmatory of the Algo-lichen hypothesis. The fungal-mycelium of which he speaks as occurring so frequently on the leaf of Michelia, &c., and coming into contact with the assumed algals, that is the “ platygonidia "' of Strigula (in which even when fertile there are scarcely any or no licheno-hyphe present), at length, as he acknowledges, destroys the *algals," which it invests with its meshes. The same phe- nomenon also takes place in our British species of Strigula, viz. S. Babingtonii, when similarly invaded by the proto-hyphæ of Cap- nodium Footii, with which it is often associated on the same leaf. But apart altogether from such eonsiderations relating to lichen-culture, there are two fatal objections to the hypothesis, either of which is quite suffieient for its subversion. The first of these has reference to the very peculiar nature of the parasitism involved in the theory that the fungal-hyphe are nourished by the captive algals. Other plants, from which parasites draw their nourishment, usually become speedily exhausted and finally perish, often involving in their death that of the parasite itself. In some cases, indeed, where the host is of sufficient size and vigour to supply food alike for its guest and itself, it may, for a more or less protracted period,sutfer but little injury. But here we have a parasite exceeding in size and number of cells by many hundred times the nourishing plant which it invests, and yet, so far from exhausting, only invigorating its host—a phenomenon which certainly nowhere else occurs in nature (vide Caspary; “ Ueber die neueren Ansichten in Betreff der Flechten,” &c., 1872). Such parasitism is indeed acknowledged by Schwendener himself, in the last of the memoirs above cited, as being unique in the Vegetable Kingdom. Atthesame time he fancied that the JVostoc-chaplets recorded by Strasburger as occurring in the leaves of Azorella, by Reinke on the stems of some species of Gunnerd, by Janczewski in the interior of Anthoceros, and by Cohn in the tissues of Lemna, have brought to notice cases of adaptation in a * Strange to say, however, according to the observations of M. Bonnier (vide Bull. Soe. Bot. Fr. 1883, p. 87) a lichen does not require any aigal whatever for its fabrication, This new discoverer has himself made lichens from the proto- plasm of “ Mosses” associated with spores,—a new kind of Commensalism of which even Schwendener never dreamed. REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 269 certain sense analogous to the parasitism involved in this theory. Such entophytie algals, however, are evidently veritable parasites, and prove nothing as to any algo-fungal parasitism in the lichen. lt has, also, more recently been alleged (though surely not seriously !) that Schwendenerism has received a certain kind of support from the supposed existence of some unicellular Algw, Zoochlorella and Zooxanthella, as they have been termed, which occur as parasites in some families of Zoophytes, and which, nourished at the expense of the protoplasm of their hosts, at the same time contribute to the synthetical formation of the organic substances utilized by these inferior animals. But these chloro- phylloid bodies have not yet been proved to be true Alga, and even if they were so, there would be no analogy whatever be- tween such parasitism and that of the Algo-lichen theory. Reverting, therefore, to Schwendenerism, and granting that the presumed slave not only feeds its master, but also in doing so flourishes and multiplies all the more, * Upon what," as Mr. Bentham so pertinently asks (Anniv. Address, Linn. Soc. 1872, p. 28), “do the gonidia themselves feed?” This is a very im- portant point in the physiology of Lichens, which Schwendenerism cannot satisfactorily explain. Detained in a dark and narrow prison, and deprived of the free life they formerly led by the tyrant who has surrounded and enclosed them in his meshes, they are evidently cut off from all commuuication with that outer world, from whieh they could receive such nourishment as they themselves require, and the much larger quantity their master exacts from them. Whence, then, and how is this nutriment derived? Now it is a well.established faet that lichens obtain their whole nourishment from water, with the different ingre- dients contained in it and necessary for their existence, accord- ing as that water comes from the clouds, from dew, from rivers and lakes, or from the sea. It has sometimes indeed been affirmed (and distinctly so by Schwendener) that the nutriment of lichens is also partly furnished by means of the substratum. This, however, is a mistake, for, as experiment shows, they thus derive nothing, and, in the case of crustaceous species, evidently can derive nothing in the shape of nutrition (ofr. N yl. in F lora, 1874, p. 60). Anything that they can in this way derive js simply through immixtion with their elements of certain particles, such as iron and chalk, dissolved by the water which is poured forth around them and mechanically received. Their true nutri- ment, derived, as we have said, from water, chiefly rain-water, 18 270 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. directly poured forth upon their surface, and penetrates the cortical stratum (when this is present) till it reaches the goni- dial stratum, around which the active vegetative life especially has its seat. The gonidia, therefore, are not special nutritive organs, although (as the instruments of absorbing carbonie acid) they greatly aid in promoting the nutrition, but act rather as vivifying stimulants,—the nutrition taking place through the corticali-gonidial stratum, in which the cellules also play their part, and serve in the slow nutrient process which is so charac- teristic of lichens. Consequently neither the poor prisoners nor the tyrant master act the parts assigned to them in Schwen- denerism; and the strange parasitism which it assumes is found on investigation to have no actual existence. The second objection, and one of even greater validity, is that there are neither fungal-mycelia nor algal-colonies in the struc- ture of Lichens. Indeed, were there so, we should naturally expect to find Lichens most abundant in those habitats which are most frequented by Fungi and Algz, and there perceive the whole process of manufacture going on before our eyes. Every field- lichenist, however, is perfectly aware that he need not look for lichens where these other Cryptogams have their special haunts. Similarly and conversely, lichens in all stages of growth are mót with plentifully in situations, e. g. granitic detritus and boulders towards the summits of lofty mountains, where fungal-mycelia and terrestrial algals are unknown. But, notwithstanding a certain superficial resemblance between the hyphe of Lichens and of Fungi, their structure and character, as Nylander has repeatedly shown, are entirely different. The hypha, or rather “myelohyphe,” Nyl., of Lichens, are perennial, firm, with thick walls, penetrated by lichenin, imputrible, and not dissolved by hydrate of potash. On the other hand, the hyphoid mycelia or “protohyphe” of Fungi are caducous, very soft, with thin walls, not at all amylaceous, readily putrifying on maceration, and on the application of hydrate of potash immediately becoming dis- solved. Moreover, the licheno-hyphe are rigid though elastic, nowhere flexuose and contorted *, but straight or straightish, so * M. Stahl, in the work already referred to (Heft i., “ Ueber die geschlecht- liche Fortpflanzung der Collemaceen ") represents the apothecia as originating from filamentose contortions or contorted hyphx. This, however, as shown by Nylander (in ‘Flora,’ 1859, p. 625, and 1879, p. 304), is entirely erroneous, and consequently is of itself sufficient to cast great doubt upon all Stabl's other observations. REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 271 that they possess no * involving " power whatever, as the Schwen- denerians assume. The hyphæ of Fungi, on the contrary, being soft and flexible, readily involve and enclose foreign substances, and indeed morphologically require the faculty of repeated con- tortions, in order to form the cellulose nucleus of the receptacle (cfr. Osw. Kihlman, Entwickel. der Ascomyceten, 1883). There is evidently, therefore, no fungal mycelium in the lichen. Again, with respect to the gonidia, notwithstanding their general resemblance to certain of the lower algals, it by no means follows that similarity is identity,—the two things being logically very different. The circumstance of the lichen-gonidia being subsimilar to such Algæ in form and structure only shows that there is a certain “parallelism” between them, and does not militate against both being regarded as belonging to distinct classes of plants*. Indeed the chlorophyll, or, phycochrom, in Lichens originates in the same way as in the cellules of other classes of plants (Phanerogams and Cryptogams) in which it occurs ; the only difference being that the gonidia are often seen as discrete cellules, although many forms variously composed are also not uncommon. Moreover, many genera and species for- merly supposed to be algals, and regarded by Schwendenerians as furnishing gonidia to lichens, have of recent years, in consequence of the discovery of their fructification, been proved to belong to lichens. Thus Sirosiphon, Hormosiphon, Scytonema, Stigonema, Cora, Dichonema, Chroolepus or Trentepohlia, Nostoc, and Gleo- capsa (at least in part), Gongrosira, and probably Phyllactidium, have now to be removed from the class of the Algæ, and conse- quently can no longer be pressed into the service of the Algo- lichen hypothesis t. In other respects the various. forms of lichen-gonidia are so well-marked as to admit of a distinct classi- fication and nomenclature of their own. These have been elabo- rately expounded by Nylauder, with diagnoses of the different * Such parallelism was long ago well pointed out by Thwaites in paper “On the Gonidia of Lichens” (Ann. & Mag. Nat. Hist. 2 ser. vol. ii 130) He rightly saw in the similarity of the gonidia to the Algæ, not identity, bu only “a typical character of essential structure binding together numerous species of various forms, and enabling us to distinguish at once in other species resemblances of analogy from those of affinity. MEE acid t Nor does the evolution of zoospores in free gonidia (i. e. in unstratifie identify such gonidia with algals, since in this t i in C t all halli), as in Chroolepus, a s contrary to the nature o f there would again be only a similarity, and nothi lichens. 272 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. types, in * Flora, 1877, pp. 358-9; from which it is evident that while some gonidia and gonimia bear a resemblance to certain algals, others are so different as never to have found a place amongst these latter. This had already been observed by Koerber in his valuable essay ‘Zur Abwehr der Schwendener-Bornet’- schen Flechtentheorie ' (1874), where, towards the close of his argument, he says that if the gonidia were all true Alge, they would all have been met with in the free state, whereas this is by no means the case; the gonidia, for instance, of Vetrocymbe, Phylliscum, Melanornia, and others, having not yet been found elsewhere than within the lichen-thallus *. Nylander also, and more distinctly, affirms (7. c. p. 356) that the gonidia of lichens do not in nature at the same time occur within the thallus, and living free without it. "This disposes of an alternative which was at one time put forward in opposition to Schwendenerism, viz. that many of the presumed algals of the hypothesis were pro- bably only erratic gonidia vegetating separately. Numerous observations, however, entirely homologate N ylander's statement ; so that we may regard it as well established that the gonidia of lichens in their different types and forms are nowhere seen save within the lichen-thallus f, and therefore cannot be Algz. Hence, as there is no fungus-mycelium, so neither is there any algal-colony in the lichen. From the considerations now adduced, it is evident that the three propositions laid down by Koerber in the essay referred to are correct, and rest upon even better grounds than he himself * Professor Cohn (in ‘‘Conspectus Familiarum Crypt." &c., in * Hedwigia,' 1872, p. 17) also affirms that algals from which Usnea, Cladonia, &c. could proceed are not known. As also directly bearing upon this point, Mr. Archer (in Quart. Journ. Mic. Sc. n. s. vol. xiii. p. 234) puts the very pertinent question, “ Why do not several other aerial types, quite as accessible to an intruding parasite as other species, play their part as gonidia-formers? t This of course does not apply to the “soredia,” or the rounded pulveru- lent eruptions which frequently occur on the cortical stratum, and which con- sist of gonidia intermingled with medullary hyphze protruded along with them. These, being but loosely adherent to the surface of the thallus, are readily dis- seminated outside the lichen, though occasionally, as in Alectoria nidulifera and the specimens exhibited of Usnea ceratina var. scabrosa, they originate propagula while still adherent to the thallus. In these soredia the gonidia constitute a “syngonidium,” and are in reality fragments of the thallus serving, like bulbils or gemmze, for the propagation of species which occur only ina sterile condition, REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 273 imagined. These propositions are (1) that the tissue in which the gonidia of a lichen are imbedded is not of a fungoid nature ; (2) that the gonidia themselves are not true Alge ; and (3) that lichens are not the expression of a condition of parasitism. But though such considerations are quite sufficient for the subversion of the Algo-lichen hypothesis, the crucial point of the origin of the gonidia and their relation to the rest of the thallus still remains to be expiscated. Itis clearly not enough to show that Schwen- denerism is impossible, for the mere rejection of one erroneous theory would not bring us nearer to the actual solution of the problem. Now the whole of the hypothesis proceeds on the assumption that there is no genetic but only a parasitical relation between the filamentose hyphe and the gonidia. The former part of the assumption is quite correct, and the latter is entirely wrong. Indeed, all researches in the way of proving a connexion, whether genetic or parasitic, between these two elements is, as will pre- sently appear, neither more nor less than going in quest of “a mare’s nest,” —is an attempt to discover what has no actual exist- ence. And yet the gonidia, equally with the hyphz, belong to and are essential organs of the autonomous (not composite) plant called a lichen. How, then, do the gonidia originate in, and how are they related to, the thallus ? On this point, as I have already hinted, previous to the date at which Speerschneider adopted and endeavoured to illustrate the Bayrhofferian origin of the gonidia, there appeared in 1852 a valuable work by the celebrated Tulasne, entitled ‘ Mémoire pour servir à l'histoire des Lichens.’ In the appended plates by which he illustrates his researches on the morphology of Lichens, there are several figures which show sufficiently wellthe origin of the gonidia in the thallus at a very early stage of its growth. These may be seen in tab. iii. ff. 1-3, in the case of Lecanora cinerea, and in tab. xi. f. 17, in the case of Cladonia coccifera ; though in the text the explanation which he gives of their origin is occasionally in some respects rather confused. In p. 36, however, when treating of the young squamules of growing Cladonia pyxidata, he, with sufficient accuracy, indicates the origin of the gonidia as follows :—“ Here and there upon the elementary filaments (i.e. the hypothallus) some small sessile cellules are "ol lourless, which, after multiplying generated, spheroid and colo , themselves, give origin in their turn to larger cellules in which the chlorophyll is collected together.” So also in p. 22, when 274 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. speaking of the gonidimia of Endocarpon miniatum, he says :— “In these cells (7. e. of the cortical stratum) is produced the chlorophyll.’ Similarly in Nylander’s ‘ Synopsis Methodica Lichenum’ (1858), tab. i. ff. 1, 3, 4, 6, the sections of the dif- ferent thalli there given, belonging to several tribes and genera, show, even at first sight, the gonidia enclosed in the thalline cellules under conditions in which they could not have entered from without, but must have originated in the cellules themselves. It is rather singular that both in the case of Tulasne and of Nylander these statements and figures should have until recently been entirely overlooked, and their import not comprehended either by the supporters or the opponents of the Algo-lichen theory. In them, however, we obtain the clue which is to guide us in our further inquiries upon the subject. As already shown, lichens can scarcely in ordinary circum- stances be successfully cultivated from the spores beyond a certain stage (hypothalline) of growth. This, however, is the less to be regretted, because in nature itself (and nature is always its own best interpreter) we can in some instances behold the whole process of the evolution of the thallus. More especially is this the case with respect to crustaceous species growing pure upon such substrata as quartz rock, flint stones, and young bark of trees*. In the various Lecanoras and Lecideas which occur on the surfaces of these, we can readily trace the life-history of a lichen, in so far as the vegetative organs are concerned, through every stage from the germinating spore to the perfect plant, in the manner now to be described. On germination, as may easily be seen in spore-culture (and indeed not unfrequently even in the apothecia themselves), the spore sends forth from the endospore, in various ways T, a germinating filament or filaments called the prothallus. This speedily passes into * Where the substratum is not pure but shows various heterogencous growths, as is usually the case on the trunks of old trees, we often find, especially in the case of Opegraphas, Arthonias, and Verrucarias, different forms of gonidia immixed and confused, apparently constituting the same thallus. Hence Bornet, as already noticed, and more recently Almquist (in * Monographia Arthoniarum Seandinavie, 1880), erroneously assume the presence of diverse gonidia in the same lichen—a polymorphism which is entirely owing to “ the struggle for existence " amongst different species. T Numerous illustrations of germinating spores belonging to various tribes and genera of Lichens may be seen in the Tables appended to the ‘Mémoire’ of Tulasne already cited. REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 275 the hypothallus, and between the two there is in reality no distinct line of demarcation. The myelohyphe thus produced contain lichenin from the very first, and in other respects present the distinctive characters already mentioned. The hypothallus, as developed, is closely agglutinated to and very thinly expanded upon the substratum, becoming in some cases soon evanescent, and in others undergoing various modifications *. In its earlier stage it differs somewhat in figure, being at one time either dendritically divergent or otherwise effigurate, and at another time more continuous and only at the circumference plumoso- radiating. The numerous specimens exhibited of Lecanora gibbosa (silicole), Lecidea geographica (quartzicole), and Lecidea alboatra (silicole) show the gradual evolution of the thallus from its earliest hypothalline condition. In these the hypothallus is blackish and radiating ; but in other lichens, such as the vitricole specimen of Lecanora galactina, also exhibited, it is white, con- sisting of byssine filaments laterally apposite and contiguous. This hypothalline stratum, as will be apparent from what has now been said, constitutes the foundation of the lichen-thallus by which it is affixed to the substratum. Now evidently is the time when, according to Schwendenerism, we ought to find the hyph: going out in quest of alg, which they might lay hold of and imprison in their meshes, as “ the spider does the fly” in its web. But what is the actual state of the case? In all the young hypothalline formations before you, there is not the slightest vestige on the specimens of any algal, Protococcus or other, which they can thus entrap, nor of any gonidia developing either from the branches or from any other part of the hyphe f. These, it will also be observed, whether * Sometimes, asin foliaceous lichens, it appears in the form of rhizing ; and at other times, in erustaceous lichens, it is visible only as a thin line limiting the thallus. t Dr. Minks (a disciple of Bayrhoffer), in his treatise ‘ Das Microgonidium , (1879), supposes that he had detected in the hyphx, and indeed in every portion of the lichen-thallus, whether vegetative or reproductive, certain minute cor- puscles which he terms “ microgonidia,” and which constitute the initial state of gonidia into which they are afterwards developed. As has elsewhere, how- ever, been shown (vide Nyl. in Flora, 1879, p. 206), these so-called gonidia are simply the “ molecular granulations " which are found in the different cellules, and which do not at all undergo the wonderful transformations indicated by Dr. Minks. This is also proved by his subsequent alleged discovery of them ina large number of species belonging to the Ascomycetes (vide ' Symbolæ Licheno-Mycologice,’ 1881), in which, as is well known, real gonidia never occur X LINN, JOURN.—BOTANY, VOL. XXI. 976 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. simple or branched, are constantly directed straight forwards, without any twisting about to hunt for algals, even were any such present. And yet at a slightly more advanced stage of growth, you would, on microscopical examination, detect plenty of gonidia in these specimens. Whence, then, and how, do these originate ? If the hyphe are not parasitical on the gonidia, may it not be that the gonidia are parasitical on the hyphae ?—an alternative at one time suggested, but which evidently is as untenable as Schwendenerism. The matter, however, lies in a nutshell, and admits of a very easy explanation. Ona further inspection of the specimens, you will readily perceive upon the surface of the hypothalline stratum the presence of a number of small, variously coloured glomerules, with which it is more or less sprinkled. These are formed at an early stage of the evolution of the hypo- thallus, at or towards its centre, and in immediate proximity to where the spore first germinated*. On anatomical dissection it is found that these glomerules consist of minute cellulose granules, in the cellules of which are to be seen the gonidia in various stages of evolution. From the fact of their thus occur- ring in a growing condition, and the impossibility of their enter- ing from without through the closed walls of the cellules, it is evident that the gonidia originate in the glomerules themselves (just as the spores doin the apothecia), and are consequently self- developed organs of the lichen. Now these glomeruli constitute the first parenchymatous tissue which is directly developed upon the hypothallus. Like all the other elements of the thallus, this parenchyma is formed, as Nylander has well expressed it (vide * Flora,’ 1879, p. 333), “ by an innate power or impulse of procrea- tion which is inherent in the spore itself, and is not communicated from any foreign source, the only assistance rendered being, as already noticed, the materials derived from the atmosphere, espe- cially in the form of rain-water." The glomerules, of which (as may be seen from the specimens) one or more are put forth at the same time, gradually become more numerous and contiguous, as the process of development goes on, until at length a continuous cortical stratum is formed upon the hypothallus. It certainly seems very strange that these cortical cellules in which the gonidia arise should be either * Specimens illustrating the earlier stages of lichen-growth (which of course never appear in herbaria) seem to be entirely unknown to the supporters of Schwendenerism. REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS, 277 quite ignored or not rightly discriminated by Schwendenerians, who speak as if only two elements entered into the structure of the lichen, viz. hyphæ and gonidia, whereas to these must unques- tionably be added a third element, viz. the cellular cortical tissue *. But not only in young crustaceous thalli do we thus perceive the gonidia originating in the cellules of the first-formed paren- chymatous tissue, but even in adult thalli amongst the higher foliaceous lichens we can sometimes observe growing gonidia in the cortical cellules. This is the case in such species as Physcia pulverulenta, Ph. lithotea, Umbilicaria pustulata, Psoroma hyp- norum, &c., in which the thalli have the cellules very distinct and evident (not confused or obliterated), and are consequently best adapted for examination. So also the intracellular origin of the gonidia is equally apparent in the thalline margin of lecanorine apothecia, and even of biatorine apothecia which have gonidia intruded into their proper margin, where they are seen in a young or growing state towards its external and youngest portion. As to the origin of the gonidia in the gonidial stratum, this also admits of a ready and simple interpretation. “ The cortical stra- tum," as observed by Nylander in ‘ Flora,’ 1877, p. 354, “ gradu- ally inereasing and expanding, is at the same time in like propor- tion dissolved (or resorbed, as it is termed in physiology) beneath, and the gonidia consequently become free." In this condition, whether enclosed in eellules or variously conjoined or discrete, they constitute the narrow gonidial stratum, which is situated between the lower portion of the cortical and the upper portion of the medullary strata. This latter stratum 1s produced from the cortical cellular tissue and arises in the thallus either before or at the same time as the gonidial stratum, so that though situated beneath this, it is not posterior in its formation. When the medulla is filamentose (for it is frequently crustaceous, as it subsequently becomes in the specimens exhibited, and is also * This origin of the gonidia, which was first, as we have already stated, observed by Tulasne, was further confirmed by Arcangeli (‘ Sulla questione n gonidi ") in Nuov. Giorn. Bot. Ital 1875, where, after recording various ots r- vations made by him upon several lichens, he says that “ the gonidia arise ro organs which have a more or less pseudo-parenchymatous strut i however, to Dr. Nylander, whose long services to lichenologica science are A valuable and extensive, that we owe the actual demonstration of the origin ° the gonidia and of their relation to the other elements of the thallus (vide ‘Flora’ from 1877, passim). x2 278 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. occasionally cellulose *), the gonidia are often seen attached to the radicles of the lichenohyphe of its exterior portion. This attachment, however, is in no way genetic or parasitic, nor does it argue any kind of “copulation " whatever, as Schwendenerism affirms. The gonidia are in fact neither adnate to nor penetrated by the myelohyphe t, but only adherent to them by means of the lichenin, with which all the different elements of the thallus are penetrated. This holds good equally in the case of unstratified thalli, where lichenohyphe are present, as amongst the Collemacei and the lower genera of the Lichenacei. In all such instances, the apparent copulation is simply an amylaceous adherence, and the faneied penetration the result of erroneous observation, or of an abnormal condition of the gonidia. Such, then, is the undoubted origin of the gonidia in, and their relation to, the rest of the lichen-thallus. To trace this in every instance from the germinating spore to the perfect thallus throughout the various families and tribes, is by cultivation apparently impossible, and in nature itself by no means an easy matter. There can, however, be no doubt that what is true in the above instances is true also in allothers. For as Schwen- dener, while predieating the same phenomenon witnessed by Reess for all the lichens in general, whose evolution he himself had studied, rightly affirms (in Erór. zur Gonid.), * It is incon- ceivable that a plant should become developed at one time in one way, and at another time in one completely different." Hence in the Collemacei, the stronghold of Schwendenerism, where it is more difficult to trace the progressive evolution of the thallus, it is evident that the assumed algo-fungal-lichen phenomena seen by its supporters must be interpreted in accordance with the well ascertained facts now brought under your notice. But even in these we are not quite left without the means of satisfactorily * In the cellulose medulla also, as in Psoroma hypnorum, the gonidia are seen to be produced in the interior of the cellules, just as they are in those of the cortical stratum. t Treub, who specially investigated this point, never saw the hyphal ex- tremities penetrating into an uninjured gonidium—still with contents; nor from their nature can they have any such power of penetration. Indeed, as has been weli observed by Mr. Archer (Trans. Journ. Micr. Se. n. s. xiii. p. 234), Bornet's researches on the subject, which showed in certain cases the hypha destroying the gonidium (vide p. 263), prove too much in so far as Schwen- denerism is concerned, inasmuch as Schwendener's parasite does not destroy, but rather stimulates, REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. 279 overcoming the difficulty. The real state of the case here may— as pointed out by Nylander in a paper “ On the Gonimic Evolu- tion of the Collemacei " (in ‘Flora,’ 1868)—" be best demonstrated by attending to normal physiological conditions rather than by observations founded on textures disrupted and torn asunder anatomically.” Now in the isidia, with which the upper surface of the thallus of many species is more or less sprinkled, and which are evidently propagula and entirely comparable with the bulbils of Phanerogams, “ we have the whole history of the evolu- tion of the thallus from its first origin from a cellule containing a single gonimium to a minute true Nostoc, and ultimately to the perfect Collemaceous texture.” Similarly, on this point he states more explicitly in the ‘ Flora,’ 1874, p. 5, that “ the cellule at first is observed to be empty, and then, by the aid of secretion, there gradually is produced the green matter, and it assumes its definite form." It follows, therefore, that Nostoc, instead of being an independent algal, which according to Schwendeuerism grows normally at one time as such, and becomes at another time the host of a parasitic fungus, is in reality to be viewed as consti- tuting the initial or rudimentary state of Collema or Leptogium*. This notion had at different times been held by many crypto- gamists, though the actual transition, which is best seen in the isidia, had not been witnessed by them. In connection with this point it has been alleged by Schwen- denerians that the * cephalodia" which occur in many species of Lichens, e. g. in Stereocaulei, owe their origin to certain so-called * algals "—. Nostoc, Stigonema, Sirosiphon, and Scytonema —insinu- ating themselves into the lichen-thallus and so producing these morbid excrescences. Apart, however, from other considerations, it has just been pointed out by Nylander, in ‘ Flora,’ 1884, p. " (note), that the syngonimia of cephalodia (as inde ed also the thalline syngonimia) give no reaction with hydrochloric (muriatic) acid, whereas in Nostoc, Scytonema, and Sirosiphon an eruginose reaction is always thus produced. “This farther M be observed that in the Collemacei, where there is scarcely any distinct hypothallus, this being r ep" esented omiy in some species by a few rhizinæ, the gonimia arise (as ‘Flora’ (1868), has also pointed out that certain Glæocapse ous other Collemacei, e. g. Synalissa picina, Bryophagus glascapsa, which have been * Nylander, in similarly constitute the thallus of vari S. meladermia, Collema evilescens, and | adduced by Schwendener as lending support to his theory. 980 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. well seen in the isidia) in the gelatinous thalline substance before the medullary hyphz, amongst which they are seen lying either scattered, or usually conjoined, or variously conglomerated, and constituting, as also each isidium does, what Nylander has terined a ‘“syngonimium.” The lichenohyphe of the medulla, when they are present (for here also, as in the Lichenacei, the medulla is sometimes entirely cellulose), most probably originate from the eortieal stratum, which is more or less evolute, though this point requires further investigation. But however this may be, their relation in every instance to the gonimia is in no way one of copulation or parasitism, but simply and solely, as already stated, of adhesion (where such exists) by means of the lichenin which permeates the thallus. In the Ephebacei again, the whole thallus, 7. e. each single fila- ment from the base to the apex, or each single fruticule more or less branched from the base to the apices, constitutes also a syn- gonimium. The gonimia here, which are tunicated, or involved in a gelatinous membrane, originate within the cellules, and after the formation of the cellules, in the manner so fully described by Nylander when treating of JVematonostoc (vide Bull. Soc. Bot. 1873, pp. 263, 264). Since this evidently elucidates also the evolution of the gonimia in the Collemacei, his observations may here with propriety be transcribed, subversive as they are alike of Schwendenerism and Minksianism. “In a physiological respect,” he says, “it is to be noticed that this Nematonostoc (as is usual in analogous thalli) is propagated by gemmules, or minute isidiomorphous globules. These gemmules are at first globulose, simple, agonimie cellules, which in growing become oblong and are divided by a septum: afterwards by the addition of a cruciate septum they are seen to be quadrilocular, and gradually pluri- cellulose. But at the same time, in these gemmules, while still in a young state, are produced the gonimia, at first a single gonimium in each loculus, and later two and more (through the division by constriction of the first gonimium), so that monili- form gonimia are produced, and in maturity the internal cellulose texture disappears. Thus from the very beginning the gonimia are produced within the cellules, and after the formation of the involving cellules." Such briefly stated is unquestionably the origin and the subsequent evolution of the gonimia alike in the Ephebacei and the Collemacei, as demonstrated on a microsco- pical examination of the isidia, or so-called Nostocine conditions REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS, 281 which many of them present, especially when sterile. If, then, the gonimia originate, as they certainly do, in these cellules, they necessarily belong to the lichen-thallus, of which, as we have already seen, they are special and important organs. Conse- quently they cannot be algals, as Schwendenerism erroneously assumes. Moreover, as the species of which the Ephebacei are composed possess no lichenohyphe whatever, the theory is in their case evidently impossible. From all these various considerations and illustrations, to which others, though of minor importance, might easily be added, it is clear that the Algo-lichen hypothesis rests upon no solid basis whatever, but simply and solely upon imagination, and that it is merely a plausible attempt to explain certain phenomena which its author and adherents supposed to be otherwise inexplicable. Notwithstanding the laboured arguments by which it has been sought to deprive them of their autonomy and intrude them amongst the Ascomycetes, Lichens still remain a distinct class of plants, intermediate between the Alge and the Fungi, related to the former through the Ephebacei and to the latter through various species of their lower genera. It is no doubt difficult at times to draw a definite line of demarcation between some of them and these other two classes; but this is only what often happens elsewhere, till more accurate observation at length solves the difficulty *. At the same time, however, with respect to Lichens, all correct observations and logical deductions show that the words of Acharius in the conclusion of the Intro- duction to his * Lichenographia Universalis" (p. xiv) are still as true as when he wrote them:—.... . * Uti ratum habeo: Lichenes ordinem naturalem peculiarem et a reliquis plantis Cryptogamis distinetum constituere." Distinct they certainly are alike in the structure of their thallus and the mode of their nutrition, and more peculiar still in their very slow and inter- * Indeed, as every systematist well knows, it is constantly becoming more and more difficult to draw any such definite lines, especially in several of the lower Orders of plants. Living nature in its wild luzuriance (whether as the result of evolution or degradation of primordial types) refuses to be bound by Hence the limits of Classes (or Orders, &c.) must strictly mathematical laws. as our biological researches are more ex- to a certain extent. necessarily vary i i in many ¢ ascertain such limits wi tended and minute. In enabling us in many cases to ascertain sut h limits with any precision, chemistry will no doubt play a still more important part in the near future than it has hitherto done. 989 REV. J. M. CROMBIE ON THE ALGO-LICHEN HYPOTHESIS. mittent manner of life, and especially in their extreme longevity (vide Crombie, Art. * Lichens" in Encyclopedia Britannica, 1882, p. 558)*. Lichens therefore are Lichens and nothing else—neither Fungi nor Algæ, nor any intermixture of these ; but everywhere and constantly preserving their own distinct type, and distinguished by many important characters peculiar to them- selves. Such of these as have a more direct bearing upon the solution of the problem before us have been more or less fully noticed in the preceding observations. From these it is suffici- ently evident that Schwendenerism, whether viewed anatomically or biologically, analytically or synthetically, is, instead of being irue science, only the * Romance of Lichenology," as I have elsewhere termed it (vide Pop. Sc. Rev. 1874, p. 276). And thus also the origin of the gonidia in, and their relation to, the rest of the lichen-thallus belong to the very rudiments of Morphological Botany, and constitute the ABC of Lichenology. DESCRIPTION OF THE PLATES. Prate VIII. Fig. 1. Hypothalline state of Cladonia coccifera, magnified, showing the first cortical glomerules developed upon the hypothallus, in the cellules of which the gonidia originate. After Tulasne. Fig. 2. Similar hypothalline condition of Lecanora gibbosa, magnified. 2a. Gonidia as seen enclosed in the cellules of the pseudo-parenchyma, magnified about 270 diameters. Fig. 3. Similar hypothalline state of Lecidea alboatra, with dendritically arranged filaments and the first minute cortical glomerules, nat. size. 3a. A portion magnified. Puate IX. Fig. 4. Lecidea geographica, hypothalline condition, portion, of natural size, as shown growing on a small piece of quartz rock. 4a. A portion of the same enlarged about four times its nat. size. Fig. 5, Thin section of the cellulose cortical stratum and of the medullary hyphe of Physcia pulverulenta, showing the gonidia originating within the cellules, x 300. Fig. 6. Vertical section of thallus of Umbilicaria pustulata, magnified about 250 diameters, showing the genesis of the gonidia within the cellules of the lower portion of the minutely cellulose corticali-gonidial stratum, from which it is evident that they are not in any way pro- duced from the medullary hyphæ (cfr. Nylander in ‘Flora, 1875, p. 303). x H . . . B B : i . This longevity is to be measured in many cases, not merely by centuries, but apparently by millenniums. ON THE FLORA OF THE PHILIPPINE ISLANDS. 283 Fig. 7. Section of thallus of Psoroma hypnorum, highly magnified, showing the gonidia enclosed in the cellules of the cellulose medulla. 7a. Sepa- rated gonidia enclosed in the cellules, magnifled 275 diameters. After Nylander. Fig. 8. Isidiose globules of Collema furvum in different stages of growth, showing that the gonimia originate from the very first in the isidia themselves. 8a. A syngonimium fully developed, magnified. On the Flora of the Philippine Islands, and its probable Deri- vation. By R. A. Rorfz, Herbarium, Royal Gardens, Kew. (Communicated by Prof. Ortven, F.R.S., F.L.S.) [Read 1st May, 1884.] (Prate X.) Tux Philippines are a large group of islands situated to the north- east of the great bank (for the most part less than 50 fathoms below the surface) which stretches out from the Siamese and Malayan peninsulas to Sumatra, Java, and Borneo. They extend over nine degrees of latitude and fourteen of longitude, and much of their present configuration is directly due to volcanic agency. The northern island of Luzon is the largest of the group, being about the size of Ireland, though of a very different shape. Mindanao, the southern island, is of almost equal extent. Negros, Samar, Panay, Mindoro, and Palawan are each about one eighth to & tenth as large as Luzon; Leyte, Cebu, Bohol, and Masbate are still smaller; while a large number, gradually decreasing m size, are scattered about in various directions. The Philippines are for the most part surrounded by deep sea, but there are several submerged banks, indicating connections with neighbouring islands at some former period. One of these banks stretches away from the northern point of Luzon in the di- rection of Formosa, and on it are situated two small groups of islets —the Bashees and the Babuyanes. North of these a deep channel extends to the southern point of Formosa. From the south- western corner of Luzon there is a very distinct connection with the northern point of Borneo, by way of Mindoro, Busuanga, me long island of Palawan, and the small island of Balabac ; t e deepest channel along this bank being the Mindoro Straits which separate Mindoro from Busuanga. ‘The Sulu archipelago stretches from the southern point of Mindanao to the north- western point of Borneo. This bank is separated from the former by the Sulu Sea—a deep sea, much of it reaching to over two 284 MR. R. A. ROLFE ON THE thousand fathoms. The deepest channel which traverses this bank is near to Borneo, and connects the Sulu Sea with another deep basin of over two thousand fathoms—the Celebes Sea, north of Celebes. The fourth and last bank is not quite so well marked as the previous ones, though sufficiently distinct ; it extends from the south-eastern point of Mindanao to the north-eastern point of the euriously-shaped island of Celebes. The Philippines may, indeed, be said to extend over sixteen degrees of longitude, poli- tically at least ; for the archipelago as far as the Bashees on the north and the Sulu archipelago on the south both now belong to Spain, like the remainder of the group, with which they can more easily be classed than with any other. From their position and surroundings we are led to infer that the fauna and flora of these islands must reveal some curious and interesting facts, and it is fortunate that sufficient materials are now to hand to enable us to examine these features. Some branches of the fauna have been pretty well worked out. Mr. Wallace, in his ‘ Geographical Distribution of Animals’ (vol. i. pp. 845-349), and later in his ‘Island Life’ (pp. 361-362), has given a detailed account of what is known respecting the fauna. From these sources we gather that eighteen genera and twenty-one species of terrestrial Mammalia are known, of which nine tenths of the species are peculiar to the islands. "There are also seventeen genera and twenty-four species of Bats. The Birds number 288 species, belonging to at least 117 genera. Of land birds six genera are peculiar to the islands, and two thirds of the total number of species; while of water birds one tenth of the species are peculiar. The proportion of peculiar species, as pointed out by Mr. Wallace, is much larger than is found on any other Malay island. Quoting from a paper read by Lord Walden before the Zoological Society, Mr. Wallace points out that 31 of the Philippine birds occur in the Papuan subregion, 47 in Celebes, 69 in India, and 75 in Java, while one species is confined to the Philippines and Java. Of the genera 50 are of very wide range; 40 are almost exclusively of the Oriental region; 2 exclusively Malayan; 5 are typically Palearctic, and have reached the islands by way of North China, two of them extending southward to the Moluccas ; and 12 belong to the Australian region, of which one is chiefly confined to Australia, one to the Papuan group, and one to Celebes. The Philippines are also peculiarly rich in terres- trial Mollusca, about 400 species being known. Helix and Bulimus FLORA OF THE PHILIPPINE ISLANDS. 285 abound in species, the large genus Cochlostyla is almost peculiar to the Philippines, and Pfeifferia is found also in the Moluccas. These facts are interesting as showing the affinities with sur- rounding countries and the large proportion of endemie species. Mr. Wallace has somewhere remarked that the flora is not suf- ficiently well known for comparison. Since that time, however, a considerable addition has been made to our knowledge. The ‘ Novissima Appendix, containing an enumeration of what is known of the flora, with a census of genera and species under each Order (published in the fourth volume cf the third edition of Blanco’s * Flora de Filipinas ’), the ‘Sinopsis de Familias y Gene:os de Plantas Lefiosas de Filipinas' (a work containing figures of no less than 465 species of 386 genera) by Don Sebastian Vidal, Conservator of Forests in the Philippines, and a collection of about 900 species of plants received at Kew, from the author of the last-named work, have all contributed largely to our know- ledge. The 468 coloured plates accompanying the third edition of Blanco’s work must also be taken into account, though a con- siderable proportion of them represent cultivated plants. It is in working up the aforementioned collection, in conjunction with Don Sebastian Vidal, that a large proportion of the data have been accumulated on which the present paper is based, and which I now offer to the Linnean Society as a brief outline of the flora of the islands. I am specially indebted to Don Sebastian Vidal for a large amount of information respecting the islands generally, and their vegetation in particular, The first paper of any importance on the flora of the Philip- pines appeared as early as 1704, as an Appendix to the third volume of Ray’s ‘ Historia Plantarum.’ It is entitled « Herbarium aliarumque Stirpium in Insula Luzone Philippinarum primaria nascentium, a Rev? Patre Georgio Josepho Camello, S.J., Observatum et Descriptarum Syllabus: ad Joannem Raium transmissus; Additis etiam plurimarum Iconibus, ab auctore propria manu ad vivum delineatis; quas ob sumptum in chalco- graphos erogandorum defectum impresentiarum emittere non licuit.” This work occupies 96 folio pages, and contaius descrip- tions of a large number of plants, with their native names. Linnæus appears to have overlooked it, and indeed every one else, as I cannot find that anything has been done with it up to the present time. The drawings, however, exist, and a consider- able number of the plants, both being preserved in the Sloane 286 MR. R. A. ROLFE ON THE Collection at the British Museum. Father Camell was a Jesuit who for some time resided in the island of Luzon. A number of Philippine plants were collected by Nee during the Malaspina Expedition, and these were described by Cavanilles in his * Icones, published between 1791 and 1801. In 1837 Blanco's ‘Flora de Filipinas’ was published, an octavo volume of 887 pages. This work was published under peculiar circumstances : away from Europe, and without opportunities of consulting European Herbaria, Blanco had to rely solely upon books for his determinations, and of these he had only a limited number. His library consisted principally of some Linnsean works, Aublet's * Histoire des Plantes de la Guiane frangoise, and some of the early volumes of De Candolle's * Prodromus.’ The geographical distribution of plants does not appear to have been taken much into account, for the bulk of the species were made to fit existing descriptions, some of them belonging to American plants. The work is written in Spanish, and, as no specimens were preserved, great uncertainty has been felt respecting them; but as the native names are given, and many of the plants are from a limited area round Manila, it will yet be possible to identify nearly the whole of them. Many, indeed, are already identified, thanks to the authors of the new edition of the * Flora de Fili- pinas,' and to Don Sebastian Vidal, who has taken great pains to obtain specimens from the original localities for the purpose of identification. A second edition appeared in 1845, after Blanco's death, but from his MSS. Several corrections were made, but it is noteworthy that some species described as new in the first edition are here referred to Linnean names, and not always correctly. A number of Philippine plants, collected by Henke, were described by Presl in his * Reliquis Hænkeanæ, published in 1830; and also a number from Cuming's collection by the same author, in his * Epimeliæ Botanice,’ published in 1849. In 1843 the ‘Plante Meyeniane’ was published, as a supplement to the nineteenth volume of the * Novorum Actorum Academie Cæsareæ Leopoldino-Caroline Nature Curiosorum. The Orders were worked up by various authors, and a number of Philippine species described. Meyen was for a month in the islands during the rainy season, and his plants are chiefly from Jala-Jala (pronounced Hala-hala), a promontory running into the large lake known as Laguna Bay, near Manila. In 1851 a small work of 116 pages was published by Father Llanos, entitled * Fragmentos de algunas FLORA OF THE PHILIPPINE ISLANDS. 287 Plantas de Filipinas no includas en la Flora de las Islas de la l° ni 2* edicion) This was also published at Manila, and the species are even more obscure than in Blanco’s work. Besides the above, a large number of Philippine plants have been de- scribed in various Monographs, in DeCandolle's ‘ Prodromus, and elsewhere, from the collections of Callery, Lobb, and others, but especially from the large collection made by the late Mr. Hugh Cuming, and which has been so widely distributed. It is unfortunate that a number of Cuming's plants, collected in Malacca, have been distributed and published as from the Philip- pines *; also that the numbers in various Herbaria do not always correspond, owing to the loose numbers getting astray. How much of this migbt be prevented by the number being attached to the plant when gathered, as in Burchell's collections, and the acknowledged uncertainty of the system of quoting numbers prevented. From a list of Cuming's plants preserved at Kew, it appears that numbers 1 to 2242 are from the Philippines, 2243 to 2251 from Sumatra, 2252 to 2399 from Malacca, and 2400 to 2436 from Singapore. It is important, however, that his pre- vious collections from America and the Pacific islands should not be confounded, as these also run from 1 to 1499. A new edition of Blanco's ‘ Flora de Filipinas’ has recently been published, but with some considerable differences from the original one. This third edition comprises four folio volumes, besides 468 plates, equivalent to three volumes more. There are two editions of the plates, one of ordinary blaek impressions, the other of coloured chromo-lithographs. The first three volumes of text contain a reprint of Blanco's work (the two previous editions compared, and the corrections of the later one retained), with a Latin trans- lation by Father Andrés Naves and Father Celestino Fernandez Villar. The fourth volume contains a reprint of the afore- * In this way two genera of Myrtaceæ (Beckea and Rhodamnia) and six M Melastomaces (Allomorpha, Sonerila, Marumia, Anplectrum, Pternandra, an Kibessa) have erroneously found their way into Villar’s enumeration. ) Phi p pines" is written on a sheet of each of the above in the Kew Herbarium, an : this led to the Philippines being included in the extra-Indian distri ution o the same in Sir J. D. Hooker's * Flora of British India, whence they were ox tracted by Villar. They are, however, all from Malacca, as proved by m o MS. list preserved at Kew, and also by the specimens in Bentham s Herbarium. Still worse is it when this circumstance leads to the imposition of an erroneous specific name, as in the case of Henslovia philippinensis, A. DC., founded on Cuming n. 2255, which is really from Malacca. 288 MR. R. A. ROLFE ON THE mentioned * Fragmentos' by Llanos, also with a Latin translation ; a paper on medicinal plants, entitled * Libro de Medicinas de esta Tierra y declaracions de la Virtudes delas Arboles y Plantas que estan en estas Islas Filipinas," by Father Ignacio de Mereado, occupying 58 pages ; and a “ Novissima Appendix,” compiled from various sources, which is intended to include all that is known of the flora up to that date. The latest contribution to the flora is the ‘Sinopsis de Familias y Géneros de Plantas Lefosas de Filipinas,’ by Don Sebastian Vidal, mentioned above, a work in- tended for the Forest Service under his charge. Omitting minor pamphlets and scattered descriptions, to which references will be found in the aforementioned * Appendix, the foregoing summary represents the bulk of what has been written on the flora of the islands, and we now turn to the * Appendix’ itself. This is compiled from various sources, but the references to Philippine plants in various monographs and scattered papers form the groundwork of it. Next in importance are the plants from Blanco’s work, with those of Llanos and Mercado, an attempt being made to refer them to their correct genera and species. Then come the additional species seen growing in various locali- ties; and, lastly, plants cultivated in gardens for ornament or utility, or introduced in various ways. A census of the genera and species under each Order is appended; but the great draw- back to it is that everything is included, whether indigenous or merely cultivated. In the text the letter “c” stands for culti- vated, but as there is no other distinguishing sign, it is a very slow process to separate them, amongst the mass of letters and figures. In this Appendix the Dicotyledons are by Fernandez Villar, and are made uniform with the ‘Genera Plantarum’ of Bentham and Hooker. The Monocotyledons were commenced by Naves, and carried on to p. 307 (Fimbristylis nutans, Vahl), at which point the work was resumed by Fernandez Villar, who carried it on to its completion. The last part of the ‘Genera Plantarum’ was not published in time to be utilized for the Monocotyledons. The plates of this work were also determined by Naves; but a considerable number of the determinations are incorrect ; not so much the names of Blanco as the extra- Philippine names to which they are referred. Many of these are corrected in the Appendix, but some yet remain to be dealt with as materials come to hand. A great difficulty which the authors FLORA OF THE PHILIPPINE ISLANDS. 289 had to contend with in the determination of their plants, was the absence of authentic specimens, consequently they had to fall back upon books—Miquel’s * Flora Indie Batav: and Hooker's * Flora of British India’ being two of the principal ones; and in several of the Orders their determinations must be considered rather as approximate than real: for instance, in the genus Eugenia 83 species are identified with Indian ones, but in 23 species from Vidal’s collection scarcely one tenth are identical. It is much to be hoped that a set of the types of this work may be deposited in some European Herbarium, thus enabling the errors to be eliminated; for until this is done it will be impossible to obtain a correct idea of the flora. in the following paper a census of genera and species are given under each Order, with a column showing the endemic ele- ment; then an attempt is made to trace the origin of the flora by a comparison with surrounding countries; this is followed by an outline of the distribution of some of the more characteristic species within the group, and the physical causes which have in- fluenced this distribution; and, lastly, a list of some corrections made in working up the aforementioned collection, with the descriptions of a new genus and a few new species. In the following Census the left-hand column shows the number of indigenous genera and species under each Order, and the right- band column shows the endemic element. Villar’s census in- cludes the cultivated and introduced plants ; consequently it does not give a fair idea of the flora. My own is chiefly based on that of Villar, with the introduced species excluded and some recently detected species added; a few which have crept in by error are also excluded. The endemic element is considerable, but the following numbers must be taken as approximate, and as repre- senting the present state of our knowledge; in Orders requiring revision, the numbers are under-estimated for obvious reasons. 290 MR. R. A. ROLFE ON THE Census oF GENERA AND SPECIES. InDIGENOUS. ENDEMIC. Natural Order. Genera. Species. Genera. Species. 1. Ranunculacez ......... 2 4 2. Dilleniacez ............ 5 10 1 4 3. Magnoliacer ......... 2 5 /— d ee 2 4. AnonaeesS ........... 15 49 |} ..... 24 5. Menispermaces ...... 9 11 6. Nympheacee ......... 2 3 7. Crucifers ............... 1 1 8. Capparidex ............ 5 14 ] wea 4 9. Violariez ............... 3 3. p e 1 10. Bixineæ ............... 5 1]1 .] .... 2 11. Pittospores ............ 1 4 qoe l 12. Polygalem ............ 4 6 Josen 2 13. Caryophyllex ......... 2 2 14. Portulacezm ............ 1 2 15. Elatinese ............... 1 2 Jo 1 16. Hypericiner ......... 1 4 ow. 2 17. Guttiferm ............ 6 22 jf ..... 9 18. Ternstroemiacee ...... 6 18. qp e. 10 19. Dipterocarpese......... 5 21 |] eee 13 20. Malvaces ............... 11 31 6 21. Sterculiaceæ............ 9 26 J ..... 14 22. Tiliacesg ............... 7 27 1 14 29. Malpighiacem ......... 3 6 ff e 3 24. Zygophyllee ......... 1 1 25. Geraniacese 4 6 ft... 1 26. Rutaeem ......... ..... 16 44 ] ..... 11 27. Simarubee ............ 5 6 28. Ochnaceæ ............ 3 3. p .-- 2 29. Burseraces ............ 6 WB | | ..... 12 30. Meliacese ............... 13 43 ] ...... 27 3l. Chailletiacez ......... 1 6 flo... 4 32. Olaciner ...... ........ 8 11 | ...... 4 83. Iicinem ............... 1 5 j| ... 4 34. Celastrinez ............ 6 Wo}. 10 35. Stackhousiaces ...... 1 1 36. Rhamnez............... 7 12 j| ... 3 37. Ampelidezm ............ 2 22 | ..... 9 38. Sapindacese ............ 20 48 ]| ..... 22 99. Anacardiaces ......... 7 21 6 40. Connaracez ............ 4 14. ]| .... 10 41. Leguminose............ 65 197 ff .... 33 42. Rosacesm ............... 5 19 ]| .... 9 48. Saxifragacem ......... 5 5 1 44. Crassulacer............ 1 2 45. Droseracee ............ 1 4 j| .... . 1 46. Holoragezm ............ 1 1 47. Rhizophorem ......... 5 10 48. Combretacee ......... 6 21 |. ]| ...... 12 49. Myrtacese ............... 11 95 || ... 40 50. Melastomacem ...... 8 42 1 15 51. Lythrariem ............ | 6 15 | ree 3 FLORA OF THE PHILIPPINE ISLANDS. 291 InpiGENovs. Natural Order. Expzauc. Genera. | Species. || Genera. | Species. 52. Onagrariem ............ 3 6 53. Samydacez ............ 3 15 8 D4. Paseiflorem ............ 1 4 )h 55. Cueurbitaces ......... 11 22 1 56. Begoniaceæ ..........-- 1 Bi} ee 7 57. Datiscaceæ ............ 1 ] d» "7 58. Ficoldem ............... 3 4 59. Umbelliferz............ 2 4 60. Araliaees2 ..........- 9 18 6 61. Cornacez ............... 2 2 Foo 62. Caprifoliacez ......... 3 3. pp e 1 63. Rubiaces ..............- 41 140 1 76 64. Composite ..........-- 29 63 tw... 6 65. Goodenovier ......... 1 2 66. Campanulacee.......-. 2 4 67. Vacciniaceg ..........-- 1 3 p ... 9 68. Ericaeem ........... A 3 5 2 69. Epacrideæ ..........-- 1 2 70. Plumbaginee ......... 1 1 71. Myrsinez .......... 8 35. qo ee 23 72. Sapotaces seese 5 97 / | ee 5 13. Ebenacese .......... 2 15 |}... 7 74. Styraceæ eere 2 5 Jp] wee 3 75. Oleaceæ | ......... e 2 13. hn ... 7 76. Salvadoraces ........- 1 1 77. Apocynaceæ — .......-- 17 4 ] .-- 20 78. Asclepiader........ s 20 5l d e 12 79. Loganiacee ....... MT 5 ] Jp 5 80. Gentianem ............ 3 T 81. Hydrophyllaceæ ...... 1 1 2. Boraginese ..........-- 8 20 | o 9 83. Convolvulacez......... 9 42. |} ... 6 84. Solanacem ..........-- 6 21 85. Scrophularinem ...... 14 40 ]| eee eee 1 86. Orobanchacese......... 1 3 87. Lentibulariee ........- 1 5. Jp e 2 88. Cyrtandracee .......-- 9 18 | .—- 14 89. Bignoniacem ........- 4 10. ]| ..- 4 90. Acanthacee ..........-- 27 "65. | ee 11 91. Verbenacez ..........-- 12 56 Jo ree 31 92. Labiatm ...... 15 40 |} ..- 2 93. Plantaginez...........- 1 3 94. Nyctaginese ........--- 2 4 95. Amarantace® ......--- 12 26 96. Chenopodiace ...... 1 1 97. Polygonaees& ......-. ] 8 98. Podostemacez ........- 1 | qo ee 1 99. Nepenthacez ......--- 1 2] . 2 100. Cytinaceg — .... 2 4 101. Aristolochiaces ...... 2 4 jJ sean 3 102, Piperaceæ ee 3 384. ooe 8 103. Choranthacez ........- 1 2 104. Myristicacer .......-- 1 | J e 6 Y LINN. JOURN.—BOTANY, VOL. XXT. 292 MR. R. A. ROLFE ON THE INDIGENOUS. Enpemic. Natural Order. ee Genera. | Species. Genera. Species. 105. Laurinez............... 8 SY / do e 10 106. Proteaces ............ 1 4 [| .-e- 3 107. Thymelzacez ......... 3 Y do ee 3 108. Elwagnaces............ 1 1 109. Loranthacez ......... 3 20 ] e 14 110. Santalaces ............ 2 2 111. Balanophoree......... 1 1 112. Euphorbiacez ......... 29 124 || ..... 40 113. Urticacez............... 25 127 [| ae 43 114. Juglandem ............ 1 2. Jj oe 1 115. Myrieacem ............ 1 l qi eee 1 116. Casuarinee ............ i 3 117. Cupuliferæ ............ 2 92 | We 12 118. Salicinez ............... 1 1 119, Ceratopbyllex......... l 1 120. Gnetacez............... 1 5 121. Coniferee ............... 4 8 j| ...- 1 122. Cycadacese ............ 1 5 123. Hydrocharidee ...... 9 12 124. Burmanniacer ...... 3 7 125. Orchidacez ............ 67 460 | .... 110 126. Scitamineæ ............ 15 97 1 7 127. Hamadoracez ......... 1 1 128. Amaryllidem ......... 5 14 ]| .-e 1 129, Taceacese ............... 1 5 130. Dioscoreaces ......... 3 19 1 4 131. Liliaceæ ............... T 31 MAE 1 132. Pontederiacesze ......... 1 6 133. Philydraceæ ......... 2 2 194. Xyrideæ ............... 1 4 135. Commelynaeesm ...... 7 42 136. Flagellariez............ 1 1 197. Juncacesg ............... 3 3 138. Palme .................. 28 8S8 ] .... 10 139. Pandanes MEN 2 21 |.) ..... 2 140. Typhacee............... 1 1 141. Aroideæ ............... 21 (olo M | ..... 6 142. Lemmaeem ............ 2 6 143. Triuridez............... 1 1 144. Alismaceæ ............ 3 3 145. Naiadacem ............ 6 10 146. Eriocaulem ............ 1 3 147. Cyperacem ............ 15 1660 Joo 3 148. Graminem ............ 67 254 ]|| ..... 1 Lycopodiacee ......... 4 12 Marsdleaces —......... 2 4 Filiees ........... ...... 50 467 ]| .... 59 Dicotyledones ......... 723 2108 4 769 Monocotyledones 273 1340 2 145 Gymnosperms ...... 6 18. j| ... 1 Cryptogamia (Vascu- lares) ............... 56 483 J| ... 52 FLORA OF THE PHILIPPINE ISLANDS. 293 Of the foregoing Orders the Anonacee, Sapindacew, Myrtaceæ, Rubiaceæ, Verbenaces, and Loranthacem are greatly in need of revision. The Orchidacee have been much sought after for hortieultural purposes, but have been deseribed in such a desul- tory manner that the number of endemie species, though cer- tainly considerable, is very doubtful. The Palmew and Pandanew are so poorly represented in herbaria, as well as the Bambusacew (of which a number are enumerated in the ‘ Flora de Filipinas’), that, so far as the endemic element is concerned, the figures given must not be taken for much. A number of Grasses, it is true, have been described as endemic ; but most of those which I have seen must be referred to more widely spread species. The following Orders enumerated by Villar have been excluded as not indigenous :—Papaveracex, Moringex, Cacteæ, Pedalines, Phy- tolaccacew, Bromeliacee, and Iridew; while Tlicincæ has been added. It will thus be seen that 119 out of 165 Orders of Dicotyledons are represented, 26 out of 34 of Monocotyledons, and all three of the Gymnospermee, the whole Phenogamic vegetation consisting of 3466 species, belonging to 1002 genera. The proportion of Monocotyledons to Dicotyledons is more than one half, a very large proportion indeed if correctly stated, as the proportion in tropical insular vegetation is seldom over one fourth, unless over very limited areas. This proportion will probably be very much reduced as our knowledge of the islands extends. The Gymnospermee are poorly represented here, as in the Malayan region generally. The proportion of Vascular Cryptogams to the Phenogamic vegetation is nearly one eighth, chiefly Ferns. There is perhaps no extensive tropical Order of which the distribution is so well known as the Ferns, and moreover the individual species have usually a wider distribution than most plants; yet here we have 52 species not known from elsewhere, or a proportion of over one tenth of the Ferns indigenous to the islands. This alone is sufficient to stamp the islands with a marked individuality. The endemic Phenogamic vegetation consists of 915 species, or a proportion of over one fourth en- demic; the Dicotyledons showing a proportion of over one MN endemie, and the Monocotyledons of a little over one tent ; v i consists chiefly of Orchids. The foregoing figures will doubtless able modification as our knowledge extends ; but emic species y2 require consider l I believe rather to increase the proportion of end 294 MR. R. A. ROLFE ON THE than otherwise, for we know sufficient of the floras of Formosa and Borneo to see that each contains a large number of species distinct from those of the Philippines. To Borneo especially we may look for additional connections. One of the principal reasons why the connection with Borneo is not more marked may be attributed to the fact that it is the southern islands which are so little known—Balabac, Palawan, the Sulu Archipelago, and Mindanao. When these have been well explored the number will probably be very largely increased. Celebes, too, and the Moluccas may be expected to yield several additional connections, more especially with Mindanao. Sufficient, how- ever, is known to show the leading features of the flora; and we may now briefly consider the endemic element in it. One of the most peculiar features which strike us is the large number of endemic species and the very small number of en- demic genera. At present six only of the latter are know from the group. A glance at surrounding islands renders this more apparent. Java has 30 endemie genera, Borneo 28, and Sumatra 14. Further away, we find Ceylon with 17, New Cale- donia with 42, New Guinea with 10, and Fiji with 9. One endemic genus only is recorded from Formosa on the north; but the little island of Hong Kong has four. One only is recorded from Celebes; but this island is little known, and more will pro- bably be found. In most of the foregoing islands the proportion of endemic species is not so much in excess of the Philippines, if indeed, some do not actually fall behind them. Turning to the endemic species we find a number of Orders, represented by a proportion of at least half endemic, as Rubiaces, Myrtacez, Verbenacew, Meliacew, Anonacesm, Myrsinex, Cyr- taudraces, Loranthacer, Cupulifere, Ternstroemiacee, Diptero- carpez, Sterculiaeez, Tiliacez, Burseracez, Celastrinee, Conna- racee, Combretacee, Malpighiacew, Ochnacem, Chailletiacez, Vacciniacee, Ericaces, Aristolochiaces, and a few other small Orders. This estimate is based on the material in the Kew Herbarium, not on Villar’s enumeration. The only species of Myrica* yet known is endemic; llicinee has four endemic * Myrica luzonica, S. Vidal, * Synopsis’ (Atlas), p. 40, t. xc. fig. B, is not a Myrica. It belongs to Euphorbiacez, and may prove to be a species of Sapium ; but at present only the male flowers are known, FLORA OF THE PHILIPPINE ISLANDS. 295 species out of five, and Proteacem has at least three endemic species *. The general and preponderating features of the flora are decidedly Malayan; most of the genera are those which spread over the great Indo-Malayan region, while others are of more restricted range. Thus the Meliaceous genus Dascycoleum has four species, two of which are peculiar to the Philippines and two to Borneo; the Verbenaceous genus Symphorema has three species, two of which are Indian, the third peculiar to the Philip- pines. Alleanthus, in Urticacee, has two species, one peculiar to the Philippines and one to Ceylon. The monotypic genus Octo- meles, a remarkable genus of Datiscemw, appears to be confined to the Philippines, Borneo, and Sumatra. On the other hand, a large number of typical Malayan genera have not yet been detected in the Philippines, though many of them occur in the neighbouring island of Borneo. This fact is a remarkable one, and points to some peculiar conditions in the past history of the islands. Mr. Wallace attributes the absence of such a large number of widely spread Malayan genera of animals to the large amount of submer- gence to which the Philippines have undoubtedly been subjected, and their consequent extinction. This, no donbt, is one of the causes; but when the geographical position and peculiar sur- roundings of the Philippines are borne in mind, it seems probable, and, indeed, almost certain, that a large number both of animals and plants never migrated so far in this direction. The large numbcr of Malayan types present, with the proportion of en- demie species, as well as a considerable boreal and Australian element, to be presently noticed, all seem to point to the fact that submergence alone will not account for the present peculia- rities, and that the former distribution of land aud sea, permitting these successive migrations to reach the islands, while at the same time preventing others, must all be taken into account. The next striking, and perhaps in some respects the most re- markable, feature in the flora is the presence of a considerable * Of the six species of Helicia enumerated by Villar one is based on Cuming no. 2338, which is from Malacca, and must consequently be excluded ; another sent to Kew as H. castaneefolia, Meisn., is H. philippinensis, Meisn., one of the three known endemic species; the remaining one I have not seen. This instance shows the difficulty of estimating the number of endemic species in the face of Villar's identifications. 296 MR. R. A. ROLFE ON THE Australian and Austro-Malayan element. The small Order Stackhousiacez consists of eleven species, one endemic in New Zealand, and the remainder all Australian; nine of them en- demie, but one (S. muricata, Lindl.) occurs also in Luzon, though not yet found in the intervening region. The Liliaceous genus Thysanotus has nineteen species, all Australian; but one species also occurs in the Philippines and in China. Osbornia octodonta, F. Muell., formerly believed to be endemie in Australia, has now been found indigenous in Luzon, where it is common at Lagui- manoc, in the province of Tayabas. Of other Myrtaceous genera are Xanthostemon and Leptospermum: the former has thirteen speeies, two of whieh are endemie in Australia, ten in New Cale- donia, and one (X. Verdugonianus, Naves) in Mindanao; the latter twenty Australian species, eighteen of which are endemie, one (L. scoparium, Forst.) occurring in New Zealand on the one hand, the remaining one (L. flavescens, Sm.) through the Archi- pelago as far as Malacca on the other. Psoralea has eleven Australian species, ten of them endemic, the other (P. badocana, Blanco) occurring again in New Guinea, Panay, aud Luzon. The Epacridaceous genus Leucopogon has 118 Australian species, all en- demie ; two oceur in New Zealand, also endemie ; and a few others occur in the Malay islands, of which one (L. suaveolens, Hook. f.) is peculiar to Borneo and Mindanao. The large Australian Order Proteacee is represented by Helicia, an outlying genus with about twenty-five species, many of them of very restricted range. The four Australian species are all endemic; Luzon has three endemic species, Java two or three, Ceylon one, Amboyna one, and Japan one; one occurs in China, Hong Kong, and Cambodia, two or three are confined to India, besides which there is some undetermined material from Formosa, Borneo, and Sumatra. This genus shows the western migration most distinctly—from Australia to the Moluccas, Java, Sumatra, Malacca, Ceylon, and India; also the northern one, through Borneo and the Philip- pines to Formosa and Japan. It must also have occurred at a comparatively remote period, for so large an amount of specific differentiation to have taken place. Again, Buchnera urticafolia, R. Br., is only known from Australia and Luzon; while Gano- phylivm faicatum, Blume, which occurs in these two places, is found also in the intermediate island of New Guinea. The ditypie Urticaceous genus Aphananthe has one species peculiar to Aus- FLORA OF THE PHILIPPINE ISLANDS. 297 tralia and Luzon, the second species being endemic in Japan. There is also a connection with some of the small Pacific islands. Carruthersia, in Apocynacem, has two Species, one endemic in Fiji, the other in Cebu; while Polypodium simplicifolium, Hook., aud Asplenium Brackenridgei, Baker, are only kuown from the Philippines and Fiji. Garuga mollis, Turez., is only known from the Philippines and Samoa; while Davallia repens, Desv., occurs only in the Philippines, Samoa, and Fiji. Melia Candollei, A. Juss., and Vitex littoralis, Decne., are only known from the Philippines and Timor; Asplenium persicifolium, J. Sm., from the Philippines and Sandwich Islands; while the Urticaceous genus Paratrophis has four species, one endemic in New Zealand, another in Tahiti, a third in Fiji, and a fourth (P. philippinensis, F. Villar) in Luzon. A connection with New Guinea has already been detected, though so little is yet known of this large island. Schizocasia, in Aroideæ, has two species, one confined to the Philippines, the other to New Guinea ; while Odina speciosa, Blume, Epithema Benthami, Clarke, and Asplenium scandens, J. Sm., are peculiar to New Guinea and the Philippines. Species which are at present only known from the Philippines and Australia may be expected to oecur in this intermediate area. The northern element now remains to be noticed, and although not extensive, it will be seen to be of a very marked character. Jsanthera discolor, Maxim., is peculiar to Formosa and Luzon ; Croton Cumingii, Muell. Arg., is peculiar to Luzon, Formosa, and the Loo Choo Islands; Vernonia philippinensis is an endemic species closely allied to V. Cumingiana, Benth., a peculiar Hong- Kong species ; Ligustrum Cumingianum, Deene., also endemie, is closely allied to a group of Chinese species ; Scutellaria luzonica is an endemic species whose nearest ally seems to be $. indica, L., a species reaching from the Himalayas to J apan ; Ophiopogon spicatus, Ker, reported as indigenous, but of which I have not seen a specimen, is a native of the same region. There are also three endemic species of Carex (C. fibrata, Boott, C. Cumingii, Boott, and C. Cumingiana, Steud.), a ty pical northern genus. The same may be said of the genus Pinus, represented here by P. insularis, Endl., and P. Merkusii, Jungh.—the former endemic in the province of Nueva Ecija, in north-eastern Luzon the latter believed to be limited in Luzon to the province of Zambales 298 MR. R. A. ROLFE ON THE (also on the eastern coast, but further south), but occurring again in Sumatra. Lilium Wallisii, Baker, and Viburnum luzoni- cum, both endemic in Luzon, are outlying species, closely allied to others from the opposite Chinese coast, where, and in Japan, both genera are well represented. Clerodendron intermedium, Cham., from Luzon, is closely allied to a Formosan species, and again to a species from Celebes but the material is not suffi- cient to prove their identity. Guettardella has one species endemic in Bohol, another in Hong Kong, and a third in Aus- tralia ; this genus, however, is now made a section of Antimonius, a more widely diffused genus: so that the two last-named ex- . amples may be taken to represent northern outliers of Malayan genera which have extended through the Philippines. Clero- dendron is strongly represented here, having nine endemie species. Lactuca brevirostris, Champ., occurs in Luzon, Formosa, Hong Kong, China, Japan, and Northern India. To return now to the Malayan element, we may trace the connections with the principal Malay islands. First, then, from its position we should suspect that Borneo would have the closest affinity with the Philippines. This, I believe, will prove to be the case ; but at present its flora is very imperfectly worked out. Dasycoleum, with two Bornean and two Philippine species, has already been mentioned; also Lewcopogon suaveolens, Hook. f.; the latter confined to Borneo and Mindanao. -Besides these, Myristica guatteriafolia, A. DC., formerly only known from the Philippines, has now been found in the little island of Labuan off North-east Borneo (Mottley 139). Dipterocarpus grandiflora, Blanco, Pipturus asper, Wedd., Cyrtopera squalida. Reichb. f., and Lindsaya concinna, J. Sm., are also limited to Borneo and the Philippines. Of Celebes so little is known that it would be difficult as yet to point out many very clear affinities ; some, however, undoubt- edly exist. Semecarpus Perottetii, March, Aglaia macrobotrys, Turez., and Momordica ovata, Cogn., are only known from Celebes and the Philippines; but besides these a number of plauts, collected at Gorontalo by Mr. Riedel, are either identical with, or closely allied to, Philippine species, but they require more thoroughly working up. Java is better known than any of the Malayan islands, and has a considerable affinity with the Philippines. Pangium edule, Reinw., formerly considered endemic in Java, has now been found FLORA OF THE PHILIPPINE ISLANDS. 209 in the Philippines. Rhododendron javanicum, Blume, Andrachne australis, Zoll, Antidesma montanum, Blume, Phyllanthus buxi- Jolius, Muell. Arg., Conocephalus ovatus, Trécul, Asplenium aniso- dontum, Presl, A. woodwardioides, Baker, Polypodium papillosum, Blume, and Helionitis Zollingeri, Kurz, are at present only known from Java and the Philippines; the last-named, if correctly loca- lized, is from the island of Panay. Again, Agleonema simplex, Blume, occurs in Java, and again in the small island of Malamaui, close to Basilan Island, south of Mindanao in the Sulu Archi- pelago. Sumatra has a considerable number of Philippine plants which occur over a limited area, besides which Pothos inequilaterus, Presl, Pinus Merkusii, Jungh., and Davallia decurrens, Hook., are at present only known from Sumatra and the Philippines ; while Octomeles sumatrana, Miq., from Sumatra, Borneo, and the Philippines, has not yet been found in Java. Litsea cinnamomea, Blume, originally described from Ternate in the Moluccas, has now been found in Luzon; Procris grandis, Wedd., from Samar *, has been found in Buru; Cyathea integra, J. Sm., is peculiar to the Philippines and Amboyna; Nephro- dium obscurum, Hook., to the Philippines and Tavoy ; Poly- podium splendens, Hook., to the Philippines and Singapore ; and Lygodium hastatum, Desv., to the Philippines and Marianne Islands. Several species appear to be peculiar to the Philippines and the Malay Peninsula, as:— Lindsaya scandens, Hook., Pteris ludens, Wall., Nephrodium crassifolium, Hook., Polypodium sessilifolium, Hook., P. stenophyllum, Blume, P. longifolium, Mett., P. palmatum, Blume, and Nephrodium giganteum, Baker; while Asplenium sor- sogonense, Presl, and Polypodium flaccigerum, Mett., extend to the Himalayas. The Indian connection is also seen in Asplenium Wightianum, Wall., Nephrodium recedens, Hook., and N. Otaria, Baker, which at present are only known from the Philippines, * Weddell (Monogr. Urtie. p. 337) says :—“ Hab. In Nova Guinea —(¥. 8. in Herb. Hook.)”. The only specimen in Herb. Hook. is Cuming n. 1730, from the island of Samar; on this sheet Weddell himself has written “ Procris grandis, Wedd. ;” consequently it is the type. T cannot account for the locality given unless Weddell accidentally omitted to copy it at the time, and supplied it from memory (with the “?”) afterwards, when he had not the sheet to refer to. A plant collected by Riedel at Buru, in the Moluccas, appears to be the same species, though the cyme is a little more pubescent. 300 MR. R. A. ROLFE ON THE Ceylon, and India; while, on the other hand, a large number of species which are scattered about the Malayan Islands do not reach India. Besides a number of species mentioned earlier in this paper, the connection between China and Formosa and the tropical Malayan Islands by way of the Philippines is supported by Alsophila tomentosa, Hook., and Polypodium lomarioides, Kunze, which occur in Formosa on the one hand and in the Malay Islands on the other, the former limited to Java in the south. There are also several species which are scattered about the Polynesian Islands, occurring both in Java and the Philippines, but not found further westward. So little is known of a considerable number of the islands that little can be said of them individually, but from what few indi- cations we have it may safely be inferred that many interesting problems of distribution will yet come to light. The Sulu Archipelago, the islands between Mindanao and Celebes, also the islands of Balabac and Palawan, are all likely to yield connecting links with Celebes and Borneo. Between Luzon and Formosa the islands are so small that the connecting links are less likely to have survived the changes which must have taken place since the period when the migration southward occurred—probably during the cold of the Glacial Epoch. Mindanao has the highest mountain in the group, the volcano of Apo reaching to over 8000 feet. Xanthostemon Verdugonianus, Naves, and Gyrinopsis Cumingiana, Decne., are endemic here ; while Leptospermum flave- scens, Sm., and Leucopogon suaveolens, Hook. f., are not known from further north. Leyte is separated from Mindanao by the Surigao Straits ; it is about a twelfth as large as Luzon. Cuming collected some plants here, of which Ilex philippinensis, Melastoma pencillatum, Naud., and Osmelia conferta, Benth., have not yet been found elsewhere. Samar lies to the east, a little north, and is slightly larger than Leyte; it is separated from it by the very narrow San Juanico Straits, while the San Bernardino Straits separate it from the southern point of Luzon. Of the plants collected here by Cuming, Grewia eriopoda, Turez., Begonia quercifolia, A.DC., Osmelia philippinensis, Benth., Buchanania nitida, Engl., Lepi- dagathis laxa, Nees, Dracontomelum Cumingianum, Baill., and Cyclostemon Cumingii, Baill., are only as yet known from Samar. FLORA OF THE PHILIPPINE ISLANDS. 301 Bohol lies south of Leyte, near to Mindanao, and is about as large as Leyte. Of plants collected here by Cuming, Ryssopteris microstemma, A. Juss., Guettardella philippinensis, Benth., Melo- dinus Cumingii, A. DC., Aischynanthus philippinensis, Clarke, Eria vulpina, Reichb. f., E. retroflexa, Lindl., Arundina speciosa, Lindl., Calanthe conspicua, Lindl. and Cypripedium Argus, Reichb. f., are not known from elsewhere. Cebu is a long narrow island north-west of Bohol, and slightly smaller. Cuming also collected here, and of his plants Kayea philippinensis, Planch, Rysso- pteris dealbata, A. Juss., Carruthersia pilosa, F. Villar, and Carex Jibrata, Boott, are not yet found elsewhere. Negros is sepa- rated from Cebu— which lies almost parallel to it—by the Tanon Straits; it is about as large as Samar, and contains a volcanic mountain 7500 feet high. Cuming visited the island, and of the plants obtained by him Carex Cumingiana, Steud., and Voacanga Cumingiana appear to be endemic. Panay lies north-west of Negros, and about equals it in size. Cuming collected some plants here; a few also occur in Vidal’s collection. Of these Utricularia rosulata, Benj., is not known from elsewhere. The island of Luzon has been considerably explored, though some parts of it are very little known, especially the Pacifie side of the northern half of the island; but sufficient data have been collected for a brief outline of its physieal conditions and vegetation. Commencing at the southernmost point, we find Bulusan, the most aetive voleano in the island, reaching to 5000 feet; a little further north is Sorsogon: then comes Mayon, a less active volcano, barren at the summit, and reaching to over 7000 feet above sea-level. Still going northward we come to Iriga, reaching 5000-6000 feet, and clothed almost to the summit with vegetation; and then to the quiescent volcano of Isarog, 6000 feet high. Just to the north of this isa depression running right across the island, and consisting entirely of raised coral- reefs. Before the period of elevation set in, this formed a sepa- rate island. North-east of this is Labo, nearly 5000 feet high ; and a little further on in the same direction a bay may be seen on either coast, the intervening area consisting of raised coral- reefs. This also marks the limit of a former separate island, of which the previously mentioned reef formed the southern limit. This district, reaching to the southern point of Luzon, is the pro- vinee of South Camarines. Some of the endemic species of the 302 MR. R. A. ROLFE ON THE island are confined to this province, doubtless owing to their former isolation. It isalso interesting to notice that the volcanic activity gradually increases southward to the extreme point, showing which way the subterranean forces are moving. North of this second depression the mountains rise again, and may be traced in an irregular line for quite half the remaining distance to the northern point of the island, when they fork, and are con- tinued in two distinct branches, the voleano of Cagua, at the extreme northern point on the eastern brauch, reaching over 3500 feet above sea-level. The Pacifie side of the island is much moister than the opposite or western side, and the different zones of vegetation on the mountains descend much lower. In the damp dense forests of this Pacific coast a considerable number of Palms occur; but the rainfall is so heavy, and the natives so hos- tile in many parts, that comparatively little has been collected. On the contrary, the western side has been considerably explored ; the forests are dense, but drier than those of the Pacific coast. The genus Pinus is confined to this western side, the endemic P. insularis, Endl., occurring in the province of Nueva Ecija, and not found further south than 15° north latitude. Pinus Merkusit, Jungh., is found further south, in the province of Zambaies, in a somewhat restricted area, but occurs again in the islaud of Sumatra. The greater part of the northern element also is limited to this western district, some of the species only occurring in the north of the island. The general character of the vegeta- tion is similar to that of the other Malay islands ; large dense forests cover the lower elevations. Anonacee, Guttiferz, Ster- euliaceze, Tiliacew, Rutacee, Burseracer, Meliaceæ, Sapindacee, Leguminose, Myrtacew, Rubiacex, Verbenacee, Huphorbiacea, Urticacee, and Palms are all abundant; while a considerable number of other Orders are also well represented. — Diptero- carpez and Sapotacez are chiefly gigantic trees occuiring deep in the forests. Climbing-plants are abundant—Menispermacea, Phytoerenez, Ampelidez, Cucurbitacez, and the genus Calamus of the Palms being the principal ones. The epiphytal Orchids are also very numerous; and in the Mangrove-swamps (oeeupy- ing a considerable area) Loranthacee are abundant. Ascend- ing a short distance up the mountains, Ternstroemiacee, Melastomacez, Myrsinew, and Cupuliferz are well represented, together with a number of smaller Orders. ‘Towards the summit E! FLORA OF THE PHILIPPINE ISLANDS. 303 a veritable mountain flora is found, including several species of Ericaceæ and Vaccinem. Gunnera macrophylla, Blume, hitherto known only from Java and Sumatra, has recently been dis- covered, occupying a considerable area at the extreme summit of the extinct volcano of Banahao. Herbaceous plants are tolerably numerous, a few Orders (as Cyrtandracee) having a consider- able proportion of endemic species, but for the most part consist- ing of widely diffused plants. To these may be added a few mere weeds of cultivation, and accidentally introduced plants. Arge- mone mexicana, L., is now a great pest. Muntingia Calabura, L., an American plant of recent introduction, is now spreading rapidly. Dalea nigra, Mart. et Gal, is a Mexican plaut, now very abundant. This plant is probably an introduction of very ancient date, brought by shipping at the time when Spain an- nually sent ships direct from Mexico to the Philippine Islands, more than one of which was captured by English sea-captains in those disturbed times. Prosopis juliflora, DC., occupies a large area round the Bay of Manila, but is evidently an Ameri- ean introduction, although nothing is known of it except that it is now so abundant as to appear truly indigenous. Carica Papaya, L., is an American introduction of the 17th century. There are also several other American plants, of which all traces of the introduction are completely lost, besides a number of plauts of yet older iutroduetion. Before the islands were taken posses- sion of by the Spaniards, successive. Malayan invasions took place, and with these were introduced many of the fruit-bearing trees and other plants which are universally cultivated through- out the whole Malayan Archipelago, The more characteristic features of the flora have now been traced ; and it remains to be seen how far this will help us in arriving at its origin and in tracing the past history of the islands. It is probable that during the period while the present genera were becoming differentiated, a considerable portion of the islands was under water. The fact that nearly all the genera have their headquarters to the south and south-west, as well as the absence of such a large number of typical Malayan genera, both point to the fact that the flora did not originate where it now exists, but reached its present location by migration northward. The paucity of endemie genera also tends to confirm this view. Whether previous to this period the Philippines formed part of the great 304 MR. R. A. ROLFE ON THE Asiatic continent or not, there does not seem sufficient evidence to show ; but at length a period of elevation set in, owing to volcanic agency ; the ancient sea-bottom emerged from beneath the water, and the subterranean forces burst through the sedimentary rocks, burying them with the ejected scorie and lava. A considerable period would be required for this process ; and it is evident that at length the islands— probably very different in configuration and extent from what they are now— became connected in some degree with the land which undoubtedly existed to the southward, and where the typical Malayan flora probably originated. Mean- time a more complete connection through the Archipelago than at present had been favouring a migration from Australia west- ward. The then existing species gradually migrated northward through the Philippines, some reaching Mindanao, others Luzon, and some passing through to Formosa, a few of which even extended to Japan by way of the Loo Choo Islands. Some others appear to have spread across from Formosa to South China and Hong Kong, as Thysanotus chrysantherus, F. Muell., the only outlying species of a genus all Australian and with eighteen others endemic in Australia. It is not necessary to assume that the Australian and Malayan migrations were contemporaneous, or that they passed over by one single connection ; on the con- trary, a few species seem to have reached the Philippines by direct communication with the Moluccas, at a time when circum- stances did not allow them to reach either Java or Borneo. The typical Malayan genera, however, seem to have passed direct through Borneo, and perhaps some from Borneo itself. Besides these connections there are evidences of a former direct connec- tion with the north-eastern point of Celebes, by way of a sunken reef, of which a few small islands are now the only remaining trace. In support of this view it may be noted that the volcanoes of Celebes are confined to the north-eastern portion; and it seems probable that the islets stretching away to the south- eastern point of Mindanao are only voleanic connections between the two. The present condition of the volcanic forces in the Philippines probably differs only in intensity from that of the past; and, if this view be correct, it is evident that periodical eruptions have taken place since the time of their emergence ; these eruptions having been to a great extent local, sometimes in one locality, sometimes in another, but formerly of gigantic dimensions,—as is proved by the enormous volcanie mountains S FLORA OF THE PHILIPPINE ISLANDS. 305 everywhere to be found throughout the group. During these up- heavals, with accompanying subsidences in adjacent areas, it is probable that the various islands were severed and reconnected in different directions, not only with adjacent regions, but with each other, and that the various plants and animals diffused themselves about as cireumstances permitted, the various con- nections serving as stepping-stones from one island to another. Possibly they were once more continuous; but, on the other hand, the island of Luzon was formerly cut into three, and there is abundant evidence to prove that the southern portion was long separated from the northern. Coral-reefs, too, abound in various directions throughout the group, and afford certain evidences of former submersion; but it seems probable that these subsidences were of a local nature rather than a recent general subsidence, which must have destroyed a large portion of its endemie vegetation. At the time when the land-connection with the Chinese region permitted a partial migration northward, a certain commingling of the two floras took place, some of the Chinese plants moving southward. After a considerable period of stability this south- ward migration was probably hastened by the cold of the early Glacial Period, during which the connections were being gradually severed, some of the species reaching Formosa, some the northern part of Luzon, and others getting still further southward. The connections to the southward were also being severed; and thus the various plants and animals became isolated, some of them with a wide dispersion within the group, and others with a considerably restricted range. A period of comparative sta- bility at length set in, which has continued down to the present time; and during this period of long isolation the species have become modified into distinct local species, some of them having a very restricted range within the group. And, lastly, since the period when the connections with surrounding countries were cut off, a number of additional plants have reached the islands—by marine currents and other causes known to effect the dispersion of plants, and by man’s agency—some of which are so thoroughly established or so widely diffused as to appear truly indigenous. Mr. Wallace, in 1876 (‘Geographical Distribution of Animals,’ vol. i. p. 344), speaking of the Philippines, remarked :—‘ They are, in fact, truly insular, while the other [Malay] islands are really continental in all the essential features of their natural 306 MR. R. A. ROLFE ON THE history." He also says (p. 846) :—' We find apparently two sets of animals—a more remote series. . .in which the species are distinct from any others, and a more recent series ... . identical with common Malayan animals. The former indicate the earliest period when these voleanie islands were connected with some part of the Malayan subregion. . . . The latter may indicate either the termination of the period of union or merely the effects of introduction by man. The reason why a larger number of mammalian forms were not introduced and established was probably because the union was effected only with some small islands, and from these communicated to other parts of the Archipelago; or it may well be that later subsidences extin- guished some of the forms that had established themselves." Four years later (‘Island Life, p. 361) he gives a somewhat different explanation of the peculiar phenomena; he says :—“ It is evident that the Philippiues once formed part of the great Malayan extension of Asia, but that they were separated con- siderably earlier than Java.... The reason of their comparative poverty in genera and species of the higher animals is, that they have been subjected to a great amount of submersion in recent times." If the Philippines had once formed part of the great Malayan extension of Asia, whieh was separated earlier than Java, it would have required an almost total submersion to have caused such wholesale extinction of typical Malayan tribes and genera; while, on the other hand, many of the survivors are not more fitted, and some much less so, to survive such a submersion, than many of those believed to be absent at the present time. Geo- logical evidence will probably in future throw much light on this point. Meantime the present evidence appears to support the former of these two opinions. It has been shown that nearly all the genera have their headquarters elsewhere, the few which are endemic being principally monotypic and closely allied to those of surrounding countries; thus proving that the present flora has been chiefly or altogether derived from these sources, and is consequently of more recent origin, and that in many cases the lines of migration can easily be traced. It has also been shown that, with the exception of a large number of species of very wide distribution, or that have been directly or indirectly iutrodueed by human agency, this migration (whieh occurred after a period of stupendous volcanic upheaval) was sufficiently remote to FLORA OF THE PHILIPPINE ISLANDS. 307 allow of a large amount of specific modification under altered circumstances; also that with the waning of the volcanic forces a period of comparative stability set in, allowing a limited amount of specialization in the different islands themselves. I think it will therefore be conceded that the foregoing outlines show that the Philippines are truly insular in the essential features of their natural history ; not so much through being an early sepa- ration of the Asiatic continent which has since had a dip under the sea, as through their being largely of volcanic—and, geo- logically, of somewhat recent—origin, as in the case of other islands of admitted oceanic origin. The following alterations, with descriptions of a few novelties, have been made during a study of the Philippine collections. A number of additions to Villar's ‘ Novissima Appendix’ had been written out, also species included by error, or that will have to be reduced as synonyms, had been indicated ; together with afew erroneously reduced synonyms, or that will have to be trans- ferred to other species; but as in the absence of a complete set of authentic specimens the Appendix would still be incomplete, it has been thought better to suppress them, at least for the present, though they are taken account of in estimating the numbers for the foregoing Census. DILLENIACEE. DILLENIA PHILIPPINENSIS, Rolfe.—D. indica, Blanco, Fl. Filip. ed. 1, p. 472; ed. 3, vol. iv. Nov. App. p.3; non L.—D. speciosa, Blanco, l.c. ed. 2, p. 329 ; ed. 3, vol. ii. p. 244, tab. 199 ; non Thunb. —Riefferschiedia ?, Llanos in Mem. de la Rl. Ac. de Cienc. de Mad. iv. ser. 3, tom. ii. part 3, p. 508. Nom. vern. “ Catmon.” Differs from D. indica, L., in the shorter, more elliptical leaves without attenuate bases, the much fewer and more distant nerves, and other characters, MAGNOLIACER. Tatauma VILLARIANA, Rolfe—T. mutabilis, F. Villar, in Blanco, Fl. Filip. ed. 8, vol. iv. p. 8, tab. 148, excl. syn., non Blume. Differs from T. mutabilis, Blume, in the more numerous, more oblique nerves, longer buds and sepals, and other characters. LINN. JOURN. — BOTANY, VOL. XXI. z 808 MR. R. A. ROLFE ON THE MALYACEF. THESPESIA CAMPYLOSIPHON, Rolfe.—Hibiscus campylosiphon, Turez. in Bull. Soc. Nat. Mosc. xxxi. pt. 1, p. 193; F. Villar, l. c. p. 25.—H. Vidalianus, Naves, in Blanco, Fl. Filip. ed. 3, t. 333.—H. grewiefolius, F. Villar, l. c. p. 24, excl. syn. plur., non Hassk. F. Villar enumerates this species twice; once wrongly referring it to H. grewiefolius, Hassk., which has not been found in the Philippines, and must therefore be excluded. STERCULIACEX. l SrERCULIA Brancot, Rolfe.—S. alata, Blanco, Fl. Filip. ed. 2, p. 525; ed. 3, vol. iii. p. 165 ; vol. iv. Nov. App. p. 27, t. 401; non Roxb. TILIACER. CotumsBia Brawcor, Rolfe.—C. floribunda, Naves in Blanco, l. c. ed. 3, t. 312; F. Villar, 1. c. vol. iv. Nov. App. p. 30, non Wall. This plant is referred to C. floribunda, Wall, which has five primary nerves and is otherwise different. It must be excluded. ILICINES. Of this Order, not yet recorded from the Philippines, the following species are known :— ILEX cymosa, Blume.—Luzon, Mt. Caraballo, 2000 ped. alt. ; S. Vidal n. 434! I. CUMINGIANA, sp. nov.—Rami glabri teretes, foliis oblongis. acuminatis integris, basi cuneatis coriaceis nitidis, venis obliquis, panieulis terminalibus quam folia duplo brevioribus, calycis lobis triangularibus, corolle tubo brevi, segmentis ovato-oblongis, staminibus petalis qualibus, ovario glabro, stylo bipartito, stig- matibus simplieibus; fructu ignoto. Luzon, Prov. Albay, Cuming n. 1241! Leaves 3-43 by 11-13 in., petioles 4—3 in.; primary nerves about nine pairs, a little prominent on both sides. Panicles 2 in. long, somewhat lax. Flowers 2 lin. in diameter. I. PHILIPPINENSIS, sp. nov.—Rami glabri teretes, foliis late ellipticis abrupte acuminatis, integris, basi obtusis coriaceis, venis patentibus, eymis axillaribus multifloris quam folia vix breviori- bus, calycis lobis triangularibus, corolle tubo brevi, segmentis ovato-oblongis, staminibus petalis excedentibus, ovario glabro, stylo breviter 3-fido, stigmatibus crassiusculis ; fructu ignoto. Philippines, island of Leyte, Cuming n. 1748! FLORA OF THE PHILIPPINE ISLANDS. 309 Leaves 2-3 by 14-2 in., petioles 4-6 lin. ; primary nerves 7-10 pairs, a little prominent on both sides. Cymes 2 in. broad and nearly as long, dense. Flowers 2 lin. in diameter. I. LoBBIANA, sp. nov.—Rami tetragoni dense rufo-pubescentes, foliis elliptico-ovatis obtusis crenulatis coriaceis nitidis, venis obsoletis, cymis axillaribus brevibus, calycis lobis rotundatis, corollz tubo brevi, segmentis rotundato-ovatis, staminibus petalis fere equalibus, stigmate sessili; fructu ignoto. Luzon, Lobb! Leaves 1-1} in. by 7-9 lin. petioles 2-3 lin. Cymes j in., few-flowered. Peduncles2 lin. Flowers 2 lin. in diameter. I. tuzonica, sp. nov.—Rami angulati subalati glabri, foliis ellipticis obtusis crenulatis basi cuneatis coriaceis, stipulis lanceo- latis, floribus ignotis, sed pedunculis axillaribus, medio bibracteo- latis supra crassiusculis, calycis lobis late rotundatis, fructibus globosis, stylo brevi integro, stigmate capitato. Luzon, Lobb! Centre of Luzon, S. Vidal, n. 66! Leaves 6-11 by 4-5 lin., petioles 1 lin. Fruiting peduncles 5 lin. Fruits 2-21 lin. in diameter. Readily distinguished from the last by the glabrous, almost winged branches and smaller leaves. SAPINDACEE. ZOLLINGERIA TRIPTERA, Zojfe.—Melicoeca triptera, Blanco, l. c. ed. 2, p. 203; ed.3, vol. ii. p. 16.—Z. macrocarpa, F. Villar, l. c. p. 53; S. Vidal, Synopsis, t. 35. fig. C; non Kurz. This species is distinct from Z. macrocarpa, Kurz, which must be excluded. LEGUMINOS#. [Dalea nigra, Mart. et Gal.—D. alopecuroides, Blanco, l. c. ed. 2, p. 389, ed. 3, vol. ii. p. 251; vol. iv. Nov. App. p. 58, excl. syn.; non Nutt.—The plant referred to D. alopecuroides, Nutt., is this species ; but all the synonyms except that of Blaneo must be excluded. It is a Mexican plant, introduced long ago, pro- bably by shipping, and now very abundant.] PrerocarPus VipALIANUS, Rolfe.—P. erinaceus, F. Villar, l. c. vol. iv. Nov. App. p. 68, excl. syn.; S. Vidal, Synopsis, t. 40. fig. B, non Poir. A most distinct species. Crupra Braxcor, Rolfe.—C. spicata, Blanco, l. c. ed. 2, p. 261 ed. 3, vol. ii. p. 121; vol. iv. Nov. App. p. 71, t. 244, excl. syn. ; non Willd. Hitherto confounded with an American species. Z2 310 MR. R. A. ROLFE ON THE COMBRETACEX. TERMINALIA CaLaMANSANAY, Rolfe.—Gimbernatia Calaman- sanay, Blanco, l. c. ed. 2, p. 266; ed. 3, vol. ii. p. 129.— T. bialata, F. Villar, l. c. p. 80, excl. syn. plur., non Kurz. This species is distinct from T. bialata, Kurz, which must be excluded. MELASTOMACES. CARIONIA TRIPLINERVIA, Sp. nov.—Rami teretes glabri, foliis lanceolato-elliptieis breviter acuminatis triplinerviis coriaceis glabris, cymis terminalibus 5-floris, floribus bracteolatis, calyce campanulato, lobis 6 minutissimis acutis, petalis 6 contortis obovatis, staminibus ovarioque ut in generibus ; fructu ignoto. Luzon, Volcano Mayon, Prov. Albay, 5100 ft. alt. ; S. Vidal, n. 779! Leaves 11-21 by 3-1 in., petiole 1 in. Peduncles 12 in., pedi- cels 4 in., with 4-6 peculiar and extremely narrow linear brac- teoles, 2 in. long, at their bases. Calyx4lin. long. Petals 8 lin. long. I have not seen a specimen of C. elegans, Naud., the type of the genus, but the floral structure agrees with Naudin’s descrip- tion and figure. It is, however, evident that this is a second species, for Naudin describes the leaves of his plant as “ quintu- plinerviis," and the cymes as “ 3-floris aut paucifloris.”” Naudin does not mention anything about bracteoles in his plant, and the figure represents the calyx-lobes as over a line long, while in the present species they are excessively minute. ARALIACER. PANAX CuurNGIANA, Rolfe.—Paratrophia Cumingiana, Presl, Epim. Bot. p. 250.—Polyscias Cumingiana, F. Villar, l. c. p. 102, excl. syn. F. Villar refers this plant to Polyscias, but it must go to Panaz. CAPRIFOLIACEX. VIBURNUM LUZONICUM, sp. nov.—Rami teretes ferrugineo- tomentosi, folis ovatis acuminatis denticulatis v. subintegris, junius utrinque rufo-tomentosis demum supra puberulis, cymis terminalibus densis rufo-tomentosis, calycis lobis brevibus rotun- dato-ovatis, corolle tubo brevissimo, lobis oblongis extus pubes- centibus, staminibus styloque brevibus, stigmate capitato, fructu ovoideo compresso. Luzon, Prov. Albay, Cuming n. 1345 ! FLORA OF THE PHILIPPINE ISLANDS. 311 Leaves 2-3 by 1-14 in.; veius 5-7 pairs, very oblique; teeth small and distant, sometimes almost obsolete; petioles } in. Cymes 12-2 in. diam. Flowers llin. Fruit 21-3 lin. RvuBIACES. MUSSENDA GRANDIFLORA, Rolfe. — Calycophyllum grandi- florum, Meyen, Reise, ii. p. 234; Walp. in Pl. Meyen. p. 356.— M. frondosa, Blanco, l. c. ed. 1, p. 167; ed. 2, p. 118; ed. 3, vol. i. p. 211; vol. iv. p. 107, t. 58, excl. vars. ; non L. The varieties enumerated by F. Villar scarcely belong here. It is doubtful, however, if they belong to M. frondosa, L., as I have not yet seen this species from the Philippines. VILLARIA, gen. nov.—Calycis tubus infundibularis ; limbus 5- dentatus. Corolla hypocraterimorpha, face villosissima ; limbi lobi 5, patentes, breves, stricte contorti. Stamina 5, fauci coroll; inserta, filamentis brevissimis ; antherz subsessiles, dorso affixi:e, lineares, utrinque acute, incluse. Discus annularis, parvus. Ovarium 1-loculare; stylus basi gracilis, apice fusiformis, villo- sissimus, vertice integro; ovula 4-8 in utraque placenta, hori- zontalia; placentis 2 parietalibus. Fructus ignotus.—Frutex glaber, inermis, ramulis teretibus. Folia opposita, coriacea. Stipule intrapetiolares, triaugulares, persistentes. Cyme axil- lares, pedunculate, pauci- v. multiflore. Bracteæ triangulares, basi connate. Flores parvi. V. PHILIPPINENSIS, sp. unica.—Frutex glaber, ramo tereti, folis elliptico-oblongis, breviter acuminatis, integris, coriaceis, basi cuneatis, 2-3 poll. longis, 1-14 poll. latis, petiolo 4 poll. longo, pedunculis gracilibus, 1-12 poll. longis, pedicellis 1 lin. longis, calyce 4 lin. longo, lobis rotundatis brevbius, coroll: tubo 2 lin. longo, lobis rotundato-ovatis. (Plate X. figs. 1-8.) Hab. Minalabat, Luzon, S. Vidal n. 836! Nom. vern. * Lasgas." The position of this genus appears to be next to Gardenia, with which it agrees in its one-celled ovary ; it is also much allied to some of the immediately following genera, but the ovary and other characters readily distinguish it from them. Since the above was written I have seen some additional material belonging to the genus. Cuming n. 874, from Prov. Albay, Luzon, is a vigorous branch with broader stipules, larger leaves, and longer cymes, but apparently otherwise quite ideutical with the fore- 312 MR. R. A. ROLFE ON THE going. Cuming n. 1271, from the same locality, is a vigorous branch with very broad stipules, obovate, very obtuse leaves with cuneate bases, 4—5 in. long by 21-4 in. broad, short cymes only one third as long as the leaves, and slightly larger flowers ; but I suspect it will prove to be only aform of the above, and not a distinct species.. Neither of these plants is in the Kew Herba- rium. The genus is named in commemoration of Father Fer- nandez Villar, to whose energy we are largely indebted for the third edition of Blanco’s ‘ Flora.’ ANTIRRH@A PHILIPPINENSIS, Folfe.—Guettardella philippi- nensis, Benth. in Hook. Kew Journ. iv. p. 197.—Antirrhea sp., F. Villar, l.c. p. 109. This is enumerated without a specific name. Psycnorri1a Taceo, Rolfe—Pæœæderia Tacpo, Blanco, l.c. ed. 1, p. 160; ed. 2, p. 113 ; ed. 3, vol. i. p. 202, t. 55.—P. malayana, F. Villar, l. c. p. 112, excl. syn. plur., non Jack. Blanco’s plant is referred by Villar to P. malayana, Jack, which is distinct, and must be excluded. COMPOSITE. VERNONIA PHILIPPINENSIS, sp. nov.—Fruticosa, ramulis ferru- gineo-tomentosis striatis teretibus, foliis petiolatis elliptico-ob- longis acuminatis, basi cuneatis, supra puberulis, subtus tomen- tosis, paniculis terminalibus amplis, capitulis pedicellatis circa 32-floris, involucro globoso squamis ovatis interioribus oblongis obtusiusculis, achznio pubescenti 10-striato, pappi setis rufis quam involucrum duplo longioribus. Luzon, Prov. Albay, Cuming n. 1092! This species is much allied to V. Cumingiana, Benth., in Kew Journ. of Bot. iv. p. 238, from Hong Kong, with which the Indian V. Andersoni, Clarke, has been identified; but differs in the broader heads with more numerous flowers, also in the more striate branches, narrower leaf-bases, and the more pubescent, less strongly striate achenes. These two species have been, and still are, somewhat confused. Bentham, in his *Florula Hongkongensis,” published in the ‘Kew Journal of Botany,’ /. ¢.,in describing the Hong-Kong plant, gave the specific name Cumingiana, on the assumption that the Philippine plant was identical. Hooker (Fl. Brit. Ind. iii. p. 241), in identifying the Hong-Kong plant with V. Ander- sont, Clarke (Comp. Ind. p. 27), says:—' V. Andersoni, Clarke, V. Cumingiana, Benth., Fl. Hong Kong, p. 170, not of Kew Journ. iio NR n nomenen y FLORA OF THE PHILIPPINE ISLANDS. 313 Bot. iv. p. 233. Bentham is mistaken in referring the Hong- Kong plant to his V. Cumingiana of the Philippines." But both Bentham's descriptions are made from the Hong-Kong plant, and his name must consequently supersede V. Andersoni, which is more recent. The Philippine plant, until now undescribed, may be called F. philippinensis. BLUMEA TETRAPTERA, Aolfe.—Conyza tetraptera, Turez. in Bull. Soc. Nat. Mose. 1851, pt. 1, p. 178.— F. Villar, l. c. p. 115. This plant must go to Blumea. APOCYNACER. VoacaANGA CUMINGIANA, sp. nov.—Rami glabri, foliis obovato- oblongis abrupte breviterque acuminatis, basi attenuatis, glabris, cymis 3-floris breviter peduneulatis, calyce tubuloso segmentis rotundatis, corolle tubo angusto, segmentis rotundato-ellipticis patentibus, stigmatibus styloque inclusis. Island of Negros, Philippines, Cuming n. 1806! Leaves 2-4 by 1-1} in., veins spreading about 6 pairs, petiole 6-7 lin. Peduncle and pedicels 3 in. long. Calyx 1 in., its lobes l lin. long. Corolla-tube 14 in., lobes 1 in. long. A marked species ; but as the two specimens are small I have not dissected a flower to examine the stamens &c. IcuwocanPus Navest, olfe—I.? frutescens?, Naves in Blanco, l. c. ed. 3, tab. 97, non R. Br.—l. ovatifolius, F. Villar, l. c. p. 131, non R. Br. Villar refers this plate to T. ovatifolius, A. DC., but it is quite distinct. i KickxrA Braxcor, Rolfe—Anasser floribus axillaribus soli- tariis, Blanco, l. c. ed. 1, p. 114; ed. 2, p. 81; ed. 3, vol. i. p. 149 (in nota).—K. arborea, F. Villar, l. c. p. 132, t. 428, non Blume. F. Villar refers this. to K. arborea, Blume, which, however, is quite distinct, and must be excluded. BIGNONIACER. STEREOSPERMUM QUADRIPINNATUM, F. Villar, l. c. p. 151, t. 252. —Millingtonia quadripinnata, Blanco, l. c. ed. 1, p. 501; ed. 2, p. 351; ed. 3, vol. ii. p. 286, t. 252. Prov. Albay, Luzon, Cuming n. 1003!; Santa Cruz, Prov. Zambales, Luzon, S. Vidal n.479! This is evidently the plant which Blanco had in view, and the only one to which his term “ Flores paniculato-racemosi” will apply. It is not the plant figured by Vidal (Syn. t. 73. fig. A), nor yet Radermachera quadripinnata, Seem. (Journ of 314 MR. R. A. ROLFE ON THE Bot. 1870, p. 147), if Cuming n. 996, from Prov. Albay, Luzon, 1 in the Kew Herbarium is the same as in the British Museum. S. prnnatum, F. Villar, l. c. p. 151.—Millingtonia pinnata, Blanco, l. c. ed. 1, p. 501; ed. 2, p. 351; ed. 3, vol. ii. p. 285.— Radermachera pinnata, Seem. Journ. of Bot. 1870, p. 147, partim, excl. syn. Bur. Of the four species in the Kew Hebarium the only one with pinnate leaves is Cuming n. 1517, from Prov. Batangas, Luzon ; and as it agrees in other respeets with Blanco’s description, I think it may be taken as belonging to this species. Cuming n. 1182, quoted by Seemann, belongs to the next species. S. Baxar2asar, Rolfe.—Radermachera Banaibanai, Bureau im Adansonia, ii. p. 194. Prov. Albay, Luzon, Cuming m. 1182! Bureau’s plant is based on Callery n. 50, which I have not seen. Seeman confounds it with the previous species, which is clearly an error, as Bureau says, “foliis 2-pinnatis.” “ Ramuli sub- velutini" is also very characteristic of this plant; and as it otherwise agrees with the description, I think both Cuming's and Callery's specimens must represent the same species. I suspect that the vernacular name Banaibanai is applied to more than one species. S. SEEMANNIIL, sp. nov.— Rami striati puberuli, foliis 2-pinnatis, foliolis elliptico-oblongis longe acuminatis integris basi attenu- atis glabris, paniculis terminalibus amplis v. nonnunquam ab- breviatis, calyce campanulato 3-dentato, lobis brevibus late rotun- datis, eorolla basi angusta supra inflata, lobis late rotundatis; fructu longissimo angustissimo subterete.—Radermachera qua- dripinna, Seem. Journ. of Bot. 1870, p. 147, non Millingtonia quadripinnata, Blanco. | Luzon, Prov. Albay, Cuming n. 996! Leaflets 3-5 by 13-13 in. ; petiolules j-1in. Panicles up to a foot atleast long, much branched and many-flowered ; but some- times as short as 3 inches, when they are more dense and with much fewer flowers. Calyx 4 lines. Corolla 1 inch long, the interior much damaged by insects. Fruit immature, nearly 1 ft. long; calyx persistent. The specimens are much broken up. Seemann's plant is Cuming n. 996, which he refers to Milling- tonia quadripinnata, Blanco ; it, however, does not agree with the one figured by Naves, and which agrees with Blanco's deserip- tion. The plant figured by Vidal as “ Stereospermum quadri- FLORA OF THE PHILIPPINE ISLANDS. 315 pinnatum, F. Villar, Millingtonia quadripinnata, Blanco" (Synop- sis, t. 73. fig. A), is a puzzle to me; it does not represent the coloured plate quoted by Villar, which I believe is the true plant of Blanco; and none of the Philippine specimens which I have seen have the calyx as represented by Vidal. The only specimen received from him (n. 479) agrees with the coloured plate of Naves, t. 252. A good series of specimens of this genus is much wanted. VERBENACER. GuNsrA CUMINGIANA, Rolfe.—CallicarpaCumingiana, Schauer, in DC. Prodr. xi. p. 644; F. Villar, l. c. p.158. The five-parted calyx and corolla and five stamens necessitate the species being transferred to G'eunsia. CALLICARPA Brawoor, Rolfe.—C. americana, Blanco, l.c. ed. 1, p. 517; ed. 2, p. 360; ed. 3, vol. ii. p. 300, t. 427 bis, non L.— C. bicolor, F. Villar, l. c. vol. iv. Nov. App. p. 158, non Juss. The plate and Blanco’s synonym are referred to C. bicolor, A. Juss., which, however, must be excluded. LABIATS. SCUTELLARIA LUZONICA, sp. nov.; caule basi procumbente, ramis ascendentibus puberulis, folis parvis petiolatis ovatis obtusis subrepando-crenatis, utrinque parce pubescentibus, race- mis secundis laxis bracteis minutis, calycibus floriferis puberulis, pedieello brevioribus, corollis gracilibus puberulis. Luzon, without precise locality, Lobb! Rami pedales v. ultra, tetragoni, pilis brevissimis. Folia 3-7 lin. longa. Squama calycis fructiferi valde aucta, membranacea. Corolla puberula, 5 lin. longa, tubo basi angustato, fauce dilatata. Nucule valde tuberculose. This species belongs to the section Stachymacris, and may stand next before S. indica, L., from which it differs in its smaller, more entire leaves, more lax racemes, and its smaller and more slender flowers. It is also a more straggling plant, with totally different pubescence. | EvuPHORBIACES. Crerstantuvs Buancor, Rolfe.—C. ferrugineus, F. Villar, l. c. p. 187, t. 353, non Muell. Arg. This species is distinct from C. ferrugineus, Muell. Arg., which must be excluded. 316 ON THE FLORA OF THE PHILIPPINE ISLANDS. MYRICACEE. Myrica VIDALIANA, sp. nov.—Rami glabri, foliis cuneatis v. oblanceolatis subobtusis serratis, basi attenuatis coriaceis, venis subtus reticulatis, petiolis brevibus; amentis masc. compositis brevibus, staminibus 2 (an semper ?) filamentis coalitis, bracteis minutis; amentis foem. longioribus, ovario ovoideo, stylo brevi, stigmatibus patentibus paullo dilatatis; fructu ignoto. Luzon, Voleano Mayon, Prov. Albay ; S. Vidal n. 926! Leaves 1-21 in. by 3-8 lin.; male catkins Àj in., female 1- 13 in. A most distinct species belonging to the section Faba. A supposed second species, M. luzonica, S. Vidal, is not a Myrica at all, but the male of some Euphorbiaceous plant, possibly Sapium. ScITAMINES. ALPINIA PARVIFLORA, Rolfe.— Amomum parviflorum, Presl» Relig. Henk. ii. p. 112, t. xix. figs. 1-13 ; F. Villar, l. c. p. 224. This species must be referred to Alpinia. A. eraciiis, Rolfe—Renealmia gracilis, Blanco, l. c. ed. 1, p. 1—R. exaltata, Blanco, l. c. ed. 2, p. 1; ed. 8, vol. i. p. 2, non L.—Kolowratia elegans, Presl, Relig. Henk. i. p. 118, t. xx. figs. 1-10.—A. gigantea, F. Villar, l. c. p. 225, excl. syn. plur., non Blume. F. Villar refers these synonyms to Alpinia gigantea, Blume, which is quite distinct, and must be excluded. Blanco, in his first edition, made a new species of it, but in the following one wrongly referred it to an American plant. DESCRIPTION OF PLATE X. Fig. 1. Villaria philippinensis, Rolfe. Branch, natural size. . Bud, showing contorted estivation, x2 diam. Expanded flower, showing the villose corolla-throat, x2 diam. . Corolla laid open, showing the linear anthers, x2 diam. . Anther, showing the extremely short filament, x 2 diam. . Section of calyx and ovary, showing the villose fusiform style, x 2 diam. . Longitudinal section of ovary, showing the ovules, x 4 diam. . Transverse section of ovary, showing the parietal placentation, x 4 diam. ONDUA to t ON THE FLORA OF MADAGASCAR. 317 Further Contributions to the Flora of Central Madagascar. By J. G. Bain, F.R.S., F.L.S. [Read 19th June, 1884.] Part I. POLYPETAL.E. 'THALAMIFLORE. In December we received at Kew a box, which, owing to the un- settled state of political affairs, had taken nearly a year in transit, containing the botanical results of the explorations during 1882 of our energetic correspondent, the Rev. Richard Baron, F.L.S., in Central Madagascar. In the present paper I have described the principal polypetalous Dicotyledons which it contains which appear to be new, and of which the material is complete enough; and I hope next Session to deal with the remaining part of the collection. CLEMATIS LAXIFLORA, n. 8p. Volubilis, obscure pubescens, foliis longe petiolatis deltoideis pinnatim trifoliolatis, foliolis ovatis acutis obscure cuspidato-dentatis membranaceis utrinque viridibus glabris, floribus laxe paniculatis pedicellis elongatis, alabastris globosis, sepalis oblongis sordide purpurascentibus tomentosis, filamentis deorsum pilosis, stylis dense albo-plumosis. A woody climber, with slender terete obscurely pubescent branches. Fully-developed leaves deltoid, trifoliolate, with a petiole 2-21 in. long; leaflets ovate, membranous, glabrous, 2-21 in. long, with a few shallow teeth with minute cusps; petiolules often bent, the end one nearly an inch long. Panicles deltoid, peduncled, as long as the fully-developed leaves; pedi- cels slender, the end one 1-12 in. long. Sepals oblong, dull lilac, 3 in. long. Anthers oblong, 4 line long, a quarter the length of the filament. Styles densely plumose.— Baron 2448 ! Leaves like those of C. mauritiana, but flowers much smaller and more numerous. Of the suberect section I am afraid that C. trifida, Hook. Ic. t. 72, and C. oligophylla, Hook. Ic. t. 80, are mere forms of C. Bojeri, Hook. Ie. t. 10. CLEMATIS MICROCUSPIS, N. sp. Volubilis, ramulis pubescentibus, foliis longe petiolatis deltoideis pin- natim 5-foliolatis, foliolis cordato-ovatis facie subglabris dorso dense 318 MR. J. G. BAKER ON THE pubescentibus, dentibus latis minute cuspidatis, floribus copiose panicu- latis, alabastris globosis, sepalis oblongis griseis tomentosis, staminibus pistillisque calyce vix brevioribus. A woody climber, with densely pubescent, strongly-ribbed stems. Fully-developed leaves pinnately 5-foliolate, with a blade 4-5 inches long and broad, and a common petiole half as long; leaflets cordate-ovate, 2-3 in. long, inciso-crenate, with a minute cusp to each tooth, moderately firm in texture, green and glabrous above, persistently grey-pubescent all over below. Axillary panicles as long as the leaves; pedicels 2-j in. long. Fully- developed sepals j in. long, dull purple, pubescent. Stamens À lin. long., one fourth the length of the filaments. Fruit not seen.— Baron 2333! 2336! CLEMATIS EDENTATA, n. 8p. Volubilis, ramulis parce pubescentibus, foliis longe petiolatis deltoideis pinnatim 5-foliolatis, foliolis ovatis acutis integris firmulis glabris, floribus parvulis laxe paniculatis, alabastris oblongis, sepalis lanceolatis pur- purascentibus extus tomentosis, pistilis ealyce aquilongis, stylo dense plumoso. A climber, with slender terete woody branches, thinly pubescent towards the top. Fully-developed leaves pinnately 5-foliolate, with a blade 4-5 in. long and broad, and a petiole half as long; leaflets ovate, entire, acute, firm in texture, glabrous, green on both sides, 2-23 in. long. Flowers in sparse leafy panicles at the end of the branches and from the axils of the developed leaves; pedicels pilose, 4-2 in. long. Bud oblong. Sepals lanceolate, 8-1 in. long, grey-purple, densely pubescent, acute, not more than yz in. broad. Stamens and fruit not seen. Styles densely pilose, reaching nearly to the top of the sepals before they fall. —Baron 2297! Like the two other species now described, this is allied to C. simensis, Fres., and C. grata, Wall. PoLYALTHIA LUCENS, n. sp. Arborea, glabra, foliis breviter petiolatis oblongis obtusis rigide coriaceis facie lucentibus basi deltoideis, floribus axillaribus solitariis cernuis atro- purpureis, pedicello flore zquilongo, petalis lanceolatis calyce duplo longioribus, carpellis fructiferis 10-12 monospermis oblongis longe pedi- cellatis. A tree, glabrous in all its parts, with slender subterete woody branchlets. Leaves alternate; petiole $ in. long; blade thick, rigidly coriaceous, 2-3 in. long, 1-1} in. broad, obtuse, deltoid at FLORA OF MADAGASCAR. 319 the base, finely veined. Flowers solitary in the axils of the leaves, on very cernuous black pedicels j in. long. Calyx } in. long; sepals deltoid. Petals 6, lanceolate, 3 in. long, narrowed from base toapex. Fruit-torus woody, hemispheric, alveolate, } in. in diam. ; carpels oblong, blackish, coriaceous, indehiscent, 1-seeded, 3 in. long, 3 in. in diam., narrowed into a cylindrical pedicel of the same length.—Baron 3116! Allied to P. Chapellieri, Baillon, Adans. viii. p. 349. In his latest paper Dr. Baillon reduces Popowia, Clathrospermum, and Polyalthia, all three to Unona. THYLACHIUM LABURNOIDES, n. sp. Fruticosum, glabrum, foliis longe petiolatis pinnatim trifoliolatis, foliolis oblongis, floribus paucis dense corymbosis breviter pedunculatis, alabastro globoso, calyce viridi herbaceo, filamentis toroque subpollicaribus. A small erect tree, glabrous in all its parts, with crowded branchlets. Leaves alternate; petiole slender, 1-13 in. long ; leaflets 1-14 in. long, all three shortly stalked, obtuse or deltoid at the tip, minutely cuspidate, moderately firm in texture, green and glabrous on both surfaces, obscurely veined. Flowers 6-8 in a dense corymb at the end of the branchlets ; pedicels 1-3 in. long. Bud globose, green, glabrous, } in. in diam. Calyx green, herbaceous, the top splitting off from near the base. Stamens 40 or more, about an inch long; anthers small, oblong. Torus rather shorter than the stamens; ovary small, oblong.— Baron 3263! THYLACHIUM LAURIFOLIUM, n. sp. Arboreum, glabrum, foliis simplicibus oblongis obtusis rigide coriaceis longe petiolatis, floribus paucis corymbosis longe petiolatis, alabastro globoso, calyce brunneo coriaceo, filamentis 2-24 poll. longis, toro tri- pollicari. A tree, glabrous in all its parts, with slender woody subterete branchlets. Leaves rather crowded, alternate, ascending ; petiole about an inch long, articulated at the tip; blade rigidly coria- ceous, dark green on both surfaces, obscurely veined, quite entire, deltoid at the base, 2-3 in. long, 1-1} in. broad. F lowers 4—6 in a corymb at the end of the branchlets; pedicels 13-2 in. long, without bracts or bracteoles. Bud globose, glabrous, coriaceous, 4 in. in diam., the calyx splitting off in a cap below 2 J the middle. Stamens about 30, with long glabrous filaments and 820 MR. J. G. BAKER ON THE small oblong anthers. Ovary small, ovoid, with a filiform torus 3 in. long.— Baron 2862! ONcoBA CAPREEXFOLIA, n. sp. Arborea, inermis, ramulis pubescentibus, folis magmis petiolatis late oblongis obtusis facie obscure dorso conspicue pubescentibus venulis exsculptis, floribus solitariis lateralibus breviter pedunculatis, sepalis del- toideis persistentibus, fructu globoso duro multisuleato griseo-velutino, placentis circiter 20, seminibus multis nigris turgidis oblongo-reni- formibus. A tree, with pubescent branchlets. Leaves alternate; petiole about 4 in. long; blade broad oblong, entire, 3-4 in. long, 14-2 in. broad, moderately firm in texture, green and obscurely pilose above, conspicuously grey-pubescent beneath, with all the veins and veinlets raised. Flowers solitary, axillary, with very short pedicels. Fruit-calyx 3 in. in diameter. Sepals deltoid, pubescent. Corolla and stamens not seen. Fruit woody, globose, indehiscent, 1-11 in. in diam., with about 20 placentas and the same number of ribs outside, persistently grey-pubescent, tipped with a cusp surmounted by a capitate black stigma. Seeds black, oblong-reniform, 4 in. long, with a crustaceous testa.— Baron 2355! Allied to O. tettensis and Petersiana. PITTOSPORUM STENOPETALUM, n. sp. Glabrum, folis petiolatis obovato-cuneatis obtusis rigide coriaceis lucentibus, floribus in cymos umbellatos paucifloros pedunculatos verticil- latos vel paniculatos dispositis, sepalis minutis deltoideis, petalis parvis oblanceolatis sordide luteo-viridibus, staminibus flore vix brevioribus, fructu bivalvi biloculari magnitudine pisi, seminibus in loculo solitariis. A tree, glabrous in all its parts, with slender terete woody branchlets. Leaves alternate, narrowed into a petiole } in. long, 2-3 in. long, 1-12 in. broad above the middle, obtuse, narrowed gradually from the middle to the base, firm in texture, green on both surfaces, shining, with fine distinctly visible erecto-patent branching veins. Cymes umbellate, few-flowered, peduncled, simply verticillate at the end of the branchlets, or forming a sparse panicle; pedicels finally {-j in. Petals } in. long, deciduous. Stamens with a versatile oblong anther, one third the length of the glabrous filament. Ovary ovoid, 2-celled, narrowed into an entire style fin. long. Capsule brown, rigidly coriaceous, globose, lin. in diam., split down to the base into 2 valves; dissepi- FLORA OF MADAGASCAR. 321 ment complete. Seeds solitary in the cells.— Baron, 417! 2218! ; Hildebrandt 3670 ! PoLYGALs PILOSA, n. sp. Perennis, pilosa, caulibus gracilibus prostratis ramosis, foliis alternis sessilibus lineari-oblongis obtusis uninerviis, floribus in racemos terminales densos dispositis, pedicellis brevissimis, bracteis minutis caducis, sepalis exterioribus ovatis herbaceis, interioribus petaloideis albis viridi venosis obovato-unguiculatis petalis zequilongis, carina curvata appendiculata. A perennial herb, with densely pilose, trailing, slender, branched stems 4-6 in. long. Leaves alternate, sessile, linear-oblong, flat, pilose, green on both surfaces, ]-l in. long. Flowers greenish, yz in. long, many in a peduneled terminal éapitate raceme ; pedicels very short; bracts very minute, deltoid. Calyx + in. long; outer sepals ovate, green, erecto-patent ; inner suborbi- culate, white, with about 5 flabellate green veins anastomosing in arches near the margin. Keel jJ; in. long, curved, appendicu- late. Capsule not seen.— Baron 939! Allied to our European P. amara, L. SPHEROSEPALUM, genus novum Guttiferarum. Sepala 4 orbicularia rigide coriacea valde imbricata. Petala 4 ab sepalis vix dissimilia tenuiora magis oblonga. Stamina perplurima uniseriata hypogyna, infra discum inserta, filamentis filiformibus, antheris parvis subglobosis versatilibus. Discus tenuis annularis. Ovarium latum planum biloculare leviter bilobum, ovulis in loculo circiter 3 axillaribus; stylus filiformis; stigma capitatum. Fructus ignotus.— Frutex Madagascariensis inermis, glaber, foliis stipulatis petiolatis alternis magnis oblongis obtusis rigide coriaceis crebre venulosis, floribus in cymas umbellatas copiose pani- culatas dispositis. Differs from normal Guttifere by its alternate stipulate leaves. It may perhaps have an affinity with Caraipa and its neighbours in Ternstremiaces, but as yet the fruit and seeds are unknown. SPHJEROSEPALUM ALTERNIFOLIUM. An erect shrub or tree, glabrous in all its parts, with slender terete woody branchlets. Leaves alternate; stipules deciduous; petiole 1-3 in. long; blade oblong obtuse, entire, rounded at the base, rigidly coriaceous, green on both sides, rather glossy above, 4-5 in. long, 2-3 in. broad at the middle, all the veins and vein- lets raised, the main ones ascending. Flowers in copious cymes forming an ample lax deltoid panicle at the end of the branches 322 MR. J. G. BAKER ON THE mixed with the top leaves; pedicels 1-3 in. long, thickened upwards. Flowers seen only in an early stage. Sepals 4, orbi- cular, much imbricated, blackish, glabrous, decussate, } in. long. Petals scarcely different trom the sepals, more oblong, thinner in texture. Stamens very numerous, inserted beneath the disk in a single row; filaments filiform, } in. long; anthers small, orbi- cular, versatile. Ovary flat on the top, 2-celled; ovules about 3 in a cell, axile. Style slender, cylindrical, j; in. long. Stigma black, capitate.— Baron 2412! SYMPHONIA (CHRYSOPIA) ACUMINATA, D. Sp. Glabra, ramulis gracilibus, foliis petiolatis obovato-oblongis acuminatis, venulis crebris erecto-patentibus, umbellis terminalibus 4-8-floris, pedi- cellis flore triplo longioribus, calycis parvi rubelli segmentis ovatis, petalis latis rubellis calycem triplo superantibus, disco patelleformi, filamentis monadelphis, antheris 15 lanceolatis tubo 2-3plo brevioribus, lobis stig- matosis stellatim patulis. A tree, with slender subterete woody branchlets, glabrous in all its parts. Leaves opposite, shortly petioled, moderately firm in texture, green on both surfaces, 2-3 in. long, $-1 in. broad above the middle, remarkably acuminate, deltoid at the base, with close fine erecto-patent veins clearly visible on the lower surface. Flowers 4-8 in an umbel from the axils of the leaves at the end of the branchlets; pedicels 2-2 in. long. Calyx lin. in diameter, bright red; sepals ovate, much imbricated, 2 in. long. Petals bright red, 3 in. long. Anthers 15, lanceolate, } lin. long; fila- ments all united in an ampulleform tube; disk patelleform ; stigmatic lobes 5, cylindrical, stellate, 1 line long.— Baron 2890 ! 2921! Closely allied to S. pauciflora and eugenioides. RHODOLENA ACUTIFOLIA, n. sp. Arborea, glabra, foliis breviter petiolatis oblongis rigide coriaceis acutis, floribus 1-2nis terminalibus, bracteis caducis, pedicellis brevibus incras- satis, bracteolis 2 minutis ovatis adpressis persistentibus, sepalis magnis coriaceis oblongis obtusis glabris, petalis obovatis purpureis calyce sesqui- longioribus, genitalibus petalis paulo brevioribus. An erect tree, glabrous in all its parts, with slender, terete, woody branchlets. Stipules deciduous before the flowers appear ; petiole short, rugose; blade 2-3 in. long, 1-13 in. broad at the middle, thick, rigidly coriaceous, lucent, with fine erecto-patent veinlets. Flowers 1-2 at the end of short branchlets; bracts FLORA OF MADAGASCAR. 323 deciduous before the flower expands; pedicels thick, under } in. long. Calyx glabrous, firm in texture, brown, an inch long, with a pair of small ovate bracteoles at its base; sepals very obtuse, imbricated, 3-3 in. broad. Petals 14 in. long, 1 in. broad, bright dark lilac. Stamens numerous; anthers small, subglobose ; fila- ments an inch long. Stigma just overtopping the anthers — Baron 2427! PsoROSPERMUM TRICHOPHYLLUM, n. sp. P. ramulis ferrugineo-pubescentibus, foliis petiolatis oblongis integris facie glabris dorso pubescentibus, cymis multifloris terminalibus brevissime pedunculatis, pedicellis brevibus, sepalis ovatis dense pilosis, petalis oblongis calyce sesquilongioribus, staminibus circiter 15 pentadelphis. A shrub, with very slender ferrugineo-pubescent branchlets. Leaves distinctly petioled, 14-2 in. long, moderately firm in texture, green and glabrous above, ferruginous on the midrib beneath, the rest covered with persistent grey pubescence. Cymes 10-12-flowered, terminal, shortly peduncled; pedicels 3-4 in., densely pubescent. Calyx 4l in., densely pubescent. Petals brownish. Stamens in 5 bundles of 3 each, as long as the calyx. Styles very short in the flowering stage. Fruit not seen.— Baron 3016! PsoROSPERMUM DISCOLOR, n. sp. P. ramulis castaneis apice ferrugineo-pubescentibus, foliis oblongis acutis integris membranaceis glabris dorso albidis, cymis terminalibus paucifloris brevissime peduneulatis, pedicellis ferrugineo-pubescentibus flore zqui- longis, sepalis ovatis, petalis oblongis ealyce longioribus, staminibus circiter 15 pentadelphis, stylis brevibus. A shrub, with castaneous very slender branchlets, glabrous, except at the tip. Petiole 1-1 in. ; blade oblong, acute, deltoid at the base, thin in texture, 2-3 in. long, bright green above, whitish beneath. Cymes terminal, corymbose, 6-10-flowered. Calyx 4; in. long, ferrugineo-pubescent, like the pedicels and peduncle. Petals l in. long, oblong. Phalanges of stamens pilose, as long as the calyx. Ovary ovoid, with 5 short styles, with capitate stigmas. Fruit not seen.— Baron 2222! 2825! Nearly allied to P. leptophyllum. PsonosPERMUM LEPTOPHYLLUM, D. 8p. P. ramulis glabris apice ferrugineo-pubescentibus, foliis oblongis acutis integris membranaceis glabris dorso albidis, cymis multifloris corymbosis LINN. JOURN.—BOTANY, VOL. XXI. 2^ 824 MR. J. G. BAKER ON THE terminalibus pedunculatis, pedicellis ferrugineo-pubescentibus flore aqui- longis, sepalis ovato-oblongis, petalis calyce duplo longioribus, staminibus pentadelphis, stylis brevibus, stigmatibus penicillatis. A shrub, with very slender reddish-brown branchlets, glabrous and shining except at the very tip. Petiole 4-3 in. long; blade 3-4 in. long, 14-1} in. broad at the middle, acuminate, deltoid at the base, thin in texture, entire, green on the upper surface, whitish beneath, the veins fine and inconspicuous. Cymes of 30- 40 flowers on a peduncle about an inch long. Calyx g in. long, thinly elothed, like the peduncle and pedicels, with ferruginous pubescence. Petals nearly lin. long. Phalanges of stamens pilose, as long as the calyx, 8 in each bundle. Styles short; stigmas capitate, penicillate. Fruit not seen.— Baron 2857! PsoROSPERMUM CERASIFOLIUM, n. 8p. Glabrum, foliis petiolatis oblongis obtusis subintegris firmulis utrinque viridibus, cymis 6-12-floris terminalibus pedunculatis, pedicellis flore 3-4plo lougioribus, sepalis oblougis, petalis oblongis acutis calyce sesqui- longioribus, staminibus 20-25 pentadelphis, stylis elongatis falcatis. A shrub, with woody brown glabrous branchlets, flattened and thickened below the nodes. Petiole lin. long; blade 2-3 in. long, 1-1} in. broad at the middle, rather rounded at the base obscurely crenulate, moderately firm in texture, green and glossy above, paler and duller beneath, with 6-8-jugate distinct main veins. Peduncle flattened, 1-2 in. long; pedicels 1-3 in., erect, glandulose. Calyx 3 in. long ; sepals glabrous, subobtuse, much dotted with black. Petals } in. long. Stamens sometimes 5 ina bundle. Immature fruit black, globose, the size of a pea, tipped m curved styles $ in. long; stigma capitate.— Baron 3033! HIBISCUS PALMATIFIDUS, n. sp. Fruticosus, ramulis ferrugineo-pubescentibus, foliis longe petiolatis orbicularibus subcoriaceis 5-nervatis facie calvatis dorso pubescentibus apice leviter trilobatis, floribus solitariis breviter pedunculatis, bracteolis 5 deltoideis basi connatis, calycis segmentis deltoideis petalis magnis latis dorso pubescentibus, fructu duro globoso piloso. j A shrub, with terete woody branchlets, densely coated with stellate brown pubescence. Petiole 14-3 in. long ; blade cordate- orbicular, 3-6 in. broad, shallowly 3-lobed at the top, moderately rigid in texture, with 5 strong ribs radiating from the base to the pc aoon FLORA OF MADAGASCAR. 925 margin, obscurely pilose in an early stage above, covered with persistent brown stellate pubescence beneath. Flowers on soli- tary axillary peduncles about } in. long. Epicalyx coriaceous, adpressed to the calyx, $ in. long and broad. Calyx as long as the epicalyx. Petals obovate-cuneate, much imbricated, 2 in. long, red-brown inside, coated with drab tomentum externally. Column as long as petals. Capsule 5-celled, globose, an inch in diameter, very hairy inside, splitting up into 5 woody valves.— Baron 2548! Connects Euhibiscus and Paritium. DoMBEYA MACRANTHA, n. sp. Fruticosa, dense pannosa, foliis longe petiolatis cordato-ovatis serru- latis, floribus terminalibus parce corymbosis et solitariis axillaribus longe peduneulatis, bracteolis magnis ovatis profunde laceratis, sepalis magnis ovatis pannosis, petalis obovato-cuneatis maximis scariosis persistentibus, staminibus fertilibus circiter 20 antheris lanceolatis, staminodiis longissimis petalis æquilongis, ovario globoso dense piloso. A shrub a yard high, with dense pubescent woody branchlets. Stipules caducous; petiole 1-2 in. long; blade 3-4 in. long, very thick in texture, and densely coated with brown tomentum on both sides, cordate at the base, obscurely denticulate. Flowers solitary, from the axils of the leaves on pedicels as long as the petiole, and a few at the end of the branch. Bracteoles ovate, per- sistent, an inch long, deeply lacerated. Sepals 13 in. long, ovate acuminate. Petals 2 in. long, 1} in. broad, brown, glabrous. Anthers lin. long. Staminodes twice as long as the stamens, their free strap-shaped points 14 in. long. Fruit woody, globose, densely pilose.— Baron 710!, also Hildebrandt 3895! from the Betsileo country. DoMBEYA FLORIBUNDA, n. sp. Arborea, ramulis adpresse lepidotis, foliis magnis petiolatis oblongis integris rigide coriaceis dorso venulosis, venis primariis intra marginem anastomosantibus, cymis decompositis multifloris longe pedunculatis, pedicellis elongatis, sepalis lanceolatis lepidotis, petalis parvis persistenti- bus cuneatis, staminibus fertilibus 10, staminodiis petalis duplo brevioribus, ovario globoso lepidoto. A tree, with woody branchlets coated with drab lepidote scales. Petiole 1-1 in. long ; stipules caducous ; blade oblong, obtuse, entire, 6-8 in. long, 2-24 in. broad, rigid in texture, glossy on both sides, the main veins anastomosing beneath by intramar- ginal arches. Cymes very compound, dichotomous, on lateral 242 326 |. MR. J. G. BAKER ON THE peduncles 2-3 in. long; pedicels about j in. long. Calyx jin. long; sepals lanceolate, lepidote like the branchlets. Petals } in. long, seariose, brown, persistent. Stamens very small. Staminodes not more than half as long as the corolla. Ovary small, globose, lepidote, narrowed into a long style.— Baron 2373! Habit and flowers like those of Melhania laurifolia, which only differs from Dombeya by its isomerous stamens. DoMBEYA REPANDA, n. 8p. Arborea, ramulis adpresse lepidotis, foliis magnis petiolatis oblongis re- pandis subcoriaceis glabris, cymis paucifloris axillaribus pedunculatis, pedi- cellis lore brevioribus, sepalis lanceolatis lepidotis, petalis parvis persis- tentibus cuneatis, staminibus fertilibus 10, staminodiis petalis duplo brevioribus, ovario globoso lepidoto. Calyx, corolla, stamens, and staminodes just as in D. flori- bunda ; but cymes sparser and leaves quite different, with 6-8 pairs of erecto-patent main veins produced nearly to the margin, and not anastomosing by distinctly raised crosa-arches a little within the margin.— Baron 2599! GREWIA MACROPHYLLA, D. Sp. Arborea, ramulis dense ferrugineo-pubescentibus, folis magnis petio- latis cordato-ovatis acutis denticulatis subcoriaceis facie calvatis dorso pubescentibus, cymis lateralibus et terminalibus pluribus corymbosis, pedicellis pedunculisque dense ferrugineo-pubescentibus, sepalis lanceo- latis dense ferrugineo-tomentosis, petalis dense pilosis, ovario globoso dense piloso. A tree, with slender terete woody branches, densely clothed with ferruginous tomentum. Petiole 4—4 in. long; blade sometimes 2 ft. long, 3 in. broad, moderately firm in texture, acute, finely inciso-crenate, green and glabrescent on the upper surface, clothed with ferruginous pubescence beneath ; the erecto-patent main veins and intermediate veinlets raised. Cymes lateral, and forming an ample end-corymb; peduncles and pedicels i-i in, densely ferruginoso-tomentose. Calyx above in. long; sepals lanceolate. Petals linear-oblong, densely pilose outside, as long as the calyx. Ovary globose, 4-celled, clothed with soft pale- brown hairs; style j in. long; stigma capitate.— Baron 2420! 2549! Allied to G. ferruginea, Hochst. GREWIA CUNEIFOLIA, N. sp. G. ramulis gracilibus lepidotis, foliis. subsessilibus obovato-cuneatis obtusis inciso-crenatis subcoriaceis glabris trinervatis, cymis axillaribus FLORA OF MADAGASCAR. 327 pedunculatis 2—3-floris, pedicellis flore longioribus, sepalis lanceolatis dorso lepidotis, petalis lanceolatis calyce equilongis, fructu tricocco rugoso. A. tree, with slender branchlets clothed with adpressed lepidote pubescence. Leaves rigid in texture, 2-3 in. long by about an ineh broad, narrowed gradually from the middle to the base, with 3 strong main veins continued from the base to the very obtuse apex, minutely toothed throughout, green and glabrous on both surfaces. Cymes axillary; peduncles and pedicels about } in. long. Calyx } in. long; sepals lanceolate, lepidote on the back. Petals lanceolate, about as long as the calyx. Stamens equalling the calyx. Fruit of three small hard rugose one-seeded glabrous lobes.— Baron 2251! 3223! Allied to G. trinervata, Baker in Trimen’s Journ 1882, p. 47. ERYTHROXYLUM FIRMUM, n. sp. Glabrum, petiolis brevissimis, foliis oblanceolato-oblongis rigide coriaceis venulis faciei inferioris pulchre anastomosantibus, floribus l- 3nis axillaribus et supra folios productis, pedicellis flore longioribus, bracteis sepalisque ovatis, petalis oblongis calycem duplo superantibus. A shrub, glabrous in all its parts, with rather flattened grey branchlets. Leaves 14-2 in. long, obtuse, narrowed to the base, very rigid in texture, green on both surfaces, the fine erecto- patent main veins beneath connected by distinct cross-arches just within the edge. Flowers in copious clusters both in the axils of the leaves and on the short leafless ramuli; bracts ovate, shorter than the stiffly erect pedicels, which are iz-à in. long. Calyx campanulate, 1 lin. in diam. ; sepals deltoid, connate at the base. Petals twice as long as the sepals, with an obovate scale across the claw, with an incurved free apex. Stamens shorter than the petals ; anthers oblong, equalling the filaments. Ovary depresso-globose, with 3 short styles, with capitate stigmas.— Baron 2740! 2848! Near the Mauritian E. laurifolium, Lam. i P Ruopoctapa, genus novum Linacearum ! Calyx campanulatus, segmentis 5 oblongis obtusis imbricatis coriaceis persistentibus reflexis. Petala 5 oblonga decidua calyce paulo longiora. Stamina 10 leviter perigyna, filamentis filiformibus basi in cupulam brevem connatis, antheris globosis. Ovarium globosum sessile biloculare, ovulis pluribus in loculo ab apice pendulis ; stylus filiformis, stigmate capitato. Fructus ignotus.— Arbor Madagascariensis, foliis alternis exstipulatis obovatis rigide coriaceis integris subsessilibus articulatis, floribus parvis 828 MR. J. G. BAKER ON THE copiose panieulatis, panicule ramis ferrugineo-pubescentibus, pedicellis brevibus basi articulatis, genitalibus breviter exsertis. This very distinct-looking plant differs from Linacem by its slightly perigynous stamens, 2-celled ovary, several ovules in a cell, and simple style. Of known genera it appears to come nearest to Asteropeia; but the fruit has not yet been found. The flowers resemble those of an Erythroxylon, and leaves those of a simple-leaved species of Rhopala. RHODOCLADA RHOPALOIDES, Baker, sp. unica. An erect tree, with crowded branchlets, clothed with fine bright ferruginous pubescence. Leaves alternate, simple, obovate- cuneate, entire, about 2 in. long, narrowed gradually from the middle to a very short petiole, which readily disarticulates at the base, glabrous on both surfaces; the erecto-patent side-veins fine and immersed. Flowers in deltoid terminal panicles 2-3 in. long, with spreading or erecto-patent branches, clothed with fer- ruginous pubescence; pedicels very short, articulated at the base, originating from the swollen nodes of the rhachis. Sepals jy in. long, oblong, persistent, reflexing, imbricated. Petals oblong, è in. long. Stamens and style slightly exserted.— Baron 3094! 3096! Oxaris (§ BroPHYTUM) MACROPODA, n. sp. O. caule elongato lignoso simplici, foliis dense congestis breviter petio- latis linearibus pinnatis, foliolis confertis 40-50-jugis quadratis basi auricu- latis apice oblique cuspidatis, pedunculis folio 2-3plo brevioribus, bracteis multis minutis congestis, pedicellis brevibus glandulosis, sepalis ovato- lanceolatis conspicue nervatis fructu globoso brevioribus, stylis hispidis. Leaves in a dense cluster at the summit of a simple woody stem 2 ft. long, linear, 3-4 in. long, with a slender pilose rhachis and short petiole. Leaflets }-1 in. broad, crowded, sensitive, reflexing, rigid in texture, green on both sides, with an excentric midrib, distinct erecto-patent veinlets, a cusp directed upwards, and a basal auricle close to the rhachis. Peduncle slender, much shorter than the leaves; pedicels 1 in. long, rugose, with pro- minent black glands. Sepals i in. long, with 3 strong ribs. Petals not seen. Capsule globose, % in. in diam. ; styles hispid, half as long as the capsule.— Baron 2307 ! Near O. myriophylla, O. Hoffm. in Rel. Rutenb. iv. 213. FLORA OF MADAGASCAR. 329 TODDALIA PILOSA, n. sp. Inermis, ramulis dense pilosis, foliis petiolatis trifoliolatis, foliolis oblan- ceolato-oblongis integris subcoriaceis dorso pilosis, cymis axillaribus sessilibus 3-5-floris, pedicellis brevissimis, sepalis orbicularibus minutis persistentibus, fructu globoso glabro nigro biloculari haud lobato. An erect tree, with unarmed slender terete densely pilose branchlets. Petiole -lin.; leaflets 2-3 in. long, under an inch broad, narrowed gradually from the middle to the base, mode- rately firm in texture, green on both surfaces, distinctly dotted over with black glands beneath. Cymes in the axils of the leaves aldown the branchlets. Calyx minute, persistent. Petals not seen. Fruit the size of a pea, black, entire, globose, with copious glandular hollows in the epicarp; all the three layers thin.— Baron 3093 ! OcHNA VACCINIOIDES, n. sp. Glabra, foliis oblongis parvis brevissime petiolatis rigidulis crenulatis, cymis lateralibus 2-3-floris, pedicellis flore equilongis, sepalis persistenti- bus inzqualibus patulis obovatis vel oblongis, stylo elongato. A shrub, glabrous in all its parts, with slender terete branch- lets. Leaves bright green on both surfaces, moderately firm in texture, 3-2 in. long, subobtuse, rounded at the base, di stinetly crenulate, with fine immersed veins. Flowers 2-3 together in nearly sessile lateral cymes; pedicels 4—3 in., ascending or cer- nuous. Fruit-calyx 2 in. in diam., spreading horizontally ; outer sepals obovate, inner oblong. Petals and stamens not seen. Style straight, gynobasic, } in. long.—Baron 3028 ! OcHNA SERRATIFOLIA, D. 8p. Glabra, foliis oblongis parvis brevissime petiolatis rigidulis serrulatis dentibus ascendentibus cuspidatis, cymis lateralibus paucifloris, pedicellis flore longioribus, sepalis persistentibus oblongis patulis, stylis apice liberis. A shrub, glabrous in all its parts, with slender woody terete branchlets. Leaves 1-1 3 in. long, subobtuse, deltoid at the base, moderately rigid in texture, dark green on both surfaces, with fine immersed veins, minutely toothed, with small ascending deltoid teeth tipped with a cusp. Cymes few, lateral, nearly sessile; pedicels 3 in. long, ascending or cernuous. Fruit-calyx spreading, nearly an inch in diameter; sepals oblong, rigid, brown, the inner much narrower than the outer. Petals not seen. Stamens very short. Ovary globose; style with 5 short 330 MR. J. G. BAKER ON THE spreading branches.—Baron 811! Allied to O. Wightiana, Wall., of India and Ceylon. GoMPHIA PERSEXFOLIA, D. sp. Glabra, ramulis angulatis, foliis petiolatis oblanceolato-oblongis obtusis emarginatis rigide coriaceis, racemis subdensis subpaniculatis, pedunculo ancipiti, pedicellis flore æquilongis, sepalis persistentibus oblongis reflexis, fructu globoso. A tree, with slender woody branchlets, glabrous in all its parts. Leaves alternate, not crowded, shortly petioled; blade 4-6 in. long, 1-1? in. broad above the middle, narrowed gra- dually to the base, entire, obtuse, and distinctly emarginate at the apex, thick and very rigid in texture, glossy, with fine im- mersed veinlets. Racemes dense, on axillary or terminal slender ancipitous ascending peduncles 14-3 in. long; flowers in shortly- peduncled cymes at some of the lower nodes; pedicels 4 in. long. Fruit-calyx 4 in. long, reflexing ; sepals brown, oblong. Petals and stamens not seen. Fruit small, globose. Style very slender, 4 in. long.— Baron 2226! Near G. obtusifolia, DC. GoOMPHIA LANCEOLATA, n. Sp. Glabra, ramulis gracilibus teretibus, foliis petiolatis lanceolatis vel ob- longo-lanceolatis integris rigide coriaceis, racemis axillaribus vel termi- nalibus pedunculatis erectis vel cernuis, pedicellis flore longioribus, sepalis oblongis persistentibus diu ascendentibus, fructu globoso. A tree, with slender terete woody branchlets, glabrous in all its parts. Leaves not crowded, thick and rigid in texture, entire, obtuse with a minute cusp, 2-4 in. long, 3-1 in. broad at the middle, green and rather glossy on both surfaces, with fine im- mersed veinlets. Racemes terminal or from the axils of the upper leaves; peduncles slender, erect or cernuous, as long as or shorter than the leaves; pedicels finally 4—3 in. Fruit- calyx 3 in. long, ascending ; sepals brown, narrow oblong. Petals and stamens not seen. Style very slender, } in. long.— Baron 2960! 2984! 3076! Allied to G. obtusifolia, DC. Mém. t. 8. G'OMPHIA ANCEPS, n. sp. Glabra, ramulis gracilibus angulatis, foliis magnis subsessilibus oblongo- lanceolatis subintegris rigide coriaceis, racemis pendulis cylindricis pedun- culo elongato ancipiti, pedicellis flore æquilongis, sepalis oblongis persis- tentibus reflexis, ovario turbinato, stylo elongato. A tree, with slender angled woody branchlets, glabrous in all FLORA OF MADAGASCAR. 881 its parts. Leaves subsessile, 6-9 in. long, 11-2 in. broad at the middle, entire or obscurely erenulate, narrowed to the base and a subobtuse point, firm but not very rigid in texture, green and rather glossy on both surfaces, with fine immersed veinlets. Peduncles lateral, slender, pendulous, ancipitous, lft.long. Ra- cemes cylindrical, moderately dense, 4-6 in. long ; lower flowers 2-3nate; pedicels articulated, 3 in. long. Calyx persistent, 4 in. long; sepals oblong, reflexing, brownish in the fruiting-stage. Petals and stamens not seen. Fruit turbinate, containing 2-3 subglobose polished brown-black seeds the size of a pea. Style finally 1 in. long.—Baron 2376! Allied to G. dependens, DC. Mém. t. 6. OLAX EMIRNENSIS, n. sp. Glaber, foliis ovatis acutis integris rigidulis uninerviis, cymis axillaribus sessilibus 6-8-floris, pedicellis brevibus basi bracteatis, calyce campanulato parvo truncato, petalis 5 lanceolatis, antheris fertilibus 3, sterilibus 6 ligu- latis, ovario globoso, stylo elongato. An erect shrub, glabrous in all its parts. Leaves alternate, dull grey-green, moderately firm in texture, 2-3 in. long, obtuse or subacute, the veins fine and entirely hidden except the midrib, the petiole very short and articulated at the base. Cymes Ses- sile, axillary, deltoid ; pedicels 3-4 in., with a minute deltoid per- sistent bract at the base. Calyx campanulate, yyin. in diam., quite truncate. Corolla reddish, cylindrical, } in. long. Stamens 3, perfect, with small oblong anthers; staminodia 6, ligulate, twice as long as the anthers. Ovary ovoid, sessile ; style half as long as the petals; stigma capitate.— Baron 3078! A near ally of the Mauritian O. psittacorwm, Vahl (Fissilia, Lam.). PYRENACANTHA CHLORANTHA, D. 8p. Volubilis, ramulis obscure pilosis, foliis breviter petiolatis obovato- oblongis rigidulis glabris subintegris, floribus foemineis racemosis, pedicellis brevissimis, bracteis linearibus, perianthii segmentis 4 obovatis pilosis, fructu oblongo compresso acuto. A shrub, with climbing slender terete pubescent branchlets. Leaves shortly petioled, bright green, moderately firm in texture, obtusely euspidate, narrowed to a slightly rounded base, 2-3 in. long, 2—1 in. broad above the middle, entire or obscurely toothed. Racemes axillary, shortly peduncled, shorter than the leaves ; rhachis, short pedicels, and linear bracts pilose. Bud globose, 332 MR. J. G. BAKER ON THE green, 4 lin. in diam. ; sepals 4, almost orbicular, green, free to the base. Fruit oblong, blaek, compressed, glabrous, indehiscent, coriaceous, above 3 in. long, narrowed to an acute apex ; epicarp thin, brown; mesocarp thin; endocarp thin, white, densely spiny inside the cavity, filled with one large seed.— Baron 1374! (fruit), 3018! (female flower). Baron 1348, with oblong-acute glabrous strongly toothed leaves, is very likely another new species of this genus ; but the flowers are too immature to be certain. DESMOSTACHYS ACUMINATA, n. Sp. D. ramulis gracilibus pilosis, foliis breviter petiolatis oblongis acuminatis rigidulis dorso hispidulis, floribus in spicas axillares fastigiatas dispositis, calycis minuti pilosi segmentis deltoideis, petalis lanceolatis dense pilosis, staminibus inclusis, ovario ovoideo piloso. An erect shrub, with slender terete woody pilose branchlets. Leaves 3-4 in. long, 1-1} in. broad, rounded at the base, pale green, moderately firm in texture, the raised main veins of the under surface united by distinct simple cross-arches. Spikes axillary, clustered, 1-11 in. long, with a very hairy slender rhachis. Calyx } lin.in diam.; sepals deltoid. Petals yellowish, densely pilose, jl; in. long. Stamens with a linear-oblong anther equalling the filament. Ovary ovoid, very hairy, narrowed into a cylindrical style as long as itself.— Baron 2622 ! DeEsMOSTACHYS DELTOIDEA, D. sp. D. ramulis gracilibus pilosis, foliis breviter petiolatis oblongis acutis rigidulis glabris, floribus in racemos paniculatos dispositis, calycis minuti pilosi segmentis deltoideis, petalis oblanceolatis viridibus dense pilosis, staminibus inclusis, fructu globoso. An erect shrub, with very slender terete pilose branchlets. Leaves shortly petioled, pale green, moderately firm in texture, glabrous, 13-2 in. long, 2-1 in. broad, subdeltoid at the base, narrowed gradually to an acute apex, the veins fine and incon- spicuous. Racemes under an inch long, forming a deltoid panicle at the end and side of the branches; axis pilose; pedicels very short; bracts minute, deltoid. Calyx campanulate, green, 4 lin. in diam. Petals very hairy, greenish, jl in. long. Disk with 5 lobes. Immature fruit globose, rather hairy, the size of a pea, tipped with the short cylindrical style—Baron 2926! D. Renschii, O. Hoffm. & Hildeb. Sertum, p. 11, is the same species as D. Planchonianus, Miers, Contrib. i. p. 68, tab. 9. FLORA OF MADAGASCAR. 338 ELÆODENDRON NITIDULUM, n. sp. E. ramulis pubescentibus, foliis alternis breviter petiolatis obovatis vel oblongis parvis rigidis nitidulis viridibus integris glabris, cymis ad ramos supra folios dispositis, floribus parvis pentameris, pedicellis brevibus, calycis segmentis deltoideis, petalis orbicularibus, staminibus inclusis. A much-branched erect shrub, with slender pubescent terete branchlets. Leaves about an inch long, obtuse, truncate or emarginate, firm in texture, deltoid at the base, bright green, the veins fine and immersed. Cymes crowded on the branchlets above the leaves; pedicels 4l; in. Calyx campanulate, 4 lin. diam.; segments deltoid. Expanded corolla jy in. diam. Stamens half as long as the petals. Fruit unknown.— Baron 3084! ELEODENDRON VACCINIOIDES, n. sp. E. ramulis pubescentibus, foliis parvis alternis breviter petiolatis oblon- gis crenatis rigidis lucidulis glabris, cymis paucifloris axillaribus, floribus minutis pentameris, pedicellis brevibus, calycis segmentis deltoideis, petalis orbicularibus, staminibus brevibus, fructu globoso parvo magnitudine pisi, semine solitario. A much-branched erect shrub, with slender terete pubescent branchlets. Leaves rigid in texture, bright green, about an inch long, obtuse, deltoid at the base, crenulate, with fine slightly raised erecto-patent main veins. Cymes copious, axillary ; pedi- cels 3; in. long. Calyx campanulate, } lin. in diam.; segments deltoid. Expanded corolla 4'; in. diam. Stamens shorter than the petals. Fruit brown, glabrous, the size of a pea, 1-seeded.— Baron 3147! ELXODENDRON TRACHYCLADUM, n. Sp. E. ramulis gracilibus dense pilosis, foliis alternis breviter petiolatis orbi- cularibus vel ovatis crenulatis rigidulis obscuris dorso obscure pubescenti- bus, cymis paucifloris axillaribus, floribus minutis pentameris, calycis segmentis deltoideis, petalis orbicularibus patulis, staminibus brevibus. A shrub, with slender terete densely pubescent branchlets. Leaves about an inch long, nearly as broad as long, rounded at both ends, dull grey-green, moderately firm in texture, glabrous above, hairy principally on the midrib beneath, the fine erecto- patent main veins slightly raised. Cymes few-flowered, axillary ; peduncle and pedicels both very short. Calyx . campanulate, } lin. in diam., pubescent. Expanded corolla 1 lin. in diam. Sta- mens half as long as the petals. Fruit unknown. — Baron 2878! 334 MR. J. G. BAKER ON THE ELXODENDRON GRISEUM, D. 8p. E. ramulis pubescentibus, folis breviter petiolatis alternis ovatis vel orbicularibus subintegris rigidulis obscuris obscure pilosis, cymis paucifloris axillaribus, floribus minutis pentameris, pedicellis brevibus, calycis seg- mentis deltoideis, petalis orbicularibus patulis, staminibus brevibus. A shrub, with slender terete pubescent branchlets. Leaves about an inch long, suborbicular or ovate, obtuse, entire or obscurely toothed, moderately firm in texture, dull grey-green, hairy mainly on the midrib beneath, the veins fine and immersed. Cymes copious, axillary, on short pubescent peduncles. Calyx campanulate, pubescent, 4 lin. diam. ; segments very small, broad, deltoid. Expanded corolla 1, in. in diam. Immature fruit ovoid, glabrous, tipped with the short style.— Baron 2650! SALACIA OLEOIDES, n. sp. Glabra, foliis crassis rigide coriaceis lucidis integris breviter petiolatis ovatis vel oblongis, cymis axillaribus paucifloris corymbosis, pedicellis flore 2-3plo longioribus, sepalis petalisque orbicularibus, fructu globoso coriaceo. A shrub, 8-16 ft. high, glabrous in all its parts, with slender terete rugose branchlets. Leaves opposite, shortly petioled, very thick and rigid in texture, quite entire, 2-3 in. long, 2-1j in. broad, obtuse, deltoid at the base, lucid, only the midrib dis- tinctly visible. Cymes copious, axillary, corymbose ; pedicels finally 1-3 in. long. Calyx campanulate, j4 in. in diam. Petals only seen immature. “ Fruit dark brownish-green whilst on the tree,” in the dried specimens globose, under an inch in diameter, very coriaceous in texture.— Baron 2837! Also gathered by Dr. Parker in the forest of Andrangaloaka. SALACIA DENTATA, D. sp. Glabra, ramulis rugosis, foliis brevissime petiolatis obovato-oblongis vel oblongis rigidis lucidis inciso-crenatis, cymis axillaribus multifloris umbellatis, pedicellis flore 2-3plo longioribus, sepalis petalisque orbicu- laribus, disco magno pentagono, filamentorum tubo ampulleformi, fructu magno turbinato. A shrub, glabrous in all its parts, with terete woody branchlets. Leaves opposite, shortly petioled, 3—4 in. long, 1-11 in. broad at the middle, acute, deltoid at the base, rigid in texture, glossy, distinctly crenate or dentate, the fine main veins raised beneath. Cymes copious, sessile, axillary, umbellate; pedicels finally 4—4 FLORA OF MADAGASCAR. 835 CF in long. Calyx campanulate, Į; in. in diam.; sepals much y imbrieated. Expanded corolla } in. in diam. ; petals orbicular- unguieulate. Disk pentagonal, lin. in diam. Stamen-column ampulleform, half as long as the petals. Immature fruit coriaceous, turbinate.— Baron 1256! 2184! 2866! &c. A near ally of S. Calypso, DC. (Calypso, Thouars, tab. 6). TINA POLYPHYLLA, n. sp. T. ramulis dense brunneo-pubescentibus, foliis pinnatis longe petiolatis, foliolis 23-27 oblongis acutis serratis subcoriaceis facie parce dorso dense pubescentibus, floribus in paniculam deltoideam dispositis, sepalis ovatis pilosis, petalis oblongis, ovario turbinato obscure piloso. A tree, with the branchlets, peduncles, and leaf-rhachises densely clothed with short brown velvety pubescence. Petiole ‘8-4 in. long; lamina a foot long ; leaflets 3-4 in. long, an inch broad, oblique at the base, distinctly serrated and stalked, obscurely pilose and green on the upper surface, brown-velvety especially on the raised very numerous main ribs beneath. Panicle half as long as the leaves, the lower branches compound ; peduncle equalling the petiole. Calyx campanulate, 3 line long, densely pilose. Petals and stamens twice as long as the sepals. Ovary turbinate, black, pilose, seated on a thick disk and nar- rowed into a shortstyle.— Baron 2447! Differs from all the other species as described in Radlkofer's Monograph of the Cupaniee, Pp. 661-663, by its very numerous leaflets. DODONÆA MADAGASCARIENSIS, Radi. in Brem. Abhand. viii. 470. Glutinosa, pilosa, foliis imparipinnatis rhachi petioloque alatis, foliolis 8-10-jugis lanceolatis ciliatis, floribus dioicis in paniculas densas terminales dispositis, sepalis lanceolatis, staminibus circiter 20 antheris oblongis apiculatis filamentis brevissimis, fructu orbiculari compresso coriaceo biloeulari alis deltoideis erectis. A glutinose shrub, 4-5 feet high, with thick woody pubescent branchlets. Leaves imparipinnate, with a blade 4-6 in. long ; rhachis winged between and above the leaflets, which are erecto- patent, sessile, acuminate, 1-2 in. long, ciliated, moderately firm in texture, unequal-sided towards the base. Flowers of both sexes in a dense terminal panicle, with pubescent branches ; pedicel pubescent; bracts linear, persistent. Sepals lanceolate, lin. long. Stamens 20, with oblong apiculate anthers 4 in. long and very short filaments. Fruit an orbicular com pressed bilocular 336 MR. J. G. BAKER ON THE coriaceous pubescent capsule j in. long, with an erect deltoid wing on each side, and a very short style.—Sent twice by Mr. Baron. Also Hildebrandt 3604, from Andrangaloaka. CALYCIFLORE. RoUnREA PLATYSEPALA, n. Sp. Arborea, glabra, foliis petiolatis pinnatis, foliolis 5-7 rigidulis ovatis acutis, floribus laxe copiose paniculatis, pedicellis flore longioribus, sepalis ovatis obtusis valde imbricatis, petalis oblongo-spathulatis calyce duplo longioribus, staminibus inclusis biseriatis antheris parvis, carpellis sessili- bus ovoideis, stylo elongato, stigmate capitato. A tree, glabrous in all its parts, with terete branchlets. Petiole 1-14 in. long; leaflets 5-7, ovate, acute, petiolulate, 1-1} in. long, rigid in texture, glossy, with a distinct brown midrib, the other veins fine and immersed. Panicles of’ numerous crowded lax racemes 1-2 in. long; pedicels slender, flexuose, 4 in. long. Calyx campanulate, jl; in. long; sepals ovate, obtuse, much imbricated, the two inner quite hidden by the three outer. Petals oblong-unguiculate, 4 in. long. Stamens with long filaments and short globose anthers. Carpels more than one, with a produced style and capitate stigma. Fruit not seen.— Baron 2528! Nearly allied to the East-Indian R. santa- loides, Wight & Arn. NEOBARONIA, genus novum Dalbergiearum. Calyx parvus, campanulatus, dentibus minutis deltoideis. Corolla papilionacea, petalis equilongis; vexillum obovatum obtusum unguicu- latum ; ale angustiores; carina navicularis recta subacuta. Stamina 10 diadelpha, filamentis 9 in tubo apice fisso connatis, supremo libero fili- formi, antheris parvis globosis. Ovarium sessile lineare, ovulis paucis, stylo brevi abrupte incurvato, stigmate capitato. Legumen coriaceum indehiscens turgidum 1-2-spermum, apice et basi attenuatum.—Arbor Madagascariensis, ramulis in phyllocladiis rigidulis oblanceolatis triplo compositis mutatis, foliis propriis nullis, floribus parvis ad dentes phyllocladiarum impositis, pedicellis brevibus, bracteis minutis deltoideis. Mr. Baron has been at last successful in proeuring flowers and fruit of a very curious plant, whieh has puzzled us for a long time. We had specimens without either, long ago from Bojer and Meller, consisting of triply compound phyllocladia with dis- tantly-toothed margins. In my last paper (Journ. Linn. Soc. vol. xx. p. 249) I described it as a doubtful Zwocarpus, and it was named in manuscript by Bojer Xylophylla ensifolia. Now A-——— FLORA OF MADAGASCAR. 337 Mr. Baron (No. 3139) has obtained full material, from which the following notes are made. It is well known to the natives under the name of Harahara, and the wood is very hard and valuable. I fear that the genus which I named Baronia in Trimen's J ournal, 1882, p. 67, will have to be merged in Rhus, and se I have called this very characteristic and peculiar Madagascar type after Mr. Baron again, with a slight variation in the name to avoid confusion. NEOBARONIA PHYLLANTHOIDES, Baker, sp. unica. A high tree, glabrous in all its parts, with very hard wood. Branchlets woody, subterete, widening gradually into the primary phylloclades, which are oblanceolate, 4-5 inches long, under an inch broad, rigid in texture, green on both surfaces, distantly minutely toothed on the margin, with fine immersed uniform anastomosing vertical veins. From the edge of primary phyl- loclades spring others, and these are sometimes again com- pound. From the teeth of the margin spring little clusters of flowers on short pedicels, with a minute deltoid bract at the base. The persistent glabrous brown calyx is à in. long and broad, with minute deltoid teeth. Corolla papilionaceous, bright purple, }in.long. Stamens diadelphous, the upper filament free, and the others united in a sheath; anthers minute, globose. Ovary linear, sessile, with a short abruptly incurved style and capitate stigma. Pod coriaceous, indehiscent, not at all com- pressed, 4 in. in diam., 14-14 in. long, 1-2-seeded, narrowed into à stalk at the base and a beak at the apex. DarnrRGrA BARONI, n. sp. Arborea, ramulis pilosis, foliis pinnatis breviter petiolatis, foliolis 15-25 oblanceolato-oblongis rigidis 1-nervatis dorso nervatis, floribus in paniculas parvas deltoideis dispositis, calycis dentibus latis, petalis ealycem duplo superantibus, staminibus monadelphis, legumine tenui rigido 1-2-$permo. A much-branched erect tree, with pilose branchlets. Leaves 2-4 in. long, exclusive of the short petiole ; leaflets opposite or alternate, with a short petiolule, obtuse, j-i in. long, firm in texture, the margin rather recurved, pilose only on the prominent midrib beneath, the other veins invisible. Flowers in small del- toid panicles with pubescent branchlets; pedicels very short. Calyx campanulate, 4; in. long, with an ovate bracteole at the base; teeth half-orbicular. Corolla 4 in. long. Filaments all united in the lower half in a sheath slit along the top; anthers 338 MR. J. G. BAKER ON THE minute, globose. Pod straight, oblong, thin, rigid, brown, 1-2- seeded, stipitate, 1-2 in. long, }-3 in. broad, very little thickened against the seeds, but with copious veins radiating from them. — Baron 2598! A valuable timber. Native name “ Voamboana." CADIA PEDICELLATA, n. sp. Arborea, ramulis glabris, foliis breviter petiolatis imparipinnatis, foliolis 11-17 oblongo-oblanceolatis obtusis rigidulis oppositis vel alternis, racemis axillaribus paucifloris, pedicellis longissimis, bracteis linearibus, calycis magni dentibus deltoideis, petalis obovatis calyce sesquilongioribus, sta- minibus petalis equilongis, ovario lineari stipitato. A tree, glabrous in all its parts, with angled slender branchlets. Leaves erowded, 3-4 in.long; leaflets about an inch long, shortly stalked, firm in texture, green and glabrous on both surfaces, entire, minutely emarginate at the apex. Racemes 2-3-flowered, axillary, much shorter than theleaves; pedicels slender, glabrous, ascending, 1-13 in. long; bracts minute, linear. Calyx cam- panulate, greenish, glabrous, Zin. long; teeth 4-7 as long as the tube. Petals obovate, imbricated, an inch long. Stamens as long as the petals; anthers minute, oblong, versatile. Ovary linear, glabrous, stipitate. Legume not seen.—Baron 2248! Near C. pubescens, Bojer, described in my last paper, Journ. Linn. Soc. xx. p. 135. MIMOSA DASYPHYLLA, n. sp. M. ramulis validis angulatis aculeis erebris minutis sparsis armatis, foliis amplis bipianatis, pinnis 7-8-jugis, foliolis 10-12-jugis oblongis petiolu- latis utrinque pubescentibus, floribus copiose paniculatis, calycis dentibus deltoideis, corolla segmentis lanceolato-deltoideis, staminibus exsertis, legumine curvato tenui stipitato 6-8-spermo suturis inermibus. Branehlets stout, woody, strongly angled, brown, shining, armed with copious scattered small hooked prickles. Leaves half a foot long, including the 1-11 in. petiole, with a prickly rhachis; pinne 2-23 in. long; leaflets 4 in. long, moderately firm in texture, with an excentrie midrib, grey and densely pubescent beneath, green and less pubescent above. Flowers in an ample deltoid terminal panicle with prickly branches. Calyx one third as long as the funnel-shaped corolla, which is ;!; in. long. Pod thin, strap-shaped, curved, distinctly stipitate, 4—5 in. long, 1-3 in. broad, thickened a little opposite the 6-8 seeds.—Baron 2426! Allied to M. nissobiensis, Benth. FLORA OF MADAGASCAR. 339 MIMOSA MYRIACANTHA, n. sp. M. ramulis validis lignosis aculeis crebris sparsis minutis uncinatis arma- tis, foliis amplis bipinnatis, pinnis I-10-jugis, foliolis sessilibus oblongo- quadratis 8-9-jugis dorso pubescentibus, floribus copiose paniculatis, calycis dentibus deltoideis, corolle segmentis lanceolato-deltoideis, staminibus exsertis, legumine oblique oblongo obtuso tenui glabro 3-4-spermo suturis aculeis minutis stramineis uncinatis copiosis armatis. Branchlets terete, woody, armed with copious minute scattered prickles. Leaves about half a foot long, shortly petioled, imbri- cated; rhachis pubescent, slightly prickly ; pinne 11-2 in. long; leaflets 1-1 in. long, obtuse, moderately firm in texture, cut away obliquely on the upper side at the base. Flowers arranged in an ample terminal panicle, with pubescent prickly branches. Calyx one third as long as the corolla, which is funnel-shaped, ;; in. long, with teeth nearly as long as the tube. Legume sessile, thin, brown, crustaceous, rather oblique, about 2 in. long, $ in. broad, finely veined transversely, not much thickened opposite the seeds, both sutures armed with copious hooked straw-coloured priekles.— Baron 2597! A near neighbour of the East-Indian M. hamata, Willd. WEINMANNIA MINUTIFLORA, N. Sp. Glabra, foliis petiolatis trifoliolatis, foliolis rigide coriaceis obtusis serratis lateralibus oblongis terminali obovato basi longe attenuato, flori- bus minutis in racemos subspicatos longissimos dispositis, calycis dentibus deltoideis, ovario ovoideo sericeo, stylis ovario æquilongis. A shrub or small tree, with glabrous branchlets and leaves. Petiole 1-2 in. long; leaflets rigid in texture, green on both surfaces, very obtuse, minutely dentate, with parallel raised erecto-patent main veins beneath, the side ones oblong, sessile, erecto-patent, 1-14 in. long, the end ones considerably larger and narrowed very gradually into a winged petiolule. Racemes copious, shortly peduncled, 4-6 in. long; axis stout, angled. Calyx campanulate, Į lin. in diam., sessile or obscurely pedicellate ; teeth 5, deltoid. Corolla not seen. Pistil py 1n. long, the styles equalling the ovoid densely-silky ovary.—Baron 2547! WEINMANNIA FRAXINIFOLIA, Nn. Sp. | m . . " " - - f Is an is arratis Glabra, foliis pinnatis, foliolis 7-9 oblongo lanceol tis acutis sc i 5 rigidis nitidulis, floribus in spicas elongatas cylindricas dispositis, calycis dentibus magnis, ovario magno ovoideo sericeo, stylis ovario æqulongis. . v 2B LINN. JOURN.— BOTANY, VOL. XXI. 340 MR. J. G. BAKER ON THE A tree, with glabrous branches and leaves. Leaves distinctly petioled, 4-5 in. long; leaflets 7-9, firm and rigid in texture, glossy, green on both surfaces, 1j-2 in. long, 1-3 in. broad, strongly inciso-crenate, sessile, narrowed to the base, the ereeto- patent main veins distinct beneath. Spikes copious, cylindrical, peduncled, 3-4 in. long. Calyx campanulate, j in. in diam., with 5 large ovate obtuse segments. Petals not seen. Pistil j in. long, the slender styles as long as the ovoid densely drab-silky ovary.— Baron 3148! KITCHINGIA SCHIZOPHYLLA, n. sp. Perennis, glabra, caule tereti, foliis petiolatis lanceolato-deltoideis carnosis pinnatis segmentis distantibus linearibus deflexis, cymis pauci- floris in paniculam laxam deltoideam dispositis, pedicellis flexuosis calyce longioribus, calycis tubo campanulato segmentis deltoideis cuspidatis, corolla rubra calycem 4plo superante, stylis elongatis. A glabrous perennial. Fully-developed leaves 4-5 in. long, including the petiole, consisting of a few deflexed entire linear segments, the lowest about an ineh long, the upper growing gradually shorter. Cymes few-flowered, arranged in a very lax large terminal panicle; pedicels slender, flexuose, 1-3 in. long. Calyx campanulate, 4 in. long and broad; segments deltoid- cuspidate, longer than the campanulate tube. Corolla bright red, only seen in a withered state. Fruit-carpels twice as long as the calyx; styles as long as the carpels. Anthers minute.— Baron 3132! MYRIOPHYLLUM AXILLIFLORUM, n. sp. M. caulibus floriferis elongatis, folis verticillatis seepissime quaternis ascendentibus firmulis lanceolatis superioribus serratis inferioribus pec- tinato-pinnatifidis, floribus omnibus axillaribus, superioribus masculis petalis 4 oblongis, inferioribus foemineis ovario acute 4-angulato vittis tuberculatis petalis 4 parvis. Leafy floriferous stems simple, branching off from a wide- creeping rootstock. Leaves rather firm in texture, sessile, ascending, four in a whorl, lanceolate, 4-3 in. long, the upper faintly toothed, the lower pectinate-pinnatifid. Flowers in the axils of all the whorls down nearly to the bottom of the branches, those of the upper whorls all male, of the lowest female. Male flowers with 4 oblong petals and 4 linear-oblong anthers, with short filaments. Female flowers with 4 small petals and an acutely-angled globose 4-celled ovary, rugose on the four keels.— — FLORA OF MADAGASCAR. 841 Baron 3325! Also Hildebrandt 4030! Allied to M. verticil- latum, L. Mr. Baron has also gathered lately in Central Mada- gascar the Cape Gunnera perpensa, L. (2238!) WEIHEA SESSILIFLORA, n. sp. Glabra, foliis petiolatis rigidulis obovato-oblongis acuminatis sursum serratis, floribus ad nodos 2-4-nis axillaribus sessilibus, bracteolis cori- aceis orbicularibus, calycis tubo campanulato segmentis parvis semi- orbicularibus, petalis dorso sericeis, antheris oblongis. A tree, glabrous in all its parts, with woody branchlets dilated below the nodes. Leaves opposite, distinctly petioled, rigid in texture, green on both surfaces, 4-5 in. long, 13-2 in. broad, lengthened out at the point, serrated in the upper half, entire in the lower half, deltoid, the base distinctly penninerved. Flowers 2-4 in a sessile cluster at the nodes, seen only in an early stage ; bracteoles orbicular, coriaceous, entire or cleft. Calyx-segments semiorbicular. Corolla globose in bud, silky on the outside. Anthers numerous, oblong.— Baron 2583 ! Evcenta (§ SYZYGIUM) LOISELEURIOIDES, n. sp. Glabra, folis parvis petiolatis obovatis rigidis, cymis terminalibus parce paniculatis, pedicellis brevissimis, calycis tubo cuneato segmentis latis rotundatis brevissimis, petalis parvis orbicularibus, staminibus circiter 30 calycis tubo brevioribus. A much-branched shrub, glabrous in all its parts, with slender square branchlets. Leaves shortly petioled, about 2 in. long, very obtuse, deltoid at the base, rigid in texture, the veins of the under surface fine, erecto-patent, reticulated. Cymes few- flowered; pedicels very short. Calyx brown, glabrous, cori- aceous, obconical, 4l, in. long and broad. Petals minute, or- bicular. Stamens not above y in. long.— Baron 2641! Homatium (§ BLACKWELLIA) CONFERTUM, n. Sp. Glabrum, folis obovato-oblongis sessilibus rigidulis serratis obtuse cuspidatis, floribus in spicas ascendentes strictas paucifloras dispositis, bracteis coriaceis oblongis, calycis tubo globoso piloso segmentis 5-6 oblanceolatis rigidulis, petalis 5 oblongis sericeis, staminibus isomeris. A tree, with the leaves crowded on the glabrous branchlets. Leaves firm in texture, green on both surfaces, 23-3 in. long, 1- lj in. broad, distinctly serrated, shortly euspidate at the tip, narrowed to a rather rounded base, the secondary veins distinct, erecto- patent, parallel. Flowers few, in shortly-peduncled stift : . ; rsistent spikes, each subtended by an orbicular coriaceoas "n B 342 MR. J. G. BAKER ON THE braet. Calyx with a eampanulate tube, and usually 5 oblanceo- late rigid persistent brown segments 4 in. long. Petals cucul- late, tomentose, much shorter than the calyx-segments, each enclosing a single stamen.— Baron 3185! 3256! VEPRECELLA HISPIDA, n. 8p. Fruticosa, ramulis gracilibus dense hispidis, foliis petiolatis ovatis acutis membranaceis integris trinervatis utrinque hispidis, cymis terminalibus laxe paniculatis, pedicellis elongatis, calycis tubo campanulato hispido dentibus minutis, petalis rubellis oblongo-spathulatis, antheris eequalibus obscure calcaratis, stylo antheris longiore. A shrub, 6-10 feet high, with slender square branchlets, densely clothed with deflexed brownish bristly hairs. Leaves distant, distinctly petioled, ovate, acute, entire, membranous, 2-3 in. long, bright green above, pale green beneath, thinly bristly on both surfaces and the margin. Flowers in a lax terminal panicle with cymose branches, both peduncles and pedicels densely bristly. Calyx with a campanulate densely bristly green tube i in. in diam., and an obscurely-toothed narrow collar-like limb. Petals bright red, 4 in. long. Stamens 8, shorter than the petals; filaments filiform; anthers yellow, sausage-shaped, ob- scurely spurred by the connective on the back at the base. Style clavate, overtopping the stamens.— Baron 420! 3257! Also, forest of Andrangaloaka, Dr. Parker! PHORNOTHAMNUS, genus novum tribus Oxysporearum, ordinis Melastomacearum. Calyx glaber, tubo campanulato persistente ovario adnato, limbo brevi erecto obscure lobato. Petala 4-5 oblongo-spathulata, obtusa. Stamina 8-10 petalis breviora, filamentis applanatis, antheris lineari-oblongis apice obscure cuspidatis uniporosis basi connectivo producto dorso obscure calearatis. Ovarium globosum 4—5-loculare basi calyci adnatum. Cap- sula globosa coriacea ad basin loculicide 4—5 valvis, valvis persistentibus rigidis oblongis, axi septis persistentibus late alato.—Suffrutex Mada- gascariensis ramosissima humifusa, ramulis foliisque lepidotis, foliis parvis oblongis petiolatis rigidis obscure trinervatis, floribus terminalibus purpurascentibus breviter pedunculatis sepissime solitariis, setis nullis. P. THYMOIDES, Baker, sp. unica. A densely-matted trailing undershrub, with the habit of Loi- seleuria procumbens. Branchlets and leaves copiously dotted over with minute brown scales. Leaves opposite, oblong, entire, coriaceous, shortly petioled, j-l in. long, with three obscure FLORA OF MADAGASCAR. 343 nerves arising from the base, of which only the midrib is pro- duced to the tip. Flowers generally solitary at the end of the branchlets on a short pedunele. Calyx brown, glabrous, cam- panulate, j in. in diam., with an obscurely lobed narrow erect limb. Corolla glabrous, conical in bud, bright red-purple; fully- developed petals oblong-spathulate, 7-3 in. long. Stamens equal, as long as the petals; filaments as long as the linear-oblong anthers, which have a single pore at the top, and are obscurely spurred on the back at the base by the produced connective. Capsule globose, l in. in diam, split down to the base into 4-5 oblong rigid valves, girt at the base by the persistent truncate cup- shaped coriaceous brown calyx.—Baron 1314, in fruit! Also Humblot 535, in flower! Allied to Veprecella aud Rousseauaia, but in habit resembling one of the Brazilian Microliciee. MEMECYLON OLEEFOLIUM, n. sp. Glabrum, ramosissimum, foliis subsessilibus rigidulis oblongis obtusis obscure penninerviis basi deltoideis dorso nigro punctatis, cymis axillaribus paucifloris pedunculatis, pedicellis brevibus, calycis tubo campanulato limbo lato obscure dentato, petalis ovatis, staminibus inclusis, antheris basi acutis, stylo elongato. A much-branched erect shrub, glabrous in all its parts. Leaves contiguous, nearly sessile, 1-14 in. long, deltoid at the base, rigid in texture, brownish above when dried, bright green beneath with copious minute immersed black dots, with a distinct midrib from base to apex, and obscure erecto-patent parallel secondary veins. Cymes copious, axillary, shortly peduncled, 3-5-flowered ; pedicels glabrous, 4-4 in. Calyx brown, glabrous, coriaceous, with a campanulate tube and a broad eoriaceous obscurely toothed collar-like limb Jin. indiam. Bud conical; petals greenish, ovate, acute, 2 iu. long. Stamens 8; filaments filiform. Anthers | in. long.—Baron curved, aeute at the base. Style filiform, 4 2233 ! 3195! Allied to M. Eleagni, Blume. MEDINILLA LEPTOPHYLLA, n. Sp. n Glabra, ramulis gracilibus, foliis sessilibus oblongis acummatis mem- branaceis trinervatis, cvmis laxis paucifloris pedunculatis, pedicellis elongatis bracteolatis, calycis tubo campanulato limbo integro, petalis oblongo: spathulatis pallide rubris, staminibus inclusis antheris basi cal- caratis, A shrub, glabrous in all its parts, with slender Woody branch- lets, with a few lenticels. Leaves sessile, about 2 in. long, unde: 344 MR. J. G. BAKER ON THE an inch broad, acuminate, thin in texture, tinged with purple when young, distinctly 3-nerved. Cymes axillary, 3-5-flowered, distinctly peduncled ; lateral pedicels 4-4 in. long, very slender, with a pair of minute bracteoles at the middle. Calyx green, with a campanulate tube 4 in. in diam., and a broad truncate limb. Petals oblong-spathulate, 7 in. long. Stamens 4 in. long; anthers linear-oblong, distinctly spurred at the base— Baron 3221! MEDINILLA LANCEOLATA, n. 8p. Glabra, ramulis gracilibus lenticellis rugosis, foliis breviter petiolatis lanceolatis rigidulis acutis uninerviis, eymis axillaribus paucifloris, pedi- cellis brevibus, calycis tubo campanulato limbo lato obscure dentato, petalis pallide rubellis latis obtusis, staminibus inclusis antheris basi obscure calcaratis. A shrub, glabrous in all its parts, with very slender woody branchlets, rough with large lenticels. Leaves shortly petioled, lanceolate, firm in texture, 12-2 in. long, 1-3 in. broad, narrowed gradually from the middle to a long point, with no veins visible except the distinct midrib. Cymes sessile or shortly peduncled, axillary ; pedicels 4 in., with a pair of small bracteoles. Calyx green, with a eampanulate tube 4'; in. in diam., and an obscurely lobed broad collar-like limb. Petals pale red, Lin. long. Stamens shorter than the petals; anthers linear-oblong, obscurely spurred at the base. Style as long as the petals.— Baron 2983! MEDINILLA LOPHOCLADA, D. sp. Glabra, ramis crassis acute tetragonis lenticellis rugosis, foliis magnis oblongis coriaceis sessilibus 7-nervatis, cymis sessilibus axillaribus multi- floris, pedicellis flore brevioribus, calycis tubo campanulato, limbo lato obscure lobato, petalis latis obtusis, staminibus 10 inclusis zequalibus antheris basi calcaratis. A shrub, glabrous in all its parts, with acutely-angled tetra- gonous woody branchlets j in. in diam., rugose between the raised ribs with copious lenticels. Leaves rigid in texture, green on both sides, 5-6 in. long, 2-3 in. broad at the middle, obtuse or subobtuse, narrowed to the base, with 7 thick ribs produced from the base to the apex. Cymes many-flowered, axillary ; pedicels 5-4 in. Calyx bright red, glabrous, with a campanulate tube j in. in diam., and a broad collar-like obscurely lobed limb. Petals 5, red, oblong-spathulate, obtuse, 4-3 in. long. Stamens FLORA OF MADAGASCAR. 345 10, shorter than the petals; anthers linear-oblong, shortly spurred on the back at the base. Style filiform, as long as the petals.— Baron 3249! AMMANNIA CRYPTANTHA, n. sp. Perennis, glabra, caulibus tetragonis, foliis sessilibus oblongis integris, floribus racemosis, singulis bractea ovata cucullata cinctis, calycis tubo campanulato dentibus 5 deltoideis, petalis 5 parvis oblanceolatis, sta- minibus in tubo calycis inclusis, stylo elongato. Stems ascending from a trailing perennial rootstock, tetra- gonous, simple in the lower half, copiously branched upwards. Leaves opposite, oblong, sessile, 1-1 in. long. Panicle of few or many ascending rather lax racemes 2-1 in. long, each flower hidden by a clasping ovate foliaceous bract, with a short pedicel inside the bract. Calyx 2 in. long, with a campanulate tube and 5 deltoid segments. Petals 5, oblanceolate, twice as long as the calyx-lobes. Stamens inserted near the base of the calyx-tube ; filaments filiform; anthers minute, orbicular. Ovary ovoid; style j in. long.—Lyall 213! Baron 476! 1902! Near the East- Indian A. rotundifolia, Wight. EPILOBIUM OLIGANTHUM, n. sp. E. caule tereti gracili superne tomentoso, foliis subsessilibus lanceolatis glabris superne denticulatis, floribus paucis laxe corymbosis, ovario cylindrieo tomentoso, calycis segmentis lanceolatis, petalis magnis obovatis, stigmate magno integro quadrato, staminibus stylo brevioribus. Stems erect, slender, terete, glabrous except towards the top, where they are white-tomentose. Leaves lanceolate, sessile, glabrous, erecto-patent, 1-14 in. long, denticulate upwards. Flowers few, in a lax terminal corymb, wavy, cylindrical, white- tomentose, above an inch long. Calyx-limb pilose, 3 in. long, with a campanulate tube and lanceolate segments. Petals obovate-cuneate, red, 4 in. long. Style shorter than the petals ; stigma large, quadrate, entire. Stamens much shorter than the style— Baron 2269! Allied to the European Æ. palustre, L. Mopercca PELTATA, n. sp. l ME Volubilis, glabra, ramulis gracilibus lignosis, cirrhis nullis, foliis sim- plicibus ovatis integris acutis membranaceis petiolo supra basin affixis, cymis axillaribus paucifloris pedicellis elongatis, floribus masculis calycis tubo infundibulari viridulo, segmentis 5 lanceolatis reflexis tubo longi- 346 MR. J. G. BAKER ON THE oribus, petalis 5 albis oblanceolatis ad faucem calycis insertis, corona nulla, staminibus 5 antheris linearibus, filamentis brevibus applanatis infra oram calycis insertis. A glabrous climber, with slender terete woody stems with a membranous brown epidermis. Leaves distant, simple; petiole slender, about an inch long, attached a little above the base of the ovate membranous blade, which is 2-8 in. long. Male flowers only seen, arranged in axillary cymes; pedicels slender, 1-3 in. long. Calyx greenish, glabrous, with a funnel-shaped tube 4 in. long, and 5 reflexing lanceolate segments rather longer than the tube. Petals 5, whitish, lanceolate, 1 in. long, inserted at the throat of the calyx. Anthers 5, linear, erect, ; in. long, with short flattened filaments inserted a little below the throat of the calyx-tube.— Baron 2827 ! MELOTHRIA ($ ZEHNERIA) EMIRNENSIS, n. Sp. Volubilis, glabra, dioica, folis longe petiolatis cordato-ovatis acutis denticulatis cuspidatis membranaceis facie obscure scaberulis, floribus masculis copiose racemosis, foemineis in umbellas axillares paucifloris szepissime dispositis raro solitariis, fructu subgloboso lacunoso, seminibus discoideis obscure marginatis. A glabrous herbaceous twiner, with slender angled smooth stems, with a long simple tendril from each node much twisted spirally towards the tip. Petiole 1-2 in. long; blade 2-3 in. long, cordate-ovate, with a deep basal sinus, irregularly patent deltoid-euspidate teeth, and a distinet mucro, thin in texture, green on both surfaces, obscurely scattered over with white points and rudimentary bristles on the upper surface. Male flowers in copious axillary racemes ; calyx minute, campanulate. Female flowers in peduncled axillary umbellate cymes, rarely solitary. Ovary oblong, narrowed into a short cylindrical neck. Calyx-tube campanulate, under j in. long and broad, densely hairy inside; teeth minute, deltoid. Petals 5, small, oblong, inserted at the throat of the calyx-tube. Stigma placed opposite the tip of the petals. Berry 3-3 in. in diam., brown when mature, conspicuously pitted all over the epidermis. Seeds whitish, obovoid-discoid, } in. long, obscurely margined.— Baron 390! 397: 2821! Parker! Mr. Baron has also rediscovered 23481 2620! 2661!) M. Rutenbergiana, Cogn. in Rel. Rutenb. ii. p. 251. | / | FLORA OF MADAGASCAR. 347 BEGoNtA ($ QUADRILOBARIA) HETEROPODA, n. sp. Acaulis, stipulis magnis ovatis scariosis persistentibus, petiolis elongatis pilis reflexis brunneis persistentibus vestitis, limbo cordato-rotundato mem- branaceo obscure palmatifido irregulariter dentato, scapo elongato glabro, floribus masculis sepalis 2 cordato-orbicularibus rubellis petalis nullis antheris 20-30 lineari-oblongis filamentis brevissimis liberis. Acaulescent. Stipules large, ovate, scariose, persistent ; petiole 3-4 in. long, densely clothed with reflexed persistent brown paleaceous hairs; blade orbicular, membranous, broader than long, 6-9 in. broad, with a deltoid basal sinus 14 in. deep, obscurely palmatifid, the margin irregularly dentate, the upper surface glabrous, the main veins beneath with a few hairs like those of the petiole. Cyme in the only specimen of 10-12 exclusively male flowers; peduncle naked, slender, fragile, 6-8 in. long; bracts ovate, membranous, folded together. Male flower- wrapper of two purplish-red cordate-orbicular sepals 3 in. broad. Stamens 20-30, with linear-oblong anthers 3 in. long and very short filaments.—Baron 3295! Allied to B. nossibea, A. DC. Mr. Baron (2448 !) has also rediscovered the very curious B. Lyallii, A. DC. RHIPSALIS HORRIDA, n. sp. Lignosa, copiose ramosa, ramulis ultimis clavatis vel cylindricis basi attenuatis verticaliter multisulcatis, pilis dense cespitosis rigidulis in- æqualibus fragilibus, floribus lateralibus sessilibus 1—3nis, fructu carnoso globoso glabro vel parce piloso magnitudine pisi. Stems woody, terete, copiously branched, calvate when very old, pale green, 4 in. in diam., fleshy when young, with irregular slender vertical ribs and densely scattered pulvilli of almost hya- line, unequal, fragile rather bristly hairs +-4 in. long. Berries l-3nate, sessile, lateral, fleshy, globose, the size of a pea, gla- brous except at the apex, or with a casual pulvillus of hairs like those of the branches. Flowers not seen.— Baron 2750! 3269! TELEPHIUM MADAGASCARIENSE, n. Sp. 2 Perenne, glabrum, caulibus decumbentibus, foliis alternis sessilibus ob- longis, stipulis scariosis deltoideis persistentibus, cymis decompositis laxis, pedicellis brevibus, sepalis 5 oblongis valde imbricatis, petalis vix exsertis, staminibus calyce æquilongis, ovario ovoideo, stylis 3 falcatis. A glabrous perennial herb, with decumbent stems half a foot long. Leaves alternate, sessile, oblong, obtuse, 2-4 in. ong, deltoid at the base ; stipules minute, persistent, scariose. Cymes 948 MR. J. G. BAKER ON THE decompound, lax, terminal, 2-3 in. in diam. ; branches arcuate; pedicels short, with a small deltoid white-edged bract at the base. Calyx campanulate, } in. long; petals 5, herbaceous, oblong, obtuse, much imbricated, whitish at the edge. Petals oblong, scarcely longer than the calyx. Stamens 5, distinctly perigy- nous; anthers small, globose, placed opposite the tip of the sepals. Pistil shorter than the calyx; ovary ovoid; styles fal- cate, as long as the ovary.— Baron 1909! This genus belongs to the Mediterranean region, and is not kaown in Tropical Africa or at the Cape. HYDROCOTYLE FILICAULIS, D. 8p. H. caulibus filiformibus glabris late reptantibus, foliis caespitosis orbicu- laribus basifixis glabris inciso-crenatis basi truncatis petiolis pubescentibus, floribus paucis in umbellas pedunculatas solitariis dispositis, bracteis ovatis, pedicellis brevissimis, fructu globoso jugis 5 inconspicuis, stylis falcatis. Stems wide-trailing, filiform, glabrous, rooting from the tufts, with several leaves, and a single peduncled umbel of flowers. Petiole pubescent, 1-3 in. long; blade orbicular, glabrous, 4-3 in. broad, almost truncate at the base, inciso-crenate. Flowers 2-6 in ar umbel; pedicels very short; bracts ovate, persistent, longer than the pedicels. Fruit orbicular, 4; in. diam.; mericarps turgid, oblong, obscurely ribbed; styles short, cylindrical, distinctly faleate.— Baron 3219! Allied to H. asiatica, L. HYDROCOTYLE SUPERPOSITA, n. sp. Glabra, rhizomate filiformi late reptante, foliis solitariis longe petiolatis peltatis orbicularibus crenatis membranaceis 9—10-nervis, floribus spicatis verticillis 2-4 superpositis, bracteis ovatis, fructu globoso, mericarpiis jugis 5 prominulis przeditis, petalis parvis ovatis, stylis rectiusculis bre- vissimis. Rhizome filiform, wide-trailing, sending out tufts of root-fibres from the nodes. Leaves solitary, glabrous, with a slender petiole 1-2 in. long; blade orbicular, peltate, 1-2 in. broad, membranous, crenate. Flowers in spikes of 2—4 superposed whorls of 4-6 flowers each, with a slender peduncle 1-14 in. long ; bracts minute, persistent. Fruit globose, laterally compressed, lin. diam.; mericarps with a distinct ridge on each side mid- way between the back and margin ; style very short, cylindrical, nearly straight. Petals ovate, minute, valvate.— Baron 1897! FLORA OF MADAGASCAR. 349 PIMPINELLA LAXIFLORA, n. sp. Perennis, glabra, caule gracili foliato, foliis petiolatis trifoliolatis, foliolis oblongis acutis dentibus latis cuspidatis, umbellis compositis iu paniculam magnam laxissimam dispositis, bracteis bracteolisque minutis lanceolatis, pedicelis longissimis, fructu oblongo glabro jugis parum prominulis, calycis limbo obsoleto, stylopodiis conicis, stylis brevissimis falcatis. A glabrous perennial, with slender terete suffruticose stems. Stem-leaves several, contiguous, long-petioled ; leaflets mode- rately firm in texture, acuminate, 2-3 in. long; side ones nearly sessile, very unequal at the base; end ones distinctly stalked, deltoid at the base; umbels 5-6nate ; peduncles very slender, 1-2 in. long; pedicels 3-Gnate, 1-11 in. long. Fruit oblong, 15 in. long, with 5 equal indistinct ribs on each mericarp. Petals not seen. Stylopodia conie; styles very short, spreading.— Baron 290! 2887! PHELLOLOPHIUM, genus novum tribus Seselinearum, ordinis Umbelliferarum. Calycis dentes obsoleti. Petala oblonga apice inflexa. Stylopodia conica in stylum brevem falcatum attenuata. Fructus oblongus teres ad commissuram haud constrictus, carpellis facie planis, dorso rotundatis, Jugis 5 crassis stramineis suberosis subsqualibus, valleculis profundis angustis. Vittæ ad valleculas solitarie. Carpophorum filiforme bifidum. Semen rectum, facie commisurali leviter bisuleato, dorso rotundato jugis 5 prominulis przdito.— Herba glabra erecta robusta orgyalis, foliis radi- calibus bipinnatis, foliolis magnis oblongis petiolo basi late dilatato, umbellis compositis copiose paniculatis, bracteis bracteolisque paucis parvis linearibus. P. MADAGASCARIENSE, Baker, sp. unica. l l An erect robust perennial herb, 5 or 6 feet high, with many- angled hollow stems. Lower leaves bipinnate, deltoid, with a petiole much dilated, and sheathing at the base; pinne 5-6- jugate; leaflets oblong, sharply toothed, 2-3 1n. long, the upper sessile, the lower shortly petioled. Compound umbels forming a panicle a foot long, with level-topped divisions; secondary umbels and flowers both about in twelves; bracts and bracteoles few, linear, entire, conspicuous. Petals not well seen ; accord- ing to Mr. Baron's note, ovate-lanceolate, greenish-white. Calyx- limb entirely obsolete. Fruit oblong, jin. long, l in. in diam., terete in transverse section, with 9 very thick straw-coloured 350 MR. J. G. BAKER ON THE neatly equal corky ribs on each mericarp. Seed aromatie, faintly bisuleate on the face, with a distinct ridge opposite each of the great corky ridges of the fruit.— Baron 60! 1814! 2227! Hilde- brandt 3868! Used as a tooth-ache remedy. Native name * Tsileondroaha." Habit of Angelica, but allied to Seseli and Fæni- culum, from which it differs by its thick corky ribs and narrow very deep vallecule, and entirely obsolete calyx-limb. CUPHOCARPUS INERMIS, n. Sp. Inermis, glaber, foliis pinnatis, foliolis oblongis obtusis crenatis, flori- bus copiose racemoso-paniculatis. A low tree, with thiek woody unarmed ultimate branchlets, glabrous in all its parts. Leaves petioled, simply pinnate ; blade 6-10 in. long; leaflets about 9, oblong, coriaceous, sub- sessile, 2-3 in. long, deltoid at the base, obtuse, conspicuously crenate. Kacemes very numerous, erecto-patent, arranged in a long panicle with a whorl at the end ; pedicels short, square, arti- culated at the apex, subtended by a minute deltoid persistent bract. Ovary globose, l-celled, 5-ribbed, crowned by the narrow entire calyx-limb. Petals 4, oblong, greenish, 4; in. long. Stamens 4. Berry globose, oblique, } in. diam.; style solitary, erect.— Gerrard 11! Baron 2748! This is an endemic genus, of which we do not possess the type species (C. aculeatus, De- caisne & Planch. in Rev. Hort. ser. 4, vol. iii. p. 109), which has never, so far as I am aware, been described in detail. GASTONIA EMIRNENSIS, n. sp. Glabra, foliis simpliciter pinnatis foliolis oblongis integris subcoriaceis oppositis petiolulatis, umbellis 6-9-floris pedunculatis in paniculam elon- gatam dispositis, pedicellis basi articulatis, ovario globoso 10-12-loculari, calycis limbo angustissimo truncato, stylis brevibus clavatis stellatim patulis. A shrub, glabrous in all its parts. Leaves a foot or more long; leaflets opposite, distinctly stalked, 4-5 in. long, obtuse, deltoid at the base, green on both sides, with parallel erecto- patent indistinct secondary veins. Panicle a foot long, with a stout woody naked subterete axis, numerous lateral scattered umbels on erecto-patent peduncles 1-2 in. long, and a whorl of five at the end; pedicels 3 in. long. Fruit globose, } in. in diam., with as many distinct ribs as there are cells. Styles forming a — —— —— ——Á— FLORA OF MADAGASCAR. 351 star $ in. in diam.— Baron 2747! Differs from the Mauritian G. cutispongia, Lam., by its almost obsolete calyx-limb. Panax ($ SPHJEROPANAX) CONFERTIFOLIUM, n. sp. Glabrum, foliis simpliciter pinnatis petiolatis, foliolis 5-7 oblongis obtusis rigidis petiolulatis, umbellis paucifloris pedunculatis paniculatis supremis verticillatis, pedicellis brevibus tetragonis, fructu globoso 5-loculari stig- matibus brevibus stellatim patulis. A shrub, glabrous in all its parts, with thick woody branchlets. Leaves crowded, ascending ; petiole 1-2 in. long ; rhachis 13-3 in.; leaflets rigid in texture, 1-2 in. long, green on both sides, with faint erecto-patent secondary veins. Panicles shorter than the leaves; end umbels 5 in a whorl; pedicels ;4—} in. long, articu- lated at the base. Ovary globose, 4 in. long and broad ; calyx- limb very narrow, subentire ; styles cylindrical, 3 lin. long, curved, spreading.— Baron 1905! Near P. zanthoxyloides, Baker. Panax (§ SPHHROPANAX) MULTIBRACTEATUM, N. sp. Glabrum, foliis simpliciter pinnatis, foliis multijugis sessilibus subcoria- ceis oblongis acutis crenato-serratis, umbellis 6-8-floris in paniculam elon- gatam dispositis, pedunculis bracteis multis minutis persistentibus preditis, ovario globoso 5-loculari, stylis subulatis falcatis. A tree, glabrous in all its parts. Leaves simply pinnate, breaking easily at the articulated nodes; leaflets sessile, multi- Jugate, 3-4 in. long, about an inch broad, acute, subcoriaceous, distantly conspicuously inciso-crenate. Umbels in a long simple panicle, with a whorl at the end; peduncles under an inch long, furnished with several minute deltoid green bracts ; pedicels } in. long. Ovary globose, 5-celled; disk conical beyond the rudi- mentary calyx-limb; styles cylindrical, reflexed, 2 line long.— Baron 2469! Panax (8$ SPHHROPANAX) AMPLIFOLIUM, D. 8p. l Glabrum, foliis magnis deltoideis bipinnatis, foliolis integris oblongis acutis subcoriaceis, umbellis 6-8-floris copiose paniculatis superioribus verticillatis, pedicellis flore longioribus, calycis tubo campanulato, limbo angustissimo truncato, ovario 4-5-loculari. A tree, glabrous in all its parts. Fully-developed leaves bipinnate, with a blade a foot long and broad and a petiole a ft. long ; leaflets paucijugate, distant, all distinctly petioled, entire, subcoriaceous, 2-3 in. long, green and glossy above. Umbels copiously panicled, those at the end of the axis several 352 MR. J. G. BAKER ON THE in a whorl; pedicel 1-1 in. Calyx with a campanulate tube y in. diam., and a narrow truncate limb. Bud green, globose. Petals and stamens 5. Ovary 4-5-celled. Berry not seen.—Baron 3233 ! PANAX (S SpH#ROPANAX) PENTAMERUM, N. 8p. Glabrum, ramulis crassis, foliis longe petiolatis deltoideis bipinnatis, foliis ovatis integris acutis paucijugis subcoriaceis, umbellis 4-6-floris paniculatis supremis verticillatis, pedicellis brevibus, fructu globoso penta- gono 5-loculari angulis acutis, stylis brevibus cylindricis stellatim patulis. A tree, with thick woody terete branchlets, glabrous in all its parts. Petiole 3-4 in. long ; blade deltoid, and bipinnate in the fully-developed lower leaves, 6-9 in. broad; pinne and leaflets paucijugate, the latter stalked, ovate, acute, entire, 2-3 in. long, firm in texture, green on both surfaces. Umbels peduncled, panicled, those at the tip of the axis copiously whorled; pedicels lin. long. Fruit globose, 5-celled, 4 in. diam., with 5 strong angles; stigma lin. diam., with 5 patent cylindrical branches. Petals not seen.— Baron 2555! 2719! MELANOPHYLLA, genus novum Cornacearum. Flores hermaphroditi. Ovarium inferum oblongum 2- rarius 3-loculare, ovulis in loculo solitariis ab apice pendulis; calycis limbo brevi minuto dentato; stylis 2-3 sursum facie applanatis. Petala 5 oblonga imbricata decidua. Stamina 5 epigyna petalis breviora, filamentis filiformibus, antheris magnis oblongis subbasifixis. Fructus ignotus.—Frutices vel arbores Madagascarienses, foliis alternis exstipulatis petiolatis oblongis integris vel serratis, floribus parvis racemosis vel racemoso-paniculatis, pedicellis basi bracteatis, apice szepe bibracteolatis, floribus foliisque siccitate nigrescentibus. Habit of Psychotria, but the leaves are alternate and exstipu- late. The fruit is unknown. There is one species represented by two numbers in Mr. Baron's last collection ; another was gathered by Mr. L. Kitching; and there are two in Humblot’s collection from the north-east of the island, Nos. 437 and 517,— in all four distinct species. M. ALNIFOLIA, Baker, n. sp.; foliis obovatis apice solum ser- ratis, floribus simpliciter racemosis, calycis limbo dentibus deltoideis. An erect much-branehed shrub or tree, glabrous in all its parts. Leaves alternate, turning black when dried ; petiole 3-3 in., articulated and dilated to clasp the stem at the base; blade FLORA OF MADAGASCAR. 353 obovate, obtuse, 2-3 in. long, deltoid at the base, moderately firm in texture, inciso-crenate towards the apex only, with a few fine erecto-patent main veins. Racemes 1-2nate at the end of the branches, shortly peduncled, 13-2 in. long; pedicels erecto- patent, § in. long, with a small persistent deltoid bract at the base, and a pair of small persistent deltoid bracteoles at the apex clasping the ovary. Ovary oblong, jl; in. long. Calyx-limb minute, collar-like, with 5 distinct deltoid teeth. Petals oblong, 4 in. long. Filaments as long as the oblong anthers. Stigmas tongue-shaped.— Baron 3097 ! 3240! M. AUCUBÆFOLIA, Baker, n.sp.; foliis longe petiolatis oblongis serratis, floribus copiose racemoso-paniculatis, calycis limbo seg- mentis brevissimis. A shrub or tree, glabrous in all its parts. Branchlets stouter than in the other species, with pale drab bark marked with close fine transverse ridges. Leaves alternate, turning blackish when dried; petiole 2-3 in. long, much dilated, amplexicaul and arti- culated at the base; blade oblong, obtuse, 5-6 in. long, distinctly serrated except towards the deltoid base, moderately firm in texture, with fine distant erecto-patent main veins. Racemes erecto-patent, forming a deltoid peduncled end-panicle; pedi- cels very short, with a persistent deltoid bract at the base. Ovary oblong, not clasped at the base by a pair of bracteoles ; limb short and collar-like, obscurely toothed. Reflexing petals 4 in. long. Stamens shorter than the petals.— Between Tama- tave and Antananarivo, L. Kitching! LINN. JOURN.— BOTANY, VOL. XXI. 354 MR. C. T. DRUERY ON A SINGULAR MODE OF Observations on a singular Mode of Development in the La y- Fern (Athyrium Filix-femina). By Cartes T. DnvERY. (Communicated by Dr. J. Murre, F.L.S.) [Read 19th June, 1884.] THe reproduction of the Filices by their spores results from sexual action taking place upon the under surface of the prothallus to which the spore gives rise. So far, I believe, no development of the perfect prothallus has been observed with- out the agency of the spore, and the following record of such a case therefore deserves special notice. Some years ago a very distinct and beautiful form of Athyrium Filiz-femina was found wild by Mr. Moule in North Devon, from whose possession it passed into that of Col. Jones, of Clifton. Many attempts were made at the time to propagate it from what were assumed to be spores, always, however, without success; and at length it was taken for granted that the peculiar growths produced by this Fern in the place of sori were merely abortive spore-cases, and that the plants, like some other abnormal forms, lacked the special vigour necessary for the for- mation of perfect reproductive spores. All further attempts at raising it were consequently abandoned; and only two divisions of the plant exist *. In the autumn of 1883 I discovered upon another Athyrium (A. F.-f., var. plumosum divaricatum) numerous proliferous bulbils occupying the place of sori on the back of the fronds ; and, reporting thisto Mr. G. B. Wollaston, he was led to re-examine A. .F.-f. clarissima, as the Fern in question had been named by Col. Jones, and came to the conclusion that these so far barren excrescences might be viviparous growths of a kindred nature, and capable of reproducing the parent form by direct bud-development. A portion of a frond was consequently sent to me, and upon examining it under the microscope I found that there were very material structural differences between the unmistakable bulbils of A. Ff. divaricatum and the singular growths upon A. F.-f. clarissima, the former being solitary bud-like growths seated in the centre of a number of brown lanceolate scales and without a trace of indusium; while the latter were com- * It is, of course, open to question whether the excrescences formed prior to 1883 were of exactly the same nature. Col. Jones inclines to the belief that they approached more nearly the character of sori, and did not in previous years present the same appearance as now described. DEVELOPMENT IN THE LADY-FERN. 855 posed of 5 or 6 or more flask-shaped bodies, each one larger than the bulbils aforesaid, and seated within an undoubted indusium. The masses were sufficiently large for their forma- tion to be clearly distinguishable by the naked eye, covering more than the space of an ordinary sorus. At this stage no signs of Spores or spore-cases could be detected, nor could any axis of growth be perceived ; so that it was impossible to form any theory as to the eventual mode of reproduction which might result; for although the tips of the flask-shaped pseudobulbils were in some cases elongated into filiform processes, no sign of circination or resemblance to fronds was visible, added to which the presence of an indusium in the place of the scales common to true bulbils led to the assumption that they were abnormal sporoid growths, and not proliferous ones likely to produce plants by direct bud- growth. To test their capabilities I laid down on Nov. 27, in a duly prepared seed-pan, numerous pinns, which I imbedded edgewise halfway in the soil, the growths being thus brought into im- mediate contact with it, lying as they did along the rhachides of the pinnzg. I then placed the pan in slight heat, with the result that the pseudobulbils immediately began to inerease in size and to develop in such a fashion, that on Dec. 24 I was able to record an evident foliaceous extension and division of the tips of the pseudobulbils, and the appearance of numerous long rigid glassy-looking rods or hairs which sprang from their bases. These rods bore a strong resemblance to the root-hairs common to the undersides of prothalli; but their decided upward growth, radiating stiffly, seemed opposed to this view, as also the fact that they sprang from the bases and sides of the pseudobulbils. It is probable, however, that they acted as aerial roots, for the growth of the tips of the pseudobulbils proceeded rapidly, until, on Feb. 10 of the present year, I recorded that they pad assumed a decided prothalloid form, while the upright rods had either become deflected or absorbed. Eventually all the tips of the flask-shaped bulbils assumed the form of perfect prothalli of the usual size and shape, the pseudobulbils themselves pens absorbed and disappearing, and the usual root-hairs deve oping under the prothalli. On March 17 several of these pre 4 were examined microscopically, both by myself and by the Rey. Mr. Aubrey, of Salisbury (to whom I am indebted for aid al stages of growth), and well-developed arche- a € o in observing the fin "T 202 956 MR. C. T. DRUERY ON A SINGULAR MODE OF gonia were found in the usual place and number, but so far neither of us was able to detect antheridia. Early in May, how- ever, I succeeded in finding a single antheridium; and it is manifest that many others must have been present on the prothalli not examined, as on May 21 the final stage was reached. Small fronds being visible in several cases, projecting from the bifur- cation of the prothallus, and evidently therefore produced from the archegonia by the ordinary sexual mode of reproduction; though the prothalli, as has been shown, had developed from growths that differed widely from spores in their form, their size, persistent adherence to the pinnae, their production of root- hairs from their surface, and, finally, the development of the prothallus from their apices by simple extension of growth. Lest it might be assumed that these prothalli may after all have resulted from true spores scattered amongst the excrescences described, it should be borne in mind, first, that no spores or spore-cases could be distinguished when the pinnæ were laid down; secondly, that all attempts to raise this Fern from spores have failed; and finally, that the entire development of the pro- thallus from the pointed tip of the pear-shaped pseudobulb— its dilation, bifurcation, and gradual assumption of the true prothallus-form—has been carefully watched and noted step by step, not merely in one case, but in many, in all of which the prothallus was evolved in the same way precisely. Where, as in this case, the whole phenomenon is new to the observer, many points of interest are apt to be overlooked, their importance being unknown until too late. Another season’s growth may therefore confidently be expected to throw more light upon this development, and especially in relation to the first appearance of the pseudobulbils themselves, which only came under close observation when already of considerable size. In framing this account of the occurrence, I have confined myself as strictly as possible to a simple and, I hope, clear record of the phenomena observed during the various stages of growth of the abnormal sporoid excrescences under observation. In con- clusion, however, I may be permitted to point out, in connection with such phenomena that, so far as formal records are concerned, the family of Athyria has hitherto been remarkable for the nonproliferous character of. the fronds, which, considering, first, its near relation to the Asplenia, so many of which are profusely proliferous, and, secondly, the protean nature of the family itself, DEVELOPMENT IN THE LADY-FERN. 357 is a singular fact. The discovery, however, of numerous pro- liferous buds which appeared upon some very small plants, which I exhibited here in 1882, led me to institute further inquiries into this subject. I then ascertained that Mr. Mapplebeck had already observed the same phenomenon, and raised plants from similar bulbils, which appeared identical in position and character with those of the Asplenia. Last year, as already remarked, I found another and very distinct form of proliferation on a mature plant of A. F.-f. plumosum divaricatum, upon which numerous bulbils were evolved in the place of the sori; this, be it observed, being on the underside of the pinne, a most unlikely place for such growths. This same transformation of the reproductive energy had already been observed on three other kindred forms of Athyrium, upon one of which the bulbils and sori were scattered almost indiscriminately over the back of the fronds, some of the sori seeming to be in an intermediate amorphous condition ; though in all other cases, so far as I could see, the sori and bulbils were distinctly differentiated by the presence in the former case of an indusium, and in the latter cf lanceolate scales arranged shuttlecock fashion around the bulbil, no trace of indusium existing. Such bulbils had, until this season, failed invariably to yield plants, and seemed incapable of forming a proper axis of growth. Mr. G. B. Wollaston has, however, succeeded in ob- taining plants this spring from 4. F-f. plumosum elegans, and one or two of those from 4. .F..f. plumosum divaricatum have developed fresh fronds with me. D. From this it will be seen that no less than three distinct forms of proliferation have now been observed on the Athyria. 1. Bulbils of the ordinary character developed in the axils and on the superior surface of the pinne, and agreeing 1m character with the ordinary bulbils of the Asplenca. 2. Bulbils formed apparently by transmuted spore-producing i. e. on the underside of e ‘upying the place of sori e ama nain A ; quite unrecorded in the pinnæ—a position so far, I believe, connection with any of the Filices. l " 3. A new form of proliferation altogether, viz. proli erous prothalli arising from pseudobulbils produced bya different trans: mutation of the reproductive force, and evolving plants only a er the prothalli have produced the usual sexual organs common prothalli resulting from spores. —. e 358 MR. C. T. DRUERY ON A SINGULAR MODE OF Further Notes on a singular Mode of Reproduction in Ath rium Filix-foemina, var. clarissima. By Cuares T. DRUÉRY. (Communicated by Dr. J. Murr, F.L.S.) [Read 20th November, 1884.] AT a meeting of the Linnean Society in June last I had the honour of bringing before your notice a record of certain pheno- mena which I had observed during the past winter in connection with the reproduction of a form of Athyrium Filix-foemina through prothalli, which were not produced from spores but from certain excrescences evolved in their stead upon the under surface of the pinne. The Athyrium in question, which was found wild in Devon, had been for many years reputed barren, the fructification which appeared copiously yielding no perfect spores, the result being that after a long period only two plants existed, the original plant having permitted but one division. In 1883 one of these plants, which had been grown under cover, was observed to produce upon the inferior surface of the pinne a large number of curious excrescences consisting of pear-shaped bulbilloid growths, attached firmly to the frond by their thicker extremities, and seated in every case within indusia, thus occu- pying the place of sporangia, to which, however, they bore no resemblance whatever. Mr. G. B. Wollaston, whose attention Fig. 1. F ig. 1. Pyriform pseudobulbils or abnormal sporoid growths of 1883, enlarged. Fig. 2. a, clavate abnormal sporoid growths of 1884 ; b, ragged indusium. was drawn to them by the previous discovery of bulbils proper upon other Athyria in the same year, which bulbils also occu- pied the place of sori, was of opinion that they were also bulbils. However, on comparing them with the bulbils produced on these other Athyria, I was struck by the fact that, while in the other cases the bulbils were seated in the centre of scales arranged shuttlecock fashion around them, in this case indusia were present REPRODUCTION IN ATHYRIUM FILIX F(MINA. 359 instead, which led me to look upon them as sporoid growths of a character essentially different from the bulbils common to many Ferns. I consequently laid down a number of pinne, with the result that I read before you in June, viz. the production of perfect prothalli by the bifurcation of the points of the’ pear- shaped pseudobulbils, which prothalli eventually developed arche- gonia and antheridia, and finally yielded plants of the same type as the parent. At the meeting in June I could do no more than lay the con- secutive record of my observations before you, since all traces of the preliminary stages had vanished when the young plants appeared, and these were then too diminutive for their character to be determined; they also afforded no evidence whatever that they had originated in other than the usual way, viz. from spores, and I consequently did not exhibit them. To-night, however, I have pleasure in exhibiting some of the plants produced as I have described. I had hoped, too, to be able to bring pinne bearing pseudobulbils as described and sketched by me last year ; but owing, partly, I believe, to the long dry summer, and partly to the fact that the parent plant (which is not under my control) was placed out of doors for a time, I am only able to produce pinne showing the fructification in a very immature state—not so immature, however, but that they afford ample evidence of abnormality. u To any one accustomed to deal with Athyrium Filiz-femina, the first glance will strike one with surprise at the presence of fresh green unripe fructification with, in most cases, unlifted indusia upon a deciduous fern in November, months after the time when Sporangia proper have ripened and scattered their spores, and when the indusia are usually in a ruinous and fragmentary state. Here and there the indusia on the pinne exhibited will be seen to be lifted and to partially disclose a number of curious clube shaped and occasionally. serpentine cellular masses which, thoug very different from the swollen’ pear-shaped bodies of last 3 nae differ as widely from embryo sori, showing no signs whatev er o annulation or of the symmetry which would characterize imma ture sporangia when sufficiently advanced to protrude from the indusium. While, however, the pear-shaped pseudobulbils are conspicuous by their absence; it will be seen that some of te club-shaped excrescences-are. larger than others. . Goo ble general appearance, I believe that, given a more favourable 360 MR. F. O. BOWER ON APOSPORY IN FERNS. season, some few would assume predominance and form the pear- shaped pseudobulbils at the expense of the weaker growths, which would abort, as in many analogous cases. I incline the more strongly to this opinion, as among the bases of the pear- shaped bodies produced last year there were numerous thin, thready, and shapeless growths, exactly such as would be likely to originate in such a way. My present object being to confirm as far as possible the data I gave in June, I would call your attention, first, to the existence of the young plants upon the table, raised as described ; and, secondly, to the manifestly non-soriferous form of fructifi- cation borne by the parent plant, an examination of which will, I think, go far to convince you that its offspring are engendered neither through spores nor by bulbils, but by some other mode of reproduction—a mode which, from constant and careful watch- ing through all its stages, I believe to be one so far unrecorded in connection with any other Fern, viz., through prothalli produced, not from spores, but by direct bud-growth from the parent frond. On Apospory in Ferns (with special reference to Mr. Charles T. Druery's Observations). By F. O. Bower, M.A., F.L.S. [Read 18th December, 1884.] (Pirates XI. & XII.) Mr. C. T. Drurry has already drawn the attention of the Society in two successive papers to Athyrium Filix-feemina, var. clarissima, ascribing to that plant a mode of transition from the sporophore generation (or Fern-plant), to the oophore (or pro- thallus), without the intervention of spores. He has pursued the subject with suecess, as far as it is possible without sub- jecting the matter to a detailed microscopical investigation. We are indebted to this observer not only for the communi- cations already received from him, but also for his generosity in supplying to the Royal Gardens at Kew material fitted for the more detailed microscopical analysis of the process. Without further recapitulation of Mr. Druery's results, I may at once proceed briefly to describe the observations which I have made on the cultures now in progress at Kew. Many minute details will be deferred for the present till the investigation is com- pleted; the chief results are, however, of such importance as to justify a preliminary notice of them. MR. F. O. BOWER ON APOSPORY IN FERNS. 361 The sori in Athyrium Filix-feemina, var. clarissima, appear in the normal position with a normal indusium. In the condition in which the specimens were when first I received them (Nov. 29), the large majority of the sporangia presented an abnormal appearance. Some few appeared of nearly normal structure, with an annulus, but were arrested at a point of development before the formation of the spores (Plate XI. fig. 1) ; others, and indeed the majority of them, showed more or less distinctly the central archespore, together with the cells which would normally form the wall of the sporangium; but there the normal develop- ment seemed to have been suddenly arrested—the archespore had not in these cases divided further to form either the tapetum or the mother-cells of the spores (Plate XI. fig. 2). The arrest of development of the archespore is, however, compensated in these cases by the more active vegetative development of the stalk of the sporangium and of the superficial cells of the head ; the result being that the arrested sporangium ultimately appears as a club- shaped body of larger size than the normal sporangium. The individual cells also are of larger size; they contain numerous chlorephyll-granules which, if present at all in normal sporangia, are relatively few in number. Further, the more rudimentary the head of the sporangium is, the more marked is the vege- tative development of the remaining parts. If pinnules showing the above characters be subjected to favourable conditions of heat and moisture, the vegetative deve- lopment of the sporangia may proceed at once. On pinnules laid on damp soil, and forced quickly in the propagating-pits at Kew, the earlier stages of this further development have been traced. The details show great irregularity ; and they are found to correspond to the greater or less completeness of arrest in the normal development of the sporangium. Thus sporangia Which show a clearly marked annulus do not usually assume any further vegetative activity ; those, however, which are arrested at an earlier stage in their normal development may produee, by a purely vegetative process, outgrowths of very irr egular form. Sometimes all the superficial cells of the club-shaped body may take part in the process, and this is most clearly seen where the arrest of the normal development is most complete (Plate XI. figs.3,5,6). In other cases the head of the arrested sporangium may be thrown off, while the stalk continues its vegetative growth (Plate XI. fig. 3). The result is in either case the formation 362 MR F. O. BOWER ON APOSPORY IN FERNS. of flattened structures, consisting only of parenchymatous, chlo- rophyll-containing cells, which, sooner or later, show at one or more points on their margins that growth with a wedge-shaped apical cell which is well known as characteristic of the Fern prothallus ; root-hairs are at the same time formed by the out- growth of individual cells (Plate XI. figs. 6,7). None of my cultures have as yet formed sexual organs: this we must wait for; but meanwhile it may be remarked that Mr. Druery’s obser- vations during the last two years show that, in the cases observed by him, sexual organs were formed, and young Fern-plants were produced by them. In any case, however, the above observa- tions show that in the Fern in question there is a transition from the sporophore generation to a structure of a prothalloid nature, without the intervention of spores, and that it is formed by a process of purely vegetative growth from more or less reduced sporangia. Finally, it may be stated that my observations do not exclude the possibility of a formation of such structures by a vegetative outgrowth of the base of the sorus itself; whether this actually occurs must be decided by further investigation *. Through Mr. Druery I have Mr. G. B. Wollaston’s permission to mention a still more interesting example of apospory, of which the latter gentleman is the discoverer, viz. Polystichum angulare, var. pulcherrimum. Here flattened organs, of undoubted prothal- loid nature, are formed by simple vegetative outgrowth of the tips of the pinnules, and without any connection with sori or spo- rangia. At first the tip of the pinnule merely extends, so as to form a flattened expansion, one layer of cells in thickness, and with a very irregular margin; while the whole tip curves down- wards, and often forms a spiral coil of one, or one and a half turns, closely covered above by the more expanded portion of the pin- nule. In other cases the outgrowth may assume very irregular forms. Ultimately the characteristic marginal growth begins at some point, sometimes terminal, but more frequently lateral. The details of this development have not yet been fully investigated. The result, however, is the formation of a flattened, often heart- shaped expansion, with a thickened cushion similar in structure to that found in normal prothalli; it bears antheridia and arche- gonia of normal structure; sometimes, as occurs also in ordinary prothalli, they are found both on the lower and upper surfaces. * Since the above was written the cultures at Kew have progressed so that prothalli bearing archegonia and antheridia may be seen, still connected at their bases with the pinnule of the parent plant (Pl. XII. fig. 8).—Feb. 11, 1885. MR. F. O. BOWER ON APOSPORY IN FERNS. 363 A point worthy of note is, that in those prothalloid structures which are taken directly from living leaves in the normal position without further cultivation, the antheridia and arehegonia have not opened; no doubt this is due to the want of water, which, as is well known, is necessary forthis process. I have not yet been able to observe the formation of youug Fern-plants on these prothalli; but it may be hoped that from cultures now in pro- gress at Kew and elsewhere further evidence may be obtained on this point. This Polystichum is thus an example of the formation of an expansion of undoubted prothalloid nature, bearing sexual organs by a process of purely vegetative outgrowth from the Fern-plant ; that is, there is a transition in this case from the sporophore generation to the oophore by a vegetative growth, and without any connection either with spores or, indeed, with sporangia or sori. It may be regarded as a still more complete example of apospory than that of Athyrium Filiz-feemina, var. clarissima. It remains to point out the bearing of these observations upon the general life-history of the Fern. The normal cycle of life of these plants may be represented graphically thus :— Diag. 1. Spore Sporangium Prothatllus or Oophore DOTO, hore SLO an / sexual OVJANS We already know various modifications of this simple cycle, which may be classified under two heads :— l. Modifications which lengthen the cycle. 2. Those which reduce it, and may therefore be regarded as short cuts. l Taking first those of the first category, they may be described as modes of vegetative reproduction, by which one generation or the other simply reproduces itself. Thus the sporophore, or Fern-plant, may form adventitious buds, which are found seated 364 MR. F. O. BOWER ON APOSPORY IN FERNS. at various points on the leaves; these may be ultimately sepa- rated from the parent plant as new individuals of the sporophore generation: as examples may be cited various species of Asple- nium. In the case of the prothallus or oophore generation, a vegetative mode of reproduction by adventitious buds, resulting in the formation of fresh prothalli, has been observed in numerous cases, e. g. Osmunda, Gymnogramme leptophylla, &c.* ; and more especially in that form of prothallus described by Cramer, but not identified with certainty: here special gemme carried on the process of vegetative reproduction of the prothallus t. Thus in each of the alternating generations of the Fern the cycle is liable in certain cases to be extended by processes of vegetative reproduction; this may be represented graphically, as in the subjoined diagram :— Diag. 2. S} ' . : "pore Sporangium Yann sous Vegetative if . Adventite Réprodecctiow \ Buds A \ | Prothatlus Sporoph ore or Oophore or Fern Plant 3j n Y Vegetative Adventitious Reproduction Sexual Organs P uds Taking now into consideration those modifications of the life- history by which the cycle may be reduced, we may find two points at which a short cut may be taken :—there may be a transition from the prothallus to the Fern-plant without the intervention of sexual organs ; secondly, there may be a direct transition from the Fern-plant to the prothallus without the intervention of spores. The former of these short cuts has already been observed: first by Farlow, in Pteris cretica; and it has been described at length by De Bary f in this plant, and also in other Ferns (Asp?- * Goebel, ‘‘ Entw. des Prothalli » i 1877 ; also Goebel, ‘ Grandzüge dec Syetomatihe QT PY BLE t Cramer, “ Ueber die geschlechtlose Vermehrung des Farnprothalliums,” Sep.-Abdr. Denksehr. d. Schweitz. naturforsch. Gesellach. Bd. xxviii., 1880. fino Quart. Journ. Mier. Sci. 1874, p. 200; De Bary, Bot. Zeit. 1878, p. 449. m ' MR. F. O. BOWER ON APOSPORY IN FERNS. 365 dium Filix-mas, var. cristatum, Aspidiun falcatum); and by Sadebeck * in Todea africana. The phenomenon now passes under the name of apogamy. The prothalli in the cases named produce either constantly or frequently, by direct budding, new Fern- plants, without the sexual organs (which may or may not be developed) taking any direct part in their origin. This may be represented graphically thus :— Diag. 3. Spore dporopiore eler Plant no observations bearing upon the converse case, viz. that of the excision of the spore from the cycle of life of the Fern, had been published, though the phenomenon had been artificially induced in certain Mosses. We now see in the case of Athyrium Filiz- Jæmina, var. clarissima, that the spores are not developed, but that there is a direct passage by a purely vegetative growth from the sporophore generation to the oophore ; and that in this plant the prothallus originates, as a rule, from the arrested sporan- gium. This may be graphically represented thus :— Diag. 4. "4 Apospory Wporanguum. Prothallts Fern Plant Sexual Organs * Schenk’s ‘ Handbuch.’ Fd. i. p. 234. 366 MR. F. O. BOWER ON APOSPORY IN FERNS. In the case of Polystichum angulare, var. pulcherrimum, there is a more extensive excision from the cycle. Here not only is the formation of the prothallus independent of the formation of spores, but it even originates quite apart from the sporangium and sorus by a purely vegetative outgrowth of the tips of the pinnules ; this may be represented graphically thus :— Diag. 5. Spore Sporangium / Apospory Prothallus Lolystichusm For Plant Sexual rgans Lastly, it remains to compare this new phenomenon of apo- spory in the Ferns with similar cases in other plants. In 1876 it was shown by Pringsheim*, and also by Stahl +, that if the stalk of the sporogonium of certain Mosses (Hypnum cupressi- Jorme, Amblystegium serpens, Bryum cespiticium, Ceratodon pur- pureus) be cut into short pieces and cultivated on damp soil, protonemal filaments spring from single cells, and ultimately produce moss-plants of the normal type. There is thus artifi- cially induced, by the prevention of the formation of spores, & direct transition from the sporophore generation to the oophore by a purely vegetative process. If we discount the artificial character of this phenomenon, it may be closely compared with the examples of apospory in Ferns, above described. A second example, but one which does not correspond exactly with the above, is that vegetative budding described by Goebel ł as occurring in plants of Isoëtes lacustris and I. echinospora. n these species a bud frequently takes the place of the sporangium, and is seated in the fovea at the base of the leaf. The bud develops as a normal shoot of the sporophore generation. lt may be concluded that in this ease there is an excision from the cycle * Monatsber. d. Akad. d. Wiss. in Derlin, 1876. t Bot. Zeit. 1876, p. 689. i Bot. Zeit. 1879, p. 1. MR. F. O. BOWER ON APOSPORY IN FERNS. 367 not only of the sporangium and spores, but also of the whole oophore generation. This may be represented graphically thus :— . Diag. 6. Isoetes Plant or Sporophore It is clear that in many points this most interesting process of aposporous reproduction in the Ferns requires further inves- tigation. It is confidently hoped that the cultures now pro- gressing in Kew, and in the private collections of Mr. Druery and others, will clear up many uncertain points. It may also be expected that, since attention is now called to the subject, other examples of the same or similar phenomena may be dis- covered which had hitherto escaped notice. DESCRIPTION OF THE PLATES. Prate XI. Athyrium Filix-femina, var. clarissima. Fig. 1. An arrested sporangium with annulus already formed, but the cell-wal)s not thickened ; the cells of the head of the sporangium have almost lost their contents, while those of the stalk (s¢) have abundant proto- plasm and chlorophyll-granules. 3825. Fig. 2. An arrested sporangium, of which the archespore (a) has not undergone division The stalk (s/) is abnormaily enlarged, and its cells have abundant protoplasm and chJorophyll. X325. Fig. 3. A sporangium, similar to fig. 2, after cultivation on moist soil for seven days. The shaded head of the sporangium does not grow, but active cell-division and growth has gone on in the stalk (st). X 325. Fig. 4. A similar sporangium with rather more enlarged head, showing super- ficial cells. X 325. Fig. 5. A similar sporangium, showing irregular growth in the stalk. Fig. 6. A sporangium after continued cultivation for twelve days; the stalk is much enlarged, and growth, with a wedge-shaped apical cell, has begun laterally from it. x13). x 825, 368 MR. C. B. PLOWRIGHT ON THE REPRODUCTION Fig. 7. Flattened expansion produced by cultivation from an arrested sporan- gium; growth, with a wedge-shaped apical cell, appears to be pro- gressing at more than one point. X 130. Piare XII. Fig. 8. Part of a pinnule of Athyrium F.-f. clarissima, with vascular bundles (vb) and one sorus. Cultivation for five weeks has resulted in the development of prothalli from some of the arrested sporangia; these prothalli bear antheridia and archegonia, but are still attached to the pinnule which bore them. X40. Polystichum angulare, var. pulcherrimum. Fig. 9. Apex of a pinnule which has grown out into a flattened expansion one layer of cells in thickness; this has not, however, as yet the distinc- tive characters of a prothallus. x 20. Fig. 10. Apex of another pinnule, which has grown out into a flattened expan- sion of considerable size; this expansion has the characteristic struc- ture of a prothallus, with marginal glandular hairs and thickened cushion ; the latter bears on its under surface organs which proved, on cutting sections of the prothallus, to be mature antheridia and young archegonia. x20. Fig. ll. One mature antheridium from a section of the prothallus shown in fig. 10; it has not yet ruptured. x325. Fig. 12. One archegonium from a section of another specimen; it bas passed the period of maturity without the neck having ruptured ; the egg-cell has become disorganized. x325. Remarks on the Reproductjén of the Heteræcious Uredines. By Cnmanrrs B. Prowdicut. (Communicated by W. T. THIsELTON Dyer, C.M.G., M.A., E.R.S., E.L.S.) [Read 6th November, 1884.] Tnx object of the present note is to draw attention to a point concerning the reproduction of the Heterecious Uredines when this takes place without the intervention of the scidiospores, which bas apparently hitherto escaped notice. This simply is that, when the heterzcious species reproduce themselves without passing through the zecidial stage, the resulting uredospores are far more abundant than in the case when they arise from the implantation upon the host-plant of the eecidiospores. My attention was first drawn to this in the summer of 1883, by some specimens of wheat received from two districts in OF THE HETERACIOUS UREDINES. 369 Australia, namely from New South Wales, communicated by Mr. A. H. Sampson, and from Queensland, communicated by Mr. James Sewell. Recently Professor Custance has kindly forwarded an extensive series of specimens, illustrating the same fact, from South Australia. These specimens were sent in com- pliance with a request made to these gentlemen for them. This arose from the pointed manner in which Australian farmers, in their letters to agricultural papers, complained of “rust” as injuring their wheat-crops; not, as our English farmers do, of “mildew.” The fungus, however, is identical in both countries (Puccinia graminis); only in Australia the barberry, if it occurs at all, does so to a very limited extent in gardens &c. Not only are the uredospores less abundant in English specimens, but the teleutospores are produced at a much earlier period in life of the parasite. A second instance is afforded by Puccinia rubigo-vera (DC.) in the eastern counties of England. Here we have an enor- mous development of the uredospores as compared with that of the teleutospores. The ZEcidium occurs on certain of the Boragi- nacee, but must be exceedingly rare in this district, for although most careful search has been made for it annually for the past ten years, I have never yet gathered a fresh specimen. This is because the teleutospores are one of the earliest to germinate ; they do so, in fact, before the foliage of the ecidial host-plant is, as a general rule, above ground, and in a condition to afford them a nidus. The ZEcidium is most frequently found upon Lycopsis arvensis, an annual which, up to the present time (May 7), has not appeared above ground in the open, but since the end of March the wheat-crop has been severely affected by the uredo- Spores of the fungus in question. A third instance is afforded by Puccinia obscura. communication to the Linnean Society, I showed that this fungus has its eecidiospores upon Bellis perennis. Professor Farlow, on learning this fact, at once wrote to me to the effect that the Puccinia occurs with him in Massachusetts, but that the æcidial host-plant does not, and is only very rarely cultivated there in gardens. He was kind enough to send me specimens of the Puccinia, which were accompanied by a far more profuse develop- ment of uredospores than 1 have ever seen associated with our English specimens. Since the above observations were made, I have received from 2D In a recent LINN. JOURN.—BOTANY, VOL. XXI. 370 MR. W. T. THISELTON DYER'S REPORT Mr. Rostrup, the eminent Danish mycologist, his paper upon * Hetereciske Uredineer'' (1884). In this I am pleased to see the same fact has been noticed by him, but with different fun gi. First with Coleosporium senecionis, which, when it occurs in dis- tricts where fir-trees are absent, consists almost wholly of uredo- spores; and, secondly, with Chrysomuza ledi. He received from Greenland a specimen of Ledum palustre, upon which the uredo- spores of this fungus were present. Now De Bary has shown that æcidiospores of this Chrysomyxa occur upon Picea eacelsa, à tree whieh does not grow in Greenland. Report on the Botany of Mr. H. O. Forbes's Expedition to Timor-Laut, by W. T. Tuisetron Dyer, F.R.S., Secretary to the British-Association Committee for the Exploration of the island; with a List of determinations of the Plants collected, by Prof. Oxrv#R, F.R.S. [Read 6th November, 1884.] Tue Timor-Laut or Tenimber Islands are a small archipelago situ- ated to the north of Australia, about halfway between the island of Timor and the Aru Islands. The natural history has hitherto been almost entirely unknown. In 1882 Mr. H. O. Forbes, aided by grants from the British Association, was enabled to spend the months of July, August, and September in the group. He landed at the islet ot Larat, which lies off the north-east coast of Yamdena (as the northern of the two portions of Timor- Laut is named), at a distance of about 15 minutes’ sail. Of the physical characters of the group Mr. Forbes gives the following aecount :— “The Tenimber Islands, as seen from the sea, are very low. There are no hills ; nothing over 400 feet on the northern island, nor on the surrounding islets, with the exception of Laibobar on the west coast of Yamdena, which rises to a height of about 1500 feet as seen from Larat across the mainland. The Tenimber group is surrounded (as I am told by the Commander of the Dutch man-of- War ‘ Samarang ’) by a very deep sea. The islands are entirely of coral formation. On the eastern shore of Yamdena there are coral-cliffs of about 100 feet in height, from which immense stalactites hang down. Along the beach are here and there blocks of tideworn sandstone ; but nowhere have ON THE BOTANY OF TIMOR-LAUT. 371 I been able to find any sedimentary rocks. In the interior of Yamdena the coral lies a few inches below the surface, being covered only by a very thin layer of dark mould. There are absolutely no traees of sedimentary strata, with the exception of one small nodule of a fine calcareous limestone. Along the shore low coral-cliffs alternate with sandy baylets (the sand is almost entirely of fine particles of coral and minute shells and broken fragments of Echini &c.), which are studded also with worn coral boulders.” Mr. Forbes also furnished to the Committee of the British Association the following account of the general characters of the vegetation of the group as far as he was able to examine it :— “Of plants my collection is not so large as could be desired, owing to my inability to obtain any one willing to assist in felling or to climb trees, as I have always been able to do in Sumatra. The season was also unfavourable, it being during our stay the height of the dry season; besides I had the misfortune to have my drying-house, with 300 to 500 specimens, burnt to the ground, a conflagration which almost involved my house and endangered the village also.. The forest is not very tall, few trees reaching even 80 feet, and has little undergrowth; but there is a thick sward of Commelyna. On the coral-rocks by the seashore several species of orchids grow, of which I have obtained living specimens. On trees at Cape Watoe Sianga, in Larat, where it faces the wind entering from the southern sea, I found a handsome and, I imagine, a new species of orchid; but I have been able only to see half-shrivelled flowers. At few other places have I seen any orchids; none on the mainland. Ebony, from the accounts of the natives (who make much use of it in their carved work), is found in considerable abundance. A very few species of ferns and two species of Lycopodiacew were found. A great species of Sterculia with dark lake flowers, which appear before the leaves, is one of the most conspicuous objects in the vegetation. Dammar trees occur; but the natives collect only to make lamps for themselves. The largest trees are figs of the genera Urostigma and Ficus; but, though abundant, they are few in species. Artocarpus incisifolia (but not the true bread-fruit tree) is abundant; but no drtocarpus integrifolia was met with, Leguminosæ and Myrtacee occur in considerable numbers, also low Composite and shrubby Apocynaceæ. Mingling in the belt of Rhizophora, where it is somewhat rocky, grows a species of 202 372 MR. W. T. THISELTON DYER'S REPORT Pandanus. Of Balanophora one species was obtained, and a pretty species of Aristolochiacee ; but both went in the confla- gration. I have seen in the hands of the natives a species of wild mangosteen ; the capsules were very thick, with almost no fleshy pulp. Tho natives eat the former, and throw away the latter. Of palms, only cocos, Areca, Borassus, and Sago were seen ; neither Eucalyptus, phyllode-bearing Acacias, Melaleuca, Casuarinas, nor Melastomacee were met with, and no sandal- wood. * Timor-Laut seems, from our present rough survey, to have great affinities with the Moluccan (Amboina) region ; perhaps more than with the Timor group. The Insecta seem very closely to resemble those of Amboina; but the Lepidoptera and Coleoptera are excessively few.” Mr. Forbes’s botanical collections (in the absence of more definite localities) were presumably made in Larat and Yamdena. They have not proved, on examination, so interesting as the zoological specimens brought back by him. But this may, no doubt, be attributed in great measure to the unfortunate accident by which so large a proportion of his dried plants appears to have been destroyed by fire. Still, in the face of our total ignorance of the nature of the flora of the group, it seems desirable to put on record a list of the species actually determined, even if it must be admitted that the general facies of the vegetation they indi- cate is somewhat common-place. A more thorough investigation of the forest of the interior than Mr. Forbes could perhaps afford time for would very probably yield more interesting results. The determinations in the following list of species have been made by Prof. Oliver, F.R.S., who has added a few notes on such species as appear of more special interest. One of these is closely allied to a Queensland plant, and another belongs to a genus hitherto regarded as endemic in New Caledonia. As far, there- fore, as the gathering indicates any distinctive peculiarities in the flora, they are with countries to the west of the Malayan archi- pelago. It is noteworthy in this connection that Mr. Sclater finds the affinities of the avifauna to be preeminently Papuan, with only a slight element from Timor*.—(W. T. T. D.] * Proc. Zool. Soc. Feb. 20, 1883, p. 56. ON THE BOTANY OF TIMOR-LAUT. Amamirta Cocculus, Wight et Arn. (In fruit.) Portulaca oleracea, L. Ochrocarpus ovalifolius, 7. And.? (Calysaccion obo- vale, Miq.) Sida humilis, Willd., var. repens. Abutilon indicum, Don. Hibiscus surattensis, Z. Thespesia populnea, Corr. Sterculia footida, L. Melochia velutina, Bedd., var. glabrata. Micromelum pubescens, Blume. Murraya exotica, Z., var. Glycosmis pentaphylla, Corr., or possibly G. sapindoides, but flowers too young. Tristellateia australasica, A. Rich. Owenia sp., may be O. cerasi- folia, F. Muell.; but in fruit only. Carapa,anC.moluccensis,Zam. ? Too young for identification. Strombosia ? Fruit only. Erioglossum edule, Blume, forma. Flemingia strobilifera, R. Br. Desmodium umbellatum, DO. Pongamia glabra, Vent. Mucuna (§ Stizolobium) sp. Canavalia obtusifolia, DC. ? Vigna lutea, 4. Gr. Dolichos Lablab, Savi? Cynometra ramiflora, L. (C. bijuga, Span.) Cassia javanica, L. ? Cesalpinia pulcherrima, Sw. Bauhinia Blancoi, Benth. ? 373 Peltophorum ferrugineum, Benth. | Bruguiera caryophylloides, Blume. Eugenia javanica, Lam. ? Momordica Charantia, Z. Zehueria, aff. Z. mucronate, Miq. Delarbrea sp.? Interesting ; a New-Caledonian genus. Randia sp. Ixora sp. Ixora? Fruit only. Ixora, aff: I timorensi, Dene. Psychotria ? sp. Vernonia cinerea, Less. Weddelia biflora, DO. Mesa, sp. Allied to a species from the Indian archipelago. Diospyros maritima, Blume. Tabernemontana parviflora, Dene. (T. orientalis, R. Br.) Marsdenia? Fruit only. Gymnema? vel Sarcolobus ? sp. Cordia subcordata, Lam. Tournefortia sarmentosa, Lam. Hewittia bicolor, Wight et Arn. Solanum verbascifolium, L. Capsicum frutescens, L., forma. Acanthus ilicifolius, L. (Dilivaria ilicifolia, Juss.) Hypoestes floribunda, R. Br., var. Eranthemum, sp. Asystasia, an À. chelonioides, Nees? Premna obtusifolia, R. Br. Vitex trifolia, L. Form very near V. Negundo. | Leucas decemdentata, Sm. | Deeringia celosioides, R. Br. ——————————— 374 MR. E. M. HOLMES ON CINCHONA LEDGERIANA. ZErua scandens, Wall. (Æ. ve- determinable from the lutina, Mog.) material.) Piper, aff. P. canino, Dietr. Fatonia pilosa, Gaud. (F.lan- Myristica, cf. M. insipida, R. ceolata, Dene.) Br. Lycopodium carinatum, Desv. Loranthus (§ Dendrophthoe), Phlegmaria, L. sp., aff. L. rigido, Wall.? Polypodium irioides, Lam. Ficus, off. F. acanthophylle, Pteris tripartita, Sw. Miq. Asplenium falcatum, Lam. ——— spp. (Three species in- | Vittaria elongata, Sw. This collection, so far as it goes, is made up in great part of the more widely diffused species of the Indian archipelago. The most interesting plants appear to be :— A plant, in fruit only, which I would refer to the Meliaceous genus Owenia, perhaps indeed to O. cerasifera, F. Muell., of Queensland. A fine Mucuna in fruit, of the section Stizolobium, which I have not identified. A Delarbrea, an Araliaceous genus hitherto only received from New Caledonia. A plant, in fruit only, which may be something new, though possibly a Strombosia (Olacineæ).—[D. O.] Remarks on Cinchona Ledgeriany as a Species. By Epwarp Moretti Hormtes, F.L.S. [Read 20th November, 1884.] Tur name Cinchona Ledgeriana appears to have been first used in Cinchona plantations in the East Indies, to distinguish the trees grown from seed collected in the northern portion of Bolivia by an Indian servant of Mr. Ledger’s, and which was subsequently distributed to Java, various plantations in India, and Ceylon. When the plants flowered, Mr. J. E. Howard figured in his magnificent work ‘The Quinology of the East-Indian Planta- tions’ three forms of Cinchona which he had received from Java, as the produce of Ledger’s seedlings, under the name of Cinchona Calisaya var. Ledgeriana, and gave a brief botanical description of the plant, by Dr. Weddell, in the accompanying text. MR. E. M. HOLMES ON CINCHONA LEDGERIANA. 375 These illustrations represented respectively the male, female, and neutral forms of the plant. N otwithstanding the publica- tion of these excellent coloured plates, there seems to have been ever since a considerable doubt among planters as to the cha- racters by which the Cinchona Calisaya var. Ledgeriana might be recognized. In consequence of this difficulty, Dr. Trimen published in the ‘Journal of Botany’ for Nov. 1881 figures and a description of what he considered to be the typical plant, and erected it into à species under the name of Cinchona Ledgeriana, Moens. To this description Mr. Howard objected that the plant figured did not correspond with what he considered to be the typical Ledgeriana plant, as described by himself, and expressed the opinion, judging from the illustration alone, that Dr. Trimen’s plant might be C. micrantha var. calisayoides. He also came to the conclusion, without seeing specimens of the tree, that the Ledgeriana described by Mr. T. N. Christie, of Ceylon, was probably C. Calisaya var. microcarpa of Weddell; but that those grown on the Yarrow Estate in Ceylon, by Mr. Laurie, were the true plant as described hy himself under the name of C. Calisaya var. Ledgeriana, Howard. Dr. Trimen, however, states positively that the three plants alluded to were all raised from the same small quantity of seed in the same nursery beds, and at the same time, and that they are all positively identical. For my own part, I should have been content to have accepted the statement of either authority as final; but having received for the Museum of the Pharmaceutical Society three specimens of Cinchona-bark labelled * Ledgeriana," one of which was sent from Darjeeling by Dr. King, another from Ceylon by Mr. T. N. Christie, through the Planters' Association there, and a third from Java presented by Mr. Howard himself ; and finding that all three differed in their external characteristics, only the Darjeeling one presenting the typical characteristics of Calisaya bark, I could come to no other conclusion than that several different varieties or forms, one or more of which are probably hybrids, are now grown in plantations under the name of Cinchona vic id Dr. Trimen, I believe, holds that the bark of Cinchona- irees does not present sufficient character for determination of species or affinity. My experience, however, on this point accords better with that of Mr. Howard, viz. that each species 876 MR. E. M. HOLMES ON CINCHONA LEDGERIANA. when mature presents a bark distinguishable both by external and internal eharaeters, and that hybrids generally give some indication of the species to which they belong by the characters of the bark. My grounds for this belief are the following. Specimens of the different varieties of Calisaya from wild trees and from cul- tivated trees in Bolivia, and from cultivated trees in Darjeeling, can be easily recognized as belonging to one type; and the same holds good with the typical forms of C. officinalis and C. succi- rubra *. As a further illustration that the bark of trees of nearly allied species is easily distinguishable, I may appeal to a specimen of the flowering Ash, Fraxinus Ornus, L., now in the Botanic Gardens at Regent’s Park, which was grafted many years ago on the trunk of the common Fraxinus excelsior, L. Here the difference in the two barks, above and below the line of juncture, is easily recognized. The chemical analysis of Cinchona-bark also gives some clue to the species from which it has been derived, in the relative quantities and character of the alkaloids and colouring-matters prevalent in it. So far as I am able to judge from the point of view of the physical characters of the bark, I am prepared to state positively that the Cinchona Ledgeriana from Darjeeling is undoubtedly that of a form of Cinchona Calisaya; and that in my opinion the specimen sent by Mr. Christie isa hybrid, apparently between C. Calisaya and C. officinalis, and that the one I received from Mr. Howard approaches more nearly in appearance to the Cali- saya type than any other, although bearing some traces of hybridization with C. officinalis. The history of the collection and distribution of Ledger's seeds also seems to support the view that the Cinchona Ledgeriana now in cultivation is not one well-marked species or variety, but embraces several varieties, some of which may be hybrids. Thus Mr. Clements Markham, in his interesting work on Peruvian Bark (p. 214), distinctly states that the seeds were collected from about fifty trees. It is hardly to be supposed that these trees, in a district where the tree abounds, and of a species so variable as Cinchona Calisaya, would all consist of * Some specimens were placed on the table to illustrate this point. MR. E. M. HOLMES ON CINCHONA LEDGERIANA. . 377 exactly the same form or variety *. The instructions given by Mr. Ledger to the Indian, on his return with the seeds, viz. to obtain more seeds of the roja, morada, and naranjada varieties of the Calisaya, indicate that these were the forms that he would have endeavoured to collect. Indeed Mr. Van Gorkom states that the Indian assured Mr. Ledger that the greater part of the seed came from * Roja" trees. Consequently, when these seeds became distributed, the seedlings should have possessed the cha- racters of this and of other varieties of the Calisaya stock. The further history of these seeds indicates that such was the case. Thus, Dr. Trimen remarks (Journ. Bot. Nov. 1881), “ In India the young plants were not distinguished from other yellow bark trees” (p. 322). Again, “ The progeny of the original seeds Shows a good deal of variation." “The upper surface usually has a velvety sheen or reflect" (p. 324). Again, Mr. Howard describes the first plants he received from Mr. Moens as a variety of C. Calisaya, referring them at first to the var. microcarpa. He afterwards (Quin. E. I. Plant. p. 85), however, considered that they possessed sufficient differences in the small size of the flowers and fruits to be made a distinct variety, in which he was supported by the celebrated botanist Dr. Weddell, who, it must be remembered, had himself collected Cinchona Calisaya in its native haunts. Mr. Howard moreover remarks that the micro- Scopical structure of the bark presents very distinctly the Cali- saya type. He had noticed, indeed, among specimens sent from Java, subvarieties differing somewhat in the shape and tint of the flowers and leaves, but presenting no features to separate them as other than varieties of the Calisaya type. He remarked, however, that the bark is proportionately thicker in some of Ledger's plants than in other varieties of Calisaya. Dr. Kuntze (Journ. Bot. 1883, p. 6) speaks of the Ledgeriana of Mungpo as a large shrub with divaricately-panicled inflorescence with slender ramification, like that of C. micrantha, whereas the descendants of Bolivian Ledgeriana in Java and Southern India are trees. Mr. Van Gorkom (p. 92) remarks that the Ledger seeds sown in Java developed into handsome young trees without any sign, at least on a hasty glance, of showing themselves distinct from * Dr. Trimen remarks that there were some very bad trees of quite another type among those from the original sowing (Pharm. Journ. Jan. 19, 1884, p. 578, ftnote). 378 MR. E. M. HOLMES ON CINCHONA LEDGERIANA. the indubitable O. Calisaya, the offspring of Java seed, but that when in 1872 they began to flower, it was observed that the flowers were smaller and of a creamy-white colour. The fruit also gave evidence of difference from the other Calisayas in cultivation. All plants of every other variety, except C. officinalis and C. succirubra, were subsequently turned out of the nursery, these being considered the most important to propagate. (This, how- ever, it will be observed, was not before it had been possible for hybridization to take place.] When the plantations were reduced exclusively to Ledger’s plants, it was noticed that the trees showed innumerable varieties of leaf; but after some experience it was found possible to point out with certainty the individuals : of the common Calisaya which had slipped in, in repairing gaps. Still Van Gorkom admits that “there are many of the older Calisaya trees introduced by Dr. Hasskarl, which do not seem different from the Ledgeriana.” From the preceding remarks it may, I think, be concluded that the seeds supplied by Ledger resulted in a variety of forms of C. Calisaya, but that these were exposed to the chance of hybridization. The probability that hybridization has actually taken place 1 ground upon the following facts: the specimens of bark already alluded to, which exhibit evidence of not being of pure Calisaya type; that Dr. Trimen states that in some plants of Ledgeriana the leaves are as broad as those of C. officinalis, var. Condaminea, so much so that it is not always easy to distinguish the two ; that the bark presents considerable variability in appearance; that the upper surface of the leaves has usually (i. e. not invariably) a velvety sheen; and that the capsules he has seen are never “nearly globular” like those described by Dr. King. Dr. Trimen also states that, both in Sikkim and Java, the Ledgeriana trees have come more true from seed since those species growing in their proximity have been cut down (Journ. Bot. 1881, p. 322). I take it for granted, then, that the Ledyeriana of the planta- tions is not a distinct form, but comprises several varieties of Calisaya as well as certain hybrids, and that the majority of these yield a large amount of quinine, this being the feature which is supposed to decide in doubtful cases (Van Gorkom, p- 93) whether or no a given tree is * Ledgeriana." lt follows, then, that Dr. Trimen has described a species from doubtful materials, since he has taken the characters given, partly from MR. E. M. HOLMES ON CINCHONA LEDGERIANA. 379 growing specimens in Ceylon, from seed obtained by Mr. Mae Ivor from trees which originated from Ledger's seed, and which, as above mentioned, may consequently have undergone hybri- dization with other species (Journ. Bot. 1881), and partly from dried specimens of the original trees from Java. The characters which he has selected from these as distinguishing the Ledgeriana do not, however, present any features which entitle it to be sepa- rated from C. Calisaya as a distinct species. The distinctive features adopted by Dr. Trimen, and empha- sized by italics in his description (Journ. Bot. Nov. 1881), are as follows :— lst. Leaves always having the broadest part at or about the middle. 2nd. Flowers small, drooping, or divaricate. Buds not at all, or very slightly, widened at the end, and never abruptly enlarged there. With regard to these characters, the first accords well with the figure of C. calisaya var. microcarpa of Weddell’s plate (Notes, p. 50), and cannot therefore be used to separate it from the Cali- saya type. The small flowers are also characteristic of Weddell’s C. Calisaya var. pallida, which on this account he would have regarded as a var. of micrantha, with smaller and narrower leaves than the type, were it not for the difference in the fruit. In Dr. Weddell’s type specimens in the Kew Herbarium the flower- buds of the Calisaya are not widened at the apex. I conclude therefore that there is not sufficient evidence pro- duced by Dr. Trimen to show that his Ledgeriana is entitled to specific rank, or is indeed anything more than a variety of C. Calisaya. If, as Dr. Trimen states, and he is confirmed in his statement by Mr. Moens and Mr. Van Gorkom, the tree can be easily recognized in all plantations by the characters he has given, and if it be also characterized by yielding a high per- centage of quinine, it is important that it should receive a di- stinetive name. As, however, Mr. Howard affirmed that the plant described by Dr. Trimen was not identical with the one described by him under the name of Ledgeriana, and as the Ledgeriana bark I have received from Ceylon is certainly not identical with Mr. Howard's bark so named, it would be advan- tageous, I think, that Dr. Trimen's plant should be distinguished as a horticultural form, belonging to the variety pallida of C. Calisaya, and differing from it chiefly in the presence of scrobi- 380 MR. F. C. S. ROPER ON RANUNCULUS LINGUA. cules (a feature which may be used to separate forms, but not varieties), until it can be ascertained definitely whether it is identical with the type of Weddell's pallida. Doubtless there may be found among the descendants of Ledger's seeds some other forms of Calisaya already described by Weddell, as well as some hybrids yielding a large percentage of quinine; indeed this seems probable from the fact that Mr. Howard's Ledgeriana yields a large percentage, and is yet different from Dr. Trimen's plant. I would suggest, therefore, to planters that herbarium speci- mens, gathered when in fully formed fruit, and accompanied by a characteristic portion of bark, should be kept for reference, of all forms that present a recognizable difference in habit of growth, appearance of bark, typical form of leaf, size and structure of flower, and shape and size of fruit. In this way only can the tangle be unravelled into which the mixture of a variety of seeds supplied by Mr. Ledger has led botanists. Note on Ranunculus Jena, Linn. By Freeman C. S. Rorsk, F.L.S., F.G.S. {Read 18th December, 1884.] (Puates XIII. & XIV.) Rayvnovtus LINGUA is a plant very generally distributed in Britain, as it was recorded from 72 counties in Watson’s Topo- graphical Botany, and is stated in the second edition to occur in 77 counties. But it appears to be local and sparingly distributed in most parts of England; and I have only seen it in three loca- lities in this part (Eastbourne) of Sussex. It is probably from this cause that the early primordial submerged leaves appear to be very little known, or generally overlooked ; at all events, they are very rarely noticed by botanical writers. The specific de- scriptions given by the great majority of authors have been drawn up from the aerial leaves alone; and these differ so widely from the early submerged leaves that no one would imagine that they belonged to the same plant. I think, therefore, it may be useful to direct attention to these early leaves, of which the accompanying sketches are accurate representations, drawn to scale of about half the natural size. On looking through the botanical works I have available, I MR. F. C. 8. ROPER ON RANUNCULUS LINGUA. 881 find that out of thirty authors of general or local floras, seven describe the leaves of Ranunculus Lingua simply as “lanceolate; ” three call them “lanceolate and amplexicaul;" two “lanceolate and nearly sessile ;” and eleven describe them as “ lanceolate, amplexicaul, and sessile.” Thus twenty-three out of the thirty simply describe the long, narrow, lanceolate leaves of the flower- ing stem; whilst, on the other hand, there are only seven who allude to the early submerged leaves, although these alone are to be found for some months during the early part of the year. These early submerged leaves differ in almost every respect from the aerial stem-leaves—as whilst these latter are narrow lan- ceolate, almost sessile, slightly sinuate-denticulate, frequently covered with adpressed hairs, and from three quarters to one inch in breadth,—the submerged leaves are three to four inches broad, ovate or oblong-ovate, cordate at the base, seven to nine inches long when full-grown, on petioles four to five inches long, rather membranous and semitransparent, perfectly glabrous, and not at all denticulate at the edges, and with a much more reticulate venation than the rather thick and almost coriaceous, opaque stem-leaves. I will now briefly allude to the authors I have consulted. Gerard. Herb. 813 (1597), says ^. . . long smooth leaves not unlike those of the Willow.” Johnson, in his edition says “long leaves a little hairy, not unlike those of the Willow” (Ger. em. 961, 1633). Smith in Engl. Bot. ed. i. tab. 100 (1793), merely says “leaves lanceolate,” a translation of the diagnosis in Linnsus's Sp. Plant., as used also by :—Hudson, Flor. Ang. 210, id. ed. 2. 340; Relhan, Flor. Cant. (1784) and (1820); Gren. & Godr. Fl. de Fr. 1848; Koch, Syn. Fl. Germ. 3rd ed. (1857) ; Ben- tham, Br. Fl., both editions, 1865 and 1878. In the ‘ Flore Fran- çaise? Lam. and DC. (1805) describe the leaves as * Janceolate and amplexicaul ; " and are followed in this by Reich. Fl. Germ. Excurs. (1830-82), and Bouvier, Fl. des Alpes (1878). In With. Arr. Br. Pl. 1830, the leaves are said to be “ lanceolate and nearly sessile," in which Woods, Tour. Fl. (1850), coincides. Gray, Nat. Arr. Br. Pl. 1821, is the first who describes the jouves as 6 lightly serrate, sessile, nearly embracing the stem ; wad fe tolled by Grev. Fl. Ed. 1824 ; De Candolle's Prod. 1824; Bluff and Fing. Comp. Fl. Germ. 1825; Mack. Fl. Hibern. 1836 ; Hook. Br. Fl. edit. 4 (1838); Leight. Fl. Shrops. (1841) ; Bab. Man. ed. 1 to 7, 1843-1874; Gren. FI. de la Ch. Juras. 1865- 882 MR. F. C. 8. ROPER ON RANUNCULUS LINGUA. 1875; Breb. Fl. de la Norm. ed. 4 (1869) ; and Hook. Stud. FI. ed. 1 to 3, 1870-1884. Mr. Bentham, in Br. Fl. eds. 1 and 2, also states that the leaves, that is the long lanceolate leaves, to which alone he alludes, are “glabrous, with a few nearly parallel veins ; " and Sir J. Hooker, in Stud. Fl. eds. 1 to 3, states that the plant is * glabrous," and that the leaves have “ veins parallel and reticu- lated.” These descriptions are not exactly correct, as both the aerial leaves and stem are in most cases densely covered with stiff adpressed hairs, at all events in my Sussex specimens ; and the veins, instead of being parallel, are given off at various points of the mid-rib, and again anastomose at the tip of the leaf with the main central vein. They are, however, more or less reticu- lated both in the submerged and aerial leaves, but much more so in the former than in the latter. We thus see that in such standard books as Hook. Br. Flora, Gren. et Godr. Fl. de France, Koch Fl. Germ., Benth. Br. Fl., Hook. Stud. Fl., and Bab. Man. up to the 7th edit. (1874), no notice whatever is taken of the sub- merged leaves, nor would any one be able to recognize the plant if collected before the flowering-stems have appeared. Of the seven authors who refer to the submerged leaves of R. Lingua, two are of old date, namely—Parkinson, who in his Theat. Botan. (1640), says of “R. palustris flammeus major” :— “ The Greater Marsh Spearwort hath a long joynted root, whence riseth up a thiek joynted smooth stalke, two foot high, furnished with large and long shining and smooth thinner leaves, than in the next (2. Flammula), some being more than half a foote long and two to three inches broad, but smaller up to the toppe;" and Bauhin, in his ‘ Pinax, 1623, under R. longifolius palustris major, says, * Folia prima aliquando subrotunda sunt." There can be little doubt that these descriptions refer to the early submerged leaves, and not to the lanceolate leaves of the flowering-stem. The best and most correct deseription is that by Dr. Bromfield in his Fl. Vectensis (1856), who says, referring to the flowering-stem, * Leaves linear-lanceolate, erect, clasping the stem, thick, smooth or sometimes slightly hairy beneath, their edges with distant, very narrow serratures ;” and he then proceeds to say :— The earlier primordial submerged leaves are very large, ovate-oblong, obtuse, cordate at the base, sli ghtly undulate crenate along the margins, on very long semiterete, somewhat hairy, sheathing petioles, obscurely and reticulately veined, remaining MR. F. C. 8S. ROPER ON BANUNCULUS LINGUA. 388 green throughout the winter." Again Boreau, in his Fl. du cent. de la Fr. edit. 3 (1857), says of R. Lingua, “ produisant sous l'eau des feuilles longuement pétiolés, larges, cordiformes ovales." In Symes's Eng. Bot. ed. 3, 1863, we find :—* The earliest leaves, and those of the barren shoots on long stalks, ovate or oblong- ovate, those on the flowering-stem sessile, linear-lanceolate, very acute, faintly and remotely denticulate or nearly entire." Lloyd, in his Fl. de l'ouest de la Fr, 3rd edit. (1876), says :—* Feuilles demi-embrassantes lancéolées-linéaires, à dents courtes, obtuses, éloignées, les radicales submergées en ecur-ovales." And finally, Prof. Babington, in the 8th edition of the ‘Manual,’ 1881, after describing the stem-leaves as in previous editions, adds “ early submerged leaves oblong, blunt, cordate at the base.” There is one point of interest about the submerged leaves of R. Lingua worthy of notice, and that is, that though the large, broad, early leaves are always submerged, never, as far as I have seen, even floating on the water, yet they have a considerable number of small oval stomata, as shown in Pl. XIII. fig. 2, about tbo of an inch in length ; similar in shape, but far less numerous and smaller in size than those on the aerial leaves, on which they are very abundant, and about 51; of an inch in length, as shown in Pl. XIII. fig.3. In most works on Structural Botany, stomata are said to be absent on submerged leaves. Sachs, however, states that they are occasionally found. May not their presence in R. Lingua tend to show that the submerged are modified aerial leaves, rather than that the large spreading submerged leaves have been contracted and modified by exposure to air? Plate XIII. represents a plant of Ranunculus Lingua, collected in February 1878, whilst the whole plant was entirely submerged, drawn to scale of rather less than half the natural size. Pl. XIV. shows a plant collected in May of the same year, showing the flowering-stem, with one of the submerged leaves still remaining. Pl. XIII. tig. 2 are the small stomata from the submerged leaves ; and Pl. XIII. fig. 3 are the larger stomata from the aerial leaves. DESCRIPTION OF THE PLATES. PrarE XIII. Fig. 1. Submerged leaves of Ranunculus Lingua, L., slightly less than half nat. size, from a marsh-ditch in Pevensey Levels, near the Waterworks, Eastbourne. Collected Feb. 21, 1878. 2. Stomata from submerged leaves, ;5; in. long, x 200. 3. Stomata from aerial leaves, 535 in. long, x 200. 884 MR. C. B. CLARKE' 8 BOTANIC NOTES Puare XIV. Fig. 1. Submerged leaf; Fig. 2. Four aerial leaves; both half nat. size: from one stem of R. Lingua, likewise obtained from a marsh-ditch in Pevensey Levels, near to the Waterworks, Eastbourne, May 25, 1878. Botanic Notes from Darjeeling/to Tonglo and Sundukphoo. By C. B. Cranks, F.R.S., F.L.S. [Read 15th January, 1885.] In the ‘Journal of the Linnean Society, vol. xv. p. 116, are printed some botanie notes of mine made on an excursion from Darjeeling to Tonglo in September 1876. I made a hasty excur- sion from Darjeeling on 2nd June, 1884, to Tonglo, and on to Sundukphoo, 15 miles further along the same ridge. There are now Government-furnished bungalows on this ridge between Sikkim and Nepaul, viz.:—Jore Pokri, from Darjeeling 13 miles, alt. 7500 feet; Tonglo, from Jore Pokri 10 miles, alt. 10,000 feet; Sundukphoo, from Tonglo 15 miles, alt. 12,000 feet; Phal- loot, from Sundukphoo 13 miles, alt. 11,800 feet. There is a Government pony-path the whole way, and it is now easy to ride out to Tonglo in six hours. When Sir J. D. Hooker made his memorable exeursion from Darjeeling to Tonglo in May 1848 (see Hook. Journ. Bot. ii. [1850] p. 17), there was no road round this ridge; and he followed the native path directly across the valley of the upper Rungait, having to work hard to get to Tonglo in three days. The new Government pony-path in many places winds along the side of the mountain, takes you into the gorges and over their torrents, and facilitates greatly the operations of the botanic collector. I have been along this ridge from Tonglo to Phalloot (and onwards north) on several occasions; but the old path was very much on the ridge, and so very rough and steep that one had to look continuously at one’s feet. The present frontier-path and bungalows, designed by J. Ware Edgar, Dep.-Commissioner of Darjeeling in 1876, have thus aided the botanist, as well as rendered it easy for the denizens of Darjee- ling to see the celebrated view from Sundukphoo. From the verandah of the bungalow, which is close to the summit of Sun- dukphoo, 12,000 feet alt., you look over the valleys of the Tambur aud Arun upon the great Nepaul wall of snow, in the centre of FROM DARJEELING TO TONGLO. 385 which stands Deodunga (Mt. Everest), the highest mountain in the world: somewhat to the right is Kinchinjunga, its summit about 45 miles off still, but several of its big peaks (Pundim, Kubra, &c.) much nearer. Indeed, in some states of atmospheric refraction, Kinchinjunga seems to tower above you at Sunduk- phoo. These snows are the setting of asuperb foreground. The mountain of Sundukphoo is a fine one, and its shoulders fall away so that from the summit you can follow with the eye far down its flanks ; the masses of rock are interspersed with large patches of bright-green grass, over which is scattered, in very loose order, Abies Webbiana, Lindl., the type form, with trun- cated rigid heads and black-green foliage. The Rhododendrons, trees and shrubs, of various colours, reach for miles, and look like columns of troops converging on Sundukphoo, the summit of which the heads of the columns have just reached. The Rho- dodendrons were in a blaze when I was there on 5th June, 1884. l append to this paper a list of the plants which I actually brought back from Tonglo aud Sundukphoo from the present ex- cursion. It must not be supposed to be at all a complete represen- tation of the early spring vegetation: I only attempted to collect from 7000 to 12,000 feet altitude such plants as I knew I could not easily get near Darjeeling at 7000-8000 feet altitude. More- over I was much hurried, I had poor weather (poor for Tonglo even), and all my servants fell ill. Still I do not think that I could, with every collecting advantage, obtain on this ridge in spring so many, or so many new, plants as I obtained on my former excur- sion in September. I have been surprised a am always sur- prised over again) to see how late the spring is in Sikkim. In Darjeeling itself on 8th June, 1884, there were many trees, notably the Erythrina, so common in the station, that then scarcely showed the leaves emerging; and on Sundukphoo, at 12,000 feet, Pyrus, Ribes, &c., were in flower without leaves. This mountain-ridge bas now been visited by many botanists, aud at the time the Rhododendrons are in bloom, and my appended list contains hardly a single new species. I observed with, interest the great variability of the Rhodo- dendrons within a limited area (but with variety of level and aspect) on Sundukphoo. The common species there, hodo- i shite to mauve and dendron campanulatum, D. Don, varies from white to m: | an pink, two shrubs close beside each other differing greatly in colour; moreover the corolla varies 1n size and shape, some of 9 p > X 2E LINN. JOURN.— BOTANY, VOL. XXI. 386 MR. C. B. CLARKE'S BOTANIC NOTES the trees at lower levels having smaller and wider-mouthed corolla. Rhododendron cinnabarinum, Hook. f., is in general rather a dull-coloured species, the brick-red shading off into a tawny yellow; but there were particular trees of this species, at various levels, on Sundukphoo of extraordinarily brilliant colour, the red having become a scarlet and the orange entirely disappeared. For gardening-purposes I can see that we must look to indi- viduals, not to species. The cultivators of Rhododendrons must send men to these upper levels to mark particular trees in spring, and the men must come again in autumn and collect the seed of the marked trees. I noticed the great variation with level of the Ariseemas. At 8000—10,000 feet Arisema Griffithii, Schott, has the hood of the spathe very much dilated, 6-8 inches broad, curved and ridged ; in the same species, at 12,000 feet alt., the spathe is less than one inch broad, and the dilatation and ridging are most obscure. I have called this the same species: as I walked up Sundukphoo, passing literally thousands of examples, the transition appeared perfectly gradual from the one extreme to the other. Arisema Hookerianum and A. utile (Y have looked at the numbered Hookerian specimens of these which Engler takes as types) are for me trifling varieties of A. Grifithii. If they are not, then I have several new species which differ a good deal more from A. Griffithii than they do. Whatever view is taken about the limits of species, there is no getting off this fork. Similar considerable variations I noticed in the Lilies. In Smilacina, which is sorted in herbaria very much by the shape of the leaves, I find the leaves to vary so greatly in nearly every species, that I have no confidence that I can rightly refer the herbarium fragments, except such as show the flowers tolerably. Then, again, looking to cultivators’ objects, Smilacina oleracea, Hook. f. et T. Thoms., in its commoner smaller forms, is an elegant, scarcely a striking plant; but I saw specimens with a drooping panicle of 200 snow-white flowers, which even my Bhootia coolies could not resist, several of them carrying a spray in one hand a few inches before the nose for miles. FROM DARJEELING TO TONGLO. 887 List of Plants collected 2nd to 6th June, 1884, on Tonglo and Sundukphoo, alt. 7000—12,000 feet. RANUNCULACEX. Anemone obtusiloba, D. Don ............ 9000 feet alt. Ranunculus Cymbalaria, Pursh .......... 10,000 » diffusus, DO... 0. cee ce ccc eene 9000 , Var. floribus minimis .............005 7000 , Isopyrum adiantifolium, Hook. f. et T. Thoms. (carpellis semper 2, nee 8) .......... 8000 ,, MAGNOLIACEX. Schizandra grandiflora, Hook. f. et T. Thoms. | 8000 ,, elongata, Hook. f. et T. Thoms. ...... 7500 ž ,„ BERBERIDACEA. Holboellia latifolia, Wall. .............. e. 10,000 ,, Berberis insignis, Hook. f. et T. Thoms. .... | 9000 , CRUCIFERÆ. Cardamine trifoliolata, Hook. f. et T. Thoms. 9000 , Draba gracillima, Hook. f. et T. Thoms. .... 12,000 ,, VIOLACER. Viola biflora, Linn..... 0. cece cece cece eens 12,000 , —— Hookeri, Z. Thoms. ........ ve 10,000 » --— serpens, Wall... 6... cece eee ene 7000 ., Var. canescens (sp. JWatt).. esseere. 7000 ,, distans, Wall............. serene 7000 » CARPOPHYLLACEXE. Stellaria longissima, Wall... ..........e* . 7000 , sikkimensis, Hook. f. ......... e 7000 , bulbosa, Wulf. 0.0.0 cece cere n nnn 12,000 ž ,„ TERNSTRÆMIACEÆ. Actinidia strigosa, Hook. f. et T. Thoms .... 7500 » GERANIACEX. Oxalis Griffithii, Edgew. et Hook. f. ....-- 10000 , RvTACEX. Zanthoxylon acanthopodium, DO. ....ee 7500 sy, VPE 7500 ; oxyphyllum, Edgew... .... ; Skimmia Laureola, Hook. f. ..... e 10,00 , CELASTRACEJ. E igi Wall. etn 10000 , uonymus frigidus, Wa 1500 2 Celastrus stylosa, Wall... ....«.« nns 2E2 388 MR. C. B. CLARKE’S BOTANIC NOTES VITACE. Vitis himalayensis, Brandis ....... 7500 feet alt. SaPINDACER. Acer Campbellii, Hook. f. et T. Thoms. .... 7000 ,, levigatum, Wall. ...... enn 7000 ž ,, ANACARDIACEEX. Rhus suecedanea, Zinn......... eee 7000 » LEGUMINOSA. Piptanthus nepalensis, D. Don..........-- 10,000 » 7 RosACEX. Prunus nepalensis, Seringe ....... eee 7500 ž , Spiræa bella, Sims ... 0.0.0 cece eens 9000 » Rubus ecalyeinus, Wall. (The “ Ground- Raspberry.” ). s.. eee 7000 , lasiocarpus, Smith ..... llle. 7000 , Hookeri, Focke ........ 00.0 ce eens 7500 [ — R. macrocarpus, King ; Journ. Linn. Soc. xv. p. 142. Rosa sericea, Lindl. ................ ees. 10,000 » Fragaria Daltoniana, J. Gay... eese 10,000 » Pyrus rhamnoides, DC. ............000005 10,000 » foliolosa, Wall. ................. es. 12,000 » microphylla, Wall............... Lus. 12,000 ž , vestita, Wall..........-.. lesse. 10,000 " [2 P. lanata ?, D. Don; Journ. Linn. Soc. xv. p. 142.] . . SAXIFRAGACER. Astilbe rivularis, Buch.-Ham. ............ 7000 » Saxifraga purpurascens, Hook. f. et T. Thoms. 12,000 ,, Tiarella polyphylla, D. Don .. ............ 8000 ,, Chrysosplenium lanuginosum.. Hook. f. et T. Thoms......... eee 8000 » nepalense, D. Don.................. 7000 » Ribes glaciale, Wall. ................ 8000-12,000 » CvcURBITACEX. Biswarea tonglensis, Cogniaur ............ 8000 » [= Warea tonglensis, C. B. Clarke, in Journ. Linn. Soc. xv. p. 129.] BEGONIACES. Begonia Cathcartii, Hook. f. ..... PEE 8000 » ARALIACEEF. Aralia pseudo-ginseng, Benth. ............ 10,000 bipinnatifida, C. B. Clarke .......... 10,000 » FROM DARJEELING TO TONGLO. 389 CoRNACES. Aucuba japonica, Thunb. ................ 7000 feet alt. CaPRIFOLIACES. _ Viburnum erubescens, Wall. ............ 10,000 , cordifolium, Wall. .............0000. 9000 , Leycesteria formosa, Wall. .............. 8000 » Pentapyxis stipulata, Hook. f. ............ 7500 ž , RUBIACEE. Lasianthus Biermanni, King...... lesen. 7000 , Rubia cordifolia, Ling. ....... ccc cc eee 9000 » Galium triflorum, Linn........... 0.0000 9000 ,, COMPOSITA. Ainsliæa pteropoda, DO. ................ 7500 =, CAMPANULACER. Peracarpa carnosa, Hook. f. et T. Thoms. .. 8000 ,, ERICACER. Vaccinium Nummularia, Hook.f et T. Thoms. 9000 ,, Pentapterygium serpens, Klotzsch ........ 7500 sy, Enkianthus himalaicus, Hook. f. et T. Thoms. | 9000 ,, Rhododendron campanulatum, D. Don 11,000-12,000 ,, Dalhousie, Hook. f. ...... eese 7000 , arboreum, Smith .......... eu. 9000-11,000 » barbatum, Wall. ......... esee s 11,000 ,, cinnabarinum, Hook. f. ........ 11,000-12,000 ,, Faleoneri, Hook. fe ......... eee 10,000 ž , PRIMULACES. Primula rotundifolia, Wall. .............. 12,000 , petiolaris, Wall........... eee 12,000 ž , Var.(?sp.) «II 11000 ,, Lysimachia prolifera, Klatt ............ 8000-9000 ž ,, MyRrstNace&. Embelia Gamblei, Kurz............ eee 7500 » STYRACEX. Styrax Hookeri, C. B. Clarke ........ 7500 ., LoGANIACEX. Buddleia Colvillei, Hook. f. ...... eee 9500 , GENTIANACE. Gentiana capitata, Buch.-Ham. ... eere 12,000 , BoRAGINACEX. 9000 » Paracaryum glochidiatum, Benth. .......- 890 MR. C. B. CLARKE’S BOTANIC NOTES SOLANACEZX. Mandragora caulescens, C. B. Clarke ...... 12,000 feet alt. l VERONICACEA. Calceolaria mexicana, Benth., efferata ...... 8000 » Sibthorpia pinnata, Benth. ...... ee 8500 » Hemiphragma heterophyllum, Wall. ...... 8000 ,, Veronica cana, Wall........ eene 8000 » CynTANDRACEEX. Chirita Kurzii, C. B. Clarke ...... sese 9000 » LABIATA. Ajuga lobata, D. Don... . nne 9000 , PiPERACEEX. Peperomia reflexa, A. Dietr........... ss 7000 5; LAURACEX. Machilus odoratissimus, Wees ............ 7000 » THYMELACER. Daphne papyracea, Wall. ............s ese 10,000 » URTICACER. Pilea umbrosa, Wedd. .............. eee 7000 » scripta, Wedd. 1... .. cece cece eee ae 8000 » ternifolia, Wedd. «1... eee ccc ees 7000 » Pilea sp. (A most minute species in full flower.). DI 12,000 » Elatostema diversifolia, Wedd. ............ 7000 » Chamabaina squamigera, Wight .......... 7500 yy ORCHIDACER. Liparis nepalensis, Lindl................. 11,000 , Collogyne corymbosa, Lindl. ............ 7500 ž » (§ Pleione) Hookeriana, Lindl. ...... 10,000 » ARACER. Ariseema speciosum, Mart. ............ 7000-8000 LI . » Rhizome horizontal, short cylindrie, with many rootlets on all sides. In all other Arisemas known to me the rhizome is hemispheric, the whole lower (hemispheric) surface devoid of rootlets. Griffithii, Schott ........ uuu. 8000-12,000 feet alt. nepenthoides, Mart. .......... lesu. 10,000 , Jacquemontii, Blume .............. 7000 Abundant at lower levels (4000-7000). FROM DARJEELING TO TONGLO. 891 LILIACEA. Ophiopogon Wallichianus, Kunth ........ 7500 feet alt. Tupistra amarantoides, Wall. ............ 7000 , Smilax ferox, JWall........... e 00 cee cues 7500, ——- menispermoidea, 4. DO. ............ 10,000 ,, Polygonatum punctatum, Wall. .......... 7500 ,, oppositifolium, Royle .............. 7500 ,, verticillatum, Allioni .............. 10,000 sé, Smilacina oleracea, Hook. f. et T. Thoms. .. 8000 » purpurea, Wall. .............. 10,000-12,000 . , [The plant called S. purpurea, Wall., in Journ. Linn. Soc. xv. p. 122, was S. oleracea, Hook. f. et T. Thoms.] divaricata, Wall. .......... eese. 7500 feet alt. Fritillaria cirrhosa, D. Don .............. 12,000 , Clintonia alpina, Kunth ........... sese. 12,000 __s, Disporum, sp. ........ cece ee eee 7500 » Paris polyphylla, Smith ........0. cece eee 10,000 __,, Trillium Govanianum, Wall. .............. 12,000 , [Smilacina sp., C. B. Clarke in Journ. Linn. Soe. xv. p. 122, was only seen at 8000 feet alt. in young bud, not collected. | CYPERACE®. Hemicarex Hookeri, Benth. et Hook. f. .... 12,000 feet alt. Carex nubigena, D. Don ......-. 00. sees 9000 , remota, Linn. .......e n nne 8000 » Daltoni, Boott .......... n 12,000 , polycephala, Boot£........ e 8000 , decora, Boott ....... eee 9000-11,000 » vesiculosa, Boott .......... oes 9000 » bengalensis, Rox5......... nne 7500 , ——— filicina, Nees ... ce cc nn 7000 » nepalensis, Spreng... .. esee n n n 10000 ,, —— phacota, Spreng... nnn 9000 » pellucida, Zwrez. ......... nnn 12,000 » With three other Carices, for which I have not yet names. GRAMINACER. Poa alpina, Linn. 1.0... cee eee ee eee m 9000 " n collected Glyceria tonglensis, C. B. Clarke (in Journ. Linn. Soc. xv. p. 119) at 7000 feet alt. on a lower ridge, but did not see it at this spring season at Tonglo. | FILICES. None collected. All in a very young stage, and none looked new or noteworthy. 392 SIR J. D. HOOKER ON THE PLANTS OF THE List of the Plants collected by Mr. Thomson, F.R. G.S., on the Mountai ns of Eastern Equatorial Africa, by Prof. DANIEL OLIVER, F.R.S.; with Observations on their Distribution, by Sir J. D. Hook, F.R.S. [Read 15th January, 1885.] Iw offering to the Linnean Society the accompanying catalogue, by Professor Oliver, of the small but very interesting herbarium made by Mr. Thomson in the highlands of Eastern Equatorial Africa, and presented by him to Kew, I think it may interest the Fellows if I preface it with some results in botanical geography which I have gleaned from a study of its contents. I may premise that of the mountain flora of Equatorial Africa nothing whatever was known previous to 1860, when Mr. Gustav Mann, who had acted as botanist to Dr. Baikie's Niger Expedi- tion, was (on Sir William Hooker's recommendation) instructed by the Secretary of State for Foreign Affairs to explore botani- cally the mountain-peaks of the Gulf of Guinea and its islands. Mr. Mann accordingly made several ascents of Clarence Peak, Fernando Po, alt. 9469 feet ; one of St. Thomas's Island, alt. 7500 feet ; and two of the Cameroons range, the culminating point of which he found to be 13,100 feet. The results of Mr. Mann’s admirable labours are well known to this Society, being published in the 6th and 7th volumes of our Journal. To those results the following remarks may be regarded as complementary, and consisting of an extension of our knowledge of the mountain flora of Equatorial Afriea from the western coast of the continent to the eastern. Of collections made in the highest regions of Eastern Africa prior to those of Mr. Thomson, the only one known to me is that of the enterprising missionary, the late Rev. Mr. New, who was the first to reach and ascend the great mountain Kilimanjaro (in 1871), and who, at Dr. Kirk's instigation, collected a few flowering plants, about twenty in all, in the uppermost zone of vegetation. These were named by Prof. Oliver, and are pub- lished in Mr. New’s Narrative. They are characteristic of a higher elevation than that obtained by Mr. Thomson on that mountain. Amongst them are two northern genera not collected by the latter traveller, Artemisia amd Bartsia, which I have added to the list from which the following conclusions are drawn. MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 893 The localities from which Mr. Thomson's specimens were brought are, with their elevations :— u . Lat. Long. Elevation. Species. Kilimanjaro es 8.30 0% E. 37° 30'. | 9000-10,000 feet. 35. Lykipia........ N. 1°-S. 1°. E. 36? 37. 6000-8000 ,, 58. Kapté plateau .. S. 1°-2°, E. 36? 37'. 5000-6000 ,, 34. Lake Naivaska.. S. 1° E. 36°. /000-8000 ,, 9. The subjects most worthy of comment indicated by a study of these collections may be grouped as follows :— l. The number and affinities of the plants characteristic of the European flora. 2. The number and affinities of plants characteristic of the South-African flora. 3. The comparison of the Eastern Equatorial mountain-flora with that of the western side of the continent. 4. The affinity of the flora with that of the highlands of Abyssinia. 5. Origin of the flora as assumed from these data. 1. The Northern or European Element.—Of the 107 genera and 140 species of flowering-plants, no less than 27 genera, including 37 species, are of a distinetly northern type, and com- prise, amongst others, species of Clematis, Ranunculus, Anemone, Delphinium, Cerastium, Hypericum, Geranium, Trifolium, Lotus, Epilobium, Caucalis, Galium, Scabiosa, Echinops, Artemisia, Son- chus, Erica, Swertia, Bartsia, Leonotis, Rumex, Juniperus, and Romulea. And amongst the species are Cerastium vulgatum (two forms), Caucalis infesta, Galium Aparine, Scabiosa Columbaria, Sonchus asper, Erica arborea, and Rumex obtusifolius. Of the above, the following genera have not been hitherto detected in South Africa :— Delphinium*, Caucalis, Echinops *, Artemisia*, Swertia, Bartsia, Leonotis*, and Juniperus*. Those marked with an asterisk have not been found in the mountains of Western Africa; nor have the following :—Anemone, Lotus, Epilobium, and Erica. Thus no fewer than 9 northern genera are added to the Equatorial African flora by this small herbarium alone. Of all these the Juniper is the most interesting, as indi- cating the southern limit of that wide-spread northern genus, and the fact of its actually reaching the Equator. The southern limits hitherto ascertained of the genus Juniperus are :—In Asia N. lat. 28°, in the Eastern Himalayas, where it is not found under 8000 feet elevation; in America it extends far lower down, 394 SIR J. D. HOOKER ON THE PLANTS OF THE to Guatemala and the Jamaican mountains, N. lat. about 15°. In Africa the J. procera was found by Schimper in the Tigre mountains in N. lat. 14°. Having regard to the comparatively low elevation of the Lykipia forest and its equatorial position, it is evident that a little downward extension of the range of Juni- perus would constitute it a tropical genus. 2. The Southern or Temperate South-African Element.—There are 35 genera in the above collections which are represented in South Africa, some of which there obtain their maximum, or are even almost peculiar to that region. The most notable of these are all those mentioned above as northern, with the exception of Delphinium, Artemisia, Echinops, Swertia, Bartsia, and Juni- perus. And of other southern types there are the species of Sparmannia, Calodendron, Psoralea, Alepidea, Felicia, Tripteris, Osteospermum, Berkeleya, Lightfootia, Blaeria, Selago, Struthiola, Podocarpus, Aristea, Gladiolus, and Kniphofia. Of these, Felicia, Osteospermum, and Alepidea had not been previously found north of the Tropic of Capricorn. One species of Clematis is identical with the Cape C. Thunbergiana, as is the Calodendron with C. capense, and the Alepidea with A. amatymbica; and the Anemone is very near A. capensis. Of the rest most have repre- sentatives in Abyssinia or the mountains of Western Equatorial Africa. No less than 15 of these South-African genera appear to be absent on the mountains of Western Equatorial Africa; they are:—Anemone, Calodendron, Psoralea, Alepidea, Felicia, Tripteris, Berkeleya, Lightfootia, Erica, Selago, Leonotis, Stru-— thiola, Aristea, Gladiolus, and Kniphofia. On the other hand, the mountains of the Gulf of Guinea contain many South-African genera not hitherto found in the Eastern equatorial mountains. Amongst the most notable of these are Anthospermum, Hieracium, Ilex, Lasiosiphon, Peddiea, Geissorhiza, and Hypoxis. 3. A Comparison of the Eastern with the Western Mountain Vegetation can only be profitably undertaken when the flora of the former is as diligently gleaned as was that of the latter by Mr. Mann; and we may hope for contributions towards this end from Mr. Johnston's exploration of Kilimanjaro. It is, however, worthy of remark, that of the genera found in the east and not hitherto in the west, the majority are of either Abyssinian or South-A frican types, whilst the compensating wealth MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 395 of the western flora is in European types not hitherto detected in the east. 4. The Affinity of the Flora with that of the Highlands of Abyssinia is very marked, as was to have been expected. Most of the genera are in fact Abyssinian, as are all, or nearly all, of the following species :—Ranunculus oreophytus, Viola abyssinica, Sparmannia abyssinica, Geranium simense ?, Trifolium simense, Lotus tigrensis ?, Lythrum rotundifolium, Epilobium stenophyllum, Diplolophium abyssinicum, Caucalis melanantha, Coreopsis abyssi- nica, Lightfootia abyssinica, Erica arborea, Swertia Schimperi, S. pumila, and Juniperus procera. Besides the above, the Abyssi- nian affinity is shown by the presence of an Ueberlinia, a genus hitherto known only as a monotypic Abyssinian one, and by the species of several of the other genera being more nearly allied to plants of that country than of any other. 5. On the Origin of the Flora.—The most striking feature of the flora thus first explored by Mr. Thomson is the discovery in Lykipia of three such typical forest-trees in close association as the Juniperus procera of Abyssinia, the Calodendron capense of South Africa, and the noble Podocarpus, a close ally both of the Cape P. elongata and of the eastern tropical P. Mannii, discovered on the top of the peak of the island of St. Thomas by the naturalist whose name it bears. And these three plants no doubt indicate the affinities of the flora being most strong with the countries north and south of it, and less so with that far to the east of it. This is what the configuration of the con- tinent would indieate as most probable, the loftier mountains being on the east side, ard being connected by more or less con- tinuity of high land from Abyssinia to the Cape Colony. That the flora of the latter country extended into the former was well known; and this renders the discovery of a locality in the line of continuous migration or distribution, where the most marked type of the northern flora (Juniperus) meets the most marked of the southern (Calodendron), and this at the respective limits of each, a most interesting one, and only second in importance to the general result of Mr. Thomson’s labours, which is the dis- covery of so many northern forms in the comparatively isolated equatorial tracts which he has been the first to explore. Thus I think it may be regarded as most probable that the Equatorial African mountain-flora is in the main an immigrant one from Abyssinia, possessing many genera and species that 396 SIR J. D. HOOKER ON THE PLANTS OF THE have advanced even as far as the Cape Colony, besides many others that have not gone so far, and of which latter a few have been collected by Dr. Kirk during Dr. Livingstone's second expedition, in the mountains of comparatively low elevation near Lake Shirwa in lat. 15? S. In a lesser degree it has been peopled by a return flow of South-African genera and species, of which many have in like manner advanced further and reached Abyssinia, whilst others have been arrested in their northward spread. It would be interesting, but in the present state of our knowledge fruitless, to speculate on the direction in which the wave of migration is now advancing, and whether the later northern preceded the southern, or vice versa. Yet when it is considered that the whole area over which Mr. Thomson’s collections were made is volcanic, and probably geologically modern, in its present configuration, it must be evident that the main features. of its vegetation are of no great antiquity. There is one more point of interest to which a study of Mr. Thomson’s collection invites attention, which is that, whereas the lowlands of Eastern Tropical Africa (and indeed of all Tropical Africa) abound in species and representative species of the Deccan peninsula of India, the highlands of these two regions seem to have nothing in common, botanically or zoologically. And what renders this more noteworthy is that, though they have no types in common, there are desiderata common to both, as exemplified by the absence of Cupulifere, and paucity of Conifere, Cycadee, and Palmee, all of which abound in the Eastern Archipelago and in most other tropical countries. Looking still further off, and comparing the African flora with the Australian, a singular dif- ference is observable in this, that whereas the Tropical-Australian flora is in very great measure made up of species belonging to Temperate- Australian genera, the Tropical-African is, except in the highlands, of a totally different type from the South-African. The tropical floras of both have been obtained largely from the Asiatic continent ; but whereas in Australia there is a mingling of the Asiatic and endemic southern genera and species, there is no such mingling of the elements in Africa, except at con- siderable elevations, in its tropical regions. J. D. HOOKER. MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 397 Catalogue of the Plants collected by Mr. J. Thomson in East Tropical Africa. By Prof. Oxtver, F.R.S. 1. CLEMATIS SINENSIS, Fres., forma. Lykipia. 2. C. THUNBERGII, Steud., var. SERICEA. Lykipia. 3. RANUNCULUS OREOPHYTUS, Del. Kilimanjaro. 4. ANEMONE TuHomsont, Oliv., sp. n. ($ Pulsatilloides, A. ca- pensi, L., aff.) ; foliis triternatim sectis segmentis cuneato-ob- ovatis trifidis, lobulis ultimis obtuse acutatis paginis glabratis parce ciliolatis, petiolo gracili glabro, scapis 1—2-pedalibus graci- libus parce pilosis, involuero tripartito segmentis trifidis lobis linearibus vagina equilongis, pedunculo apice dense piloso, sepalis 14—20 linearibus v. oblongo-spathulatis extus parce pilo- sulis glabratisve, carpellis (floriferis) dense hirsutis. Kilimanjaro. The leaves are detached from the scapes, but I think not mismatched. 5. DELPHINIUM MACROCENTRON, Oliv. sp. n. Herba erecta 2-3-pedalis vel ultra, pilosula, foliis caulinis palmatim 5-partitis segmentis (fol. infer.) 3-5fidis lobo centrali elongato lineari- lanceolato acuminato, fol. super. segmentis indivisis elongato- linearibus, racemis paucifloris pedunculatis, pedicellis erectis apice recurvis, floribus ceruleo-purpureis, calcare erecto sub- cylindraceo obtuso lamina 2-4plo longiore pilosulo, petalis ante- rioribus longe unguiculatis lamina oblongo-spathulata bifida parce setulosa, carpellis 3 pilosis, stylis longiusculis superne glabratis recurvis. Lykipia. Flores 11-2 poll.; calcar 3-13 poll. 6. VIOLA ABYSSINICA, Steud. Kilimanjaro. 7. UEBELINIA ROTUNDIFOLIA, Oliv. sp. n. Herbula foliosa, proeumbens, caulibus pilis reflexis seriebus duabus longitudi- naliter hirsutis, foliis suborbiculatis mucronulatis scaberulis sub- levibusve setuloso-ciliolatis erassiusculis subsessilibus v. breviter latiuscule petiolatis, floribus quasi axillaribus folio brevioribus solitariis, calyce 5fido lobis ovatis apiculato-acutatis, petalis lineari-spathulatis calyce sublongioribus, staminibus 9-10, ovario ellipsoideo, stylis distinctis, seminibus paucis (4-5), Kilimanjaro. Folia 4 poll. lata. 398 PROF. D. OLIVER ON THE PLANTS OF THE 8. CERASTIUM VULGATUM, L., var. UNILATERALE. Lykipia. 9. C. VULGATUM, L. Kilimanjaro. 10. Hyrerrcum. H. peplidifolio, Rich. aff. (an var. calycis lobis 2plo longioribus ?). Kilimanjaro. 11. H. LANCEOLATUM, Lam. Lykipia. 12. Srna SCHIMPERIANA, Hochst. Kapté. 13. AnUTILON INDICUM, Don. Lykipia. 14. PAVONIA ScurMPERIANA, Hochst. Kapté. 15. HIBISCUS cnASSINERVIS, Hochst. Lykipia. 16. H. aossyrrNvs, Thunb. Crater south of Lake Nairascha and Kapté. 17. Grewia OCCIDENTALIS, L. Crater south of Lake Nairascha and Lykipia. 18. SPARMANNIA ABYSSINICA, Hochst. Lykipia. 19. GERANIUM an G. simense ? (folia radicalia). Kilimanjaro. 20. G. SIMENSE, Hochst., var. GLABRIUS. Kilimanjaro. 21. Impatiens Tuoxsowi, Oliw., sp. n. Glaberrima, succu- lenta, foliis verticillatis ovalibus v. oblanceolato-ovalibus breviter acuminatis subsessilibus setoso-ciliatis pedunculis axillaribus uni- floris supra medium bibracteatis folio sæpius brevioribus, sepalis lateralibus petalis 2-3plo brevioribus ovatis acuminato-apiculatis, sepalo postico ovato cucullato abrupte acuminato petalis breviore in calcar elongatum gracile abrupte producto, petalo antico rotundato amplo apiculato alis bipartitis equilongo. Lykipia. Folia 33-53 poll. longa, 1-1} poll lata. Flores ut videtur purpurei, 13-2 poll. diam. 22. IMPATIENS KILIMANJARI, Oliv., sp. n. Herbula glabra, succulenta, ramosa, foliis alternis petiolatis ovatis acutiusculis Jatiuscule setoso-crenatis, pedunculis axillaribus folio ssepius lon- gioribus l-floris, sepalis lateralibus parvis ovatis lanceolatisve acuminatis, calcare cylindrico valido ineurvato acuto ore dilatato breviter apiculato, petalo antico concavo late ovato apiculato, alis bifidis petalo antico æquilongis, lobo autico obovato postico angustiore obovato-oblongo. Kilimanjaro. MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 399 Folia 2-2 poll. longa, petiolo sepius subequilonga. Flores 1 poll. longi. 23. IMPATIENS, sp. noy. (Insufficient for description.) Crater south of Lake Nairascha. 24. CALODENDRON CAPENSE, Thunb. Kapté and Lykipia. Fine tree of 30-50 feet. 25. TURRÆA MoMBASANA, C. de C. § H., forma. Crater south of Lake Nairascha. 26. HELINUS MYSTACINUS, E. Mey. Kapté. 27. Croratarta DILLONIANA, Baker. Kapté. 28. C. THomsont, Oliv., sp.n. Fructiculus, ramosus, caulibus gracilibus ultimis hirtis, foliis petiolatis 3-foliolatis, foliolis ellip- ticis obovatisve obtusis mucronatis reticulatis pilis setisve paucis appressis pagina superiore v. utrinque onustis, floribus majus- culis peduneulatis solitariis (v. geminis ?), pedunculis folio oppo- sitis vel quasi terminalibus medio articulatis bracteatis apice bibracteolatis, bracteolis lineari-subulatis tubo calycis subzqui- longis, calyce parce et appresse hirtello profunde 5fido lobis lan- ceolatis acutis, vexillo purpureo striato dorsi medio hirsuto calyce fere 2plo longiore. Kapté. Folia petiolo hirto gracili stricto } poll. longo v. breviore, foliolo centrali 4 poll. longo. Flores }-3 poll. lati. Legumen turgidum, appresse hirtellum, utrinque attenuatum, apice stylo persistente faleato coronatum. 29. PsoRALEA FOLIOSA, Oliv., sp. n.; ramulis pubescentibus dense foliosis, foliis trifoliolatis brevissime petiolatis petiolo sti- pulis erectis lanceolatis ssepius breviore, foliolis oblanceolatis obtusis mueronatis brevissime petiolulatis glandulis immersis conspicue punetatis supra glabratis subtus precipue in costa parce pubescentibus, floribus breviter pedicellatis in fasciculos terminales foliis breviores dispositis, calyce 5fido lobis acutis lobo antico subulato cxteris duplo longiore. Lykipia. Folia 3-14 poll. longa. Foliola 4-5 lin. longa. Legumen non vidi. 30. TRIFOLIUM SIMENSE, Fres., VAT. ANGUSTIFOLIA. 31. TRIFOLIUM, sp. (Inadequate. Destroyed.) Lykipia. Kapté. 400 PROF. D. OLIVER ON THE PLANTS OF THE 32. LOTUS IGRENSIS, Baker? Crater south of Lake Nairascha. 33. Viena, sp. (Inadequate fragment.) Kapté. 34. CASSIA DIDYMOBOTRYA, Fres. Kapté. 35. LYTHRUM noruxDrFOLIUM, Hochst. Kapté. 36. EPILOBIUM STENOPHYLLUM, Fres. Kilimanjaro. 37. ALEPIDEA AMATYMBICA, Ech.§ Zey.? (Imperfect.) Lykipia. 38. DIPLOLOPHIUM ABYssINICUM, Benth. j Hook. f. Lykipia. 39. Caucatis MELANANTHA, Benth. & Hook. f., var. Kapté. 40. C. INFESTA, Curt. Lykipia. 41. PENTAS SCHIMPERIANA, Vatke. Lykipia. 42. P. cARNEA, Benth., forma. Kapté. 43. P. PuRnPUREA, Oliv., forma. Lykipia. 44. GALIUM APARINE, L. Kilimanjaro. 45. ScABIOSA COLUMBARIA, L. Lykipia. 46. GUTENBERGIA CORDIFOLIA, Benth. Kapté and Lykipia. 47. VERNONIA, aff. V. Melleri, Oliv. & Hiern, an var. ? Kapté. 48. FELICIA MURICATA (Aster, Less.), var. ? Crater south of Lake Nairascha. 49. MrcRoaLossA VOLUBILIS, DC. Kapté. 50. SPILERANTHUS SUAYEOLENS, DC. Kapté. 91. S. anaciLIs, Oliv. sp. n. Herba 14-pedalis, erecta, glaber- rima, eaule aptero gracili, folis anguste linearibus utrinque attenuatis integris, capitulis compositis solitariis terminalibus peduneulatis hemisphericis braeteis exterioribus paucis lineari- lanceolatis obtusiusculis floribus :quilongis, capitulis 9-12-floris, bracteis oblongis obtusis truncatisve apicem versus denticulatis floribus subequilongis, floribus omnibus fertilibus v. 1-2 abortivis, acheniis crassiusculis leviter compressis parce hirtellis. Kapté. Folia 2-3 poll. longa, 1-2 lin. lata. Capitula 3 poll. diam., pedunculo 1-2 poll. longo. 52. DicuigocEPHALA CHRYSANTHEMIFOLIA, DC. Lykipia. 53. SPILANTHES AcMELLA, L., forma. Kapté. MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 401 91. ACHYROCLINE SCHIMPERI, Sch. Bip. Kapté. 95. GNAPHALIUM, aff. G. auriculato. Kilimanjaro, 96. G. UNIONIS, Sch. Bip. Kapté. 57. HELICHRYSUM F@TIDUM, Cass. Kapté. 58. H. ADENOCARPUM, DCO., var. ALPINUM. Kilimanjaro. 99. Coreopsis, sp. nov.? Allied to C. abyssinica, Sch., and perhaps abnormal. Not in a state for description. Lykipia. 60. BrpENs PrLOSA, Z.—B. leucantha, W. Kapté and Kilimanjaro. 61. MarANTHERA BRowNEr, Sch. Bip. Kapté. 62. WEDELIA MOSSAMBICENSIS, Oliv. ? Kapté. 63. Guizorra, an G. Schultzii, Hochst. ? Lykipia. 64. TRIPTERIS VArLLANTII, Decne. Kapté. 65. TRIPTERIS, sp.nov.? (No fruit.) Kilimanjaro. 66. Emrin, aff. E. angustifolia, DC. ? Lykipia. 67. GyNURA vITELLINA, Benth. Kapté and Lykipia. 68. CINERARIA ABYSSINICA, Sch., forma. Lykipia. 69. SENEcrO, sp. nov.? (Inadequate.) Kapté. 70. OrHonna, sp. nov.? (1 capitulum only.) Crater south of Lake Nairascha. 71. ECHINOPS AMPLEXICAULIS, Oliv. Lykipia. 72. BerKHEYA SPEKEANA, Oliv. Lykipia. 73. CnEPIS, sp.? (No leaves.) Lykipia. 74. SONCHUS ASPER, L. Lykipia. 75. Losera. (Fragment.) Lykipia. 76. LosELIA, an var. L. coronipifolie, L., foliis subintegris glabris ? Crater south of Lake Nairascha. 77. LIGHTFOOTIA ABYSSINICA, Hochst., var. tenuis, glaberrima, foliis anguste linearibus integris. Crater south of Lake Nairascha. 78. ERICA ARBOREA, L. and Lykipia. 79. JASMINUM AURICULATUM, Vahi.—J.settense, Al. 80. J. * ABYSSINICUM, R. Br." Lykipia. 81. Canrssa, an C. edulis, Vahl? Lykipia. LINN. JOURN.—BOTANY, VOL. XXI. Crater south of Lake Nairascha Lykipia. 2F 402 PROF. D. OLIVER ON THE PLANTS OF THE 82. ACOKANTHERA (Carissa Schimperi, A. DC., Strychnos abyssinica, Hochst.) Mixed with the Carissa, but no doubt the plant to which the label * Murju—a deadly poison "— applies. 83. GOMPHOCARPUS PHYSOCARPUS, E. Mey., or G. abyssinicus, Hochst. ? Lykipia. 84, SWERTIA SCHIMPERI, Griseb. Kilimanjaro. 85. S. PUMILA, Hochst. Kilimanjaro. 86. SWERTIA, ap. Lykipia. 87. EHRETIA, an var. E. abyssinice ? Kapté. 88. CxNoaLossUM LANCEOLATUM, Forsk.? (Fragmentary.) Lykipia. 89. SorANUM, sp. (Fragment.) Lykipia. 90. S. NIGRUM, L., forma. Kilimanjaro. 91. SoPUBIA, an var. S. Dregeanc, Benth. ? Lykipia. 92. ALECTRA ASPERRIMA, Benth. Lykipia. 93. RHAMPHICARPA ? or Strraa? (No fruit.) Kapteé. 94. RHAMPHICARPA ? (No fruit.) Kapté. 95. SeLaco Tmowsowr, Rolfe, sp. n. Annua, ramis erectis albido-pubescentibus, foliis lineari-lanceolatis subobtusis inte- gris puberulis, capitulis terminalibus subglobosis in fructu parum elongatis, floribus sessilibus, bracteis linearibus obtusis pubes- centibus, calyce 5-fido lobis subulato-linearibus arcte ciliatis, coroll tubo brevi lobis oblongis tubo duplo brevioribus, stami- nibus styloque breviter exsertis, fructu ovoideo-globoso sub- compresso. Kilimanjaro. Planta 1-1 ped. alt. Folia 3-6 lin. longa, 1-1 lin. lata. Ca- pitula 2-3 lin. lata. Bractew 1 lin. longe. Calyx 3 lin. longus. Corolla 1 lin. longa. Fructus j lin. longus.—-A very distinct species, approaching S. cephalophora, Thunb., in habit, but readily distinguished by its minute flowers. (R. A. Rolfe.) 96. HEBENSTREITIA INTEGRIFOLIA, L.? Crater south of Lake Nairascha. 97. H. DENTATA, L. Lykipia and Kilimanjaro. 98. JUSTICIA NEGLECTA, 7. And. Y 99. JUSTICIA, sp. Lykipia. Lykipia. MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 403 100. DrcrrPTERA, near D. maculata, Nees. (Fragment.) Lykipia. 101. Barterta, aff. B. repenti, Nees. Kapté. 102. Crossanpra, aff. C. undulefolia, Salisb. Lykipia. 103. HyPoEsTES ANTENNIFERA, S. Moore? Lykipia. 101. H. Rora, And., var. PUBESCENS ? Lykipia. 105. H. vEnTICILLARIS, R. Br. Kapté. 106. THUNBERGIA FUSCATA, T. And. Kapté. 107. LANTANA Krsr, A. Rich.? Lykipia. 108. LANTANA, Sp. Kapté. 109. LrPPrA ASPERIFOLIA, Rich. ? Kapté. 110. Ocymum, sp. Lykipia. 111. Ocymum, sp. ? Kapté. 112. OcxMvM, sp. Lykipia. 113. PrEcTRANTHUS, sp. ? Kilimanjaro. 114, LEUCAS MASAIENSIS, Oliv., sp.n. Herba decumbens, cau- libus pilis brevibus decurvatis hirtellis, foliis petiolatis obovato- rotundatis obtusis late crenatis basi cuneatim late angustatis rotundatisve, verticilastris solitariis longiuscule pedunculatis multifloris, bracteolis anguste linearibus calycem subequan- tibus, calyce tubuloso campanulato ore subequali 12-dentato dentibus brevibus subulatis, corolle labio antico 3-partito lobo centrali rotundato-obovato retuso, tubo calycem equante. Lykipia. 115. Lucas, aff. L. stachydiformi, Benth. Kapté. 116. Lroworis RUGOSA, Benth. Kapté. 117. MICROMERIA PUNCTATA, Benth., var. Crater south of Lake Nairascha. 118. Cyaruuta, an C. globulifera ? Kapté. 119. ACHYRANTHES ARGENTEA, Lam. Kapté. 120. ÆRUA LANATA, Juss. Kapté. 121. Rumex ostUsIFoLIUS? (No leaves.) Kilimanjaro. 122. TAYMELACEA, dub. ; near Synaptolepis, but without hypo- gynous disk, and 8 faucial squame half as long as perianth- segments, entire or notched. (Fragmentarv.) Crater south of Lake Nairascha. 404 PROF. D. OLIVER ON THE PLANTS OF THE 193. Srrururota TuowsoNi, Oliv., sp. n., aff. S. ovate, Thunb. Frutex ramis erectis pilosulis superne dense foliatis, foliis enerviis laxe imbricatis verticillatis sessilibus ovato-lanceolatis ovalibusve subacutis concaviusculis paginis glabris pilosulo- ciliatis, bracteolis lineari-ovalibus conduplicatis folii dimidio brevioribus, floribus folio equilongis v. leviter exsertis, tubo peri- anthii glabro apice dilatato, limbi lobis ovato-lanceolatis acutis glandulis setis subequilongis lobis perianthii 4—5plo brevioribus. Lykipia. Folia 4—5 poll. longa. A Cape type, not previously represented from Tropical Africa in the Kew Herbarium. 124. CLUxTIA LANCEOLATA, Forsk.? (C. Richardiana, M. Arg., var. P) Lykipia. 125. JUNIPERUS PROCERA, Hochst. Tree of 80-100 feet. Lykipia. 126. Popocarrus ELONGATA, D’ Hér., forma? Tree of over 100 feet. Lykipia. 127. HABENARIA PLEISTADENIA, Reichb. f., sp. nov. Ultra spi- thamea, foliis basilaribus (geminis P) cuneato-oblongis apiculatis (0:07-0:02 m.), siccis bene nervosis, caule gracili solido (excl. inflorescentia 0:21 alto, inflorescentia 0°06), inferne calvo, medio pilis glandulosis sparsis, apicem versus ae apice copiosissimis vestito, vaginis distantibus lineari-lanceis acuminato-subulatis paucis glandipilibus, racemo densiusculo plurifloro dein secundi- floro, bracteis anguste lanceo-acuminatis glandipilibus ovaria glandipilia demum curvula non equantibus, sepalis triangulis obtusis, lateralibus sublongioribus, omnibus extus parce hinc glandipilibus, tepalis ligulatis acutis superne sepalum impar versus obtuse-angulatis, labello ligulato acuto basi utrinque falcula lineari-aeuminata diametrum lamins transversum non excedente, cruribus stigmaticis abbreviatis retusis, canalibus anthere ascendentibus brevissimis. Kilimanjaro 9000-10,000 feet. Affinis Habenarie crocee, Schweinf., et Guinganje, Rchb. f.; foliis ac pilis glandulosis abunde distincta.— H. G. Reichd. 128. H. Tuomsont, Reichb. f., sp. nov. Affinis Hubenarie humiliori, Rchb. f., ultra spithamea, foliis basilaribus oblongis, caulinis iu vaginas acutas abeuntibus, racemo densiusculo, bracteis MOUNTAINS OF EASTERN EQUATORIAL AFRICA. 405 oblongis acutis ovaria pedicellata equantibus, margine et disco externo muriculatis, ovariis pedicellatis sublevibus, sepalo impari oblongo obtuso, sepalis lateralibus inequalibus, margine inferiori extrorsum obtusangulo, apiculatis, tepalorum partitione superiore lineari margine muriculata, partitione labellum attingenti lancea, medio oblique plicata, saltem apicem versus margine minute muriculata, labelli partitionibus lateralibus lanceis acutis apice recurvis partitione mediana longiori lancea porrecta, calcare a basi teneriori antrorsum ampliato obtusiuscule acuto (semper medio flexo plicato P) ovarium pedicellatum non equaute. Lykipia, 6000—8000 feet. lllustri peregrinatori magno cum gaudio dicata. Ex grege Habenarie cultrate, A. Rich., Schimperi, Hochst., ete., tepalis ac calcare egregia.—H. G. Reichb. 129. Hapenarta, aff. H. nyikane, Reichb. f. Lykipia. 130. Hapenanta, aff. H. kilimanjari, Reichb. f. Ly kipia. 131. AwanmzcUM,sp. (No leaves.) Lykipia. 132. SATYRIUM, sp. Kilimanjaro. 133. Orcurpacea, dub. Kilimanjaro. 134. ARISTEA ALATA, Baker, sp. n.; foliis ensiformibus rigi- dulis margine hyalinis, caule e basi ad apicem conspicue alato foliis 2 reductis supra medium iustructo apice furcato, spath» valvis exterioribus lanceolatis dorso firmulis margine hyalinis, interioribus hyalinis integris pallide brunneis, fructu oblongo obtuse angulato, pedicello fructui equilongo vel paulo longiori. Lykipia, Masai country, alt. 6000-8000 feet. Folia basalia pedalia et ultra. Caulis 14-2 ped. Spathæ valve exteriores 7-8lin.longs. Capsula 4—4} lin. longa. Perianthium marcescens solum vidi. Easily distinguished by its long, distinctly winged stems and small oblong obtusely angled capsule with a long pedicel. The genus has its head-quarters at the Cape, with outliers in Angola, Abyssinia, and Madagascar.— J. G. Baker. 185. GLADIOLUS QUARTINIANUS, A. Rich. Lykipia. 136, G. (§ EUGLADIOLUS) WATSONIOIDES, Baker, sp. n. Elatus, grandiforus, foliis linearibus subcoriaceis glabris, caule tereti foliis 2-3 reductis instructo, spica laxa 6-8-flora unilaterali, spathæ valvis lanceolatis firmulis magnis inæqualibus, perianthii D Lad G LINN. JOURN.—- BOTANY, VOL. XXI. 406 ON PLANTS OF EASTERN EQUATORIAL AFRICA. splendide rubri tubo anguste infundibulari limbo longiori, limbi segmentis oblongis vel oblongo-lanceolatis, genitalibus limbo distinete brevioribus, Kilimanjaro. Folia 4-6 lin. lata, basalia sesquipedalia. Caulis (spica in- clusa) 3-4-pedalis. Spathe valva exterior 14-24 poll. Peri- - anthii tubus 18-20 lin. longus, ore 3 lin. diam., limbus 15-16 lin. longus, segmentis 4-7 lin. latis. This may be the plant gathered by Von der Decken and Kersten, which Dr. Klatt has referred (Bot. von Ost-Atrika, p. 73) to G. Garnierii ; but it is, I think, quite distinct speci- fically from the original Madagascar plant so called, which is the same that was distributed long ago by Hilsenberg and Bojer under the name of G. ignescens, and which I characterized under that name in ‘Trimen’s Journal, 1876, p. 334. It is quite different from the Kilimanjaro G. Newii, Baker in ' Trimen's Journ.’ 1876, p. 334.—J. G. Baker. 137. KwrPHoriA TuowsoNt, Baker, sp. n.; foliis linearibus acute carinatis margine scabris venis inter costam et marginem 14-16, racemo cylindrieo elongato, pedicellis fructu sæpe lon- gioribus bracteis lanceolatis acuminatis pedicello longioribus, perianthio cylindrico-infundibulari segmentis brevibus obtusis, staminibus inclusis, stylo demum exserto, fructu globoso magni- tudine pisi. Kilimanjaro. Folia pedalia et ultra, basi 6-S lin. lata. Racemus demum pedalis vel sesquipedalis, pedicellis 11-2 lin. longis. Perianthium 12-13 lin. longum, ore 2 lin. diam. Capsula 2 lin. diam. This comes nearest to the Cape and Natal K. sarmentosa, Kunth (Bot. Mag. t. 744), and the equatorial K. Grantii, Baker. Half a dozen species are now known in Abyssinia; and one has lately been discovered in Central Madagascar.—J. G. Baker. 188. RoMULEA CAMEROONIANA, Baker. Kilimanjaro. 139. Cvaworrs HIRSUTA, Fisch. & Mey., var. ? Lykipia. 140. Carex (fragment only), an C. Steudneri, Böckler, vel ' C. Kóstlini, Hochst. ? Kilimanjaro. MR. J. G. BAKER ON THE FLORA OF MADAGASCAR. 407 Further Contributions to the Flora of Ma gascar.—Second and Final Part. By J. G. Baxenr, F.R.S. [Read 15th January, 1885.] Iw the present paper are described the remainder of the novelties sent home by Mr. Baron up to the end of 1883. A report on the earlier half of the same set is printed in Journ. Linn. Soc. anted, pp. 317-353. Type specimens of all of them will be found in the Kew Herbarium, and a large number of them also . at the British Museum. Fine and extensive collections have lately been sent to Europe from the north-west of the island by M. Humblot; and distribution has been made of the large gatherings made in the central provinces by the late lamented Dr. Hildebrandt; so that our knowledge of Madagascar botany at the present time is advancing rapidly. As Dr. Hildebrandt and Mr. Baron have worked over the same ground, the plants they have obtained are very often identical; but Humblot’s are mostly different. MONOPETAL.E. ScHISMATOCLADA CONCINNA, D. Sp. Arborea, glabra, stipulis deltoideis, foliis breviter petiolatis obovato- cuneatis cuspidatis subcoriaceis, floribus in paniculas terminales peduncu- latas ramulis corymbosis dispositis, pedicellis brevissimis, calycis tubo campanulato dentibus deltoideis, corolle tubo cylindrico segmentis ob- longis tubo triplo brevioribus, staminibus segmentis zquilongis, fructu oblongo ad basin bifido valvis rigidulis. A. much-branched shrub, glabrous in all its parts, with slender terete branchlets. Leaves 11-2 in. long, narrowed from the middle to the base, moderately firm in texture, green and glabrous on both surfaces, with fine ascending main veins. Flowers copious, arranged iu terminal panicles, with flat-topped branches. Calyx jj in. long; teeth deltoid, as long as the tube. Corolla- Anthers linear, inserted at the throat of the tube 4 in. long. Capsule corolla-tube ; filaments filiform, equalling the anthers. _ 4 in. long, splitting down to the base into two valves.— Baron (without number). SCHISMATOCLADA VIBURNOIDES, n. 8p. Glabra, stipulis deltoideis, foliis confertis subsessilibus obovato-oblongis obtusis subcoriaceis, floribus in corymbos densos terminales pedunculatos LINN. JOURN.— BOTANY, VOL. XXI. 2H 408 MR. J. G. BAKER ON THE dispositis, pedicellis brevissimis, bracteis lanceolatis, calycis tubo campa- nulato dentibus parvis ciliatis deltoideis, corolle tubo elongato cylin- drico, segmentis ovato-oblongis tubo 5-6plo brevioribus, staminibus segmentis equilongis. An erect shrub, with straight branchlets, glabrous in all its parts; internodes very short; stipules deltoid. Leaves ascend- ing, subsessile, firm in texture, 3-4 in. long, 14-14 in. broad at the middle, narrowed from the middle to the base, green on both surfaces, with fine indistinct main veins. Flowers numerous, arranged in dense terminal corymbose cymes ; pedicels very short; bracts small, lanceolate, persistent, ciliated. Calyx zs in. long; teeth deltoid, ciliated. Corolla-tube ? in. long, hairy all down inside; segments } in. Anthers jin. long, ver- satile. Style reaching to the top of the corolla-tube, shortly bifid. Fruit not seen Baron 3220! DANAIS VESTITA, n. sp. Fruticosa, ramulis rectis dense fusco-pilosis, foliis petiolatis oblongis acutis subcoriaceis facie scabris dorso pubescentibus, floribus in cymas corymbosas axillares dispositis, pedicellis productis, bracteis minutis lan- ceolatis, calycis villosi segmentis lanceolatis tubo campanulato longioribus, corolle tubo cylindrico apice infundibulari, segmentis lanceolatis tubo quadruplo brevioribus, staminibus exsertis. A shrub, with straight terete branchlets densely clothed with spreading brown hairs. Stipules deeply lacerated, with ciliated lanceolate segments; petiole very short, densely pilose; blade 3-4 in. long, 12-2 in. broad, acute, rounded at the base, rough with short bristly hairs above, softly brown-pubescent beneath, with 8-10 pairs of raised arcuate-ascending main veins. Flowers in copious cymes in the axils of the leaves; pedicels 3-3 in long, densely villose, as are the bracts and calyx. Calyx with a globose tube and 5 lanceolate segments 4 in. long. Corolla- tube glabrous, 3-3 in. long, funnel-shaped at the apex; seg- ments lanceolate, + in. long. Stamens twice as long as the corolla-segments. Fruit not seen.— Baron 2329! PENTAS MICRANTHA, n. sp. " Herbacea, caulibus gracilibus glabris, stipulis parvis deltoideis, folis brevissime petiolatis oblongis acutis membranaceis, floribus in cymas capitatas terminales pedunculatas dispositis, pedicellis nullis vel brevissi- mis, calycis dentibus lanceolatis foliaceis inaqualibus tubo campanulato squilongis, corolla tubo deorsum cylindrico sursum infundibulari, seg- mentis parvis patulis ovato-lanceolatis, fructu globoso, seminibus in loculo pluribus minutis. FLORA OF MADAGASCAR. 409 Herbaceous, probably annual, with slender elongated glabrous stems. Leaves in distant pairs, shortly petioled, membranous, 2-3 in.long, narrowed from the middle to both ends, slightly pubescent. Flowers a dozen or more together in congested peduneled terminal cymes. Calyx finally 4 in. long, with 5 unequal lanceolate foliaceous teeth and a campanulate tube. Corolla with a tube 4 in. long, and an expanded limb not more than 4 in. in diam. Capsule globose, projecting at the apex from the calyx, splitting into two valves at the top, with very numerous minute seeds in each of the two cells.—Forests of Tanala pro- vince, Baron 310! also 3292! OLDENLANDIA LATIFOLIA, n. sp. Herbacea, annua, caulibus diffusis e basi copiose ramosis gracilibus obscure pilosis, foliis sessilibus oblongis parvis membranaceis obtusis vel subacutis, floribus axillaribus solitariis sessilibus, calycis dentibus lanceo- latis tubo campanulato squilongis, corollz albze segmentis oblongis tubo cylindrico zquilongis, capsulis oblongis, seminibus in loculo paucis oblongis nigris. An annual, with weak slender stems, diffusely branched from the erown of the root. Leaves oblong, membranous, sessile or nearly so, i-i in. long, slightly pilose; stipules entire, deltoid. Flowers solitary in the axils of the leaves all down the stem. Calyx hispid, finally 4 in. long, with 4 foliaceous lanceolate per- sistent teeth as long as the campanulate tube. Corolla white, 4 in. long. Capsule broad oblong, splitting down from the apex Seeds few in each cell, minute, à to the base into two valves. g, Also South Betsileo, Allied to the Cape oblong, with a black testa.— Baron 307 ! in the wood of Ankafina, Hildebrandt 3941 ! O. rupicola, Sonder. HEDYOTIS TRICHOGLOSSA, N. 8p. Fruticosa, glabra, ramulis gracilimis, stipulis parvis deltoideis, foliis petiolatis oblongis acutis membranaceis, floribus in cymas laxas terminales pedunculatas dispositis, pedicellis brevissimis, calycis dentibus 4 deltoideis tubo campanulato zquilongis, corolla albæ tubo cylindrico, segmentis lingulatis pilosis tubo vix brevioribus, antheris lineari-oblongis filamentis brevissimis, stylo flori zequilongo. A shrub, glabrous in all its parts, with very slender straight terete branchlets. Leaves shortly petioled, oblong, acute, deltoid at the base, about 2 in. long by under an inch broad, bright green, glabrous on both surfaces, distinctly pinninerved. Flowers in 202 410 MR. J. G. BAKER ON THE Y lax terminal peduncled corymbose cymes; pedicéls very short. Calyx jl in. long, glabrous; tube semiglobose. Corolla white, under 4 in. long, with 4 lingulate ascending segments nearly as long as the tube, and densely pilose inside. Style slender, forked, overtopping the anthers. Fruit not seen.— Baron 2782! MUSS.ENDA FUSCOPILOSA, n. Sp. Fruticosa, ramulis lignosis dense fusco-pilosis, stipulis lanceolatis, foliis petioletis oblongis subobtusis facie parce dorso magis fusco-pilosis, flori- bus paucis terminalibus subsessilibus umbellatis, bracteis lanceolatis, calycis magni segmentis lanceolatis tubo hirsutissimo longioribus, corolle tubo cylindrico dense piloso 3-4-pollicari, segmentis oblongo-lanceolatis tubo quadruplo brevioribus, genitalibus in tubo inclusis. A shrub or tree, with terete woody branches, clothed with spreading bright-brown hairs. Leaves 4-5 in. long, moderately firm in texture, deltoid at the base, green and slightly hairy above, more hairy beneath, with 12-14 pairs of arcuate-ascending veins. Flowers 3—4 together in terminal nearly sessile umbels. Flower- calyx i-i in. long, the lanceolate teeth exceeding the densely pubescent oblong tube. Corolla-tube densely pubescent, 3-32 in. long; segments patent, #-1 in. long. Stamens and style included in the corolla-tube. Fruit not seen.—Baron 2467! 2470 ! MUSSÆNDA MACROPODA, n. sp. Erecta, fruticosa, ramulis apice pilosis, foliis longe petiolatis oblongis obtusis vel subacutis facie glabrescentibus dorso tenuiter pilosis, floribus copiose corymboso-paniculatis, pedicellis brevibus, calycis sericei dentibus lanceolatis tubo turbinato equilongis, corolla tubo dense sericeo sesqui- pollicari, segmentis oblongis cuspidatis tubo 3-4plo brevioribus. An erect shrub or tree, with woody branchlets, thinly pubes- cent towards the tips. Leaves oblong, moderately firm in texture, 3-4 in. long, green and finally glabrous above, deltoid at the base, obscurely pubescent beneath, with 8-9-jugate distinctly raised arcuate-ascending veins. Flowers in copious terminal shortly- peduncled corymbose panicles; bracts rigid, lanceolate, per- sistent ; pedicels often 3-1 in, long. Flower-calyx 1-$ in. long, the lanceolate-acuminate teeth equalling the densely grey-silky ovary. Corolla with a densely silky tube 13-12 in. long; seg- ments oblong-cuspidate, J in. long. Stamens and style not exserted from the corolla-tubes.— Baron 3088! Allied to M. trichophlebia, Baker in Journ. Linn. Soc. xx. p. 166, and the Mauritian M. Landia, Lam. FLORA OF MADAGASCAR. 411 TARENNA (§ WEBERA) MACROCHLAMYS, n. sp. Arborea, ramulis crassis apice ferrugineo-pubescentibus, stipulis magnis rigidis caducis, foliis petiolatis obovatis emarginatis coriaceis magnis facie glabrescentibus dorso venulosis ferrugineo-pubescentibus, floribus copiosis in panieulas terminales ramis cymosis primum bracteis magnis ovatis coria- ceis cinctas dispositis, pedicellis sepissime productis brevibus, calycis tubo globoso, limbo brevi truncato, corolle tubo brevi cylindrico, seg- mentis oblongis tubo zquilongis, fructu globoso levi magnitudine pisi. An erect tree, 30—40 feet high, with stout terete branchlets, clothed upwards, as are the branches of the panicle and under- side of the leaves, with short ferruginous pubescence. Stipules large, rigid, orbicular, spathulate. Leaves thick and rigid in texture, 4-6 in. long and nearly as broad, prominently emargi- nate at the apex, all the veins and veinlets conspicuously raised beneath. Panicle ample, terminal, with very numerous divari- cated dichotomously cymose branchlets. Calyx-tube coriaceous, semiglobose, 74; in. in diam.; limb short, collar-like, truncate. Corolla 4 in. long. Fruit indehiscent, coriaceous, black, smooth, l in. in diam., with several tightly-packed seeds in each cell.— Baron 423 11941! 1956! Forest of Andrangaloaka, Dr. G. W. Parker! PrEcTRONIA (§ CANTHIUM) BUXIFOLIA, n. sp. Fruticosa, glabra, stipulis parvis deltoideis, foliis breviter petiolatis parvis oblongis lucidis subcoriaceis, floribus in umbellas multas axillares paucifloras dispositis, pedicellis calyce 3-4plo longioribus, calycis tubo campanulato dentibus minutis deltoideis, corolla tubo infundibulari seg- mentis ovatis, fructu duro compresso didymo conspicue emarginato. An erect shrub, glabrous in all its parts, with slender straight branchlets. Leaves firm in texture, bright green, 1-13 in. long, deltoid at the base, shining on both surfaces, the veining fine and indistinct. Flowers 4-6 together in umbels in the axils of the leaves all down the branchlets ; pedicels }-j in. long. Calyx- tube globose, green, yy in. in diam. Corolla 4 in. long. Fruit hard, didymous, compressed, 3 in. in diam., with a deep groove between the two oblong carpels.— Baron 274! 965! 1019! 2177! 2213! 3137! PrrcrRoxrA (§ CanrHiuM) BOIVINIANA, n. 8p. Arborea, glabra, stipulis parvis deltoideis, foliis breviter petiolatis ob- longis acutis subcoriaceis, floribus in umbellas multas axillares paucifloras dispositis, pedicellis calycem floriferum 2-3plo superantibus, calycis tubo 412 MR. J. G. BAKER ON THE campanulato dentibus minutis deltoideis, corolle tubo infundibulari seg- mentis brevibus, fructü duro compresso didymo conspicue emarginato. A tree, glabrous in allits parts, with slender straight terete branchlets. Leaves moderately firm in texture, 2-3 in. long, deltoid at the base, with distant fine ascending main veins. Flowers 2—4 together from the axils of the leaves all down the branchlets; pedicels finally 1 in. long. Flower-calyx campanu- late, J in. indiam. Fruit hardy, didymous, compressed, conspi- euously emarginate, J in. in diam, sometimes one of the carpels aborted. —Baron 2942! 3071! Also gathered by Boivin. Allied to the Seychelles P. acuminata, Baker. IxoRA EMIRNENSIS, n. sp. Fruticosa, glabra, stipulis parvis deltoideis, foliis petiolatis oblongis acutis subcoriaceis, floribus in paniculas terminales pedunculatas ramulis corymbosis dispositis, bracteis minutis deltoideis, pedicellis brevissimis, calycis tubo campanulato dentibus deltoideis, corolla segmentis ovato- lanceolatis parvis tubo cylindrico 4-5plo brevioribus, filamentis brevissimis, fructu globoso magnitudine pisi. An erect shrub, 10-15 feet high, glabrous in all its parts; with slender terete branchlets. Leaves shortly petioled, 3—4 in. long, 1-14 in. broad, deltoid at the base, green and glabrous on both sides, with copious fine venation. Panicles terminal, distinctly peduncled, both end and side branches level-topped. Flower-calyx with a eampanulate tube -} in. in diam. and 4 small deltoid teeth. Corolla with a slender cylindrical tube 3 in. long and an expanded limb 1 in. in diam. Stamens shorter than the segments ; anthers linear, versatile, +4, in. long; filaments very short. Fruit brown, globose, coriaceous, i in. in diam., crowned by the persistent calyx, with one hemispherical seed filling each of the two cells.— Baron 1247 ! 2228! Forest of Andrangaloaka, Dr. G. W. Parker! Alied to the Seychelles T. pudica, and much inferior in floral effect to J. odorata, Hook., which Mr. Baron has twice collected (188! 1444 !). PSYCHOTRIA ($ GRUMILEA) MESENTERICARPA, n. Sp. Glabra vel subglabra, ramulis subcompressis, stipulis deltoideis euspi- datis coriaceis caducis, foliis breviter petiolatis obovatis cuspidatis sub- coriaceis, floribus in panieulam terminalem ramulis corymbosis dispositis, pedicellis brevibus, calycis tubo campanulato limbo brevissimo obscure dentato, corollae tubo infundibulari segmentis ovatis, fructu globoso nigro magnitudine pisi parvi, seminibus sulcatis albumine ruminato. se FLORA OF MADAGASCAR. 413 A tree, with slightly compressed woody glabrous or obscurely pubescent branchlets. Leaves shortly petioled, obovate, deltoid at the base, rounded to a cusp at the apex, 3—4 in. long, firm in texture, green and glabrous on both surfaces, with 8-10 pairs of distinct arcuate-ascending main veins. Flowers in short- peduncled deltoid terminal panicles. Calyx obconie, ps in. in diam., with a short collar-like limb. Corolla iin. long, with 5 small ovate segments. Stamens and style just exserted from the throat of the corolla-tube. Fruit black, globose, 4 in. in diam., crowned by the persistent calyx-limb. Seeds one filling each cell. — Baron 851! 1240! 2969! 2995 ! 3015! PSYCHOTRIA LUCIDULA, n. sp. Arborea, glabra, stipulis deltoideis, foliis magnis petiolatis obovato- oblongis cuspidatis subcoriaceis, floribus parvis copiosis in paniculam ter- minalem ramulis corymbosis dispositis, pedicellis brevibus, bracteis minutis deltoideis persistentibus, calycis tubo campanulato dentibus 5 minutis deltoideis, corollae tubo infundibulari segmentis ovato-lanceolatis tubo duplo brevioribus, fructu ovoideo glabro brunneo ruguloso, seminibus profunde sulcatis. A tree, glabrous in all its parts, with slender straight rather compressed branchlets. Leaves opposite, distinctly petioled, 4-6 in. long, 2-22 in. broad, deltoid at the base, acute or obtuse, with a cusp at the apex, pale green and quite glabrous on both surfaces, rather glossy beneath, with 9-10 pairs of distinct arcuate-ascending main veins. Flowers in a peduncled broad short end-panicle. Calyx campanulate, j; in. in diam; teeth minute. Corolla Zin. long; expanded limb j in. in diam., with the stamens just exserted from the throat of the tube. Fruit ovoid, brown, 4 in. long, rather compressed, glabrous, rough, with a distinct groove between the two carpels. Seeds solitary, erect, with the epicarp projecting into five longitudinal furrows down their face.— Baron 1285! 2699! GxoPHILA GERRARDI, n. 8p. G. caulibus filiformibus late reptantibus pubescentibus, foliis longe petio- latis reniformibus facie viridibus glabris dorso glaucescentibus venis pubes- centibus, floribus paucis in capitulum terminalem pedunculatum bracteis foliaceis oblongis aggregatis, calycis tubo obconico dentibus minutis del- toideis, corolle segmentis ovato-lanceolatis tubo infundibulari duplo bre- vioribus, staminibus segmentis brevioribus. A herbaceous perennial, with slender wide-trailing pubescent 414 MR. J. G. BAKER ON THE stems. Petiole 1-21 in. long; stipules simple, deltoid ; blade 1-11 in. long, moderately firm in texture, nearly as broad as long, ` furnished with two semicircular basal auricles, bright green and glabrous on the upper suríace, whitish beneath, with brown slightly pubescent veins. Flowers several together, sessile in a peduncled terminal head, surrounded by an involucre of distinct imbricated pilose oblong foliaceous bracts Lin. long. Calyx shorter than the bracts, pubescent. Corolla under 4 in. long, pubescent. Stamens shorter than the corolla-segments ; anthers linear-oblong ; filaments very short. Fruit not seen,— Gathered long ago by Gerrard (102!), and now refound by Baron 2444! and Humblot 133 : HozrocAzPA, genus novum ordinis Rubiacearum tribus Anthospermearum. Flores hermaphroditi. Calycis tubus parvus, campanulatus; segmenta 5, raro 6, magna lanceolata foliacea inzqualia. Corolla hypocrateri- morpha, tubo cylindrico fauce villosa, segmentis oblongis patulis. Ova- rium 3-5-loculare, ovulis in loculo solitariis erectis; stylus elongatus fili- formis apice bifidus. Stamina sepissime 5, infra faucem corollx tubi inserta, filamentis leviter applanatis, antheris oblongis. Fructus inde- hiscens, seminibus in loculo solitariis.— Herba perennis Madagascariensis, foliis oppositis subsessilibus obovato-oblongis, stipulis foliaceis, floribus in cymas densas terminales breviter pedunculatas dispositis. HOLOCARPA vERONICOIDES, Baker. A perennial herb, with the habit of a Hedyotis or of Veronica officinalis, with short, trailing or spreading, slightly pubescent, slender terete stems. Leaves in opposite pairs, sessile, oblong or oblanceolate-oblong, 3-} in. long, obtuse, deltoid at the base, green and glabrous on both surfaces, moderately firm in texture. No veins except the midrib obvious ; stipules large, foliaceous, persistent, linear or lanceolate, simple or compound, ciliated. Flowers few together, in congested corymbose peduncled ter- minal cymes; pedicels none or very short. Corolla 4 in. long, with a globose tube; segments lanceolate, foliaceous, unequal. Corolla lilac; tube 4 in. long, cylindrical downwards, funnel-shaped at the top; segments oblong cuspidate, iin. long. Stamens about as long as the segments ; anthers linear-oblong; filaments a little flattened, longer than the anthers. Style filiform, about as Jong as the corolla-tube, bifid at the apex. Fruit globose, inde- hiscent, $ in. in diam., crowded by the large leafy calyx-segments ; FLORA OF MADAGASCAR. 415 cells 8-5, with a single seed in each.— Allied to Otiophora, from which it differs by its hermaphrodite, usually pentamerous flowers, corolla-tube villose at the throat, and indehiscent 3—5.celled fruit. It was collected long ago in flower by Dr. Lyall, and has now been refound by Mr. Baron 736! and Dr. Hildebrandt, no. 3848! VERNONIA POLYTRICHOLEPIS, n. sp. Fruticosa, ramulis rectis dense pubescentibus, foliis subsessilibus parvis oblongis integris rigide coriaceis facie viridibus obscure pubescentibus dorso persistenter albo-incanis, capitulis parvis 6-8-floris in paniculam ramulis glomerato-corymbosis dispositis, involucro brevi campanulato im- bricato bracteis obtusis densissime villosis, achenio piloso, pappo albido setis firmulis persistentibus. A much-branched shrub, with terete slender branchlets, densely clothed with short whitish pubescence. Leaves about an inch long, nearly sessile, subacute, rounded at the base, green and finally nearly glabrous above, matted beneath with brownish- white persistent tomentum. Panicle terminal, 2-3 in. long and broad, the very numerous sessile heads crowded in clusters at the top of the branches. Involucre campanulate, j in. long; bracts obtuse, densely clothed with short whitish woolly pu- bescence, the outer very small. Corolla jin. long; segments very short. Pappus as long as the corolla-tube; bristles firm, persistent, distinctly eiliated.— Baron 2337! 2530! Allied to V. Lyallii, Baker, in Journ. Linn. Soc. xx. p. 174. VERNONIA VOLUTA, n. sp. Fruticosa, volubilis, ramulis gracilibus pubescentibus cito calvatis, foliis petiolatis oblongis acutis integris subcoriaccis facie glabratis dorso brunneo- pubescentibus, capitulis parvis multifloris in paniculas laterales rhachi valde flexuoso ramulis corymbosis dispositis, involucro campanulato piloso brac- teis paucis lanceolatis acutis subzequilongis, pappo albo firmulo. A climber, with slender terete woody stems, clothed at first with short brown pubescence. Leaves 2-3 in. long, acute, rounded at the base, subcoriaceous, green and glabrescent above, coated beneath with persistent brown tomentum. Heads arranged in copious axillary panicles, longer than the leaves, with a very zigzag rhachis, clothed with short tomentum, as are the corymbose spreading or reflexed branches. Involucre campanulate, 1 in. long, pilose, the lanceolate acute brown bracts scarcely more than uniserial. Flowers about 20 in a head. Pappus of white 416 MR. J. G. BAKER ON THE moderately firm bristles.— Baron 2375! Allied to V. apocyni- folia, Baker in Journ. Linn. Soc. xx. p. 175. VERNONIA STREPTOCLADA, n. Sp. Fruticosa, volubilis, ramulis gracilibus pubescentibus cito calvatis, foliis parvis petiolatis ovatis szepissime acutis utrinque molliter brunneo-pubes- centibus, eapitulis parvis multifloris in paniculas laterales vel terminales rhachi brunneo-velutino insigniter flesuoso ramulis corymboso-glomeratis dispositis, involucro brevi campanulato piloso bracteis pauciseriatis rigidis obtusis, achenio glabro, pappo albido ex involucro exserto. A climber, with terete woody stems, clothed at first with short brown pubescence. Leaves 1-14 in. long, nearly as broad as long, rounded at the base, rarely obtuse, not at all rigid in texture, coated on both sides with soft brown pubescence. Heads arranged in panicles, with a very zigzag pubescent rbachis, the lower branches reflexed. Involucre } in. long; bracts all rigid and obtuse. Flowers 20 or more in a head. Achene gla- brous; pappus of moderately firm persistent whitish bristles, 4 in. long—Baron 3041! 3076! Dr. Parker! Allied to V. voluta and. V. apocynifolia. Vernonia (§ DISTEPHANUS) TRICHANTHA, n. sp. Fruticosa, ramulis rectis dense incanis sursum villosis, foliis parvis sub- sessilibus oblongis integris coriaceis trinervatis utrinque albido-incanis, capitulis paucis multifloris corymbosis glomeratis, involucro magno cam- pauulato bracteis pauciseriatis adpressis lanceolatis dense albido-incanis, floribus extus apice villosis, achenio piloso, pappo rigidulo pulehre rubello. A shrub, with slender straight woody branchlets, clothed with thin persistent whitish tomentum, or towards the top with thicker, looser pubescence. Leaves thick and rigid in texture, not crowded, 1-14 in. long, three-nerved nearly from the base to the tip, matted on both sides with whitish tomentum. Heads few, crowded in dense terminal eorymbs. Involucre campanulate, nearly 3 in. in diam.; bracts lanceolate, acute, adpressed, densely clothed with white arachnoid pubescence. Corolla 4 in. long, densely villose outside towards the tip; segments oblong-lanceo- late, 4 as long as the tube. Pappus pale scarlet, rigid, as long as the corolla-tube.— Baron 607! Allied to V. ochroleuca and V. inulafolia, Baker in Journ. Linn. Soc. xx. pp. 179, 180. FLORA OF MADAGASCAR. 417 APODOCEPHALA, genus novum Compositarum tribus Eupatoriacearum. Capitula homogama 3-4-flora, floribus omnibus tubulosis hermaphro- ditis. Involucrum oblongum, bracteis 10-12 rigidis adpressis, exterioribus sensim brevioribus, extimis ovatis, intimis oblongis. Receptaculum nudum. Corolle zquales, regulares, tubo cylindrico glanduloso, seg- mentis oblongo-lanceolatis tubo squilongis. Antherz apice connectivo dilatato deltoideo appendiculate, basi auriculis acutis predite. Stylus profunde bifidus, ramis falcatis linearibus obtusis ad basin stigmatosis. Achenia linearia, angulata, glabra, apice cupulo obscuro solum coronata ; pappus nullus.—Frutex Madagascariensis, ramulis robustis lignosis, foliis alternis petiolatis oblongis subcoriaceis dorso pubescentibus, capitulis per- multis parvis glomeratis in paniculam amplam latam dispositis. APODOCEPHALA PAUCIFLORA, Baker. A large shrub, with the habit of a Vernonia or Eupatorium, with stout terete woody branchlets. Leaves oblong acute, rounded at the base, 4—5 in. long, subcoriaceous in texture, green, glabrous and rather lucent above, clothed with persistent short brown pubescence beneath, with 10-12 distinct often forking pairs of rather ascending main veins; petiole above an inch long. Panicle terminal, 6-8 in. long and broad, the woody branches clothed with short brown pubescence. Capitula glomerate in dense congested corymbs at the end of the branchlets, sessile. Involuere 4 in. long; bracts drab-brown, rigid, the outer grow- ing gradually smaller. Achene glabrous, j in. long, strongly angled, dilated into an obscure cup without any bristles or paleze at the apex. Corolla 3 in. long, with a cylindrical glandular tube and 5 oblong-lanceolate segments as long as the tube. Stamens and style much shorter than the corolla-segments.— Baron 8251! Allied to Ageratum and Carelia. HELICHRYSUM LEUCOSPHERUM, n. sp. Herbaceum, perenne, ramulis foliisque persistenter albo-incanis, foliis sessilibus lanceolatis integris erectis uninerviis, capitulis parvis sessilibus paucifloris in glomerulum globosum terminalem aggregatis, involucro in- fundibulari bracteis pauciseriatis lanceolatis erectis adpressis acutis, pappo albido. A perennial herb, with slender erect stems coated with per- sistent white tomentum. Leaves sessile, erect, crowded, lanceo- late, 1-14 in. long, tapering to a point, coated on both sides with persistent tomentum like that of the stem. Heads very numerous, 418 MR. J. G. BAKER ON THE small, sessile, aggregated in globose terminal clusters 2-4 in. in diam. Involucre } in. long, villose at the base, composed of a few rows of pure white rigid erect bracts of moderately firm tex- ture. Pappus of white bristles.—Baron 2611! HELICHRYSUM XYLOCLADUM, n. sp. Fruticosum, ramulis lignosis albido-tomentosis, foliis parvis oblongis obtusis coriaceis integris sessilibus facie tenuiter dorso dense albido- tomentosis, capitulis 6—7-floris in paniculam amplam latam ramulis divari- catis dispositis, involuero magno obconico multiseriali albido-tomentoso, bracteis apice scarioso patulo orbiculari przeditis, pappo albido. A shrub, with straight woody branchlets clothed with per- sistent whitish tomentum. Leaves under an inch long, obtuse, cuneate at the base, very thick and rigid in texture, clothed thinly above and more densely beneath with whitish persistent tomentum like that of the stem. Capitula very numerous, arranged ina panicle 6-8 in. long and broad, with divaricated corymbose branches. Involucre ł4—} in. long, tomentose in the lower part; all the bracts rigid downwards, furnished with a spreading chartaceous white orbicular tip, the outer ones gra- dually shorter. Receptacle flat, very small. Pappus of white eiliated bristles.— Baron 3208! 3324! l MELANTHERA MADAGASCARIENSIS, N. Sp. Herbacea, sarmentosa, ramulis gracilibus hispidis, foliis petiolatis lan- ceolatis membranaceis utrinque hispidulis, capitulis radiatis laxe corym- bosis, pedunculis pedicellisque erectis elongatis, involucro brevi campanu- lato bracteis subæquilongis lanceolatis, ligulis foemineis, achænio obconico angulato apice truncato setis pluribus fragilibus coronato. A sarmentose perennial herb, with slender angled slightly hispid stems. Leaves opposite, 3-4 in. long, obscurely dentate, cordate or truncate at the base, green and scabrous above, pale green and more hispid beneath; petiole 1—L in. long. Heads arranged in a very lax corymb on long slender erect peduncles and pedicels. Unexpanded heads globose, 4 in. in diam., with the rigid ovate-lanceolate paleæ of the receptacle exceeding the young corollas. Involucre broadly campanulate, 3 in. in diam., à in. long; bracts lanceolate, subequal, almost foliaceous, very hispid. Ligules 15-20, bright yellow, 3 in. long, with a perfect style but no stamens. Achene angled, truncate at the apex, ys in. long; bristles 6-10, fragile, caducous, as long as the achene.— Baron 2344! 2534! Humblot 410! A new genus for the island. Three species are known in Tropical Africa. ee FLORA OF MADAGASCAR. 419 SENECIO PURPUREO-VIRIDIS, n. sp. Fruticosus, ramulis gracilibus pubescentibus, foliis multis sessilibus am- plexicaulibus lyrato-pandurzformibus profunde inciso-crenatis medio cito angustatis basi auriculatis facie viridibus glabratis dorso purpureis pubes- centibus, capitulis paucis magnis corymbosis ligulatis, involucro campanu- lato bracteis lanceolatis, ligulis luteis, achenio glabro, pappo firmulo albo. A shrub, with long slender straight copiously leafy pubescent branchlets. Leaves moderately firm in texture, 13-2 in. long, the oblong obtuse deeply-toothed upper half 4 in. broad, narrowed suddenly halfway down into a repand wing to the midrib, which is dilated at the base into 2 round amplexicaul auricles. Heads few, arranged in a terminal corymb ; peduncles ascending, pubescent, their upper leaves reduced to minute entire lanceolate bracts. Involuere campanulate, 3 in. in diam. ; bracts about 12, pubescent on the back. Ligules 3 in. long. Disk-flowers very numerous. Achene glabrous, jl; in. long. Pappus 4 in. long; bristles pure white, ciliated, firm in texture for the genus.— Baron 3264! Nearest S. adenodontus, of the species described in the ‘ Prodromus.’ ARDISIA MYRIANTHA, n. Sp. Arborea, foliis breviter petiolatis oblanceolato-oblongis obtusis glabris rigide coriaceis venulis subtilibus, floribus in panieulas axillares breviter pedunculatas ramulis racemosis divaricatis dispositis, pedicellis brevibus pubescentibus, bracteis minutis deltoideis, calycis tubo brevissimo seg- mentis orbicularibus imbricatis, corollae tubo brevissimo segmentis orbi- cularibus, antheris parvis subsessilibus. An erect shrub or small tree. Leaves 4-5 in. long, 1-1} in. broad, narrowed gradually from the middle to the base, thick and rigid in texture, the veins fine and indistinct. Panicles shorter than the leaves ; branches few and spreading ; rhachises finely pubescent ; pedicels at most as long as the flowers; bracts minute, persistent. Expanded calyx yẹ in. in diam.; segments 5, orbieular, much imbrieated. Corolla 75 in. long. Anthers not more than half as long as the corolla-segments. Ovary orbicular ; style very short. Fruit not seen.— Baron 2312! Nearly allied to A. bipinnata, Baker, in Journ. Linn. Soc. xx. p. 201. ARDISIA OLIGANTHA, n. 8p. Fruticosa, glabra, ramulis gracilibus, foliis parvis oblongis subobtusis breviter petiolatis subcoriaceis, floribus in corymbos axillares peduncu- latos paucifloros dispositis, pedicellis elongatis, bracteis minutis, calycis tubo 420 MR. J. G. BAKER ON THE brevissimo segmentis ovatis, corolle tubo brevissimo segmentis orbicu- laribus, antheris ad faucem tubi sessilibus. A shrub, glabrous in all its parts, with very slender branchlets. Leaves 2-11 in. long, deltoid at the base, moderately firm in texture, green and glabrous on both surfaces, the veins fine and immersed ; petiole 4-1 in. long. Corymbs 2-3-flowered, about as long as the leaves; pedicels very slender, 7-3 in. long. Ex- panded calyx +; in. in diam. ; spreading segments twice as long as the tube. Corolla 4 in. long. Anthers deltoid, half as long as the corolla-segment. Ovary globose; style short. Fruit not seen.— Baron 2918! ARDISIA ? MACROSCYPHA, n. sp. Fruticosa, glabra, ramulis gracilibus teretibus, foliis subsessilibus ob- longis obtusis subcoriaceis, floribus axillaribus solitariis, pedicellis elon- gatis, calycis tubo campanulato segmentis ovatis, corolla tubo oblongo segmentis semiorbicularibus tubo brevioribus, antheris magnis acutis prope basin corollz tubi insertis. An erect shrub, glabrous in all its parts, with slender subterete branchlets. Leaves obscurely petioled, 14-2 in. long, firm in texture, green and glabrous on both surfaces, with fine indistinct veining. Flowers in the only specimen seen solitary from the axils of the upper leaves on slender ascending pedicels above an inch long, articulated above the middle. Calyx jy in. in diam. ; segments broad ovate, obtuse, ciliated, twice as long as the tube. Corolla 1 in. long, with 5 semiorbicular segments, not more than half as long as the tube. Anthers 4 in. long, inserted low down in the corolla-tube. Ovary globose; style cylindrical, jg in. long.—BHaron 9978! Differs from Ardisia by its short corolla- lobes and stamens inserted low down in the corolla-tube, and may perhaps claim to be considered a new genus. ARDISIA UMBELLATA, n. sp. Fruticosa, glabra, foliis oblongis acutis breviter petiolatis subcoriaceis, floribus in cymas axillares paucifloras pedunculatas dispositis, pedicellis elongatis, calycis segmentis ovatis tubo campanulato aquilongis, corollze tubo brevissimo segmentis orbicularibus cuspidatis, antheris parvis ovatis ad faucem tubi pene sessilibus. An erect much-branched shrub, glabrous in all its parts, with slender subterete branchlets. Leaves 2-3 in. long, deltoid at the base, moderately firm in texture, green and glabrous on both surfaces, the veins fine and indistinct. Flowers 2-6 in copious FLORA OF MADAGASCAR. 421 axillary umbels; pedicels 1 in. long; peduncles very slender, shorter than the leaves. Calyx jin. in diam. ; segments 5, obtuse. Corolla $ in. long; segments much imbricated in bud, copiously black-dotted. Stamens less than half as long as the corolla- segments. Ovary globose; style cylindrical, as long as the ovary. Fruit not seen.— Baron 2938! ARDISIA LONGIPES, n. sp. Fruticosa, glabra, foliis petiolatis obovato-oblongis obtusis subcoriaceis, floribus in racemos laxos pedunculatos dispositis, pedicellis elongatis, calycis tubo brevi segmentis ovatis obtusis, corolla tubo brevissimo segmentis orbicularibus cuspidatis, antheris ad faucem coroll: tubi subsessilibus. An erect much-branched shrub, glabrous in all its parts, with slender subterete branchlets. Leaves 2-3 in. long, 1-14 in. broad, narrowed gradually from the middle to a deltoid base, moderately firm in texture, green and glabrous on both surfaces, the veins fine and immersed. Flowers in copious racemes, some- times congested into corymbs, from the axils of the leaves, on slender peduncles 1-13 in. long; pedicels ascending, 3—4 in. long ; bracts 0, or very minute. Expanded calyx rotate, 4 in. in diam. Corolla 3 in. long ; segments much imbricated. Stamens half as long as the corolla-segments. Style cylindrical, as long as the globose ovary. Fruit not seen.— Baron 2224! ONCOSTEMUM PLATYCLADUM, n. sp. Fruticosum, glabrum, ramulis compressis acute ancipitibus, foliis bre- vissime petiolatis oblongis acutis subcoriaceis utrinque viridibus, floribus in corymbos paucifloros breviter pedunculatos dispositis, pedicellis elon- gatis, calycis et corollae tubo brevissimo segmentis orbicularibus, filamentis in urceolum parvum ampullaforme ore angusto coalitis. An erect shrub, glabrous in all its parts, with remarkably flattened acutely-angled branchlets. Leaves 3-4 in. long, narrowed gradually from the middle to both ends, moderately firm in texture, bright green on both surfaces, with fine immersed veins. Flowers about three to a corymb ; peduncle short, anci- pitous; bracts none; pedicels very slender, 1-1 in. long. Ex- panded calyx à in. in diam. Expanded corolla 4 in. in diam. Staminal urceolus not more than half as long as the corolla; the anthers, ovary, and style included inside it. Fruit not seen.— Baron 2882! ONCOSTEMUM NERIIFOLIUM, n. Sp. Fruticosum, glabrum, ramulis teretibus, foliis petiolatis oblanceolato- 422 MR. J. G. BAKER ON THE oblongis obtusis rigide coriaceis, floribus in corymbos axillares peduncu- latos dispositis, pedicellis elongatis, calycis tubo brevi segmentis orbicu- laribus, corollee tubo brevissimo segmentis orbicularibus, filamentis in urceolum fauce breviter lobata connatis. An erect shrub or small tree, glabrous in all its parts. Branch- lets terete, straight, not very slender. Leaves 3-4 in. long, i-1l in. broad above the middle, narrowed gradually from the middle to the base, very thick and rigid in texture, green and glabrous on both surfaces, the veins fine, crowded, slightly raised. Corymbs usually shorter than the leaves, 4—8-flowered ; pedicels 3-3 in. long; bractsO. Expanded calyx jin. in diam. Corolla ig in. long. Filaments united in a globose urceolus half as long as the corolla-segments, inside which the stamens, ovary, and style are included. Fruit not seen.— Baron 3036! ONCOSTEMUM VENULOSUM, n. sp. Arbusculum, glabrum, ramulis gracilibus teretibus, foliis breviter petiolatis oblanceolato-oblongis acutis subcoriaceis utrinque viridibus venulis ex- sculptis, floribus in corymbos paucifloros pedunculatos axillares dispositis, pedicellis elongatis, calycis tubo brevissimo segmentis orbicularibus imbri- catis, corollae tubo brevissimo segmentis ovatis, filamentis in tubo brevi lato segmentis rotundatis coalitis. An erect shrub, glabrous in all parts, with slender branches not at all flattened. Leaves 2-3 in. long, 2-1 in. broad, narrowed from the middle to base, firm in texture, bright green on both surfaces, all the veins and veinlets distinctly raised; petiole i-i in. long. Flowers 2-4 in a corymb ; peduncle about an inch long; pedicels 4-3 in.; bracts absent. Calyx j in. in diam., very much dotted with black ; segments orbieular, broader than long. Corolla j in. long, very much dotted with black. Staminal urceolus less than half as long as the corolla-segments, with rounded lobes, the acute tips of the anthers protruding from its open throat. Ovary globose; style short, cylindrical; stigma capitate. Fruit not seen.— Baron 2986! 2997! Andrangaloaka, Hildebrandt 4083 ! Diospyros FUSCO-VELUTINA, n. sp. Arborea, ramulis virgatis glabris, foliis brevissime petiolatis oblongis magnis rigide coriaceis lucidis, pedicellis brevissimis lateralibus solitariis, calyce fructifero segmentis 5 ovatis crassis rigide coriaceis utrinque fusco- velutinis, fructu globoso cuspidato dense persistenter fusco-velutino calyci equilongo. l A tree, with wand-like glabrous dark rather shining chestnut- FLORA OF MADAGASCAR. 423 brown branchlets. Leaves 6-8 in. long, about 3 in. broad, acute, rounded at the base, very thick and rigid in texture, glossy above, with 10-12 pairs of distinct erecto-patent raised main veins, uniting in arches before they reach the margin. Pedicels stout, solitary, densely pilose, }-3 in. long. Fruit-calyx campanulate, 1-14 in. in diam., thick and rigid in texture, densely brown-velvety both inside and out; spreading tube 4 in. long; segments ovate, acute, žin. long and broad. Fruit woody in texture, 3 in. in diam., densely brown-velvety, like the calyx.— Baron 2361 ! Drosrynos MEGASEPALA, n. sp. Arborea, glabra, foliis petiolatis magnis oblongis acutis rigide coriaceis, pedieellis lateralibus longissimis fastigiatis gracilibus interdum furcatis, calyce fructifero segmentis 4 late ovatis acutis magnis rigide coriaceis, fructu glabro ovoideo cuspidato magnitudine nucis juglandis. An erect tree, glabrous in all its parts, with terete branchlets i-lin.indiam. Leaves nearly a foot long, 3-4in. broad, rounded at the base, rigid in texture, dark drab on both sides in the dried state, with distant arcuate-ascending main veins, connected by fine raised eross veinlets. — Flowers in a cluster from the side of thiek branchlets; pedicels 5-6 in. long, slender, thiekened up- wards, in one instance forked. Fruit-calyx 23 in. in diam., rigidly coriaceous in texture, drab, marked with copious raised ana- stomosing veins; united portion an inch long ; segments 4, about 2 in. long and broad, covered with white meal inside. Fruit woody in texture, mealy, above an inch long, quite hidden by the large calyx.— Baron 2365 ! DrosPYROS SPHJEROSEPALA, n. 8p. Fruticosa, glabra, ramulis ultimis ancipitibus, foliis sessilibus oblongis rigide coriaceis, floribus foemineis solitariis lateralibus brevissime pedun- culatis, calycis tubo brevissimo segmentis 4 orbicularibus obtusis glabris coriaceis, corolla tubo campanulato segmentis 4 ovatis, ovario glabro. A much-branched erect shrub or small tree, glabrous in all its parts, with acutely-angled branchlets. Leaves 2-3 in. long, deltoid at the base, rigid in texture, green and glabrous on both surfaces, the main veins fine and erecto-patent. Female flowers solitary, lateral; pedicels very short, with 2-3 minute ovate obtuse bracts. Flower-calyx 3 in. in diam., with a short spreading tube and 4 glabrous coriaceous brown veinless suborbieular obtuse segments. Corolla shorter than the calyx ; segments 4, ovate, twice as long as the tube. Fruit not seen.— Baron 2308! LINN. JOURN.—BOTANY, VOL. XXI. 21 424 MR. J. G. BAKER ON THE DrosPYROS GONOCLADA, n. sp. Fruticosa, glabra, ramulis ancipitibus, foliis sessilibus oblongis obtusis rigide coriaceis, floribus foemineis in cymas axillares 3-4-floras dispositis, pedicellis brevibus pilosis, calycis parvi segmentis 4 ovatis acutis, coroll: tubo brevi segmentis ovatis. An erect shrub, with slender glabrous branchlets, acutely angled towards the tip. Leaves alternate, rigid in texture, 2-3 in. long, deltoid at the base, green on both surfaces, with distant fine arcuate-ascending anastomosing main veins. Flowers 3-4 in lateral cymes, with very short peduncles and pedicels; bracts minute, lanceolate. Flower-calyx campanulate, glabrous, 4 in. long, with 4 ovate acute segments 2-3 times as long as the tube. Corolla 4 in. long, glabrous; segments 4, ovate, twice as long as the tube. Fruit not seen.— Baron 2313! Of all these four new Ebonies we have only the pistillate plants, two of them in the flowering and two in the fruiting stage, so that much further material is wanted for a thorough knowledge of their characters. HOLARRHENA ? MADAGASCARIENSIS, N. sp. Volubilis, fruticosa, glabra, foliis subsessilibus obovato-oblongis subcori- aceis cuspidatis, floribus in cymas umbellatas paucifloras axillares dispo- sitis, pedicellis cernuis, calycis tubo campanulato segmentis ovatis, corolli tubo cylindrico segmentis oblongis tubo duplo brevioribus, staminibus supra medium tubi insertis, folliculis cylindricis rigidis patulis. A woody climber, glabrous in all its parts. Leaves opposite, almost sessile, 2-3 in. long, about an inch broad, rounded at the apex to a large cusp, narrowed gradually from the middle to the base, moderately firm in texture, green and glabrous on both sur- faces, with 5-6 pairs of erecto-patent main veins. Flowers 3-6 in umbellate cymes in the axils of the upper leaves; pedicels 4- 3 in. long. Calyx 4 in. long, the ovate segments twice as long as the tube. Corolla with a cylindrical tube 4 in. long and 5 oblong segments half as long as the tube. Anthers lanceolate-deltoid, inserted above the middle of the corolla-tube, on short broad filaments. Follicles woody in texture, 5 in. long, spreading at a right angle from the top of the peduncles. Seeds with a coma of soft hairs nearly an inch long.— Baron 3242! Differs from Holarrhena in the insertion of the stamens. Adenoplusia axillaris, Radlkofer, in Bremen Abhand. viii. 461, founded on Hildebrandt 3671 — Buddleia axillaris, Willd. ; Baker in Journ, Linn. Soc. xx. p. 206; Baron 1481 and 1980. ENT. puse. FLORA OF MADAGASCAR. 425 BUDDLEIA SPHEROCEPHALA, n. Sp. Fruticosa, ramulis albido-tomentosis, foliis verticillatis breviter petiolatis oblanceolato-oblongis dentatis subcoriaceis dorso tenuiter tomentosis, floribus in capitulos globosos racemosos aggregatis, calycis segmentis brevissimis, corollz tubo infundibulari segmentis parvis deltoideis, bracteis rigidis obtusis, fruetu oblongo-clavato apice piloso. An erect shrub, 6 or 7 feet high, with the branchlets coated with whitish tomentum towards the tips. Leaves crowded, ascending, mostly three in a whorl, 3-4 in. long, 1-11 in. broad above the middle, obtuse or subacute, narrowed gradually from the middle to the base, rigid in texture, glabrescent above, thinly coated with whitish tomentum beneath; veius 5-6-jugate, raised, arcuate-ascending. Heads globose, 3~} in. in diam., few or many in a short terminal raceme. Bracts oblanceolate, oblong, obtuse, rigidly coriaceous, navicular, pilose inside and out. Capsule clavate, 15-3 in. long, splitting into two emarginate rigid valves. —-Baron 2239! 3111! Also on clay soil, at Maromanga, at 4000 ft., between Tamatave and Antananarivo, Dr. Meller! Allied to the well-known B. globosa, Lam. GAERTNERA PHANEROPHLEBIA, N. 8p. Fruticosa, ramulis fusco-pilosis, stipulis membranaceis vaginantibus, foliis breviter petiolatis oblongis acutis ciliatis venis erecto-patentibus pilosis, floribus in cymas congestas terminales dispositis, calycis segmentis magnis lanceolatis, corolle tubo subcylindrico piloso, segmentis brevibus oblongis, fructu parvo globoso glabro. An erect shrub, with slender branchlets, densely clothed up- wards with soft brown spreading hairs. Stipules 2 in. long. Leaves 3—4 in. long, about an inch broad, rounded at the base, moderately firm in texture, with 8-12 pairs of raised parallel erecto- patent veins, clothed with adpressed brown hairs. Flowers in small globose sessile terminal heads. Calyx finally 1-1 in. long, with a small campanulate tube and 5 large pilose lanceolate segments. Corolla à in. long, the oblong spreading segments not more than 3—4 as long as the subcylindrical pilose tube. Fruit small, globose, glabrous.—Baron 2372 ! 2982! GAERTNERA PHYLLOSTACHYA, D. 8p. Fruticosa, glabra, stipulis magnis vaginantibus, foliis breviter petiolatis oblanceolato-oblongis acutis subcoriaceis glabris, floribus in paniculam amplam ramulis corymbosis dispositis, bracteis multis magnis foliaceis persistentibus, calyce subtruncato, corolle tubo subcylindrico, segmentis oblongis tubo brevioribus, fructu parvo globoso glabro. 212 426 MR. J. G. BAKER ON THE A shrub or small tree, glabrous in all its parts. Stipules membranous, 4-1 in. long. Leaves 4—5 in. long, 13-12 in. broad, deltoid at the base, moderately firm in texture, green and glabrous on both surfaces, with distant fine arcuate-ascending main veins. Flowers in a lax deltoid terminal panicle 3—4 in. long and broad, with eorymbose branches, mixed with copious oblong or lanceo- late acute persistent foliaceous bracts 3-1 in. long. Calyx cam- panulate, 4; in. long, with an obscurely ciliated subentire collar- like limb. Corolla above 3 in. long. Fruit globose, glabrous, 4 in. in diam.— Baron 2327! 2683! Also Humblot 510! lrowza ($ ANISEIA) PHYLLONEURA, Baker. Herbacea, volubilis, ramulis pilosis, foliis petiolatis ovato-hastatis mem- branaceis glabris, floribus axillaribus szpe solitariis, bracteis foliaceis per- sistentibus, calycis segmentis magnis inzequalibus cordato-ovatis cuspidatis acute carinatis basi dentatis, corolla tubo calyci æquilongo limbo late infundibulari tubo æquilongo, fructu globoso calyce breviori.— Aniseia hastata, Meisn., in Mart. Fl. Bras. vii. 319. Baron 2516! 2605! 2671! This is a Brazilian species, and Mr. Baron’s specimens quite agree with those gathered in Brazil by Burchell. I have changed the specific name because Aniseia is merged in Zpomea by Bentham and Hooker, and there is already an Zpomea hastata of Linneus, which = Quamoclit sagittefolia of Choisy. SOLANUM MYOXOTRICHUM, n. sp. Fruticosum, ramulis aculeatis densissime ferrugineo-hispidis, foliis petiolatis ovatis acutis utrinque molliter pubescentibus costa aculeata, cymis sessilibus 2-3-floris, pedicellis elongatis hispidis aculeatis, calycis segmentis ovatis acuminatis, corolla segmentis deltoideis tubo longioribus, antheris magnis. A much-branched erect shrub, with stems armed with copious small brown prickles and very dense spreading bright brown bristly hairs. Leaves shortly petioled, about 2 in. long, acumi- nate, subentire, broadly rounded at the base, dull green above, whitish beneath, softly pubescent on both surfaces, the midrib armed with several large pungent brown prickles. Pedicels 1-1} in. long, armed with copious bristles and aciculi. Calyx 2 in. long, prickly and very pubescent. Expanded corolla 1 in. in diam. Anthers 4 in. long. Style protruded beyond the anthers. Berry not seen.— Baron 1005! 2805! Allied to S. indicum, Linn. FLORA OF MADAGASCAR. 427 SOPUBIA STRICTA, n. sp. Herbacea, pubescens, caule stricto inferne simplici superne ramosissimo ramulis ascendentibus virgatis, foliis oppositis sessilibus linearibus uni- nerviis, floribus oppositis laxissime racemosis, pedicellis fructui zequilongis, calycis tubo campanulato segmentis deltoideis, corolla tubo infundibulari segmentis latis, fructu globoso magnitudine pisi. A perennial herb, with a stiffly erect pubescent stem about 2 ft. long, much branched in the upper half, the branchlets long, simple, ascending. Leaves in opposite pairs, invariably simple, 3-1 in. long, ascending, one-nerved, with revolute margins. Racemes sometimes half'a foot long, the flowers in pairs on short ascending pedicels, bracteated at the base by reduced leaves. Calyx 4 in. long and broad; teeth rather shorter than the tube. Corolla not well seen. Fruit black, glabrous, X in. in diam., splitting down nearly to the base into two valves.— Baron 2709! UTRICULARIA [BARENSIS, n. Sp. Limosa, parva, aphylla, caule gracili brevi, racemis laxe 2-3-floris, floribus purpurascentibus, pedicellis brevibus, bracteis 2-3nis acutis, calycis labiis oblongis, corolhe labio postico parvo erecto, antico magno latissimo cuneato, calcare deltoideo. Stem leafless, very slender, straight, erect, 2-3 in. long. Flowers 2-3, arranged in a lax terminal raceme; pedicels erecto- patent, 4-3 in. long; bracts minute, lanceolate, persistent. Calyx 2 in. long, slit down nearly to the base into two oblong lips. Upper lip of the corolla deltoid-cuneate, j-$ in. long and broad; lower lip considerably broader than long ($-4 in. broad), rounded at the apex, cuneate at the base; spur half as long as the lower lip.—Ibara country, L. Kitehing! Allied to U.spartea, Baker, in Journ. Linn. Soc. xx. 216. DrbYMOCARPUS VESTITA, n. Sp. Annua, pilosa, foliis ovatis petiolatis membranaceis serratis, pedunculis axillaribus 2-3-floris, pedicellis elongatis, calycis tubo brevissimo segmentis lanceolatis, corollæ tubo late infundibulari segmentis brevibus semiorbicu- laribus, genitalibus inclusis, fructu cylindrico piloso. An erect annual, with densely pubescent flexuose stems 3—4 in. long. Leaves opposite, 1-2 in. long, obtuse, membranous, densely hairy on both surfaces, inciso-crenate. Flowers in cymes from the axils of the upper leaves; peduncle about an inch long; pedicel finally 4-4 in., erecto-patent. Calyx campanulate, pilose, 4 in. long, cut down nearly to the base into 5 lanceolate rz i segments. Corolla j in. long; segments not more than j as 428 MR. J. G. BAKER ON THE long as the tube. Capsule 1 in. long, pilose, narrowed gradually into a curved style 4 in. long; stigma capitate.— Baron 2655! COLEA PARVIFLORA, n. Sp. Fruticosa, erecta, glabra, foliis longe petiolatis foliolis 1-5 oblongis coriaceis pinninerviis, floribus in cymas axillares paucifloras dispositis, pedicellis brevibus, calycis tubo obconico segmentis brevissimis, corollz pilosze tubo infundibulari segmentis brevibus oblongis, genitalibus inclusis, fructu oblongo acuto. An erect much-branched shrub, with glabrous terete branchlets. Petiole sometimes an inch long. Leaves rarely simple, usually compound, with 3-5 oblong obtuse coriaceous leaflets 1-14 in. long, green and glabrous on both surfaces, rather glossy above, with 5-6 pairs of distinet parallel areuate ascending main veins. Flowers few together in shortly-peduncled axillary cymes from the leafy branchlets. Calyx 2-1 in. in diam., glabrous, with small deltoid segments. Corolla $ in. long, pilose, with a funnel- shaped tube and small obtuse segments. Stamens not reaching the tip of the perianth-segments. Capsule glabrous, smooth, 13 in. long, narrowed to the point—Baron 3099! THUNBERGIA CONVOLVULIFOLIA, n. sp. Volubilis, glabra, foliis breviter petiolatis cordato-ovatis acutis, floribus axillaribus solitariis, pedicellis elongatis, bracteis oblongis foliaceis persisten- tibus, calycis dentibus lanceolatis, corolla tubo deorsum cylindrico sursum oblongo ventricoso, segmentis parvis orbicularibus, genitalibus in tubum inclusis, fructu globoso longe cuspidato. A climber, with slender shrubby angled stems. Leaves 1-1} in. long, moderately firm in texture, turning blackish when dried, with an open basal sinus and rounded auricles; petiole 1-1 in. long. Flowers solitary from the axils of the leaves on ascending peduncles about an inch long. Bracts 4 in. long, similar to the leaves in colour and texture. Calyx with a very short patelleeform tube and 10 rigid erect teeth jl, in. long. Corolla ł in. long, the ventrieose upper part of the tube j in. in diam.; segments purplish black, g~§ in. long and broad. Capsule globose, 4 in. in diam., with a beak nearly 2 in. long.— Baron 2923! Nearly allied to T. angulata, Hils. and Bojer, of which there is a figure in Hooker’s ‘ Exotic Flora,’ tab. 166. RUELLIA BREVICAULIS, n. sp. Ine" perennis, pilosa, caulibus brevibus, foliis parvis sessilibus obla - 1 à . . . T . . nceolato-oblongis obtusis vel subacutis, floribus solitariis axillaribus, FLORA OF MADAGASCAR. ` 429 pedicellis brevibus, bracteolis oblongo-lanceolatis persistentibus, calycis tubo infundibulari segmentis lanceolato-deltoideis, corolla tubo cylindrico pollicari, segmentis orbicularibus patulis, genitalibus in corolle tubo inclusis. A herbaceous perennial, copiously branched at the crown of the root, with slender ascending pubescent stems 2-3 in. long. Leaves opposite, sessile, 4 in. long, narrowed from the middle to the base, moderately firm in texture, dark green and hispid on both surfaces. Flowers solitary in the axils of the leaves all down the stem on short ascending pedicels. Calyx 3 in. long, green, pilose; seg- ments j the length ofthe tube. Corolla with a slender cylindrical tube an inch long, only dilated very slightly at the top; segments 5, orbicular-cuneate, subequal, 4 in. long and broad. Stamens 4, inserted near the top of the corolla-tube; filaments short. Cap- sule not seen.— Baron 1896 ! JUSTICIA ($ ANISOSTACHYA) TRICHOPHYLLA, n. sp. Fruticosa, ramulis dense pilosis, foliis longe petiolatis oblongis acutis membranaceis utrinque pilosis, floribus in spicas paniculatas terminales dispositis, bracteis magnis membranaceis obovato-cuneatis persistentibus, calycis tubo brevissimo segmentis oblanceolatis acutis, capsule valvis bractec equilongis. A shrub, with slender branchlets, densely pubescent towards the tip. Leaves 3—4 in. long, membranous, dark green, acuminate, rounded at the base, pubescent on both surfaces; petiole 2-1 in. long, densely pubescent. Spikes 1-13 in. long, many to a panicle, the lower branches of which spring from the axils of large leaves. Bracts obovate-cuneate, 1 in. long, persistent, green at the base and drab at the top in the dried specimens. Calyx j in. long, infundibuliform, cut down nearly to the base into oblanceolate acute segments. Capsule j in. long, split down nearly to the base into two rigid diverging valves, with a couple of seeds at the middle of each.—Baron 2142! Allied to J. haplostachya and J. Commersoni. Jusricta ($ ANISOSTACHYA) TRITICEA, N. Sp. m Fruticosa, glabra, ramulis gracillimis, foliis breviter petiolatis oblongis acutis, floribus in spicas densas pedunculatas panieulatas dispositis, bracteis rigidis oblongo-lanceolatis ciliatis dense imbricatis, calyce profunde bila- biato, corolle tubo cylindrico, limbo brevi bilabiato, genitalibus ex tubo protrusis. An erect shrub, with slender glabrous branchlets. Leaves 2- 430 MR. J. G. BAKER ON THE 3 in. long, narrowed from the middle to the base, moderately firm in texture, green and glabrous ou both surfaces. Spikes very dense, secund, 1-12 in. long, produced from the axils of the upper leaves on peduncles of about the same length. Bracts oblong- lanceolate, in. long, with 5-7 prominent ribs on the back, the margin densely ciliated. Calyx about as long as the bract, glabrous; segments oblong-lanceolate. Corolla whitish, with a cylindrical tube as long as the calyx ; limb jin. long ; upper lip obovate; lower deflexed, deeply 3-lobed. Capsule not seen.— Baron 2545! Allied to J. Bojeri, Nees. IsoarossA GRACILLIMA, n. Sp. Herbacea, glabra, ramulis gracillimis, foliis oblongis acutis membranaceis glabris, floribus in paniculas laxissimas terminales dispositis, pedicellis elongatis, calycis segmentis lanceolatis longe acuminatis, corolla tubo in- fundibulari limbo brevi, antherarum loculis disjunctis, superiori interdum basi calcarato. An erect perennial herb, glabrous in all its parts, with very slender branchlets. Leaves distant, opposite, shortly petioled, dark green, 14-2 in. long. Flowers arranged in a very lax panicle, with very slender ascending few-flowered racemose branchlets ; pedicels erecto-patent, 3-3 in. long. Calyx lin. long, slit down nearly to the base into 5 lanceolate acuminate segments. Corolla yellow, $ in. long; tube infundibuliform ; limb bilabiate, half as longasthetube. Stamens 4, included in the corolla-tube, with long filaments and superposed small oblong anthers, the upper cell sometimes but not invariably spurred at the base.—Baron 2325! Isoarossa ANGUSTA, Baker. Herbacea, ramulis gracillimis apice pubescentibus, floribus breviter peti- olatis lanceolatis glabris, floribus in paniculas laxissimas terminales ramulis corymbosis dispositis, calycis segmentis lanceolatis, corolla tubo infundi- bulari limbo brevi, antherarum loculis disjunctis oblongis.—Clinacanthus angustus, Nees in DC. Prod. xi. 511. An erect perennial herb, with very slender branchlets, shortly pubescent towards the top. Leaves shortly petioled, 14-2 in. long, 4 in. broad. Panicle erect, very long and lax; pedicels 4- 2 in. long. Calyx } in. long, cut down nearly to the base into 5 lanceolate acute segments, Corolla yellow, 4 in. long; upper lip orbicular, shorter than the lower. Stamens included in the corolla- tube; cells of the anther much separated. Capsule with 2 diverging clavate valves 4 in. long, with 2 seeds at the middle of FLORA OF MADAGASCAR. 431 each.— Baron 2198! We have long had two sheets at Kew from the herbarium of Justice Blackburn. Isoarnossa MELLERI, n. sp. Herbacea, ramulis gracillimis apice pilosis, foliis breviter petiolatis ovatis acutis subglabris membranaceis, floribus in paniculas laxissimas terminales ramulis paueifloris corymbosis dispositis, pedicellis elongatis, calycis seg- mentis lanceolatis, corolla tubo infundibulari limbo tubo sequilongo, an- therarum loeulis disjunctis oblongis. An erect fragile perennial herb 2-3 feet high, with slender fragile branchlets pilose towards thetip. Leaves distant, opposite, turning black when dried, thin in texture, rounded at the base, the lower 14-2 in. long. Flowers arranged in a very lax erect panicle, the lower branches of which spring from the axil of well-developed leaves; pedicels slender, erect, pilose, sometimes 3-3? in. long. Calyx 4 in. long, pubescent at the base; tube very short. Corolla yellow, $ in. long; lips very unequal, the lower with 3 large oblong divisions. Stamens included in the corolla-tube ; anther- cells not spurred at the base. Capsule not seen.— Between Tamatave and Antananarivo, Dr. Meller! Three sheets in the Kew herbarium, the specimens gathered in 1862, Native name Ravisara. HYPOESTES STACHYOIDES, n. sp. Herbacea, perennis, glabra, foliis parvis breviter petiolatis remotis ovatis obtusis, floribus in spicas laxas paucifloras axillares et terminales dispositis, involucro unifloro bracteis 4 rigidulis oblongis acutis, calyce involucro paulo breviori, corolla albz tubo infundibulari, limbo tubo breviori, geni- talibus ex tubo protrusis. A herbaceous perennial, glabrous in all its parts, with slender ascending stems under a foot long. Leaves j-j in. long, mode- rately firm in texture, green and glabrous on both surfaces. Spikes laxly few-flowered, erect, terminal, produced from the end of the stems and axils of the upper leaves. Involucre glabrous, } in. long; bracts much imbricated. Calyx but little shorter than the involucre. Corolla-tube 4 in. long ; upper lip oblanceolate-oblong, entire, 4 in. long ; lower nearly as long, 3-lobed. Stamens reach- ing out of the tube halfway up the limb. Capsule not seen.— Baron, 2626! Allied to H. maculosa, Nees. HYPOESTES UNILATERALIS, n. sp. Herbacea, perennis, glabra, foliis petiolatis lanceolatis, floribus in pani- culam terminalem ramulis brevibus secundis spicatis dispositis, involucro 432 MR. J. G. BAKER ON THE unifloro piloso bracteis 4 rigidulis oblongis, calyce involucro duplo breviori, coroll albz pilose tubo subcylindrico, labiis tubo duplo brevioribus, staminibus ex tubo breviter protrusis. A herbaceous perennial, with short slender erect glabrous stems. Leaves 1-12 in. long, }-} in. broad, pinninerved, tapering gradu- ally to the base, moderately firm in texture, green and glabrous on both surfaces; petiole 1—3 in. long. Panicle with the lower branches produced from the axils of well-developed leaves. In- voluere infundibuliform, 4 in. long; bracts dull green, densely pilose, much imbricated. Calyx jl in. long; segments oblong- lanceolate. Corolla-tube 1 in. long ; limb half as long as the tube; upper lip oblong ; lower deflexed, with 3 oblong lobes. Anthers just protruded from the corolla-tube. Capsule not seen.— Baron 2665! Allied to H. secundiflora, Baker. HYPOESTES JASMINOIDES, n. 8p. Fruticosa, glabra, foliis brevissime petiolatis oblongis acutis pinninerviis, floribus in spicas laxas paucifloras paniculatas dispositis, involucro unifloro glabro bracteis 4 rigidulis acutis, calyce involucro paulo breviori, corolla albz tubo elongato subcylindrico labiis brevibus latis, genitalibus in tubo inclusis. A much-branched shrub, with slender terete branchlets. Leaves 2-3 in. long, deltoid at the base, moderately firm in texture, green and glabrous on both surfaces; petiole not more than 4—4 in. long. Spikes 1-14 in. long, produced only in the axils of the upper leaves, erect, shortly peduncled. Involucre glabrous, in- fundibuliform, 1 in. long ; segments green, rigid, much imbricated. Corolla white, pubescent; tube 4 in. long, almost cylindrical; limb 1 in. long; upper lip oblong, obtuse, entire; lower with 3 small obtuse lobes. Capsule not seen.— Baron 1224, 2533! Allied to H. comorensis, Baker in Journ. Linn. Soe. xx. p. 223. HYPoESTES TRICHOCHLAMYS, n. sp. Fruticosa, ramulis pubescentibus, foliis petiolatis oblongis acutis obscure pilosis, floribus in spicas secundas subdensas paniculatas terminales dispo- sitis, involuero unifloro piloso segmentis 4 oblanceolatis obtusis, calyce quam involucrum paulo breviori, corollze pilosze albze tubo quam involucrum sesquilongiori, limbo parvo segmentis oblongis. A shrub, with slender pubescent tetragonous branchlets. Leaves distinctly petioled, moderately firm in texture, 2-3 in. long, cuneate at the base, dark green and thinly pilose on both surfaces. Spikes 1-2 in. long, ascending, arranged in small ter- FLORA OF MADAGASCAR. 433 minal panicles. Involucre narrowly infun dibuliform, 1-1 in. long ; braets firm in texture, densely pubescent, green towards the tip, drab towards the base. Calyx pale, more than half as long as the involucre ; segments very acute. Corolla white, 4 in. long, with a long tube and a short bilabiate limb with oblong segments. Stamens and stigma exserted from the corolla-tube.— Baron 2928 ! Allied to H. saxicola, Nees. ORTHOSIPHON SECUNDIFLORUS, n. sp. Fruticosus, ramulis pubescentibus, foliis oblongis acutis petiolatis pubes- centibus serratis, floribus 2-3nis in racemos laxos secundos dispositis, pedicellis calyci fructifero sequilongis, calycis labio supremo brevi lato, labio inferiori dentibus lanceolato-deltoideis, corolle tubo elongato cylin- drico, limbo tubo duplo breviori, genitalibus in tubo inclusis. Stems erect, shrubby, much branched, pubescent. Leaves 2-3 in. long, moderately firm in texture, distinctly serrated, green and pubescent on both sides. Racemes 12-2 in. long ; pedicels finally in. long; bracts minute, foliaceous, ovate. Calyx finally 4 in. long; upper lip concave, scariose, strongly veined, not more than half as long as the campanulate tube ; lower lip rather longer than the upper, the two lowest teeth the longest. Corolla-tube nearly 4 in. long; lips half as long as the tube.— Baron 1226! OnTHOSIPHON EMIRNENSIS, n. sp. Herbaceus, ramulis pilosis, foliis petiolatis oblongis obtusis crenatis mem- branaceis, floribus laxe racemosis verticillatis, pedicellis calyci fructifero æquilongis, calycis labio supremo orbiculari concavo tubo æquilongo, labio inferiori dentibus lanceolatis, corolla tubo cylindrico, limbo parvo, geni- talibus in tubo inclusis. A. perennial herb, with finely pilose erect slender stems under a foot long. Leaves 1-2 in. long, membranous, deltoid at the base, green and glabrous on both surfaces; petiole j- in. long. Racemes peduncled, terminal, at most an inch long ; flowers 3-6 in a whorl; pedicels finally 4 in. long; bracts small, foliaceous, oblong-lanceolate. Calyx very accrescent, at first not more than 4l in., finally } in. long; upper lip suborbicular, as long as the campanulate tube, concave and conspicuously veined ; lower lip with 3 rigid lanceolate teeth. Corolla with a cylindrical tube § in. long; lobes of the limb minute, orbicular— Baron 1056! 2190! 3259! ORTHOSIPHON BREVICAULIS, n. Sp. Herbaceus, perennis, caule subnullo vel brevissimo, foliis magnis oblan- ceolato-oblongis obtusis serratis glabris, floribus verticillatis in racemum 484 MR. J. G. BAKER ON THE laxum terminalem longe peduneulatum dispositis, pedicellis pubescentibus calyci fructifero zequilongis, calycis labio supremo concavo tubo zquilongo, labio inferiori dentibus parvis acutis, corollz tubo elongato cylindrico labiis parvis, genitalibus ex tubo protrusis. A perennial herb, with either all the leaves in a rosette from the crown of the root or one pair at the top of a short internode. Leaves 3-4 in. long, 1-1} in. broad, firm in texture, green and nearly glabrous on both surfaces, narrowed gradually from the middle to the base, serrated in the upper half, entire in the lower. Raceme 1-2 in. long, shorter than its pubescent erect peduncle. Pedicels finally 7 in. long. Calyx very aecrescent, finally j in. long; upper lip concave, quite as longas the tube; lower lip with 4 rigid deltoid or lanceolate-deltoid teeth. Corolla $ in. long; lips not more than 4 as long as the tube.— Baron 2656 ! PLECTRANTHUS CYMOSUS, n. Sp. Perennis, pubescens, foliis longe petiolatis ovatis subacutis conspicue crenatis membranaceis, floribus in cymas verticillatas in panieulam elon- gatam terminalem cylindricam aggregatas dispositis, bracteis nullis, pedi- cellis ascendentibus pubescentibus, calyce parvo profunde bilabiato, corollze tubo infundibulari prope basin curvato, fauce lata, labio postico parvo antico magno oblongo, staminibus ultra labium exsertis. An erect perennial herb, 2-3 feet high, with a tetragonous stem, leafy up to the base of the inflorescence. Leaves opposite, 13-2 in. long, nearly as broad, subacute, deeply crenate, dull green, thin in texture, slightly pubescent; petiole 1 in. long. Flowers in small cymes arranged inu vertieils of about four, forming a leafless cylindrical panicle nearly a foot long ; pedicels short, ascending. Calyx campanulate, j in. long. Corolla lilac, pubescent, with a broadly funnel-shaped tube 1 in. long, a small orbieular upper lip and an oblong acute concave lower lip as long as the tube. Stamens declinate, the two longer exserted beyond the lower lip of the corolla; anthers minute, orbicular.—Baron 2250! Lieer4 OLIGoPHYLLA, Baker in Journ. Linn. Soc. xx. p. 225, proves to be identical with Acrocephalus villosus, Benth. VITEX ($ CHRYSOMALLUM) TRICHANTHA, n. sp. Erecta, fruticosa, ramulis apice pilosis, folis simplicibus petiolatis oblongis acutis subcoriaceis facie glabris dorso pilosis, cymis axilla- ribus 1-2-floris, bracteolis lanceolatis pilosis, calycis tubo campanulato villoso dentibus lanceolatis tubo equilongis, corolle tubo piloso curvato infundibulari elongato, segmentis parvis, genitalibus breviter exsertis. FLORA OF MADAGASCAR. 435 A much-branched erect shrub, with branchlets densely villose towards the tip. Leaves opposite, distinctly petioled, firm in texture, 2-3 in. long, rounded at the base, glabrous when mature above, pubescent mainly on the midrib beneath. Peduncles about j in. long. Calyx j in. long, very shaggy, with a pair of persistent bracteoles at the base. Corolla-tube curved, villose, above an inch long, 4 in. in diam. at the throat ; lobes of the limb very small, orbicular. Stamens and bifid style exserted beyond the tip of the corolla-segments; anthers minute, globose. Fruit not seen.— Baron 2316! Allied to V. Bojeri, Schauer, = Baron 2972! CLERODENDRON ? BRUNSVIGIOIDES, n. sp. Fruticosa, erecta, glabra, foliis breviter petiolatis oblongis acutis, flori- bus axillaribus pedunculatis solitariis, calycis tubo infundibulari segmentis ovatis acutis tubo brevioribus, corolle tubo infundibulari calyce duplo longiori, limbi segmentis orbicularibus, genitalibus in tubo inclusis. An erect shrub, with slender branchlets, obscurely pilose towards the tip. Leaves 3-4 in. long, much narrowed to the base, simple, entire, moderately firm in texture, green and gla- brous on both surfaces. Flowers solitary from the axils of the leaves on ascending peduncles 1-3 in. long. Calyx 3-3 in. long, clothed with adpressed hairs; teeth half as long as the tube. Corolla-tube above an inch long, dilated in the upper half ; ex- panded limb an inch in diameter; orbicular subequal segments imbricated, about 3 in. long and broad. Stamens and style not protruded from the throat of the corolla-tube. Fruit not seen. — Baron 2716! Near C.? petunioides, Baker in Journ. Linn. Soc. xx. p. 230. INCOMPLETA. POLYGONIUM BRACHYPODUM, Baker in Journ. Linn. Soc. xx. p. 239—of which Mr. Baron now sends better specimens as No. 2208— proves identical with P. Meissnerianum, Cham. et Schlecht., which in America extends from Louisiana and Texas to South Brazil. Hyprostacuys STOLONIFERA, D. Sp. H. caulibus longe reptantibus, foliis pedunculisque ad nodos cæspitosis, foliis longe petiolatis deltoideis decompositis segmentis ultimis minutis lanceolatis, floribus in spicas densas cylindricas pedunculatas dispositis, bracteis ovatis. 436 MR. J. G. BAKER ON THE Stems wide-trailing on the surface of the ground, rooting at the nodes, where they bear tufts of leaves and spikes. Petiole 1-2 in. long, rough with spreading papille ; lamina reaching a length of 5-6 in.; pinne and pinnules deltoid; ultimate segments lanceolate, not more than 74-3; in. long. Spikes several to a node, simple, peduncled, cylindrical, seen only in a young state. — Baron 2628! Allied to H. multifida, A. Juss. ; Deless. Icones, vol. iii. t. 93. PIPER (§ CuBEBA) PACHYPHYLLUM, n. sp. Dioicum, fruticosum, glabrum, ramulis articulatis, foliis coriaceis ob- longis acutis basi paulo inequalibus late rotundatis vel subeequalibus, spicis foemineis densifloris cylindricis oppositifoliis breviter. peduuculatis, bracteis orbicularibus peltatis, fructibus ellipsoideis glabris distincte stipitatis, stigmatibus 3 sessilibus patentibus. Stems shrubby, terete, articulated at the nodes. Leaves 4-5 in. long, 2-21 in. broad, firm and thick in texture, green and glabrous on both surfaces, the secondary veins distant and very ascending, connected at the apex by intramarginal arches ; petiole under 4 in. long. Female spikes only seen, finally 22-3 in. long; peduncle 4-2 in. long. Ovary with a pedicel as long as itself, black, glabrous; stigmas 3, sessile.— Baron 2415! Very near P. borbonense, C. DC., which Mr. Baron has gathered several times in Central Madagascar (1280! 1539! 2207! 2335!), and which Dr. Parker has sent under the native name Ferifery. Mr. Baron has also found P. capense, Linn. fil. (25181), which extends to the Cape, the Zambesi country, Fernando Po, and the Cameroon Mountains. PEPEROMIA TRICHOPHYLLA, n. sp. P. caule simplici semipedali flexuoso piloso, foliis alternis obovato-cuneatis obtusis petiolatis membranaceis utrinque pilosis venis lateralibus paucis valde ascendentibus, spicis elongatis gracillimis laxifloris ascendentibus pedunculatis, bracteis orbicularibus peltatis, ovario ovoideo sessili, stigmate sessili terminali. Stems weak, flexuose, densely pilose. Leaf with a blade 11-2 in. long, orbicular at the apex, cuneate in the lower half, mem- branous in texture, dark green and pubescent on both surfaces, the midrib and lateral veins distinctly visible when the leaf is held up to the light, as are the anastomosing connecting veinlets ; petiole j-Z in. long. Spikes produced from the axils of the leaves on short slender peduncles, finally 3-4 in. long. Ovary ascend- FLORA OF MADAGASCAR. 437 ing, glabrous, inserted in the grooves of the sulcate rhachis; stigma globose, sessile.— Baron 500! 3190! Allied to P. Lyallii, C. DC. Driros&£rA Tuovansir, Roem. & Sch. Of this very curious endemie Proteaceous tree Mr. Baron has now (No. 3253) procured complete specimens, from which the following notes are taken. Dr. Baillon has recently described it fully in the Bulletin of the Linnean Society of Paris, p. 394, from specimens gathered in the north-west of the island by M. Humblot. A large tree with thick terete branchlets, coated with brown tomentum. Leaves alternate, rigidly coriaceous, glabrous ; blade cuneate, deeply bifid, 4 ft. long and broad, the apical lobes semi- orbicular and simple in the mature leaf, prolonged and again forked in the young leaves; petiole 3-4 in. long, flattened and winged towards the top. Flowers diccious, arranged in copious ascending axillary panicles about as long as the petiole, with a tomentose axis and branchlets; branchlets laxly spicate, each flower subtended by a minute ovate bract. Male perianth clavate, « in. long, slit down to the base into four lanceolate valvate segments, glabrous inside. Stamens a little shorter than the perianth-segments; anther j in. long, tipped with the pro- truded connective, much longer than the fleshy cylindrical fila- ment. Female panicle shorter and less compound than the male. Ovary globose, densely ferruginous, with a short thick style and 2-lobed stigma. Fruit hard, thick, 1-seeded, indehiscent, oblong, l4 in. long. Native name Vivaona. The plaut gathered by Dr. Parker, mentioned in Journ. Bot. 1882, p. 244, is a form of the same species. Viscum ($ ProroNUXIA) LOPHIOCLADUM, n. sp. Foliosum, glabrum, ramulis articulatis acute ancipitibus, foliis subsessi- libus ovatis subobtusis rigide coriaceis veuis occultis, floribus ad nodos ses- silibus glomeratis bracteis in cupulam connatis, ovario globoso haud papilloso. A shrub, with compressed woody branchlets with two raised angles, falling in pieces readily at the nodes, with internodes 3-1 in.long. Leaves very thiek and rigid in texture, 2-3 in. long, opaque, brown and quite glabrous on both surfaces when dried, rounded toa cuneate base. Flowers few in a cluster, quite sessile 438 MR. J. G. BAKER ON THE in the axil of the leaves, seen only in an early stage. Sepals deltoid.— Baron 2751 ! Visoum ($ PLoIONUXIA) RHYTIDOCARPUM, n. sp. Foliosum, glabrum, ramulis teretibus granulosis, foliis subsessilibus oblongis obtusis crassis rigide coriaceis venis oceultis immersis, floribus in umbellas subsessiles axillares 2-3-floras dispositis, bracteis in cupulam connatis, ovario globoso haud papilloso. A much-branched shrub, with terete very granular branch- lets, the internodes 4-1 in. long. Leaves very thick and rigid in texture, 3-1 in. long, obtuse, narrowed at the base to an indi- stinct petiole, opaque and brown with a much-wrinkled epidermis on both surfaces when dried. Flowers 2-8 together in the axils of the leaves in a small oblong coriaceous cupule. Ovary not seen fully mature, but not at all papillose in an early stage.— Baron 3110! Near V. triflorun, DC. Viscum ($ PLOIONUXIA) GRANULOSUM, n. sp. Foliosum, glabrum, ramulis teretibus granulosis, foliis oblongis subob- tusis brevissime petiolatis basi attenuatis rigide coriaceis venis occultis, floribus 2-4nis in umbellas axillares brevissime pedunculatas dispositis, bracteis in cupulam connatis, ovario ovoideo granuloso, stylo brevissimo cylindrico, stigmate capitato. A much-branched shrub, with slender terete very granular branchlets, breaking up easily at the nodes. Leaves 11-2 in. long, obscurely petioled, very thiek and rigid in texture, brown, rugose and opaque on both surfaces when dried. Umbels some- times 3—4 toanode. Bracts ovate, spreading, coriaceous, slightly connate at the base. Mature ovary l in. in diam., shortly pedi- cellate, black, minutely papillose, tipped with a short cylindrical style, with a capitate stigma. Sepals 4, minute, oblong-lanceolate. —Baron 3115! Viscum (§ PLOIONUXIA) CUNEIFOLIUM, n. Sp. Foliosum, ramulis granulosis, foliis oppositis subsessilibus orbicularibus vel obovatis basi cuneatis crassis rigide coriaceis venis occultis, floribus 2-3nis in umbellas axillares brevissime pedunculatas dispositis, bracteis in cupulam connatis, ovario globoso levi, stylo brevissimo cylindrico, stigmate capitato. A much-branched shrub, with nearly terete slender scabrous branchlets, the upper internodes uot more than } in. long. FLORA OF MADAGASCAR. 439 Leaves opposite, very obtuse and very thick and rigid in texture, opaque, brown when dried, 1-14 in. long, 3-1 in. broad, nar- rowed very gradually at the base. Flowers in axillary umbels on very short peduncles; bracts coriaceous, connate into a small oblong cupule with a minutely ciliated margin. Sepals 4, minute, oblong, connivent. Mature ovary black, i in. in diam., crowned with a very short style and capitate stigma.—Baron 2807! Viscum ($ProrowuxiA) Rapvuna, n. sp. Foliosum, glabrum, ramulis apice acute tetragonis, foliis subsessilibus oblongis parvis obtusis vel subacutis rigide coriaceis rugosis venis immersis occultis, floribus 1-3nis in umbellas axillares brevissime pedunculatas dis- positis, bracteis in cupulam connatis ovario globoso scabro haud papilloso, stylo brevissimo cylindrico, stigmate capitato. A much-branched shrub, glabrous iu all its parts, with very slender terete acutely-angled branchlets, with internodes 4—4 in. long. Leaves }—8 in. long, 3-3 in. broad, thick and coriaceous in texture, acute or subobtuse, much narrowed at the base, opaque and nearly black when dried, very rough on both surfaces. Flowers 1-3 on a very short peduncle; bracts spreading, ovate, obtuse, minutely ciliated. Mature ovary black, j'; in. in diam. ; sepals lanceolate, very minute; style very short; stigma capitate. — Baron 3072! Allied to V. triflorum, DC. Viscum (§PLOIONUXIA) APODUM, n. sp. Foliosum, glabrum, ramulis teretibus, foliis oblongis acutis subsessilibus basi attenuatis rigide coriaceis obscure trinervatis, floribus axillaribus soli- tariis bracteis minutis ovatis, ovario globoso densissime papilloso-muricato, stylo conico brevissimo, stigmate capitato. A much-branched shrub, glabrous in all its parts, with slender subterete branchlets, with internodes j-1l in. long. Leaves opposite, thick and rigid in texture, brownish and scabrous on both surfaces when dried, 2-3 in. long, #-1 in. broad at the middle, acuminate at both ends, the veins very obscure. Female flowers solitary and sessile in the axils of the leaves all down the branches; bracts minute, ovate, ciliated. Mature ovary globose, Lin. in diam., brown, densely coated with hard prominent papille ; style very short.—Baron 3012! Allied to V. tuberculatum, A. Rich., and V. multicostatum, Baker. Viscum (§ASPIDUXIA) TRACHYCARPUM, N. Sp. Aphyllum, ramulis teretibus glabris ad nodos articulatis, floribus 1-3nis [3 LINN, JOURN.—BOTA NY, VOL. XXI. 2K 440 MR. J. G. BAKER ON THE ad nodos sessilibus, ovario globoso nigro dense papilloso-muricato, stylo brevissimo, stigmate capitato. A much-branched shrub, with terete many-ribbed divaricating branchlets, breaking up readily at the nodes, with internodes 1-14 in. long. Flowers 2-6 in sessile verticels at the nodes. Mature ovary black, globose, 4; in. in diam., covered all over with prominent hard black papille; style very short, cylindrical ; stigma capitate. Male flowers and sepals not seen.—.Baron 2408! EUPHORBIA TETRAPTERA, n. 8p. Arborea, glabra, foliis alternis petiolatis oblongis integris acuminatis subcoriaceis, capitulis in paniculam decompositam terminalem ramis dicho- tomiter cymosis dispositis, bracteis parvis lanceolatis persistentibus, invo- lucro eampanulato appendicibus 5 orbicvlaribus integris carnosis, dentibus minutis quadratis emarginatis, ovario biloculari, styli ramis brevibus apice emarginatis, carpellis fructiferis globosis dorso bialatis. A much-branched erect tree, with woody terete not at all fleshy branchlets. Leaves 2-3 in. long, acuminate, entire, deltoid at the base, moderately firm in texture, mostly alternate but the upper opposite, green and glabrous on both surfaces, with fine distinct main veins anastomosing in intramarginal arches. Pa- nicles 2-3 in. in diam., at the end of the branchlets, overtopped by the upper whorled leaves. Involucre with a campanulate tube z's in. in diam. and 5 orbicular entire spreading segments, with a minute square erect tooth between each. Fruit 4 in. in diam., conically narrowed at the base into an erect gynophore a in. long ; earpels never more than two, globose, coriaceous, with two deltoid wings from the upper part; styles unusually short and notched at the apex only.— Baron 2775!3037! A very curious and distinct species. UAPACA MYRICAEFOLIA, n. sp. Arborea, glabra, foliis parvis brevissime petiolatis oblanceolato-oblongis obtusis rigide coriaceis, floribus masculis in capitulos globosos bracteis 4 coriaceis glabris persistentibus reflexis cinctos dispositis, fructu globoso glabro 4-loculari. A much-branched erect tree, glabrous in all its parts. Leaves ascending, 13-2 in. long, 3-3? in. broad above the middle, very obtuse, narrowed gradually from the middle to the base, firm and rigid in texture, green and glabrous on both surfaces, with fine indistinct erecto-patent main veins. Male heads globose, } in. in FLORA OF MADAGASCAR. 441 diam., girt with 4 persistent cuneate reflexing coriaceous glabrous bracts ; peduncle about an inch long. Fruit hard, glabrous, glo- bose, 3 in. in diam., with a thin coriaceous epicarp and 4 pyrenes. — Baron 2209 ! 2864! 2961! Uapaca CLUSIOIDES, n. sp. l Glabra, ramulis percrassis, foliis magnis obovato-cuneatis petiolatis rigide coriaceis obtusis, floribus fæœmineis ignotis, floribus masculis in capitulos globosos axillares pedunculatos dispositis, bracteis reflexis coria- cers, perianthio obconico coriaceo ore piloso, staminibus 5 exsertis. A tree, glabrous in all its parts except the perianth, with the leaves and male heads produced from terete woody branches half an inch in diameter. Leaves 6-8 in. long, 4-5 in. broad above the middle, very thick and rigid in texture, green and glabrous on both surfaces, with 6-8 pairs of parallel arcuate-ascending main veins distinct from the midrib to the margin; petiole 1-14 in. long. Male heads globose, 4 in. in diam., enclosed in a whorl of very much imbricating persistent reflexing orbicular-cuneate bracts; peduncle rugose, erecto-patent, 1-14 in. long. Male perianth 4i, in. long, formed of 5 coriaceous segments. Stamens 5, much longer than the perianth; anthers oblong. Fruit un- known.— Baron 2546 ! BRIDELIA COCCOLOBJEFOLIA, n. Sp. Arborea, glabra, foliis petiolatis ovatis vel oblongis cordatis sub- coriaceis lucidis, floribus foemineis in paniculam terminalem ramulis spi- catis dispositis glomeratis, perianthii segmentis ovatis, disco cupulari, fructu ellipsoideo coriaceo venoso apiculato, semine solitario. A tree, glabrous in all its parts, with terete woody branchlets. Leaves alternate, 3-4 in. long, 1}-2 in. broad, obtuse or sub- obtuse, shortly cordate at the base, firm in texture, green and rather glossy above, opaque beneath, with 10-12 pairs of very distinct arcuate-ascending veinlets. Female flowers arranged in several dense spikes 2-3 in. long, which form a terminal panicle, sessile and clustering on the branchlets. Fruit 6 in. in diam., dark brown, coriaceous in texture, with a distinctly veined epicarp, girt at the base by the persistent campanulate perianth, with its ovate segments, and inserted on a much-raised disk. Male flowers not seen.—Baron 2330 ! 2450! Allied to B. angolensis, Müll. Arg. ACALYPHA HOLOGYNA, N. Sp. . MN ilosis, folii longis Fruticosa, monoica, ramulis junioribus obscure pilosis, foliis oblong 2K2 442 MR. J. G. BAKER ON THE acutis membranaceis serratis subglabris petiolatis, floribus masculis in spicas graciles cylindrieas dispositis glomeratis bracteis ovatis scariosis persistentibus, floribus foemineis 1-2ms ad foliorum superiorum axillas peduneulatis, bracteis unifloris orbicularibus hispidis serratis, stigmatibus capillaceo-multifidis. A shrub with slender woody glabrous old branchlets, those of the year zigzag and obscurely pubescent. Leaves alternate, 3-4 in. long, about an inch broad, acute, distinctly serrated, thin in texture, green on both surfaces and very nearly glabrous ; petiole about 4 in. long. Male flowers in copious cylindrical spikes 2-9 in. long from the older leafless branchlets ; flowers several in a cluster; bud green, globose, glabrous, 34 in. in diam.; segments ovate; bracts brown, scariose, persistent, y in. long. Female flowers one to a cluster, enclosed in a foliaceous toothed bract which is orbicular when folded up and # in. in diam., the clusters usually solitary, rarely geminate, on a short peduncle from the axil of a leaf on the young branchlets. Stigmas j-j in. long, cut into very numerous capillary segments.— Baron 2889! MACARANGA MYRIOLEPIDA, D. Sp. Arborea, ramulis gracilibus apice lepidotis obscure pilosis, foliis parvis confertis brevissime petiolatis obovatis obtusis subcoriaceis facie viridibus glabris dorso densissime albido-lepidotis, floribus masculis ignotis, floribus foemineis iu spicas axillares breviter pedunculatas dispositis, perianthii segmentis parvis oblongis, fructu globoso magnitudine pisi parvi lepidoto uniloculari, stylo brevissimo. A much-branched erect shrub or tree, with terete branchlets glabrous except at the very tip. Leaves about an inch long, very obtuse, rounded at the base, moderately firm in texture, green and glabrous above, covered beneath with minute greenish- white lepidote scales, the 5—6-jugate arcuate-ascending main veins distinet from midrib to margin. Female spikes of 4-8 crowded flowers on short peduncles from the leaf-bearing branch- lets. Perianth-segments minute, persistent. Fruit lin.in diam., densely lepidote, finally splitting open to the base, filled up with the single brown seed.— Baron 3133! MACARANGA RIBESIOIDES, n. sp. Arborea, glabra, ramulis validis, foliis longe petiolatis obovatis cuspi- datis rigide coriaceis utrinque viridibus, floribus masculis ignotis, floribus foemineis in racemos copiosos axillares pedunculatos dispositis, perianthii segmentis parvis ovatis, fructu globoso magnitudine pisi parvi uniloculari, stigmate elongato, FLORA OF MADAGASCAR 443 A much-branched tree, glabrous in all its parts. Leaves 2-3 in. long, $-1} in. broad above the middle, more or less cuspidate, narrowed below the middle to a rather rounded base, moderately firm in texture, green and glabrous on both surfaces, with 6-8 pairs of arcuate-ascending main veins distinct from midrib to margin. Female flowers in racemes of 6-8 on short peduncles from the flowering branchlets ; bracts minute, deltoid ; pedicels finally 35 in. long. Fruit globose, 4 in. in diam., lepidote, girt at the base by the persistent perianth; style j., in. long, subulate. —Baron 2898! CHHTACME MADAGASCARIENSIS, N. sp. Arborea, glabra, aculeata, foliis breviter petiolatis late oblongis rigide coriaceis lucidis, floribus foemineis solitariis axillaribus brevissime pedicel- latis, fruetu globoso stigmatibus 2 subulatis coronato. A tree, glabrous in all its parts, with the slender terete branches armed with an erecto-patent pungent spine 3-3 in. long from the axil of each leaf. Leaves alternate, very thick and rigid in texture, 2-21 in. long, 1-14 in. broad at the middle, glossy on both surfaces, subobtuse, unequal at the base, with a strong midrib and fine side-veins; petiole 4 in. long. Male flowers not seen. Drupes globose, 3 in. in diam., solitary on short pedicels in the axils of the leaves, with a large endocarp. Stigmas subulate, persistent, $ in. long.— Baron 2397! This is a genus allied to Celtis, of which only a single species is already known, which extends from the Cape to Angola and Niam-Niam Land. Frous (SUnosrIGMA) TILLEFOLIA, n. sp. . Glabra, stipulis parvis coriaceis, foliis alternis petiolatis cordato-orbicu- laribus subcoriaceis scabris, receptaculis magnis solitariis sessilibus glo- bosis. A shrub, with suleate moderately stout brown glabrous branch- lets. Leaves 4—5 in. long and broad, deltoid at the apex, entire, decidedly cordate at the base, green and glabrous on both sur- faces, scabrous above, with 5-6 pairs of erecto-patent main veins reaching the margin; petiole 1-12 in. long; bracts small, rigid in texture. Receptacles 1 in. in diam., very coriaceous, glabrous, sessile from the side of small leafy branchlets.— Baron 3285! Ficus (SUnosTIGMA) SPILEROPHYLLA, n. Sp. mE F. ramulis glabris, stipulis membranaceis acuminatis, foliis alternis petio- latis magnis cordato-orbicularibus integris cuspidatis tenuibus glabris, 444 MR. J. G. BAKER ON THE receptaculis globosis glabris magnitudine mediocribus axillaribus breviter pedunculatis, bracteis 3 patulis deltoideis. A shrub, glabrous in all its parts, with moderately stout terete branchlets. Leaves 6-8 in. long and broad, deeply cordate, acutely cuspidate at the apex, thin in texture, green and glabrous on both surfaces, with about 5 pairs of parallel erecto-patent main veins; petiole about 1 in. long; stipules acuminate, brown, sca- riose, about the same length. Receptacles 4 in. in diam., on pedun- cles of about the same length, from the leaf-bearing branchlets ; bracts 8, small, spreading. Perianth-segments lanceolate. Style eylindrical.— Baron 2381 ! Ficus (SUnRosTIGMA) PODOPHYLLA, n. Sp. Glabra, stipulis parvis, foliis alternis longe petiolatis cordato-ovatis integris acutis subeoriaceis, receptaculis globosis glabris magnitudine mediocribus longe pedicellatis ad ramos haud foliatos productis, bracteis 3 deltoideis suffultis. A shrub, glabrous in all its parts, with slender terete woody branchlets. Leaves like those of the balsam-poplar in shape and size, 3-4 in. long, very acute, quite entire, slightly cordate at the base, moderately firm in texture, green and glabrous on both sur- faces, with distant erecto-patent main veins; petiole slender, 2-3 in. long. Receptacles globose, 4 in. in diam., geminate on pedicels about an inch long from thickened leafless branchlets ; bracts 3, deltoid, persistent, squarrose. Style twice as long as the young ovary. Perianth-segments lanceolate, as long as the ovary and style.— Baron 3323! Ficus ($UnosTIGMA) MEGAPODA, n. sp. Glabra, stipulis magnis coriaceis acuminatis, foliis alternis longissime petiolatis cordato-ovatis acutis integris subcoriaceis, receptaculis axil- laribus solitariis pedunculatis globosis glabris magnitudine mediocribus, bracteis parvis deltoideis. A shrub, glabrous in all its parts, with moderately stout woody branchlets. Leaves like those of the balsam-poplar, 3—4 in. long, cuspidate at the apex, quite entire, broadly rounded or cordate at the base, moderately firm in texture, green and gla- brous on both surfaces, with distant erecto-patent main veins; 1 a —91 i . 1 l1 . petiole 2-33 in. long; stipules 14 in. long, glabrous, acuminate. Receptacles £ in. in diam., glabrous, with a thick very coriaceous skin, on a peduncle of about the same length from the axil of a leaf; bracts 3, small, persistent. Perianth-segments lanceolate, FLORA OF MADAGASCAR. 445 twice as long the chestnut-brown fruit, which is tipped with a cylindrical style.— Baron 2536 ! 3305! Ficus ($UnosTIGMA?) TRICHOPHLEBIA, n. Sp. F. ramulis glabris, stipulis parvis pilosis coriaceis, foliis alternis longe petiolatis cordato-orbicularibus cuspidatis integris subcoriaceis facie scabris dorso pubescentibus, receptaculis globosis pilosis parvis ad ramos efoliatos paniculatis, bracteis minutis deltoideis. A shrub, with moderately stout suleate woody glabrous branch- lets. Leaves 5-7 in. long and broad, deeply cordate at the base, deltoid with a small cusp at the apex, green on both sides, very scabrous above, clothed all over with short fine brown pubescence beneath, with about 5 pairs of strong erecto-patent main veins; petiole 22—3 in. long; bracts under an inch long. Receptacles globose, pilose, 4 in. in diam., mammillate at the apex, arranged in a copious terminal panicle on leafless branchlets ; bracts 3, minute, deltoid; pedicels 1-3 in. long. Ovary with a long unilateral style. Perianth-segments lanceolate, as long as the ovary and style. Fruit subglobose, dark brown.— Baron 2417! Ficus (SUnosTIGMA) APODOCEPHALA, n. sp. F. ramulis validis glabris, stipulis magnis lanceolatis membranaceis, foliis alternis petiolatis oblongis pedalibus rigide coriaceis utrinque glabris, venis primariis 8-10-jugis ascendentibus intra marginem anastomosantibus, receptaculis sessilibus axillaribus parvis globosis pilosis, bracteis 3 magnis deltoideis coriaceis suffultis. A shrub, with strongly-angled glabrous branchlets 2 in. in diam. Stipules brown, membranous, 14 in. long. Petiole varying in length from 1 to 8 inches; blade nearly a foot long, 4-44 in. broad, obtuse, rounded at the base, rigid in texture, green and glabrous on both surfaces, with 8-10 pairs of strong ascending main veins anastomosing in arches within the margin. Recep- tacles globose, sessile, geminate in the axils of the leaves, z in. in diam., subtended at the base by three large coriaceous persistent deltoid bracts. Perianth-segments lanceolate. Ovary with a long style.— Baron 2521! URERA SPHJEROPHYLLA, N. Sp. MM Sarmentosa, ramulis gracilibus glabris apice parce setosis, foliis petio- latis orbicularibus cuspidatis modice firmis glabris ad costam faciei infe- floribus foemineis in paniculas axillares ramulis rioris parce hispidis, ovario glabro ovoideo obliquo, stigmate divaricatis spicatis dispositis, magno capitato. 446 MR. J. G. BAKER ON THE A subscandent shrub, with slender terete woody glabrous branchlets, bristly only at the young tips. Leaves moderately firm in texture, 12-2 in. long and broad, bluntly cuspidate, entire, broadly rounded at the base, glabrous except a few minute bristles on the midrib beneath. Male flowers not seen. Female flowers in sessile axillary deltoid panicles about an inch long, with a few spreading slender slightly bristly spicate branches. Fruit glabrous, oblique ovoid, 4l; in. long, crowned with a large orbicular sessile stigma.—Baron 3179! Nearly allied to the Mauritian U. acuminata, Gaudich. PILEA CAPITATA, n. sp. Herbacea, perennis, inermis, dioica, foliis oppositis breviter petiolatis valde inzqualibus oblongis acutis serratis, floribus in capitulos axillares longe pedunculatos dispositis, perianthio masculo tubo campanulato seg- mentis 4 parvis deltoideis, perianthii foeminei segmentis valde inzequalibus, fruetu ovoideo compresso acute marginato. An erect herbaceous perennial, with simple glabrous unarmed stems under a foot long. Leaves in opposite pairs, shortly petioled, very unequal in size, moderately firm in texture, di- stinctly serrated, triplinerved from base to apex, green and slightly hispid on both surfaces, the larger ones 14-2 in., the smaller 3-$ in. long. Flowers of both sexes in dense capitate cymes on long axillary ascending peduncles. Male perianth 3g in. long, gamophyllous, with 4 small deltoid segments. Fruit ig in. long, brown, flat, glabrous, not at all oblique, tipped with a minute capitate stigma.— Baron 2528! 2621! Belongs to the small section Heterophylle, of which only two Old-World species are already known. . PILEA LONGIPES, n. sp. Herbacea, perennis, inermis, dioica, foliis longe petiolatis ovatis æquali- bus serratis triplinerviis, floribus in paniculos cymosos longe peduncu- latos axillares dispositis, perianthio masculo tubo gamophyllo campanu- lato, segmentis 4 parvis deltoideis. An erect unarmed herbaceous perennial with much-branched stems under a foot long. Leaves moderately firm in texture, 1-13 in. long, triplinerved from base to apex, subobtuse, di- stinctly serrated, rounded at the base, obscurely hispid on both sides; petiole 4-2 in. long. Male cymes corymbose at the tip of a long peduncle ascending from the axils of the upper leaves. Male perianth 4!; in. long, with a campanulate tube and 4 small ud A FLORA OF MADAGASCAR. 447 segments ; pedicel about as long as the perianth, articulated at the tip. Female flowers and fruit not seen.— Baron 3261! Allied to P. umbellata, Weddell, of Bourbon. PopocanPus ($EUPODOCARPUS) MADAGASCARIENSIS, n. Sp. Arborea, glaberrima, ramulis apice tetragonis, foliis sessilibus lanceolatis rigide coriaceis uninerviis, floribus foemineis solitaniis axillaribus sessilibus, receptaculi bracteis oblongis, fructu globoso magnitudine pisi. A much-branched tree, with verticillate final branchlets. Leaves alternate, very rigid in texture, reaching a length of 5-6 in. and a breadth of 4 in., acuminate, narrowed to the base, drab and rather glossy on both surfaces, flat, furnished with a distinct midrib, all the other veins hidden, the margin very narrowly recurved. Male flowers not seen. Female flowers solitary in the axils of the leaves. Receptacle 4; in. long, the oblong bracts free at the obtuse tip. Fruit hard, globose, brownish-black, smooth, glaucous, 4 in. in diam.— Baron 2794! 3129! Parker! * Wild and abundant in the forests of the interior. Wood ex- tensively used in house-building, for flooring &c. Native name ‘Hetatra.’” Nearly allied to the Cape P. Thunbergii, Hook., in London Journ. Bot. i. 657, t. 221. MONOCOTYLEDONS. PANDANUS (§SussEA) MICROCEPHALUS, n. sp. P. folis ensiformibus sesquipedalibus rigidulis aculeis crebris ascen- dentibus marginatis, capitulis fructiferis subsessilibus solitariis parvis globosis, drupis 50-60 oblanceolato-hexagonis 1- raro 2-locellatis dimidio superiori liberis, apice libero oblongo-hexagono, stigmate magno reni- formi adpresso. Leaves densely crowded at the summit of the peduncle, 1i ft. long, an inch broad at the base, tapering to the point, not very rigid in texture, margined with close minute erecto-patent prickles. Fruit-head globose, solitary, nearly sessile, #-1 in. in diam. Drupes 50-60, usually 1-celled, free in the upper half, 4 in. long, j in. in diam.; free apex oblong, angled ; solitary stigma horny, reniform, flat, 4'; in. in diam., occupying the whole tip.— Baron 2321! Native name, *Isirihoalavo." Allied to Sussea conoidea, Gaudich., Atlas Bonite, tab. 24 ; but that species has an oblong head 4-5 in. long and only the upper third or quarter of the drupes free. 448 MR. J. G. BAKER ON THE PANDANUS (§SUSSEA) OLIGOCEPHALUS, n. Sp. P. folis ensiformibus sesquipedalibus aculeis crebris minutis erecto- patentibus marginatis, capitulis 2-3 parvis globosis, drupis 20-30 1-lo- cellatis ohlongo-hexagonis dimidio superiori liberis, apice libero conico, stigmate sessili parvo reniformi. Leaves erowded at the top of the peduncle, ensiform, not very rigid in texture, 13 ft. long, an inch broad at the base, margined by minute close erecto-patent prickles. Fruit-heads 2-3 on a short peduncle, 1-1} in. in diam. Drupes 20-30 to a head, all 1- celled, obloug-hexagonal, 4 in. long, 4—4 in. in diam., free in the upper half; free apex conic, angled; stigma flat, sessile, reniform, js in. in diam., occupying the whole of the tip.—Forest 40 miles from the coast, Baron 1666! PANDANUS (§VINSONIA) CONCRETUS, n. sp. P. foliis 5-6-pedalibus ensiformibus rigide coriaceis, capitulis fructiferis solitariis magnis oblongis, drupis 40--50 hexagono-cuneatis apice rotundato solum liberis, stigmatibus 6-12 parvis planis reniformibus in areolam paulo elevatam circulariter dispositis segregatis. Fully-developed leaf 5 or 6 ft. long, very rigid in texture, 5 in. broad at the base, tapering gradually to the apex; prickles lanceolate-deltoid, erecto-patent, sometimes } in. apart. Fruit-heads solitary, oblong, 3-4 in. in diam. Drupes j-1 in. broad each way at the top, 1 in. in diam. at the base, connate ex- cept the top 7-3 inch; apex rounded, with a rather polished brown smooth epicarp, in the centre of which the 6-12 small flat reniform stigmas are disposed in an irregular circle on a slightly raised areole occupying the central half of the free apex; epicarp of the concrete portion splitting up into copious fine fibres ; endocarp thick and woody. Peduncle 1 in. in diam. just beneath the head.— Baron 2778! Differs from P. utilis (Gaudich., Atl. Bonite, tab. 17. figs. 1-5) by the drupes being free at the very top only, and tho stigmas not crowded but forming an irregular ring 1-1 in. road. PANDANUS (§VINSONTA) CERATOPHORUS, n. sp. P. folis ensiformibus rigide coriaceis 3-4-pedalibus aculeis marginalibus parvis crebris ascendentibus, capitulis fructiferis solitariis oblongis, drupis 90-60 hexagono-cuneatis 5-6-locellatis quarto superiore liberis, apice libero conico, stigmatibus confertis stylo producto uncinato deciduo. Fully-developed leaves ensifo igi iac : i 3 in. broad at the base, tapering M - n Ml § gradually from base to apex, FLORA OF MADAGASCAR. 449 margined with close small ascending prickles. Peduncle 4ft. long, its leaves at the top linear, not more than a foot long and lin. broad. Fruit-heads solitary, oblong, erect, 3 in. long, 2 in. in diam. Drupes 50-60 to a head, hexagonal, } in. in diam. at the top, 1 in. at the base, scarcely compressed laterally ; free apex conical, the 5-6 stigmas crowded in a small flattened space in its centre; epicarp breaking up into fine parallel fibres; endocarp brown and horny. Styles columnar, } in. long, horny, tapering to the apex, deciduous.— Baron 2320! Heads like those of Vinsonia humilis and Pervilleana, as figured by Gaudichaud, Atl. Bonite, t. 17 and 31. Styles like those of P. sechellarum, Balf. fil. DRACJOENA XIPHOPHYLLA, N. Sp. Arborea, folis sessilibus ensiformibus subcoriaceis viridibus 3-4-peda- libus eosta obscura, floribus in panieulam amplam deltoideam dispositis, ramis patulis dense racemosis bracteis primariis magnis lanceolatis, pedi- cellis dense glomeratis apice articulatis bracteis propriis minutis deltoideis, perianthii segmentis tubo cylindrico subzquilongis, staminibus segmentis :equilongis. Leaves 3-4 ft. long, 11 in. broad at the middle, tapering gradually to an acute point, narrowed to 3-1 in. above the dilated clasping base, firm in texture, closely veined, the midrib visible only when looked at from beneath and lost some distanee from the apex. Branches of the panicle shortly peduncled, the lower a foot long ; lower bracts 3—4 in. long; rhachis stout and deeply grooved; pedicels usually 6-8 in a cluster, § in. long, articulated at the apex. Perianth 2-3 in. long.; segments ligulate, equalling the cylindrical tube. Anthers small, oblong, placed opposite the tip of the perianth-segments. Berry usually with one seed, the size of a pea— Baron 2455 ! 2729! 2804! A fine plant, midway between D. fragrans and D. floribunda. DIOSCOREA ACUMINATA, n. 8p. Volubilis, glabra, caulibus gracillimis bulbilliferis, foliis petiolatis cordato-ovatis integris acuminatis viridibus, floribus foemineis solitariis in spicas laxas axillares dispositis, bracteis minutis deltoideis, pedicellis flori zquilongis, ovario obconico, perianthii segmentis viridibus orbicularibus, perianthio masculo fructuque ignotis. A herbaceous climber, glabrous in all its parts, with very : 3 illa the size of a pea in the slender stems, bearing globose bulbillx the size pe: e €) € ipe ` axils of some of the leaves. Leaves 2-3 iu. long, moderately : : > s, deeply cordate, with firm in texture, green on both surfaces, deeply cordat 7 450 MR. J. G. BAKER ON THE main ribs radiating from the apex of the petiole. Female racemes solitary in the axils of the leaves, few-flowered, simple, 1-15 in. long, shortly peduncled; pedicels solitary, y} in. long; bracts minute, deltoid. Expanded perianth greenish, j in. in diam. ; segments orbicular, equalling the obconic ovary.— Baron 2654! ERIOCAULON FLUITANS, Baker in Journ. Linn. Soc. xx. p. 277 (Baron 926), proves conspecific with E. melanocephalum, Kunth CE. aquaticum, Sagot.), of Guiana and North Brazil. (Sagot 1330! Burchell 4208 !) CYPERUS ALTERNIFOLIUS, Linn. Of this, which is common in gardens, Mr. Baron has now sent wild specimens (2707 !), the first we have received. CYPERUS (§Papyrus) IMERINENSIS, Boeckl. in Engler Jahrb. 1884, p. 500. Founded on Hildebrandt’s 3798. Very nearly allied to the Egyptian Papyrus.— Baron 3274! CYPERUS DIVULSUS, Ridley in Journ. Bot. 1884, p. 15. Hil- debrandt’s collecting number should be 4020, not 4080 as printed. Dr. Boeckler has described the same plant under the name of C. paucispiculatus, in Engler's Jahrbuch for 1884, p. 497. DicHRONEMA CANDIDA, Baker.—Psilocarya candida, Nees in Mart. Fl. Bras. ii. p. 117.—Rhynchospora candida, Boeckl. in Linnea, xxxvii. p. 605.—Cyperus leucostachys, Bojer MSS.— Baron 2340! Bombatoka, Bojer! Known elsewhere in West Tropical Africa and Guiana. HELEOCHARIS ($HELEOGENES) CJESPITOSISSIMA, n. Sp. _Densissime ceespitosa, stolonifera, foliis propriis nullis, caulibus gra- eillimis 13-2 pollicaribus, floribus 5-6 in spieam oblongam terminalem dispositis, glumis ovato-lanceolatis castaneis obtusis valde imbricatis margme pallidis dorso viridi carinatis, nuce oblonga, stylo trifido, setis hypogynis 6 nuci zquilongis. j Stems forming very dense masses and sending out a few rooting stolons, very slender, tetragonous, greenish, 14-2 in. long. Spike oblong, terminal, j5j- in. long. Glumes about 6 to a spike, erecto-patent, much imbrieated, obtuse, navicular, bright "etant brown, with a distinet pale keel, and becoming gradu- ally whitish towards the edge. N v, greeni "as long as the glume, enclosed in the when f aiat Live ; pale ciliated hypo- gynous bristles, which are united into a distinct whitish cup at FLORA OF MADAGASCAR. 451 the base.— Baron 2242! Nearly allied to H. chetaria, Reem. and Sch., and H. minuta, Boeckl. in Engler's Jahrb. 1884, p. 503. CrADIUM ($MACILERINA) PANTOPODUM, n. sp. Perenne, 2-3-pedale, foliis distichis linearibus rigide coriaceis, spiculis clavatis omnibus pedicellatis in paniculam angustam elongatam dispositis, glumis 6-/ castaneis inferioribus parvis ovatis, superioribus oblongis, floribus 1-2, staminibus 3 linearibus, stylo profunde trifido, setis hypo- gynis 6 subulatis ciliatis. Tufts densely cespitose. Basal leaves rigid in texture, 1-11 ft. long, 4 in. broad, tapering to a point, nearly flat on the face, rather convex on the back, with a suddenly dilated oblong brown base with a erisped edge. Stems 11-2 feet below the inflores- cence, with a few reduced leaves. Panicle a foot long, with distinct ascending corymbose branches. Spikelets l in. long, with 3—4 small ovate empty glumes at the base and 3 longer oblong ones. Styles 4 in. long, protruded from the apex of the spikelet. Hypogynous bristles very slender, about as long as the flowering glumes.— Baron 2072! 3316! Allied to the West- Indian Macherina restioides, Vahl, and M. filifolia, Griseb., and perhaps also to C. aipholepis of the Seychelles, which is known only in an immature condition. Crapium ($MacuznINA) MELLERI, n. sp. Perenne, orgyale, foliis distichis linearibus rigide coriaceis, spiculis cylindrieis omnibus pedicellatis in panieulam angustam elongatam dispositis, glumis circiter 15 castaneis inferioribus multis parvis ovatis, superioribus paucis magnis oblongo-lanceolatis, floribus 1-2, staminibus 3 linearibus, stylo profunde trifido, setis hypogynis 6 subulatis ciliatis. A near ally of the last, of which it may be possibly only a variety. It quite agrees with it in the structure of the flowers, but differs by its much more robust habit and taller growth, and much more numerous spikelets sheathed by 10-12 small closely imbrieated barren glumes at the base instead of only 3 or 4 and more acute produeed inner glumes. The whole plant reaches the height of a man, the panicle and leaves being each about 3 feet in length, the latter } in. broad at the top of the large dilated base and tapering gradually to an acuminate apex.— Baron 1026! 2846! Between Tamatave and Antananarivo, Dr. Meller ! Carex DARONT, n. sp. p Dense cæspitosa, foliis linearibus elongatis, caule gracili acute triquetrc, 452 MR. J. G. BAKER ON THE spicis foemineis 5-6 cylindricis pedunculatis cernuis superne densis inferne laxis, glumis oblongis brunneis dorso nervis 3 viridibus percursis, peri- gynio ellipsoideo lenticulari viridi obscure rostrato faciebus multinervatis, stylo bifido, spica mascula unica terminali. A densely cæspitose swamp species. Leaves linear, 2-3 ft. long, 2 in. broad at the base, erect, firm in texture, closely ribbed, glabrous, tapering to the point, prominently costate. Stem 11-2 ft. long below the inflorescence, slender, sharply 3-angled. Female spikes 5-6, cylindrical, cernuous, 4—5 in. long, 3-1 in. in diam., the lower distinctly peduncled and subtended by long leaves, the upper with a few male flowers at the top. Glumes obtuse, i in. long, brown and membranous towards the edge, firm in texture in the centre, with 3 strong ribs. Perigyne as long as the glume, greenish, distinetly margined, narrowed into a short entire beak, each face with several distinet vertieal ribs. Male spike erect, terminal, 4-5 in. long, its glumes like those of the female.— Baron 2795! Nearest C. stricta of our British species, and a near ally of C. madagascariensis, Boeckl. in Engler's Jahrb. 1884, p. 517, founded on Hildebrandt's 3758. OPLISMENUS BROMOIDES, n. sp. Perennis, vaginis dense pilosis, foliis linearibus, spiculis subcylindricis solitariis vel geminis sessilibus vel pedicellatis in paniculam elongatam ramis paucis simplicibus ascendentibus dispositis, gluma sterili exteriori in aristam longissimam scabram producta, reliquis oblongo - lanceolatis acutis. A perennial, with slender cylindrical stems 1-11 ft. long, ascending from a decumbent base. Leaves 3-4 in. long, $-4 in. broad, glabrous except towards the base; sheaths about an inch long, densely pilose both upon the surface and margin. Panicle 6-8 in. long, with a few very ascending long slender simple secund branches. Spikelets laxly disposed, usually solitary and sessile, 4 in. long. Outer sterile glume oblong- lanceolate, 5-nerved, nearly as long as the spikelet, produced into a long awn; inner sterile glume just like the outer in shape and texture, but not awned. Flowering-glume oblong-lanceolate, acute, glabrous, just like the sterile glumes in texture, distinctly 5-nerved on the back.— Baron 3213 ! Nearly allied to O. setarius, Reem. & Sch. (0. acuminatus, Nees). qa - ECHINOLENA MADAGASCARIENSIS, n. 8p. E. foliis magnis lanceolatis, vaginis margine ciliatis, spiculis dense spicatis, FLORA OF MADAGASCAB. 453 rhachi applanata, glumis 4 oblongo-lanceolatis, exteriori pilis patulis basi bulbosis dense vestito, flosculo inferiori abortivo neutro. Stems a foot or more long, slender, ascending, glabrous, terete, branched from the base. Sheaths of the numerous leaves about an inch long, densely ciliated all down the edge; blade thin, acuminate, glabrous, oblique at the base, 8-5 in. long, 4-4 in. broad. Spikes terminal or axillary, peduncled, 1-11 in. long, simple or compound at the base; rhachis flattened, produced into a long point at the top beyond the spikelets. Spikelets dense, unilateral, erecto-patent, 4 in. long. Outer sterile glume oblong-lanceolate, acuminate, as long as the spikelet, densely clothed with firm spreading hairs with a bulbous base. Lower flower small, very imperfect. Upper flower with an oblong- lanceolate 5-nerved pilose flowering-glume 4j in. long, and a pale of about the same length.— Diego Suarez, Bernier! (Herb Gay). It is fully described and figured by Gay in manuscript, but not named. The only other species of this curious genus is a plant of Guiana and Brazil. PENNISETUM (§GYMNOTHRIX) TRITICOIDES, n. sp. Perenne, dense czespitosum, foliis lineari-setaceis erectis glabris elongatis vaginis apice ciliatis, spiculis solitariis erectis in spicam cylindricam dis- positis, setis pluribus rigidis scabris valde inzqualibus unica reliquis longiori, glumis 2 exterioribus parvis ovatis, 3 oblongo-lanceolatis palez zequilongis. A densely-tufted perennial, with stems 1-2 ft. long. Leaves 1-1 ft. long, linear, } in. broad at the clasping base, tapering gradually upwards into a setaceous point. Spike simple, erect, cylindrical, 4-5 in. long, 4 in. in diam. “Spikelets solitary, erect, cylindrical, 4 in. long, tapering to the point. Bristles a dozen or more, slender, rigid, distinctly ciliated, one much longer than the rest and about twice the length of the spikelet. Outer sterile glume very small, ovate; the other larger, about half as long as the spikelet. Outer flowering-glume oblong-lanceolate, acute, but not awned, firm in texture, pale green, glabrous, 5- nerved.* Stigmas protruding from the tip of the closed spikelet. — Baron 683! 3239! 3294! Parker! Nearly allied to the Abyssinian P. riparium, Hochst.! Resembles in habit our European Triticum caninum. Paortosracuys GEMINATUM, Hackel—In Sitz. Vien. Akad. Wissen. 1884, p. 131, Hackel has described a new endemic genus of 454 MR. J. G. BAKER ON THE Madagascar grasses under the name of Pacilostachys, one species of which is the plant I described in Journ. Linn. Soe. xx. p. 300 (Baron 1061, 3226) under the name of Lophatherum geminatum, and a second (D. Hildebrandtii, Hackel), founded on Hilde- brandt's No. 3759. Eragrostis ($PrEROESSA) MAXIMA, Baker. — Megastachya maxima, Bojer MSS. Erecta, robusta, foliis lanceolatis basi cordatis, spiculis 6-12-floris oblongis complanatis in panieulam amplam laxissimam rhomboideam dis- positis omnibus pedicellatis, rhachilla continua, glumis sterilibus oblongis floriferis minoribus, glumis floriferis oblongis distichis lateralibus com- plicatis, paleis hyalinis ciliatis. Stems erect, 2-3 ft. long. Leaves lanceolate, acuminate, thin in texture, 4-6 in. long, 4-2 in. broad, cordate at the base; sheaths of the stem-leaves 2-3 in.long. Panicle erect, rhomboid, very lax, 6-9 in. long and broad; main branches erecto-patent; pedicels capillary, erect, 1-2 in. long. Spikelets oblong, flattened, 3-2 in. long; rhachis not articulated. Outer sterile glume not more than half as long as the flowering-glume. Flowering-glume oblong, 4 in. long, glossy, drab, firm in texture, with flattened sides and a scabrous keel. Pale oblong, hyaline, about half as long as the flowering-glume, with a strongly ciliated margin. -Caryopsis ovoid, brown, glossy— Baron 2551! Gathered also long ago by Bojer. One of the most robust and ornamental species of the genus. Nastus CAPITATUS, Kunth.— Baron 2591! CEPHALOSTACHYUM CHAPELLIERI, Munro.—Forest of Andran- galoaka, Dr. Parker! Baron 2564! Mr. Baron’s specimens have heads composed entirely of the barren spikelets, and show well the reduced bract-like leaves, hairy inside, which conceal the heads in an early stage, and also the curious deciduous deeply fimbriated lanceolate ligules. The three Bamboos which Mr. Baron and Dr. Parker have collected all prove identical with species known previously, FILICES. LYCOPODIUM MEGASTACHYUM, n. sp, Dendricolum, caule erecto subpedali dichotomiter fureato, foliis densis reflexis lanceolatis rigide coriaceis, costa ad faciem inferiorem perspicua, FLORA OF MADAGASCAR. 455 spicis 3-1nis magnis cylindricis pendulis simplicibus vel furcatis, bracteis ovatis ascendentibus coriaceis fructu longioribus. A robust epiphytic species, with erect forked stems about a foot long. Leaves crowded, reflexed, rigid in texture, bright green, 4 in. long, with the midrib not visible on the upper surface, but distinct beneath. Spikes 3-4 to a branch, pendulous, 1 in. diam., sometimes 5-6 in. long. Bracts quite different from the leaves in shape and direction, uniform, ovate, multifarious, 1-1 in. long, yellowish green, coriaceous. Capsule orbicular, j!; in. diam.— Baron 2840! Belongs to the group of L. PAlegmaria. I do not think that L. pachyphyllum, Kuhn, founded on Hilde- brandt’s 4141 will prove to be distinct specifically from Z. obtusi- folium, Sw. It was also collected many years ago by Mr. W. Pool. ALSOPHILA BARONI, n. sp. Arborea, frondibus deltoideis tripinnatis utrinque viridibus przter costas pinnularum glabris, pinnis petiolatis oblongo-lanceolatis, pinnulis sessilibus lanceolatis, segmentis tertiariis lanceolatis integris, venis 8-10- jugis, inferioribus compositis, soris parvis globosis costularibus. A tree-fern, with the habit of A. australis and A. excelsa. Rhachis of the pinne pale brown, unarmed, glabrous on the underside, clothed on the upper surface with minute red-brown erisped hair-like pales. Fronds moderately firm in texture, green above, pale green beneath. Pinns about a foot long. Pinnules crowded, $-3 in. broad, cut down to the paleaceous rhachis into distinet adnate areuate-ascending tertiary segments i5 in. broad, with reflexed edges; veins fine and indistinct, the lower with 3—4 branches, the central ones erecto-patent, deeply forked. Sori minute, globose, superficial.— Baron 3143 ! LINN. JOURN.—BOTANY, VOL. XXI. 456 MR. H. N. RIDLEY ON THE The s of Madagascar. By Heyry N. Rivtey, M.A., F.L.S., Assistant, Botanical Department, British Museum. [Read 15th January, 1885.] (Pirate XV.) Tur Orchidex of Madagascar, as far as they are at present known to me, belong to 30 genera containing nearly 140 species; but it is to be expected that this number will be largely increased when the botanical riches of the country are more fully explored. This paper must therefore be only considered as a prodromus, giving an account of the species hitherto described or figured, together with those novelties which have come under my personal observation in the great herbaria of the British Museum and Kew. It would at present be premature to base any arguments as to the origin of the flora of Madagascar upon the distribution of the genera and species of Orchidee as at present known; but it will be of interest to examine the list and compare it with that of Africa and Tropical Asia. The Epidendreee are represented by 6 genera, two of which, Oberonia and Cirrhopetalum, are interesting from their absence from Africa, the remainder also being more extensively deve- loped in Tropical Asia than in Africa. Of the Vande there are 11 genera, four of which, so far as is certainly known, are confined to the Mascarene archipelago ; one, Polystachya, is dis- tributed over both hemispheres ; the remainder are either exclu- sively African, as Lissochilus, or are most abundant in Southern and Tropical Africa, with outlying species in Southern Asia. The genus Acampe, however, is probably more of an Asiatic type than of an African one. The small number of Neottiex gives somewhat of an African facies to the list. There are only 4 genera: one, Gymnochilus, is exclusively Mascarene; the others consist of the two widely distributed genera Corymbis and Pogonia, and Monochilus, which is chiefly Malayan. The Ophrydew are very well represented. There are eight genera, of which two are only known from Madagascar, viz. Bicornella and Platycoryne ; one i8 found also in the other islands of the archipelago, viz. Cynorchis. Of the rest, two occur also in Africa ; and two, Disperis and Saty- rium, while occurring in India, are most abundant in Africa. ORCHIDS OF MADAGASCAR. 457 Thus, broadly speaking, we may say that the Epidendree are typically Asiatic, while the remainder are more of an African character. As might be expected, a large proportion of the species are endemic; and but few have a distribution further than the archi- pelago or neighbouring coasts of Africa. The most widely spread species are Cirrhopetalum Thouarsii, perhaps the most widely distributed of all epiphytic Orchids, extending its range as far east as the Society Islands, and Corymbis corymbosa, which is found also in West Africa. OBERONIA. O. Brevirorta, Lindley, Gen. et Sp. Orch. Pl. p. 16; Folia Orchid. no. 36; Ad. Brongniart, Voy. ‘Coquille, t. 40 B; S. Moore, in Fl. Maur. § Seych. p. 341.—Cymbidium equitans, Thouars, Orch. Iles Afrig. t. 92.—Pleurothallis disticha, A. Richard, Orch. Maur. t. 8. fig. 1; Bojer, Hort. Maur. p. 319.— Malaxis brevifolia, Reichenbach fil., Otia Hamburg. ii. p. 73. Madagascar, Imerina, Deans Cowan! Mauritius, Wallich! Sir James Macgregor! in Herb. Brit. Mus. Rodriguez, Balfour, Transit-of-Venus Expedition, no. 1223 !; Herb. Brit. Mus. Bour- bon. Comoro Islands, Johanna, at 1000-1400 metres, Hilde- brandt no. 1709. This little plant appears to be widely distributed over the East-African islands. I have not seen it, however, from the main land. It isthe most western species of the genus, which has its headquarters in Tropical Asia, especially India, and has outlying species also in Australia, New Zealand, and Polynesia. The stem varies from less than 1 inch in height to 3 inches, and is covered with equitant leaves, often as much as 13 inch long. The raceme is sometimes short and erect ; at others long, slender, and nodding, attaining a length of 4 inches. The bracts are as long as the ovary in the flower; but in fruit the pedicel as well as the ovary lengthens, so that the bract is then only as long as the pedicel. The flowers are very minute, yellowish green. Liparis. The genus Liparis has a distribution only surpassed among Orchids by Habenaria. The headquarters of the genus is to be found in Tropical Asia. None of the Madagascar species are 2L2 458 MR. H. N. RIDLEY ON THE known outside the Masearene archipelago, and all but one belong to the section Mollifoliz. Lipaurs cxsPrtosa, Lindley, Bot. Reg. sub t. 882 ; Gen. et Sp. Orch. Pl. p. 32; A. Richard, Orch. Maur. p. 93; Bojer, Hort. Maur. p. 321; S. Moore, in Flora Maurit. § Seych. p. 342.— Malaxis cæspitosa, Thouars, Orch. Iles Afr. t. 90.—L. angusti- folia, Lindley, Orch. Pl. p. 31.— Malaxis angustifolia, Blume, Bijd. p. 893. Madagascar, Imerina, Deans Cowan. Mauritius, Bojer, Herb. Brit. Mus.; Bouton. Bourbon, Bernier, Herb. Kew. Java, Forbes, 914. Sumatra, Forbes no. 2574 b (pars). Ceylon, Thwaites no. 2351 in Herb. Brit. Mus. . This little plant is allied to L. viridiflora. It is generally about 3 inches or 34 inches in height. The leaf 14-2 inches in length, and 2 lines across in the broadest part. The flowers are given as white in Thouars’s drawing ; but in the dried specimens they appear greenish. I cannot distinguish the Mascarene plant from the Malayan L. angustifolia. L. LUTEA, n. sp.—Terrestris semipedalis, caule basi dilatato, foliis lanceolatis ; scapo gracili tri- vel plus vaginato ; racemo laxi- usculo, floribus parvis flavis 8, rhachidi appressis ; bracteis ovatis lanceolatis acuminatis, ovarii dimidio æquantibus ; sepalo postico angusto lineari obtuso, lateralibus ovatis falcatulis, labellum paullo superantibus ; petalis linearibus quam sepalum posticum multo angustioribus et longioribus, labello ovato cordato obtuso, | in medio venis tribus; columna brevi erassiuscula, parum curva. Ankafana, in palude, rara, Deans Cowan. This plant attains the height of 62 inches. The leaves are 2 inches long by 3 an inch across. The flower-spike is erect, rather stiff; the flowers are not patent, but appressed to the raceme, small, bright yellow. "The column is short and rather thick, two thirds of the length of the dorsal sepal. L. BICORNIS, n. sp.—Terrestris, pseudobulbis nullis; foliis 3-5 lanceolatis acutis erectis; caule semipedali foliato, floribus circa 12, in apice caulis aggregatis ; bracteis linearibus lanceolatis acutis, ovarium sub:equantibus ; sepalo postico oblongo lanceolato obtuso, lateralibus ovato-lanceolatis obtusis ; petalis linearibus obtusis ; labello obeuneato emarginato, marginibus lateralibus — ORCHIDS OF MADAGASCAR. 409 inerassatis, callo bieornuto parvo ad basin; columna quam la- bellum breviore crassa, parum curva, basi et apice incrassata, alis parvis; anthera ovata obtusiuscula. Imerina, in paludibus, Hildebrandt 3849! in Herb. Brit. Mus. A rather small-flowered species with narrow erect leaves, a stem 7 to 8 inches in height, and a rather compact spike of about twelve flowers with stout pedicels. The column is rather stout and but little curved except at the upper part. The lip is obovate- cuneate, and emarginate, with a depression in the middle of its length, the edges at the base turning up flank the column ; there is a small two-borned callus close under the lower part of the column. LIPARIS LONGIPETALA, n. sp.—Terrestris, pseudobulbis ovatis parvis crassis; vaginis membranaceis albis tectis, foliis duobus lanceolatis acutis ; caule semipedali folia superante ; bracteis 7-10 remotis, inferioribus lanceolatis acutis brevibus, superioribus ovatis acutis vix dimidium ovarii equantibus ; floribus remotis, pedicellis tenuibus ; sepalo postico late lineari subacuto, lateralibus semiova- libus faleatis; petalis anguste linearibus longis, apice parum dila- latis, labello quam sepala multo breviore, oblongo, brevissime mucronato, basi angustato, callo parvo obtuso; columna brevi, basi angusta recta, parte summa curva, alis brevibus latiusculis acutis. In sylvis Ankafana, S. Betsileo, Hildebrandt, March 1881, no. 3980! in Herb. Brit. Mus. A smaller plant than the two preceding, with fleshy ovate pseudobulbs and lanceolate leaves gradually tapering to the base, 21-4 inches in length, 3 an inch broad in the broadest part. The stems are 5 inches in length, with from 7 to 10 scattered bracts. The flowers are smaller than those of the preceding, and 6 or 7 in number, with pedicels 4 lines in length. The lateral sepals are almost semicircular, 2 lines in length, the two together forming a circle; the narrow petals are just twice as long. The lip is much shorter than the lateral sepals, oblong, with a very short blunt process in the centre of the broad apex ; the base is much and suddenly narrowed, and in the middle line just below the apex of the anther is a short obtuse callus, apparently of an orange eolour. The column is rather short and but little curved ; the lateral wings extend but a short way along the column, but are rather broad and pointed, coming to a point in the middle. 460 MR. H. N. RIDLEY ON THE The apex of the anther is more acute than that of the two preceding species. Liparts FLAVESCENS, Lindley, Bot. Reg. sub t. 882 ; Orch. Pl. p.29; Ach. Richard, Orch. Maur. p. 51 ; S. Moore, Fl. Maur. § Seych. p. 342.—Malaxis flavescens, Thouars, Orch. Iles Afr. t. 25. Terrestris semipedalis, rhizomate diu repente, caule ebulboso basi vaginis membranaceis tecto ; foliis congestis ovatis petiolatis 12 uncia longis, 2 uncia latis, flaecidis ; racemo paucifloro ; sepalo postico anguste lanceolato, lateralibus ovato-lanceolatis faleatis ; petalis linearibus ; labello rotundato lato integro cuspidato, costis in medio tribus, ungui brevi; anthera ovata. Madagascar, Lyall! in Herb. Kew. Also found in Mauritius (Bojer), Seychelles (Horne), and Bourbon. This plant has along creeping rhizome, throwing up stems about 7 inches in height including the scape, with tufted ovate leaves. L. onNITHORRHYNCHOS, n. sp.—Terrestris, rhizomate repente ; pseudobulbis nullis; caule erecto semipedali, parte inferiore vaginis albis membranaceis tecta; foliis duobus, rarius tribus, ovatis acutis breviter petiolatis; floribus mediocribus 3-12 re- motis, pedicellis semiuncialibus ; bracteis ovatis lanceolatis, supe- rioribus lanceolatis acutis; sepalo postico lineari, lateralibus deflexis ovato-lanceolatis obtusis falcatis; petalis linearibus angustis, apice parum dilatatis; labello late cordato integro, ecalloso, basi angustato; columna paullo curva, basi angustata, alis latioribus longis ; anthera ovata rostrata. Ankafana, S. Betsileo, in sylvis, 4049 Hildebrandt; Deans Cowan in Herb. Brit. Mus. This terrestrial Liparis has some affinity with Z. Bowkeri, Harvey, a Cape plant. The lower part of the stem, about 2 inches, is covered with white membranous leaf-sheaths; above these are the leaves, generally two in number, ovate, with rather short petioles; they are 5-ribbed, and 2 inches in length by 1 in diameter. The flower-spike is rather lax, and bears from 3 to 12light-green flowers, rather large in size for a Liparis; their pedicels about 4 an inch. The linear dorsal sepalis 5 lines in length, twice the length of the column. The lateral sepals are somewhat similar in shape to those of Z. Bowkeri, rather broad aud faleate, 43 lines in length. In one specimen they were con- nate fora portion of their length, but separate at each end. The hip was broadly cordate, 5 lines long, base narrowed; a rather ORCHIDS OF MADAGASCAR. 461 broad bar runs down the centre, but there are no callosities. The column is almost straight, but slightly eurved in the base, the wings flanking the stigma. The anther is provided with a rather long green beak. LiPARIS LONGICAULIS, n. sp.—Terrestris, rhizomate repente ; caule pedali, parte inferiore (5-unciali) vaginis albis membranaceis tecta; folis ovatis acutis breviter petiolatis, caulinis ovatis acu- minatis ; floribus paucis ad 6 remotis inajusculis; bracteis lanceo- latis, 1-3 pedicellorum equantibus ; sepalo postico lineari obtuso longo, lateralibus ovatis lanceolatis obtusis faleatis, margine interiore irregulariter crenulato, venis conspicuis ; petalis angus- tissime linearibus; labello integro oblongo obtuso crenulato, quam sepala multo breviore, venis conspicuis, callo biconico obtuso parvo; columna longiuseula arcuata gracili, alis parvis tenuibus ; anthera ovata, apice obtusa. Ankafana, in sylvis, Hildebrandt no. 3979!; Deans Cowan in Herb. Brit. Mus. This is allied to LZ. ornithorrhynchos, differing in the longer basal portion of the stem covered with whitish membranous sheaths, the oblong lip with crenulate edge, and the curved slender column. The flowers are olivaceous green. The narrow dorsal sepal is 8 lines, the laterals 7 lines, and the lip 5 lines in length. L. ocHRACEA, n. sp.; rhizomate breviter repente, caule basi bulboso, 2-unciali; foliis circiter 5, ovatis vel ovatis-lanceolatis, scapo paucifloro, floribus majuseulis; bracteis ovatis acuminatis, caulem amplectentibus, ovario ferme æqualibus ; sepalo postico lineari, lateralibus oblongis quam labellum longioribus; petalis linearibus quam sepalum posticum angustioribus ; labello basi an- gustato longo, lamina abrupte deflexa oblonga truncata ; columna brevi curva crassiuscula. Ankafana, Deans Cowan (vidi iconem pictam). This plant differs from L. ornithorrhynchos, to which it is closely allied, in the bulbous base of the stem and short rhizome, the more numerous leaves, 4 inches in length by 2 inches in dia- meter, and the shape of the lip, which has a rather long narrow base and an oblong truncate lamina which is abruptly deflexed. The flowers are olivaceous ochraceous, the lip having a dark ochreous patch in the middle of the lamina. 1 have only seen a very good coloured drawing of this plant made by Deans Cowan. 462 MR. H. N. RIDLEY ON THE There is no specimen in his herbarium. The whole plant appears to be about 54 inches in height. LIPARIS PARVA, n. sp.—Epiphyta parvula pseudobulbo parvo elongato; foliis paucis lanceolatis vel ovato-lanceolatis; caule gracili paucifloro; bracteis ovatis, ovarii trientem equantibus ; floribus parvis olivaceis ; sepalo postico lineari, lateralibus lanceo- latis faleatis obtusis ; petalis linearibus, basi angustioribus; labello integro ovato obtuso; columna erassiuscula semitereti ferme recta, alis brevibus. Trees, Ankafana, Deans Cowan! in Herb. Brit. Mus. L. coNNATA, n. sp.—Terrestris, caule ebulboso ; foliis ovatis petiolatis congestis flaecidis', scapo brevi multifloro ; bracteis lan- ceolatis acutis, flores superantibus ; floribus mediocribus ; sepalis connatis ovatis obtusis, nervis conspicuis ; petalis linearibus 8- nerviis, angustis obtusis; labello oblongo-elliptico, venis promi- nulis rufescentibus (in sicco) presertim duabus medianis ad bases quorum calli cornuti duo, labelli marginibus crenulatis ; columna curva graciliuscula, alis brevibus acutis. Imerina, Deans Cowan! in Herb. Brit. Mus. It is possible that this plant is a monstrous state of some other Liparis; but all the flowers I examined on two plants seemed in the same state. Those, however, on one of the two plants were in a very young state. The bud appears to open along the upper part where the posticous sepal should be. The other two sepals form an ovate-obtuse organ bifid at the apex, the line of demarcation between them (a white nerve) being conspicuous; but I could see no trace of the posticous sepal. The veins on the perianth-segments are red and very conspicuous, especially those on the lip, two of which, running along the whole length, are very well marked, and terminate at the base in two short horn- like calli. The whole plant is about 9 inches in height, the leaves 8 inches long by 13 inch across. The flowers appear to be reddish, and are doubly saccate at the base. The leaves are also apparently tinted with red. BULBOPHYLLUM. A large genus distributed over the tropics of both worlds, being especially abundant in Tropical Asia. Besides the species here described, there are among Deans Cowan’s drawings figures of several apparently nondescript, but of which specimeus do not occur in the herbarium. ORCHIDS CF MADAGASCAR. 463 BULBOPHYLLUM wvTANS, Thouars, Orch. Iles Afriq. t. 107 ; folio t. iv. ; Lindley, Gen. and Sp. Orch. Pl. p. 52; Richard, Orchid. Maur. p. 63 ; S. Moore in Fl.of Maur and Seych. p. 345. Madagascar, fide Thouars; Ankaratra montes, Rutenberg. Mauritius, Ayres, Herb. Kew. Ins. Mascarenes, Sir J. Macgregor in Herb. Brit. Mus. B. MULTIFLORA, n. sp.; pseudobulbis quadrato-cylindricis, semiuncialibus, vaginis membranaceis tectis, nitidis rugosis flavis monophyllis ; folio lanceolato obtuso, apice bilobo, coriaceo ; ra- cemo denso nutante longo gracili ; floribus brevissime pedicellatis copiosis flavescentibus; bracteis membranaceis lanceolatis vel superioribus ovatis, acutis, ovarium superantibus ; sepalis lineari- bus lanceolatis acutis; petalis oblongis lanceolatis ; labello un- guiculato, lobo medio brevissimo carnoso linguiformi curvo, medio depresso marginibus evectis ; columna brevi crassiuscula, denti- bus clinandrii brevibus obtusis, antheram vix superantibus; anthera ovata depressa. Imerina, Ankafana, * on trees in great clusters, often covering the whole tree," Deans Cowan ! in Herb. Brit. Mus. B. Baroni, n. sp.; rhizomate longo lignoso; pseudobulbis dissitis oblongis flavis rugosis (siccis), monophyllis; foliis oblongis obeuneatis obtusis, bilobis erectis ; scapo longiusculo basi vaginis membranaceis tecto, racemo laxo, floribus pluribus parvis; bracteis ovatis obtusis vel subacutis, ovario equantibus ; sepalis ovato-lanceolatis; petalis lanceolatis obtusis multo minoribus; labello ovato subaeuto, curvo marginibus erectis minimo; columna dentibus brevibus obtusis. Andrangaloaka, E. Imerina, apud truncos arborum, floribus flavis, Hildebrandt no. 38728! Ankafana, Deans Cowan! Central Madagascar, Baron no. 1176! in Herb. Brit. Mus. ; Baron 714, Herb. Kew. This plant is nearly allied to B. nutans, Thouars, for which it has been mistaken. It is easily distinguished by its elongate pseudobulbs, half an inch in length, bearing only one rather long leaf, 2, more rarely 3, inches in length by j in diameter, aud more erect scape 33 inches or less in length. The flowers are small, about twenty in number. B. MINUTUM, Thouars, Orch Iles Afr. t. 110; Lindley, Gen. and Sp. Orch, Pl. p. 52. 464 MR. H. N. RIDLEY ON THE Madagascar, Thouars. I have not seen this plant, and only know it from Thouars's figure; it approaches B. Thompsonii, especially in the shape ot the eolumn-teeth. The whole plant is about 4 inches in height; the leaves, 4 lines in length, are called oval by Thouars, but are rather oblong in the figure. BULBOPHYLLUM TuowrsoNrt, n. sp.; pseudobulbis ovato- globosis monophyllis rugosis nitidis, in rhizomate dissitis ; folio oblanceolata obtuso coriaceo triunciali; seapo nutante 4-unciali, vaginis 5, albis membranacels ; racemo compacto ; braeteis parvis ovatis acutis, ovarii dimidio zquantibus; floribus parvis ad 20; sepalis subsimilibus deltoideis acutis; petalis linearibus acutis, subspathulatis dimidio sepalorum :equantibus; labello oblongo- lanceolato acuto carnoso flavo, apice fasce pilorum minuto ; columna brevi crassa, dentibus erectis longis acutis; capsula ovata oblonga sessili. Madagascar, Thompson! in Herb. Brit. Mus. This plant is allied to B. nutans. Its leaf is 3 inches in length by 1 across ; the scape is 4 inches high. B. enxcrUM, Thouars, Orch. Iles Afrig. t. 96; Lindley, Gen. and Sp. Orch. Pl. p. 53. I have seen no plant answering exactly to Thouars's figure, which seems to be allied to the last species, differing however in the short teeth of the column, shape of lip, and colour. It does not appear to have been found elsewhere. There is a drawing of a Bulbophyllum among Deans Cowan's sketches which much resembles B. clavatum, Thouars, a species not yet recorded from Madagascar, but differing in having but one leaf to the pseudobulb instead of two. It was found at Aukafana. B. HILDEBRANDTII, Reichenbach fil., Otia Hamb. fase. 2, p. 74, Beravi Mountains, Hildebrandt, no. 2988 a, I have not seen. Its affinity is stated to be with B. incurvum, Thouars. B. occrvsuM, n. sp.; rhizomate valido Jignoso ; pseudobulbis aggregatis paucis oblongis in vaginis magnis celatis diphyllis ; folis lanceolatis spathulatis coriaceis striatis semipedalibus, quam scapum vix brevioribus ; scapo crassiuseulo 1-2 vaginato ; NAMEN bracteis magnis triangularibus, circiter 9, flores erme includentibus ; sepalis trian ularibus acuti rali ad basin connatis; petalis spathulatis mucrovatie apice Birti, sepali dimidio «quantibus ; labello unguiculato, basi ad sepalos ORCHIDS OF MADAGASCAR. 465 laterales adnato, lobo medio linguiformi carnoso, costis tribus carnosis hispidis; columna brevi crassa; dentibus clinandrii erectis longis acutis ; anthera depressa, apice acuta. Madagascar, Ankafana, Hildebrandt no. 8985 ! in Herb. Brit. Mus. This plant is closely allied to B. occultum, Thouars, and B. variegatum, Thouars, natives of Mauritius and Bourben, which have not yet been recorded from Madagascar. It differs from those species in the fewer and more distant bracts, and the larger, apparently purple flowers, the broader lip, and acute long teeth of column. The leaves are stiff and hard, 6 inches in length by i inch in the broadest part. The scape is about the same length, bears a nodding raceme of about 9 flowers. CIRRHOPETALUM. C. Tnuovamsm, Lindley, Gen. and Spec. Orch. Pl. p. 58; Bot. Reg. 1838, t.ii.; Bot. Mag. t. 4237 (reproduced also in Moore's 1llustrations of Orchideous Plants, Cirrhopetalum, t. iii.). —Buibophyllum longiflorum, Thouars, Orch. Iles Afr. t. 97, 98; Reichenbach fil. in Walp, Ann. vi. 260; S. Moore in Flora Maurit. et Seych. p. 346.—Epidendrum umbellatum, Forster, Prodr, p. 321.— Cymbidium umbellatum, Sprengel, Syst. Veget. iii. p. 723.—Zygoglossum umbellatum, Reinwardt, Syll. Pl. Soc. Ratisb. ii. 4. Madagascar, Thouars. Mauritius, Thouars, Bojer. Bourbon, Capt. Carmichael! in Herb. Brit. Mus. Java, Reinwardt. Manilla, Cuming. Society Islands, Matthews ; Otaheite, Forster! Collie! (Zn Becchey's Voyage) in Herb. Brit. Mus. The petals and dorsal sepal in the East-Asiatic form are yellow with brown spots; the lateral sepals cinnamon or tawny, the inner side dotted with red-brown ; but in Thouars’s figure the dorsal sepal and petals are coloured green, the column and lip pink, the lateral sepals white. Perhaps there is some mistake in this colouring, for in the synopsis, he says under petals, “tr. long. rouge obscur.” He also represents the fruit as pen- dulous and stouter than in the Polynesian plant, in which it is generally at least erect when ripe. The leaf in the Mauritian plants seems to be narrower than that of the Polynesian variety. The remaining species of the genus are East-Indian, with out- lying species in China and Australia. CALANTHE. This genus extends over the tropics of the Old and New World, 466 MR. H. N. RIDLEY ON THE being most abundant in the Indo-Malayan region. The only recorded species from the Mascarene Islands is found also in Natal. CALANTHE SYLVATICA, Lindley, Gen. and Sp. Orch. Pl. 250; Fol. Orch. no. 15; Reichb. f. in Walp. Ann. vi. 914; S. Moore, Fl. Maur. et Seych. p.363.—C. versicolor, Lindley, Sertum Orchid. i t. 49; Fol. Orch. no. 16.—Centrosis sylvatica, Thouars, Orch. | Iles Afr. tt. 85, 86.— Centrosia Auberti, Rich. Orch. Maur. 45, t. 7. f. 3.—Bletia sylvatica, Bojer, Hort. Maur. 318. Madagascar, Thompson! Imerina, Deans Cowan! Mauritius, Sieber, Fl. Maur. 170! Herb. Brit. Mus. ; Bojer, Gardner, Herb. Kew. It occurs also in Bourbon, S. Africa, and the Seychelles. The whole plant is about 1 foot or 15 inches high. The leaves from 3 inches to 9 long, about 2 across. The spur 1 inch long. All the Mascarene species of this genus belong to the section Gastrorchis, Blume, which are distinguished by the saccate, ven- | tricose, not spurred, base of the labellum, which is also usually | shorter and broader and more open than in the remainder of the | genus. The petals and sepals are also usually broader. The genus is distributed over the tropics of the Old World, and in China and Japan. The Mascarene species are :— PA. villosus, PHAIvs. | Reichb. f., Mauritius ; Ph. tetragonus, Reichb. f., Mauritius and Seychelles; Ph. stuppeus, Blume, Bourbon; Ph. tuberculosus, Blume, Humblotii, Reichb. f. and Ph. pulchellus, Kranzlin, Madagascar. | Pu. TUBERCULOSUS, Blume, Orchid. Archipel. Ind. p. 13; Gard. | Chron. March 1881, p. 341 (xylogr.); Garden, July 1884, p. 46. | —Limodorum tuberculosum, Thouars, Orch. Iles Afr. t. 31.— Bletia tuberculosa, Sprengel, Syst. Veget. iii. 744; Lindley, Gen. and Spec. Orch. Pl. p. 143. Madagascar, endemic, Thouars! in Herb. Brit. Mus. It has been introduced into cultivation, but comparatively seldom flowers. In Thouars’s figure the flowers are given as | entirely pink, whieh is at variance with those of the plants flowered in this country. Pu. Hvunrori, Reichenbach fil. in Gard. Chron. 1880, ii. p- 812. —Planta tripedalis, rhizomate repente; foliis late lanceolatis plieatis pedalibus ; racemo laxo ; bracteis ovario duplo brevioribus, ORCHIDS OF MADAGASCAR. 467 anceolatis cadueis ; floribus speciosis; sepalis angustioribus ovatis lanceolatis roseis; petalis ovatis obtusis ; labello brevi ecal- carato lato trilobo, Jobis lateralibus evectis rotundatis marginibus minute crenulatis, lobo medio retuso, apice recurvo; callo bicorni carnoso, basi hispido; columna elongata gracili curva; polliniis 8. Ankafana, in damp shade on the ground in thelower part of the forest, Deans Cowan; same locality, Hildebrandt no. 3984 in Herb. Brit. Mus. A very handsome plant belonging to the spurless group, and remarkable for the shortness of the lip. The leaves are a foot long by 3 inches broad, gradually tapering to the base; the stems bear racemes of about 13 flowers. The sepals and petals are rose-pink ; the lateral lobes of the Iabellum are yellow with pink spots, the middle lobe pink with a yellow centre; the callus, which consists of two slightly curved diverging horns, is yellow ; in front of it lies a patch of yellow hairs; the column is also yellow and the bracts brown. The plant was introduced into cultivation by Leon Humblot. Pu. pucHELLUS, Kranzlin, Verhandl. Brem. vii. p. 254, I. have not seen. It was obtained at Ambaravambato by Rutenberg. VANDER. EULOPHIA. This genus, which is most abundant in the Cape and tropical Africa, extending also to India and Malaya, is very well repre- sented in Madagascar. All the Madagascar species known to me belong to the section Genuine, in which the flowering scape is leafless, 7. e. bears only sheathing-leaves, and springs from the side of the leaf-bearing pseudobulb. In most the leaves appear not to attain their full development till after flowering. In one species, E. beravensis, Reichb. f., the leaves are borne at the top of a tall stiff stem, naked except for a few sheathing-leaves. Most of the species are terrestrial, but at least one is epiphytic. Out of the nine species all but one, a native also of Mauritius, are endemic. The flowers are usually yellow, plain or variously marked. E. vAGINATA, n. sp.— Terrestris; foliis gramineis linearibus acutis recurvis quam seapus brevioribus ; seapo basi vaginis membrana- ceis pluribus ampliatis albeseentibus tecto, caulinis acuminatis ; 468 MR. H. N. RIDLEY ON THE racemo laxiusculo ; bracteis quam ovaria paullo brevioribus linea- ribus acuminatis; floribus mediocribus ; sepalis lanceolatis; petalis ovatis obtusis brevioribus et latioribus ; labello parvo, petalis sub- eequante, trilobo, lobis lateralibus ovatis obtusis, medio cuneato apice lato obtuso, lamellis tribus, duabus cristatis elevatis, inter- media minus elevata integra, calcare brevi conico-cylindrieo recto ; columna brevi, apice dilatato, stigmate grandiusculo ovato. Ankaratra montes, apud colles gramineas, Jan. 1881, Hilde- brandt no. 3864! Herb. Brit. Mus. This species is allied to E. reticulata. The leaves are 8 inches in height, the scape one foot. EULOPHIA PILEATA, n. sp.—Gracilis, pseudobulbis elongatis ; radicibus crassis longis albis; foliis anguste linearibus longis plieatis; scapo elato gracili, basi vaginis paucis tecto, racemo laxiuseulo ; floribus mediocribus ; bracteis membranaceis brevis- simis linearibus acuminatis; pedicellis gracilibus; petalis sepa- lisque ligulatis obtusis subsimilibus; labello oblongo, lobo medio profunde emarginato, carinis duabus e basi labelli orientibus ; calcare recto obtuso eylindrico-conieo ; columna parva oblonga exalata recta ; stigmate obtuse triangulari; anthera conica bicornuta, cornubus recurvis, apicibus fuscis. Ankafana, Nutongoa, Deans Cowan. A rather tall and slender plant, with a scape of 20 inches in height, long narrow leaves 12 inches long and } across, and a loose raceme of about 20 rather small flowers, orange-yellow in colour, with reddish streaks. The petals and sepals are very similar in shape, but the former have 3 veins distinct in the dry state, not visible in the sepals. The lip is rather broad and oblong, with two ridges highest at the base and gradually taper- ing away towards the median lobe, which is deeply notched. The column is rather small and narrow with straight sides. The form of the anther is rather remarkable; it is conical, the apex bifid, forming two recurved horns, each terminated by a dark-coloured knob, the whole resembling a Phrygian cap. The pollen-gland is short and triangular. The roots are remarkably long, thick, and corky. After flowering, the ripening fruits become deflexed according to Deans Cowan’s drawing. E. scrrera, Lindley, Gen. and Spec. Orch. Pl. p. 182; 8. Moore in Flor. Maur. et Seych. p. 359.—Limodorum scriptum, Thouars, Orch. Iles Afriq. tt. 46, 47; folio plate ii.; Rich. Orch. Maur. 48; Bojer, Hort. Maur. p. 313. ORCHIDS OF MADAGASCAR. 469 Madagascar, Forbes! Bourbon, Capt. Carmichael, Nov. 1813, in Herb. Brit. Mus. Mauritius, fide Bojer. Insula Masca- renes, Sir J. Macgregor! There is also an unlocalized specimen from Thouars’s Herbarium in the British Museum, of what seems to be Limodorum concolor, made by S. Moore a variety of E. scripta, Lindley; and except for the coloration I can really see no difference. The rhizome is, in one of the Bourbon specimens, very stout and thick and covered with a dense mass of the fibrovascular bundles of the old leaves. The longest leaves I have seen are from 2 to 3 inches in length and narrow, but Richard, 7. e., gives them 8 to 9 inches long and 1 across, and S. Moore 62 inches long by 3. The scape attains a length of nearly 2 feet. The lower bracts are 3 an inch long, the pedicel of the flower 1 inch, the ovary 4, The flowers are an inch across, yellowish green with large purple spots. S. Moore describes two forms—one, male, with slender curved column; the other, female, with a stout straight column and rudimentary anther. EULOPHIA MADAGASCARIENSIS, Avanzlin, Verhandl. Bremen, vii. p. 255, I have not seen. It was obtained by Rutenberg on the shores of Lake Itasi, and near Antananarivo. E. RurEkNBEnRGIANA, Aranzlin, Verhandl. Bremen, vi. p. 255. Imerina in paludibus, Hildebrandt no. 3842 in Herb. Brit. Mus.! prope Antananarivo, vulgaris, Rutenberg. l l The plant collected by Hildebrandt seems to be identical with that described by Dr. Kranzlin, an original specimen of which, however, I have not seen. He states that it has a large spatha- ceous basal leaf 5-6 centimetres in length, bifid at the apex. This is wanting in Hildebrandt's specimens, which, however, lack almost all of the pseudobulb. The plant is allied to Æ. ensata, Lindley. The flowers are yellow, the lateral lobes of the lip purple at the apex. The compact conical raceme with comose bracts, connivent perianth, and bearded lip, with two low ridges on the disk between the lateral lobes, distinguish it from the other Mascarene species. E. GALBANA, n. sp.—Epiphytica ; pseudobulbis ovoideis flavis LI nw p] LI . nitidis ; foliis paucis angustis linearibus lanceolatis acutis plicatis, ; iali, vaginis 6 anaceis scapum superantibus; scapo 7-unciali, vaginis 6 membranacei 470 MR. H. N. RIDLEY ON THE pallidis ampliatis, acuminatis instructo ; racemo 10-floro ; bracteis minimis ovato-lanceolatis membranaceis; sepalis petalisque con- niventibus subsimilibus ovato-lanceolatis acutis flavis; petalis parum minoribus et obtusioribus ; labello parvo tenui quam sepala breviore trilobo, lobis lateralibus parvis obtusis, medio oblongo ovato obtuso barbatulo, flavo ochreo-maeulato; calcare nullo. Rara in arboribus Aukafana, Deans Cowan in Herb. Brit. Mus. This species is allied to E. ensata, Lindley, a native of Sierra Leone, but differs in the absence of spur, fewer flowers, and less bearded lip. The bases of the sepals and petals are prolonged backwards and downwards, so as to form a short gibbosity, in which, however, the lip takes no share. The small lip is narrowed at the base, and its lateral lobes, when spread out, are curved outwards at the apices ; the median lobe is slightly bearded in the middle. The flowers are greenish yellow, and seem never to open widely ; the lip is yellow as if stained with ochre. The description is taken from a single specimen and an excel- lent coloured drawing by Deans Cowan. EULOPHIA RAMOSA, n. 8p. ; scapo paniculato, ramis gracilibus, bracteis lanceolatis acutis; floribus pluribus, E. concoloris sub- sequalibus ; sepalis petalisque ligulatis subspathulatis, basi angus- tatis, patentibus ; sepalis longioribus versus apicem dilatatis ; labello angusto recto trilobo, lobis lateralibus oblongis, marginibus exterioribus rectis integris, apicibus obtusis erispis, lobo medio oblongo ovato, apice emarginato, marginibus crispis, in medio cristis tribus brevibus; calcare conico obtuso curvo pendulo brevi; columna crassiuscula brevi ; stigmate oblongo ovato. Ankafana, Deans Cowan. Madagascar, no special locality, Hilsenberg and Bojer. E. BERAVENSIS, Reichenbach fil., Otia Hamburg. fasc. ii. p. 74.— Collected at Beravi, *in collibus arenosis in umbrosis frutice- torum, Juli 1879," Hildebrandt no. 3055! Forms according to Professor Reichenbaeh the type of a new section of the genus, distinguished by the leaves not springing from the pseudobulb as in the rest of the genus, but from the summit of a cylindrical stiff bare stem, which in the specimen in the British Museum herbarium is about a foot in height. The flowers are not larger than those of E. pulchra. E. RETICULATA, n. $p.; rhizomate repente, radicibus validulis albis, foliis paucis teretibus aeutis ; scapo erecto, basi vaginis ORCHIDS OF MADAGASCAR. 471 ampliatis membranaceis tecto, bracteis acutis; floribus paueis in racemo laxo; sepalis ovato-lanceolatis mueronatis ; petalis ovatis obtusis latioribus, omnibus patentibus venis reticulatis; lobis labelli lateralibus parvis ovato-obtusis, medio majore obcuneato obtuso, basi angustata, marginibus crispis, venis conspicuis, cristis 7; calcare brevi cylindrico erassiusculo obtuso; columna brevi crassa; anthera plana patelliformi oblonga ; polliniis sub- globosis. Ad flumen Sassak prov. Imani, Hilsenberg ! in Herb. Brit. Mus. “ Tandrooka-ondri-taki ” incolarum. The scape is about one foot in height, covered at the base with loose membranous sheaths; the lower ones ovate, those on the higher portions of the scape more pointed. The leaves appear to be terete, but in the only specimen which I have seen are in bad condition ; the largest is 5 inches long. The flowers are rather bigger than those of E. scripta and open wide. The narrow linear-lanceolate acuminate bracts are half the length of the ovary. The petals and sepals are curiously veined in a reti- culate manner; the former are a little shorter and considerably broader than the latter, but the difference is hardly sufficiently marked to cause me to refer it to the genus Lissochilus. The central lobe of the lip is large in proportion to the rest, and decorated with 7 crests or raised notched ridges. The pollinia are more circular in outline than usual; each has a semilunar depression at the back. LissocHILUS. Only two species of this African genus have as yet been met with in Madagascar, and none are known from the other islands of the archipelago. I have seen neither of the species which were collected by Rutenberg and described by Dr. Kranzlin under the names of L. madagascariensis and L. Rutenbergianus in the Verhandl. Bremen, vii. pp. 256, 257. CyRTOPODIUM. This genus is represented by a single species, occurring in all the larger of the Mascarene Islands The remaining species of the genus are scattered over the tropies of America, Africa, and Indo-Malaya. C. PLANTAGINEUM, méhi.—Cyrtopera plantaginea, Lindley, Gen. and Spec. Orch. Pl. p. 189; S. Moore in Fl. Maur. & Seych. LINN. JOURN.— BOTANY, VOL. XXT. 2M 472 MR. H. N. BIDLEY ON THE p. 861.—Limodorum plantagineum, Thouars, Orch. lles Afr. tt. 41, 42 ; Bojer, Hort. Maur. p. 313. Madagascar, Forbes! Ankafana, Hildebrandt no. 8956! same loeality and Fiarantosoa, Deans Cowan! in Herb. Brit. Mus. Mauritius, fide Bojer. Bourbon. This conspicuous plant seems to be most nearly allied to C. bicolor (Eulophia bicolor, Blume), from which it differs in the broader white petals and narrower leaves and colouring of the lip. The sepals are spreading, green; the petals broader and shorter, parallel with the column, white; the lip green, except the median lobe, which is white with rose-coloured crests. The spur is pinkish green. Deans Cowan, in a note to the drawing from which the above details are taken, says :—“ This Orchid is common in many places, in some it forms a bed in which several thousand plants are to be found in a few square yards.” It is called * Tenondahy ” by the natives. CYMBIDIUM. The only species of the genus recorded from Madagascar is a very doubtful plant figured by Thouars, Orchid. Iles Afriq. tt. 39, 40, under the name of Limodorum flabellatum, which was after- wards referred to the genus Cymbidium by Lindley in his * Genera and Species of Orchidaceous Plants, p. 167, under the name of C. flabellatum. His description is obviously taken from Thouars's plate. I have seen no plant answering to the figure, and have great doubts as to its belonging to the genus Cymbidium as now understood. In habit it rather resembles a Eulophia. The genus is distributed over the tropical and subtropical regions of the Old World, being most abundant in the Indo-Malayan region. Two species, however, occur in Africa, both in the south. GraMMANGIS. The species on which this genus is based is G. Ellisii, Reichb. f. (Grammatophyllum Ellisii, Lol.), a plant only known from Ma- dagascar, to which Benth. and Hook. f.,Genera Plantarum,’ p.537, would add Cymbidium Huttonii, Lindl., a species foi merly in cul- tivation which has been stated to have come from Java. The genus is allied to Ansellia and Cymbidium, from which latter it iffers in the petals being much smaller than the sepals. EIL DELIS, Reichb. f. Xenia Orchid. ii. 17.—Grammatophyllum in, Lindley, Bot. Mag. t. 5179 ; Williams, Orchid Album, ORCHIDS OF MADAGASCAR. 473 vol. iv. t. 47; Bateman, Second Century of Orchidaceous Plants, t. 176; Flore des Serres, xiv. t. 1488. Was introduced to this country by the Rev. W. Ellis and first flowered on Aug. 23, 1859, from plants obtained some years before. Since this time it has frequently flowered in this country; but I have not seen a wild specimen in either of the herbaria of the British Museum or Kew. PoLysTACHYA. This genus is widely distributed throughout the tropics of both worlds, and is well represented in Madagascar. Most of the species here mentioned are endemic, only one occurring also in the other islands. Besides these there are figures of one or two among Deans Cowan’s drawings which are probably undescribed, but without sufficient material for description. P.cunrrata, Lindley, Bot. Reg. sub t. 851; Gen. and Spee. Orch. Pl. p. 73 ; Reichenbach fil, in Bonplandia, 1856, p. 324; S. Moore in Flora Maur. § Seych. p. 361.—Dendrobium cultri- forme, Thouars, Orch. Iles Afr. t. 87; A. Rich. Orch. Maur. t. 8. f. 3. Madagascar: Imerina, Ikangosoa, Ankafana, Deans Cowan! Ankaratra montes, apud truncos sylve primeve, Jan. 1881, Mil- debrandt no. 2868 ! in Herb. Brit. Mus.; Parker in Herb. Kew! Mauritius, Capt. Carmichael! Herb. Brit. Mus., Bouton! Ayres! in Herb. Kew. Bourbon, Balfour! in Herb. Kew. The broad solitary leaf, 5 inches in length by 1 in diameter, and larger white flowers in a glabrous panicle, distinguish this plant from the rest of the Madagascar species. The scape is from 6-8 inches long, and has a solitary long sheathing-leaf 1-12 inch in length at the base. The variety nana, S. Moore, a smaller plant with usually a racemose inflorescence, occurs also in Madagascar. P. ANCEPS, n. sp.; caulibus approximatis, basi turgidis vix pseudobulbosis, basibus foliorum vetustorum tectis ; foliis sepius duobus, lanceolatis obtusis bilobis; scapo erecto glabro, vagina longissima ancipiti sepius ad basin panicule tecto; bracteis parvis triangularibus ; floribus parvis, quam P. zeylanice ma- joribus; pedicellis longiusculis ; sepalo postico lanceolato acuto, sepalis lateralibus lanceolato-triangularibus acutis ; petalis linea- ribus lanceolatis acutis, sepalo postico subaquantibus; labello 2x2 474 MR. H. N. RIDLEY ON THE trilobo, lobis lateralibus brevibus obtusis, medio oblongo obtuso ; unguiculo labelli longiusculo, in basi callo oblongo papilloso. Ankafana, Hildebrandt! no. 4222 ; Imerina, Ankafana, Deans Cowan! in Herb. Brit. Mus.; Baron 192! Herb. Kew. The flowers of this plant much resemble those of P. fusiformis, Lindley, Dendrobium fusiforme, Thouars, two flowers of which, from Thouars’s collection, I have seen in the Herbarium of the British Museum, but the habit of the plant is quite different. There are two sheathing-leaves on the scape, of which the lower one almost invariably reaches to the base of the panicle and is about 3} to 4 inches long, striate, and winged on both sides for its whole length; the other is very much shorter, being partially overlapped by the lower one. The whole plant dries black. There is a figure of what appears to be this plant in Deans Cowan’s drawings, the sepals and petals and base of the lip of which are cinnamon, the apex of the lip rose-colour. PoLYSTACHYA ROSEA, n. sp.; foliis 2-3 angustis oblongis lanceolatis, apice bilobis, patentibus vel reflexis; scapo erecto vaginis ad 3 dissitis breviusculis munito simplici, rarius ramoso ; floribus paucis roseis; bracteis minutis ovatis; sepalo postico angusto laneeolato acuto; petalis linearibus brevioribus ; sepalis lateralibus lanceolatis triangularibus curvis aeutis; labello trilobo, lobis lateralibus brevibus obtusis, medio brevi oblongo integro; capsula oblonga j-unciali. Imerina, Deans Cowan! in Herb. Brit. Mus. This plant resembles P. luteola in its general habit, but is distinguished by its simpler inflorescence, rarely branched, more numerous and shorter leaf-sheaths on the stem, broader dorsal sepal, and colour. The whole plant is half a foot high ; the leaves 4 inches long, half an inch in diameter. The bracts are 1 line long, the pedicel 9 lines. The whole of the flower is rose-pink, with a darker spot in the centre of the lip. P. VIRESCENS, n. sp.—Semipedalis, caule basi tumido; foliis ovatis obtusis 2, inferiore minore ; racemo laxo basi paullo ramoso, floribus pluribus parvis virescentibus, bracteis ovatis brevibus; sepalo postieo lanceolato, lateralibus oblique triangulari-ovatis ; petalis angustis linearibus rectis ; labello trilobo, lobis lateralibus brevibus ovatis obtusis, lobo medio oblongo obtuso. Ankafana, on trees, Deans Cowan, ORCHIDS OF MADAGASCAR. 475 The whole plant is 6 inches high, the longest leaves 2 inches in length by ?in breadth. I only know this from two coloured drawings with details and notes. It must be closely allied to P. luteola, Hook., a S. American plant. Other Madagascar species are :— PorysTACHYA ROSELLATA, Ridley in Journ. Linn. Soc. xx. 330.— Ankafana, Deans Cowan in Herb. Brit. Mus. ; and P. MULTIFLORA, fiidley, l. c., from the same locality, collected both by Deans Cowan and Hildebrandt (March 1881, no. 4220, in Herb. Brit. Mus.). P. Jussrevrana, Reichenbach fil. in Walpers's Annales, vi. p. 640, I cannot recognize, as the description is too imperfect for such a difficult genus. ACAMPE. This genus consists of about a dozen species, of which the bulk are natives of tropical Asia and China, one or two occurring in Southern Africa. The only Madagascar species yet known is A, Renschiana, Reichenbach fil, Otia Hamburg, fasc. 2, p. 77, which was discovered by Hildebrandt in Nossibé; no. 3392 of his col- lection. ANGRECUM. A large genus of about 60 known species, of which all but one are natives of Africa and its adjoining islands, the remaining one being found in China and Japan. The species differ very greatly in the form and size of the flower, and in the cau dicle and gland of the pollinium, and have been divided into genera according to the number of the latter, with no very satisfae- tory result, since the pollinia are frequently missing in her- barium speeimens, and moreover sometimes plants apparently closely allied differ entirely in the structure of the pollinia. § Macrura. A. SESQUIPEDALE, Thouars, Orch. Iles Afriq. it. 66, 67 ; folio, pl. i, i.; Hooker, in Bot. Mag. t. 5113; Lindley, Gard. Chron. 1857, p. 252 ; Flore des Serres, t. 1413 ; Illustr. Horticole, vol. vii. p. 77, t. 475; Gartenflora, t. 744.—Aeranthus sesquipedalis, Lindley, Gen. and Sp. Orch. PI. p. 214 (nomen nudum). uu Madagascar: Ankafana, apud saxa, Hildebrandt no. 3983 ! same locality, very common, Deans Cowan! in Herb. Brit. Mus. ; 476 MR. H. N. RIDLEY ON THE between Tamatave and Antananarivo, Meller no. 1617!; Central Madagascar, Baron! in Herb. Kew. It is also reported from Zanzibar, but I have not seen specimens. This plant was first flowered by the Rev. W. Ellis, at Hoddes- don, in the year 1857 ; specimens from whom, with a drawing of the plant made by Mrs. Ellis, are in the Lindleyan Herbarium. The whole plant is not more than two feet high, being smaller than A. eburneum, simple or with one or two branches. The leaves are numerous, broad, oblong, thick and fleshy, dark green, and keeled at the base, very unequally lobed, and more imbricate than those of eburneum. Several stout peduncles are produced from the axils, each bearing from 2 to 4 of the large white or greenish-white waxy flowers. The sepals and petals are lanceo- late acuminate, almost triangular, the former spreading, the latter reflexed. The lip is similar in size and shape, 3 inches in length. The spur, in spite of the specific name, appears never to attain a greater length than one foot. The bract at the base of the ovary is short and blunt, fitting closely to the pedicel, but its edges do not quite meet. The column is white, very short and broad, the clinandrum small, the rostellar lobes large and foli- aceous, horizontal, and stiff, the right-hand one overlapping the other; they are blunt and raised in the middle, white edged with yellow ; beneath them is a deep chamber, at the back of which is the stigma. The pollinia are bluntly wedge-shaped, and the caudicles are shorter than the narrow flat viscid disks. The ribs of the ovary are produced into somewhat sinuous and appressed wings. ANGRACUM GLADIIFOLIUM, Thouars, Orch. Iles Afriq. t. 535 Lindley, Gen. and Spec. Orch. Pl. y. 246; Bot. Reg. 1840, t. 68; Rich. Orch. Maur. p. 69.— Orchis mauritiana, Lamarck, Encycl. vi. p. C01.—4A cranthus gladiifolius, Reichenbach fil. in Walp. Ann. vi. p. 900; S. Moore in Flor. Maur. & Seych. p. 350. Madagascar, Thouars; Imerina! Tanala! Deans Cowan in Herb. Brit. Mus. Mauritius, Capt. Carmichael! Sieber Fl. Maur. 3. no. 171! in Herb. Brit. Mus.; Bouton! Ayres! Herb. Kew. Bourbon, Boivin! Balfour! Herb. Kew. Ins. Masearenes, specimens without specific locality, collected by Thouars and Sir J. Macgregor, are in Herb. Brit. Mus. This plant is easily recognized by its flattened, slightly flexuous stem, covered with wrinkled shortly-winged leaf-sheaths. It is ORCHIDS OF MADAGASCAR. 477 about à foot or rather more high, with narrow, acute, spreading, light-green leaves, 21 iuches in length by 4 an inch across. The flowers are solitary, on short peduncles, white or yellowish white. The ovary and pedicel are white, 14 inch in length, twice as long as the peduncle. The petals and sepals are about à inch long, the petals narrower than the sepals, and spreading, the lateral sepals deflexed. The lip is broader, with a slender spur 3 inches long, white, becoming yellowish towards the apex. The column is rather short. The caudicles somewhat long. The affinity of the plant is with A. recurvum. ANGRÆCUM RECURVUM, Thouars, Orch. Iles Afr. t. 56; Rich. Orch. Maur. p. 70.—A. fragrans, var., Lindley, Gen. and Spee. Orch. Pl. p. 246.—Aeranthus rectus, var.? recurvus, S. Moore, in Flor. Maur. and. Seychelles, p. 851. A. caule crasso erecto ; foliis distichis imbricatis, apice bilobis, 5-uncialibus; pedunculis pluribus in axillis foliorum erectis vel patentibus unifloris; vaginis ampliatis membranaceis obtusis; floribus majoribus, albis inexpansis; sepalis lanceolatis acutis; petalis subsimilibus porrectis, basibus omnium margine crispis et ad columnam et basin labelli adnatis; labello spathulato acuto, basi decurvo, lobis lateralibus erectis brevibus approximatis, medio spathulato acuto porrecto; calcare longo filiformi tereti, quadrante basali recto pendulo, reliquo abrupte curvo; columna brevi; lobis rostelli latis foliaceis obtusis ercetis; anthera de- presso-conica ; polliniis bilobis glandulis oblongis ovatis; caudi- culis minimis. Madagascar, Ankafana, Deans Cowan! in Herb. Brit. Mus. Mauritius, Bojer. Also Bourbon. This plant has been reduced to a variety of A. rectum, Thouars, by S. Moore, l. e., which it probably is. I have seen no specimen resembling Thouars's figure of the latter plant; but there is among Deans Cowan’s drawings, besides a very good figure of 4. recurvum, Thouars, a coloured drawing somewhat resembling Thouars’s figure of A. rectum. It differs from the other species in the much closer leaves, blunter sepals and petals, the lateral sepals being recurved and not thrown forward, the more angular lip, and more irregularly and slightly bent green spur. Of it Deans Cowan says :—“ Very powerful perfume at night, erect on branches, Ankafana.” A. recurvum, Thouars, has an erect stem, covered with rather 478 MR. H. N. RIDLEY ON THE close strap-shaped leaves, 5 inches in length by 2 an inch across, dark green, with the apices recurved. From the axils spring about six slender peduncles on each plant, about 2 inches long, bearing a few loose brown sheaths. The ovary and pedicel are about 21 inches in length, erect or a little spreading. The posticous sepal is erect. The laterals, together with the petals, are thrown forwards at right angles to the erect column. At the base they are adnate to the column, and the lateral lobes of the lip and their edges at this spot are crisped and wrinkled. The base of the lip is thickened and curved downwards, and the side lobes are erect, so as to form a guiding passage to the entrance of the spur. The middle lobe of the lip is spathulate in shape and acute; it is bent up, making an angle with the base so as to be parallel to the petals. The whole lip is somewhat shorter than the sepals, which are three quarters of an inch in length. The spur is 4 inches long, the first inch of which is almost parallel with the ovary ; it is then bent up abruptly, almost at right angles, taking a sigmoid curve. The column is short, but proportionally longer than in A. sesquipedale. The lobes of the rostellum are rather large, and overhang the entrance to the spur. The pollinia are oblong, grooved on the back so as to be almost bilobed, the hind lobe being the smaller one. The caudicles are very short, the glands separate, ovate, acute, and flat. ANGRECUM SPATHULATUM,n. sp.; caule crassiusculo pedali, vaginis vetustis membranaceis striatis transversim rugosis tecto ; foliis ligulatis obtusis crassiusculis uncialibus, ineequaliter bilobis ; floribus mediocribus singulis in pedunculis brevibus vaginis 3-4 tectis; sepalo postico erecto lanceolato, lateralibus obliquis lan- ceolatis acutis; petalis spathulato-lanceolatis ; labello ovato spathulato obtuso; calcare filiformi pendulo, ovarium paullo superante ; columna brevi, rostelli lobis aliformibus erectis planis ; caudiculis brevibus ; glandulis parvis distinetis; ovario recto. Ankafana, no. 3988, Hildebrandt! in Herb. Brit. Mus. The affinity of this plant is with A. recurvum, Thouars. It has a stout stiff stem, about 1 foot in height, covered with the bases of the old leaves, and terminated by a few short, blunt, unequally bilobed leaves, 1 inch long by 3 in breadth. The flowers occur singly, each supported on a slender peduncle, covered. with 3 or 4 sheathing-leaves. They are rather bigger than those of ORCHIDS OF MADAGASCAR. 479 A. gladiifolium. The dorsal sepal is erect; the others, with the petal, are thrown forwards as in A. recurvum. The petals are narrowed at the base, then dilated, and acute at the apex. The lateral sepals are obliquely lanceolate. The spur is an inch long, and hangs straight down, a little longer than the ovary. The rostellar lobes are foliaceous and stiff, as in A. recurvum, but instead of sloping upwards so as to form an arch as in that plant, they are quite vertical and the edges do not meet. The pollinia are grooved on the back; the glands like those of A. recurvum. ANGRECUM MAXILLARIOIDES, n. sp.; caule brevi; foliis ligu- latis obtusis inequaliter bilobis 9-uncialibus vel ultra; flore singulo magno albo in pedunculo semipedali ex axilla folii in- ferioris oriente ; sepalis earnosis lanceolatis acutis ; petalis simili- bus angustioribus, deorso carinatis; labello late lanceolato acuto carnoso, sepalis subzquali ; calcare filiformi pendulo, basi abrupte dilatato biunciali; columna brevi crassa; lobis rostelli cornutis erectis ; ovario longo, 6-costato. Ankafana in arboribus, Deans Cowan! in Herb. Brit. Mus.; Central Madagascar, G. W. Parker! in Herb. Kew. This plant has much the habit of Angrecum triquetrum, Thouars, but is much larger in all its parts. The leaves are strap-shaped and stiff, 9 inches in length and 1 across in Deans Cowan's specimen ; but he says, in a note to a drawing of it:— * Of this there seems to be another variety, with leaves about twice as large as this one.” From the axil of one of the lower ones rises a peduncle bearing a solitary large white waxy flower, recalling very much the habit of one of the large Maxillarias. The base of the peduncle is covered with a few short sheathing- leaves. The sepals are lanceolate and acute, 7-veined, 12 inch long by 5 lines across at the base; the petals are narrower, 6-veined, with a median ridge on the dorsal surface. The lip is in the normal position raised above the lower sepals and petals; it is broadly lanceolate-acute. The sides at the base are erect, leaving a narrow channel down the middle, and at a point about one third of its length from the base the edges are so closely approxi- mated as almost to meet, after which they diverge again. There is a low ridge running in the median line for a short way from the base of the lip. The spur is cylindrical and pendent, 2 inches where it joins the lip it is somewhat abruptly dilated long ; into a small bulb. The column is short and thick, the lobes of 480 MR. H. N. RIDLEY ON THE the rostellum erect and horn-shaped. The ovary is long, almost cylindrical, and slightly curved, with six low ribs running its whole course. The anther and pollinia I have not seen. The whole plant dries black. The old leaf-sheaths of the lower leaves break up into fibres which cover the stem. $ Euangreca. ANGRECUM EBURNEUM, Thouars, Orch. Iles Afrig. t. 65; Rich. Orch. Maur.t. 71; Lindley, Gen. and Spec. Orch. Pl. p. 245; Reichenbach fil.in Walp. Ann. v. p. 904; S. Moore, in Flor. Maur. and Seych. p. 356; Bot. Mag. 5.5170 ; Lindley, Paxt. Fl. Gard. vol. i. pp. 25, 26 (xylogr.).—Limodorum eburneum, Bory, Voy. i. p. 859, t.19; Willd. Sp. Pl. iv. p. 125.—A. virens, Lindley, Bot. Reg. 1847, sub t. 19; Past. Fl. Gard. vol. i. p. 25, figs. 9, 10. Madagascar, Forbes (fide Lindley); also in Ambongo Island, Pervillé no. 148 in Herb. Lindley. Seychelles, Perceval Wright in Herb. Kew. Bourbon, Thouars. Comoro Islands, Johanna, Kirk. This plant and the two following are very closely allied, and indeed A. eburneum and superbum were reduced by Lindley, in Paxton’s Mag., to one species. The flowers, however, of ebur- neum are somewhat smaller, and the cordate lip is so constant that it seems better at present to keep them distinct. In 4. superbum the lip is pure white, and the carina shorter and broader, the petals and sepals duller green, and the whole flower more fleshy. The variety virens (A. virens, Lindley) only differs from typical eburneum by being somewhat greenish in the middle of the lip. A. SUPERBUM, Thouars, Orch. Iles Afriq. tt. 62, 63, 64 ; Lindley, Gen. and Sp. Orch. Pl. p. 245.—A. eburneum, Lindley, Paxt. Mag. 1849, xvi. p. 90; Bot. Reg. xviii. pl. 1522; Hook. Bot. Mag. t. 4761 (non Thouars). Madagascar only, Thouars! in Herb. Brit. Mus.; inter Tamatave and Antananarivo, Meller in Herb. Kew. This species is distinguished from 4. eburneum by its broader and more rhombiform lip, with a shorter and somewhat broader cusp, thicker and blunter spur, and shorter, broader, and blunter sepals and petals. A. BRONGNIARTIANUM, Reichenbach Jil. Pescatorea, i. p. 165 Walp. Ann. vi. p. 904. Nossibé, Ankiabé, May 1879, Hildebrandt no. 2990. Also ORCHIDS OF MADAGASCAR. 481 Comoro Isles, Johanna, Hildebrandt no. 1704 !; and Mahé, Seychelles, Boivin! in Herb. Brit. Mus. This is the largest plant of the section; the leaves are more than 2 feet long and 3 inches across, the raceme is more than 2 feet in height, and bears a number of flowers as large as those of A. superbum, from which plant it is distinguished by its wider leaves and smaller bracts, lanceolate and acuminate and acute petals and sepals, and much longer and more slender, slightly curved spur. The petals and sepals are green, 2 inches in length by 2 lines across. The lip is white, and has a rhomb- shaped ridge running down the centre as in superbum, but rather more prolonged, ending in a raised line, The capsule is elongate, subcylindrical, 14 inch in length and 4 inch in diameter in the only specimen which I have seen, which is, however, not quite ripe. The bracts are ovate, short, rather blunt, ? inch in height. The column is green. The following six species, together with A. caudatum, form a well-marked group, characterized by the long terete, not folia- ceous rostellum, the larger semiterete column, and the usually ovate cuneate leaves. ANGRECUM CITRATUM, Thouars, Orch. Iles Afrig. t. 61; Hooker, in Bot. Mag. t. 5624. This plant appears to be common in Madagascar, whence it was introduced into cultivation by Messrs. Veitch. It has been collected at Ankafana “very common” (Deans Cowan, and Hildebrandt no. 3987) ; Imarina (Deans Cowan, and Baron no. 201); between Tamatave and Antananarivo (Meller in Herb. Kew); and there are specimens collected by Du Petit Thouars and Thompson in Herb. Brit. Mus., and by Langley Kitching, and Parker in Herb. Kew, without specific localities. It does not appear to have been met with in any of the other islands. The stem is short, and the leaves vary from lanceolate to obcuneate, entire at the tip, 33 inches long by 13 across in dried specimens. There is usually, in the wild plant at least, only one raceme to the plant ; but ina specimen collected by Thompson in the British-Museum Herbarium there are no less than 5, one of which attains a length of 11 inches; they bear about 18 flowers, sometimes much less, and sometimes as many as 27, rather crowded together. The flowers are about 1 inch across, white or greenish white; I have never seen them yellow as figured by 482 MR. H. N. RIDLEY ON THE Thouars. The dorsal sepal is much smaller than the laterals; it is erect, ovate, and blunt, 1 of an inch long ; the petals are wider than the sepals, j an inch long, 5-veined. The lip is almost obeuneate, ian inch long and the same across in the broadest part, 7-veined. The spur is 1 inch long, cylindrical, and hang- ing vertically downwards for the greater part of its length, and then abruptly bent at an obtuse angle, at which point it is much dilated. The column, as in its allies, is rather longer than usual (1 line), glabrous, semiterete, and slightly curved ; the stigma is oval; the anther is ovate acute. The pollinia are fixed to a single gland with an obeuneate flat caudicle. The rostellum resembles that of A. modestum, to which this plant is allied. The capsule is about 1 inch long and 2 lines thick, obscurely 3-angled, with low ribs. ANGRÆCUM MODESTUM, Hooker fil. in Bot. Mag. t. 6673. Madagascar, Ankafana, Deans Cowan! in Herb. Brit. Mus. Introduced into cultivation by the Dowager Lady Ashburton, a specimen from whom is in the Kew Herbarium. A stout plant with cuneate-lanceolate leaves, 5 inches in length ‘and lin diameter. The raceme is about a foot in length, the rhachis and bracts brown, and bears from 6 to 9 white flowers about as big as those of A. gladiifolium. The sepals are lanceo- late, § an inch long and 2} lines across; the petals a little broader and shorter. The lip is similar in shape to the petals, but broader and more acute and fleshy, § an inch long and 4 across ; it clasps the column at the base, and below the latter is a slight depression leading to the spur, over which hangs the rostellum and caudicle of the pollinia. The spur is 4 inches in length, filiform, broadest at the base, gradually tapering to a point. The column is straight, yellow, rather thick and sparingly hispid ; it is about 2 lines long. The anther is conical, with rather a long apiculus ; inside it has two small flaps or pockets which make it nearly bilocular. The pollinia are rather large, 4 a line in length, oblong, with a small groove in the back. They are attached to a long (13 mm.) caudicle, which lies in a groove of the rostellum, and bears a broad oblong gland (1 mm. long), which shows some signs of its double origin in having a notch at the point. The rostellum is subterete and horn-shaped, eurved up at the end, and overbanging the stigma and mouth of the spur. There is a groove along the upper surface, in which lies the caudicle, and the ORCHIDS OF MADAGASCAR. 483 end is clasped by the gland, which projects a little beyond it. The stigma is semioval and rather deeply sunk. The affinity of this plant is with A. apiculatum, Lindl., a native of Sierra Leone. Angrecum fuscatum, Reichenbach fil. in Gard. Chron. Oct. 14, 1882, p. 488; A. articulatum, Reichb. f. in Gard. Chron. 1872, p. 73; A. hyaloides, Reichb. f. in Gard. Chron. 1880, i. p. 264; and A. fastuosum, Jeichb. f. l. c. 1884, ii. pp. 748, 844, belong to this section. All were introduced into cultivation from Madagascar, and appear to be extinct again in England. I have seen no speci- men, wild or cultivated, in the herbaria of the British Museum or Kew. A. Eris, Reichenbach fil. Flora, 1872, p. 278; Gard. Chron. 1875, p. 277 (xylogr.) ; Floral Magazine n. s. t. 191. Madagascar, Ankafana, Deans Cowan in Herb. Brit. Mus.! This fine plant was introduced into cultivation by Rev. W. Ellis, whose name it bears. It is characterized by its completely reflexed petals and sepals, which give it the appearance of a hovering bird. It has large dark-green leaves, 10 inches in length by 2 across, broadly ligulate and unequally bilobed. The flower-spikes in cultivation attain a length of 2 feet; they are somewhat nodding, and bear from 18 to 24 flowers, very fragrant, from pure white to yellowish white; the long gracefully curved spur, 6 inches or less in length, and the ovary being of an ochreous-cinnamon colour. A. eryptodon, Reichenbach fil. Gard. Chron. 1883, ii. p. 307, was introduced into cultivation by Low. Its affinity is with A. CAailluanum, Lindley. A. FILICORNU, Thouars, Orch. Iles Afriq. t. 32; Lindley, Gen. and Sp. Orch. Pl. p. 246.—Aeranthus ? Thouarsii, S. Moore, in Flor. Maur. § Seych. p. 351. Madagascar, Thouars ; Antananarivo, Rutenberg, fide Kranzlin ? Tanala, Imarina, Deans Cowan. I only know this species from Thouars’s figure, unless the plant collected by Deans Cowan above mentioned belongs to it. This closely resembles 4. filicornw in most respects, but the lip, instead of being pandurate and rather acute, is oblong-elliptical and blunt, and the spur is only 2 inches long. The stem is flexuous, about 6 inches high; the leaves linear, spreading, blunt, 484 MR. H. N. RIDLEY ON THE and bilobed at the apex, with rather long, transversely wrinkled sheaths. The flower is open, the dorsal sepal erect, the laterals spreading, elliptical, lanceolate, and curved. The lip is 2 an inch long by 4. The spur is very slender and straight; the ovary 1 inch long; the pedicel very slender, with a short ovate bract. ANGRECUM TERETIFOLIUM, n. sp.; caule erecto flexuoso gracili rigido; foliis teretibus, demum recurvis, rigidis acutis dissitis; pedunculo gracili tereti erecto unifloro; bractea ovata; floribus mediocribus expansis ; sepalis anguste linearibus acumi- natis semiuncialibus; petalis angustissimis linearibus acuminatis subeequalibus ; labello lanceolato triangulari acuminato acuto; caleare longissimo filiformi; capsula oblonga semiunciali costis prominulis. Ankafana, Deans Cowan! in Herb. Brit. Mus. This plant has erect flexuous stems from 6 to 14 inches in height, with internodes 4 an inch in length. The leaves are terete, pungent, at first erect, afterwards recurved, 21 inches in length, subdistichous, with rugose sheaths. The flower of moderate size, with very narrow petals and sepals, } an inch in length by 1 line in the broadest part. The lip is nearly an inch long, 4 across in the broadest part, ovate, lanceolate, triangular, acute. The spur is very long and slender; the only one I have seen was 4 inches long and imperfect. The affinity is with A. Jilicornu, Thouars. A. IMPLICATUM, Thouars, Orch. Iles Afr. t.58; Lindley, Gen. § Sp. Orch. Pl. p. 246; Aerobion implicatum, Sprengel ;—and A. RurENBERGIANUM, Kranzlin, Verh. Bremen, vii. p. 257, are species not knowntome. The latter was collected by Rutenberg in the Ankaratra Mountains. It is stated to resemble A. cucul- latum, Thouars, except in the lip, which is described as “rhom- beum." A. Cowanit, n. sp.; caule semipedali curvo crassiusculo; foliis paucis dissitis linearibus obtusis oblique bilobis, striatis patulis ; pedunculo gracili 11-uneiali erecto, ex axilla folii inferioris, unifloro; flore mediocri inexpanso tenui; bractea ob- tusa; sepalis petalisque subsimilibus ovatis lanceolatis obtusis; labello ovato obtuso, sepalis subzquali; caleare filiformi obtuso pendulo semiunciali ; lobis rostelli foliaceis verticalibus; anthera brevissime apiculata; polliniis suleatis; caudieula brevissima, glandula singula; ovario parum curvo. ORCHIDS OF MADAGASCAR. 485 Imerina, Deans Cowan! in Herb. Brit. Mus. This plant has a stem 6 inches long, covered with the old leaf- sheaths, and bearing 4 or 5 narrow, linear, blunt leaves at the top. The leaves are 3-33 inches long, } inch across. The petals, sepals, and lip are all very similar in shape, the lip being the broadest ; they are ovate-lanceolate and hardly expanded. ANGRECUM CLAVIGERUM,n. sp.; caule brevi radicante ; foliis lanceolatis cbtusis, inequaliter bilobis, carnosis uncialibus; floribus singulis albis mediocribus, in pedunculis gracilibus suberectis uncialibus e medio internodii orientibus, vaginis tribus ovatis parvis; sepalis lanceolatis acuminatis acutis ; petalis subsimilibus brevioribus latioribus; labello petalis subequali, ovato-lanceo- lato plano, carina media; calcare brevi, ovario subequali, pen- dulo clavato, apice curvo; columna brevi crassa; lobis rostelli foliaceis erectis obtusis; anthera depressa operculari; polliniis parvis, caudicula brevi, glandula angusta lineari quam caudicula longiore ; ovario brevi curvo, z-unciali. Imerina, Ankafana, Deans Cowan!; Ankafana, Hildebrandt, Mar. 1, 1881, no. 3978! in Herb. Brit. Mus. A small plant with much of the habit of A. pectinatum, creep- ing on the bark of trees and emitting copious and long roots. The leaves are thick and fleshy, oblong lanceolate, blunt and unequally bilobed, 1 inch long by i in diameter. The flowers are solitary on slender peduncles, springing from the lower part of an internode and not from the axil of a leaf. They are about the size of those of A. cucullatum, Thouars, white with a green centre. The petals and sepals are rather thin in texture and hardly spreading ; the petals are 6, the sepals 8 lines in length. The lip is similar in shape and about as long as the petals, but a little broader; the median line is pinched up so as to form a central ridge. The spur is about 4 of an inch long, hanging vertieally down ; it is club-shaped, with the end turned up. The stigma is rather broad; the anther flat and rather small. The pollinia are supported on a straight short pedicel and a longer narrow gland. A. ROSTRATUM, n. sp.; rhizomate longo lignoso gracili; radieibus multis gracilibus; foliis copiosis alternis lanceolatis obtusis carnosis, vaginis (siccis) transversim rugosis; floribus singulis in axillis foliorum superiorum, viridibus mediocribus, pedicello l-uneiali gracillimo; sepalis late linearibus obtusis carnosis; petalis linearibus lanceolatis, basi dilatatis, obtusis 486 MR. H. N. RIDLEY ON THE [ carnosis, versus apicem teretiusculis; labello recto, lobis laterali- ' bus parvis ovatis erectis, ad columnam appressis, lobo medio | rostriformi carnoso tereti acuminato ; calcare longo horizontali, basi decurvo dilatato, apice clavato; columna brevissima lata ; anthera obtuse subovata obscure biloculata; ovario pedicelle | subzequali. Ankafana, in sylvis, Martio 1881, Hildebrandt no. 3976!; ibidem, Deans Cowan (icon.). This plant 1s allied to A. gladiifolium, Thou. The rhizome is about a foot long, rather slender, with numerous thick, fleshy, blunt leaves from ? to nearly 1 inch in length. The pale green flowers, rather smaller than those of A. gladiifolium, Thou., are solitary in the axils of the upper leaves on very slender pedicels 4 an inch in length. The petals and lateral sepals, which are narrow and fleshy, are apparently not patent, but carried forward horizontally. The lip is 3 an inch in length; it has two smail erect lateral lobes, the median one being long and beak-like, quite cylindrical for the greater part of its length, and tapering to a point. Between the lateral lobes is a depression, which gradually fades away in the median lobe. The spur is long and horizontal, a little dilated where it joins the lip; and the end is clavate and slightly up- curved. The ovary is about 3 an inch in length. Aeranthus Curnowianus, Reichenb. fil. Gard. Chron. 1883, ii. p. 906, a plant recently introduced into cultivation from Mada- gascar, would belong to the genus Angrecum as defined by the authors of the * Genera Plantarum. I have not seen it. ANGRÆCUM CRASSUM, Thouars, Orch. Iles Afriq. tt. 70, 71 ; Lindley, Gen. and Sp. Orch. Pl. p. 245.—Aerobion crassum, Sprengel, Syst. Veget. ii. p. 716. A. caule crasso erecto tereti 15-unciali ; foliis ligulatis loratis obtusis, inzqualiter bilobis patentibus subpedalibus; racemo horizontali, bracteis brevissimis crassis ; floribus erectis secundis inapertis magnis, albescente-viridibus; petalis sepalisque lanceo- latis, his brevioribus ; labello cucullato ovato, basi saecato; calcare erasso brevi cylindrico obtuso pendulo, quam ovarium triente breviore ; columna curva brevi ; ovario crasso ovali-oblongo, ferme sessili; costis prominulis. Madagascar, Thouars ; prope Beforon inter Tamatave et Anta- nanarivo, Meller! in Herb. Kew. “Flowers alternately white and yellow.” I ORCHIDS OF MADAGASCAR. 487 $ Pectinaria. AxamzcuM PECTINATUM, Thouars, Orch. Afriq. t. 51; Lindley, Gen. and Sp. Orch. Pl. p. 247 ; Rich. Orch. Maur. p. 68, t. 10. f. 5 ; Hook. Journ. Bot. i. 47, t. 116.—Aeranthus pectinatus, Reichb. J. in Walp. Ann. vi. p. 900; S. Moore, in Flor. Maur. et Seych. p. 350. A. rhizomate sepe longe repente; caulibus erectis vel curvis, rigidis sepius ramosis; foliis distichis coriaceis linearibus loratis obtusis, parum recurvis, J-uncialibus ; floribus singulis in axillis foliorum superiorum, inexpansis parvis albis, ferme sessilibus ; bracteis paucis ovatis obtusis ; sepalis et petalis oblongis obtusis ; labello lanceolato acuto ; calcare quam ovarium breviore rectius- culo obtuso subclavato; columna brevi; lobis rostelli brevibus subacutis ; anthera depressa; glandulis brevibus oblongis ; caudi- culis nullis; capsula ovali fusiformi 3-unciali, costis alatis. Madagascar, Thompson! in Herb. Brit. Mus. Mauritius, Chapelier! Capt. Carmichael! Dr.Wallich! in Herb. Brit. Mus.; Pouce, Ayres! Bouton! Telfair! in Herb. Kew. Also Bourbon. The main stem of this plant creeps on the tree-trunk in old specimens for as much as 4 inches, sending up stiff lateral branches, erect or more rarely nodding, closely covered with the transversely rugose leaf-sheaths, and from 5 to 6 inches in height. The leaves are about 4 an inch long, rather thick, linear, oblong, obtuse. The flowers are very shortly pedicellate, and occur singly in the axils of the upper leaves; they are white, and do not appear ever to expand. The petals and sepals are short, lanceolate, oblong, the lip acute. The spur is 3 of an inch long, straight, obtuse, slightly dilate at the apex. The pollinia sessile on two short oblong glands. The capsule is subeylindrical, 3 an inch long. MYSTACIDIUM, Lindley. The typical species of this genus are short-stemmed plants with secund racemes of rather or very small flowers, with usually long spurs, to which have been added a number of plants of very dif- ferent habit, including the genus Gussonia, Richard (JMicroca'lia, Lindley), which seems to be sufficiently distinct. Pectinaria, a section proposed by the authors of the * Genera Plantarum ' for Angraecum distichum, Lindley, and A. pectinatum, Thouars, seems better referred to Angraecum ; while to the section Gomphocentrum, distinguished by its racemes of many flowers and club-shaped or ` ponp D LINN. JOURN.— BOTANY, VOL. XXI. -~N '488 MR. H. N. RIDLEY ON THE saccate spur, must be added M. ochraceum, n. sp., with solitary flowers and a filiform spur. § Gomphocentrum. MYsTACIDIUM CAULESCENS, mihi.—Angrecum caulescens, Thouars, Orch. Iles Afr. t. 55; Rich. Orch. Maur. p. 78, t. 10. fig. 3; Lindley, Gen. and Sp. Orch. Pl. p. 247; Hook. Lond. Journ. Bot. 1834, p. 48, t. 117 ; S. Moore, in Flor. Maur. § Seych. p. 357. M. caule sepius brevi; folis pluribus lanceolatis acutis striatis carinatis; racemo flexuoso erecto paucifloro, vaginis paucis ovatis; floribus parvis albis brevi-pedicellatis expansis ; sepalis lanceolatis acutis, petalis similibus angustioribus; labello ovato cordato acutiusculo plano; calcare brevi crassiusculo clavato recto, apice curvo; columna brevi crassiuscula curva; lobis rostelli verticalibus obtusis ; anthera depressa ; ovario erecto parum curvo. Ankafana, Deans Cowan! in Herb. Brit. Mus. Also Bourbon, Balfour. Mauritius, Telfair! Ayres, Bouton, etc. A small tufted plant, with a usually short stem and a moderate number of leaves, generally lanceolate, 2-34 inches long, by 4-3 across, sometimes shorter and broader, 14 inch long by i, and a number of rather long white roots. The racemes are slender, zigzag, and few-flowered, from one to three on a plant; they rise from the axils of one or more of the lowest leaves. The flowers, from two to five on a raceme, are white, 2 or 3 lines across. The spur is short, rather thick, and club-shaped, with the apex turned up. The capsule is elliptical, 4 an inch long, rather large in proportion to the size of the flower. A. multiflorum, Thouars, t. 74, seems, as Moore (1. c.) has classed it, to be a luxuriant variety of this species. M. OCHRACEUM, n. sp. ; caule brevi; foliis angustis linearibus lanceolatis, inequaliter bilobis, pedunculos superantibus striatis carinatis; peduneulis tenuibus rectis, vagina singula, floribus singulis parvis ochraceis ; bracteis ovatis ; sepalo postico lanceo- lato acuto erecto, lateralibus similibus reflexis; petalis angus- tioribus brevioribus acutis; labello eymbiformi acuto; calcare filiformi, apice clavato; columna brevissima crassa ; anthera de- pressa biloba; ovario a latere torto, pedicello brevi. Ankafana, Deans Cowan! in Herb. Brit. Mus. The affinity of this plant is with M. caulescens. The stem is as ORCHIDS OF MADAGASCAR. 489 short as in that species, and the habit is much the same. The leaves are narrow, 4 inches long by 1 in diameter, narrowed at the base; the peduncles very slender, 3 inches long, with a single dark- coloured close-fitting sheathing-leaf, and bearing at the apex a single yellowish ochreous flower, rather larger than that of M. caulescens. The petals and sepals are rather narrow and acute, the latter erect or reflexed, the former carried forwards over the lip. The lip is not flattened but concave and somewhat boat-shaped, acute at the tip, a little longer than the petals. The spur is long and slender, pendulous or spreading, filiform, with the apex a little dilated. The ovary is short and remarkably twisted side- ways; the spur running across the twisted portion. MyysTACIDIUM INAPERTUM, mihi. — Angrecum inapertum, Thouars, Orch. Iles Afr. t. 50; Lindley, Gen. § Sp. Orch. Pl. p. 247; S. Moore in Flora of Maur. and Seych. p. 358. JM. caule semipedali; folis paucis lanceolatis angustis acutis, apice bilobis, 23-uncialibus ; pedunculis paucis tenuibus brevibus paucifloris ; bracteis minimis ovatis; floribus parvis inapertis viridibus; petalis sepalisque angustis lanceolatis aeutis; labello subsimili subsquali; calcare brevi curvo, quam ovarium multo breviori, clavato obtuso horizontali; ovario angulato oblongo. Madagascar, Ankafana, Deans Cowan! in Herb. Brit. Mus. Mauritius, Thouars, Bojer. Bourbon, Thouars. This curious plant is represented by a couple of stems with leaves and a single flower, together with a coloured drawing, in Deans Cowan’s collection ; and I have seen it nowhere else. The stems are about 6 inches high, and emit several long roots; at the top are a few, three or four, narrow leaves, 25-2 inches long and about 4 inch or rather less across. The peduncles are about ) an inch long, and bear one or two (?) small flowers. The colour of the flower, as given by Thouars, is white. Deans Cowan figures it as green. It does not appear to open entirely, and is possibly self-fertilized. The spur is short and club-shaped, parallel with the ovary. The ovary is straight and rather sharply angled. The whole flower is half an inch in length. It seems to be most nearly related to M. caulescens. — § Eumystacidium. M. TENELLUM, n. sp.—Pusillum, caulibus 2-3-phyllis ; foliis viridibus flaecidis obtuse lanceolatis, quam racemus brevioribus; racemis tenuibus erectis, demum nutantibus folia superantibus ; 9 N 9 a L - 490 MR. H. N. RIDLEY ON THE floribus minutis dissitis; bracteis ovatis obtusis quam pedicelli brevioribus ; petalis sepalisque subsimilibus, ovatis obtusis brevi- bus; labello ovato obtuso carnosulo, lateribus erectis; calcare erasso obtuso cylindrico, quam ovarium breviore ; columna brevis- sima ; anthera depressa, stigmate oblongo-angusto ; polliniis duo- bus, caudicula singula tenui elongata; glandula minima triangula; capsulis parvis oblongis ellipticis. Ankafana, Deans Cowan ! in Herb. Brit. Mus. This little plant is allied to M. pusillum, Lindley, a native of the Cape. It has white flowers, according to a drawing made by Deans Cowan. The leaves are ¿$ an inch long and 2 lines across, broader and more flaccid than M. pusillum. The racemes are an inch long, the flowers 2 millim., the capsule 1 line. MYSTACIDIUM GRAMINIFOLIUM, n. sp.; rhizomate brevi repente, caulibus erectis crebris foliosis ; foliis ligulatis coriaceis obtusis bilobis striatis carinatis, marginibus minute crispis; vaginis membranaceis striatis; floribus minutis ad 6, in racemis gracillimis flexuosis laxis in axillis foliorum superiorum ; bracteis ovatis vaginantibus ; petalis sepalisque subsimilibus, lanceolatis acuminatis, petalis angustioribus; labello lanceolato acuminato, sepala subæquante; calcare cylindrico brevi obtuso, columna brevissima; lobis rostelli parvis obtusis; polliniis duobus; glandula parva ovali ; stigmate ovali ; ovario gracili erecto. Ankafana, Martio 1881, no. 3977, Hildebrandt!; same loc., Deans Cowan! in Herb. Brit. Mus. A very small-flowered plant, remarkable for its numerous narrow grassy leaves, somewhat like those of an Zsochilus with an unequally bilobed apex and erisped margin. The small flowers are laxly arranged on a very slender rhachis. "The petals, sepals, and lip are very similar in shape, triangular lanceolate, acuminate, giving the flower a star-shaped appearance. The spur is rather short, straight, cylindrical, and blunt. Gussonia, Richard, Orch. Maur. p. 76.—Mierocolia, Lindley, Orch. Pl. p. 60.— Mystacidii species, Bentham § Hook. f., Gen. Plant. jii. p. 584.—Angreci species, auct. plur. The original species of this genus was G. aphylla, Rich., which was afterwards reduced to Anyrecum. This plant and several others closely allied to it have so many remarkable characters in common not possessed by any others in the genus Mystacidium, that it seems more satisfactory to restore the old genus Gussonia ORCHIDS OF MADAGASCAR. 491 for their reception. The extensive development of roots, formiug in G. aphylla a dense mass, the absence of leaves at least during the flowering season, the short stem, the minute flowers, are the most important characteristics of the genus. Besides the species here described, Angrecum globulosum, Hochst. (Gussonia globulosa, mihi) appears to belong to the genus. It is a native of Abyssinia, and is remarkable for bearing a little bud-shaped cone of. leaves in the centre of the mass of roots. Epiphytz pusille, aphylle (sub anthesin), radicibus smpius copiosis, longis; caule brevi; floribus minutis, petalis sepalisque ovatis obtusis; calcare brevi obtuso quam pedicelli breviore; columna recta brevi; anthera apice producta acuta; polliniorum eaudicula unica lineari, glandula parva, lobis rostelli linearibus porrectis. Flores parvi, aurantiaci, calcar ferme semi- unciale, cylindricum ........... ... G. Gilpine. Flores minutissimi, labellum ovatum obtusum, calcar scrotiforme, racemus laxus 3- uncialis ........ Lecce eee eee G. exilis. ealear clavatum, racemus brevis com- pactus 2... eee eee eee eee G. aphylla. labellum trilobum, lobus medius linearis.. — G. physophora. GussowrA GirPINEX, mihi.—Angrecum Gilpine, S. Moore, Journ. Linn. Soc. xvi. p. 206. G. radicibus longis validulis subteretibus; caule brevi, semi- unciali ; racemis 11-uncialibus plurifloris; floribus fulvis parvis, pedicellis gracilibus rectis; bracteis ovatis cueullatis obtusis patentibus, inferioribus vaginantibus; petalis sepalisque brevi- bus ovatis obtusis, sepalis paullo majoribus et acutioribus ; labello subequali ovato obtuso; calcare recto crassiusculo tereti, quam pedicelli parum breviore; columna recta tenui; anthera magna ovata acuminata; polliniis parvulis, caudicula lineari, basi dilatata, longa; glandula parva oblonga; rostelli lobis linearibus ascen- dentibus porrectis. Antananarivo, Miss Gilpin!; Betsileo, Baron no. 199! Herb. Kew; Ankafana, Deans Cowan! Herb. Brit. Mus. This plant is distinguished from the rest of the genus by its larger orange-coloured flowers, with a straight thick spur nearly The stem is usually short, 3 an inch in length à an inch in length. at most, and emits numerous rather thick roots, sometimes 492 MR. H. N. RIDLEY ON THE more than 6 inches in length, but far fewer than in either G. exilis or aphylla. Each stem bears one or two short flower- racemes, and the withered remains of some of previous years' growth. The flowers, 11 or more to a raceme, are supported on very slender pedicels 3 an inch in length, with small ovate spreading bracts. The column is larger than usual, and of very thin consistency. The anther is large and firmly attached to its filament, and, as in the other species of the genus, is prolonged in front into a sharp beak, which lies over the long rostellum. The pollinia are small, but the caudiele is long and slender, slightly dilated at the upper part, and bearing a small oblong gland. The rostellum is oblong and bent downwards over the stigma, and ends in two flat narrow lobes projecting outwards over the entrance of the spur, and slightly curved up at the end. GussoNIA APHYLLA, A. Richard, Orch. Maur. 76, t. 11. f. 1.— Angrecum aphyllum, Thouars, Orch. Iles Afr. t. 73; S. Moore, in Flor. Maur. and Seych. p. 358; Reichb. f. in Walp. Ann. vi. p. 907. —Saccolabium aphyllum, Lindley, Gen. and Sp. Orch, Pl. p. 223. G. caule longo, radices validulas longas emittente; racemis uncialibus gracilibus; floribus minutis aggregatis; bracteis mi- nutis ovatis acutis patentibus, pedicellis gracillimis brevibus; sepalis, petalis et labello ovatis oblongis subacutis subzequalibus ; petalis paullo minoribus; calcare recto clavato, basi dilatato, apice abrupte saccato decurvo; caudicula oblonga; glandula subtri- angulari ; lobis rostelli linearibus porrectis. Madagasear: Nosi-Komba in arboribus supra ora, Dec. 1879, no. 3256, Hildebrandt!; no special locality, Thouars! in Herb. Brit. Mus. Mauritius, on seashore round Flacq, and on Mon- tagne Longue, Bojer! Bouton. This plant is distinguished from the nearly allied G. exilis by the shorter and denser racemes, longer stem, fewer and stronger roots, and larger flowers, with a clavate spur. The pedicels of the flowers are 3 millim. long, the petals and sepals 13 millim., equalling the spur. Thouars figures the flowers as white; Moore states that they are reddish. G. PHYsoPHORA.—Angrecum physophorum, Reichenbach fil. Otia Hamburg. fase. ii. p. 78. Resembles G. aphylla, but has larger flowers with a three-lobed lip. Tt was obtained by Hilde- brandt in Nosi-Komba, no. 3255. ORCHIDS OF MADAGASCAR. 493 GussowrA EXILIS.—Microceelia exilis, Lindley, Gen. and Sp. Orch. Pl. p. 61. G. caule brevi crassiusculo ramoso, radices copiosissime emit- tente; racemis pluribus elongatis gracillimis triuncialibus pluri- foris; vaginis pluribus acutis; floribus minutissimis ; bracteis baud vaginantibus, membranaceis ovatis acuminatis; sepalis ob- longis lanceolatis obtusis; petalis subsimilibus minoribus parum angustioribus; labello brevi oblongo ovato obtuso; calcare gib- boso-saccato ; columna brevi; anthera depressa conica; polliniis parvis oblongis; caudicula oblonga; lobis rostelli magnis tri- angulis. Madagascar, Forbes, no. 28, in Herb. Brit. Mus.! and Herb. Lindley! This singular little scrambling epiphyte is remarkable for the immense extent of its roots in proportion to the size of the rest of the plant. It has a short stout stem but little branched, l inch long and about 2 lines thick. The racemes are much longer and laxer than those of G. aphylla, and bear at the base a few acute sheathing-leaves. I have seen no other leaves at- tached to the plant, but among the roots are entangled a number of very small leathery oblong leaves, which may possibly belong to it. The bracts are not sheathing at the base, they are ovate acuminate. The flowers are among the smallest of Orchideous plants, hardly 1 millim. in length. The sepals and petals are blunt and oblong-lanceolate ; the lip very short, its base forming a rather large saccate spur. The anther is rather large and somewhat firmly attached to the column. The pollinia are very small and orange-coloured. The lobes of the rostellum are foliaceous, triangular. AERANTHUS, Lindley. This genus has been removed from the neighbourhood of An- grecum and referred to that of Aerides by the authors of the ‘Genera Plantarum, apparently on the supposition that the column has a long foot like the latter genus. This is not the case in any of the specimens which I have seen, which include one of Lindley's types of the genus. The comparative scarcity of spe- cimens in satisfactory condition for examination is probably the cause of the error. Among Deans Cowan's spirit-specimens is, however, a single flower, from the examination of which I make 494 MR. H. N. RIDLEY ON THE the following notes. The column is very short and footless, longer, however, than in most Angreca. The base of the lip is swollen and gibbous, forming a kind of pouch, the bottom of which is perforated vertically by the spur, so that the entrance to the spur is not placed close to the base of the column, but at some distance below, and it seems that it is the portion of the lip between the aperture and the base of the column which has been taken for the foot of the column. The structure and venation show clearly that it is a part of the lip. The genera Mystacidium and Afonia were formerly added by Reichenbach to Aeranthus, which, however, seems better confined to the two original species, distinguished by the peculiar form of the labellum alluded to above, the tailed perianth, and the few-flowered slender peduncle rising from the lowest leaves. AERANTIIUS GRANDIFLORUS, Lindley, Bot. Reg. t. 817; Gen.and Sp. Orch. Pl. p. 213 (pars). Madagascar, St. Marys, Forbes! in Herb. Lindley. Introduced in a living state and cultivated at Kew, 1824. The only specimen that I have seen of this plant is a peduncle with a single flower in the Lindley Herbarium, taken from the plant introduced by Forbes. -It is about 8} inches in length, and covered with 7 or more dry, long, striate acute sheathing-leaves, each averaging two inches in length. Lindley, in the ‘ Genera and Species of Orchidaceous Plants,’ reduced it to A. Arachnites, Lindley, from which it appears to be distinguished by the larger size, shorter and broader undulate leaves, more numerous lax sheaths on the peduncle exceeding the internodes and quite covering the peduncle, and the paler yellow-green petals and sepals and white lip. In Deans Cowan's spirit-collection, however, is a peduncle with a flower which probably belongs to this species. The flower is about as large as that of A. grandiflorus, but the tails of the petals and sepals are much longer and the peduncle-sheaths much fewer, the lip and petals shorter than the sepals, which are 3 inches long, of which the tails form the greater length, 2} inches. The lip is 1 inch in length and nearly 2 in the broadest part. The pedicel has only one or two sheathing-leaves, which are rather long. The rostellar lobes, probably the wings of the column in Lindley’s description, are small, acute, and deflexed, the stigma rather long and oblong in shape. But Lindley's plant had “alis semi- ORCHIDS OF MADAGASCAR. 495 ovatis conniventibus supra stigma horizontaliter porrectis, stig- mate subrotundo excavato." AERANTHUS ÁnACHNITES, Lindley, Bot. Reg. sub t. 817 (Acran- thes, lapsu) ; S. Moore in Flor. Maurit. 4 Seych. p. 352.—Aeran- thus Arachnitis, Hook. Bot. Mag. no. 6034.—A. grandiflorus, Lindley, Orch. Pl. p. 243, ex parte.—Dendrobium Arachnites, Thouars, Orch. Iles Afr. t. 88. Madagascar, low down on trees, Ankafana, Deans Cowan (icon.). Mauritius, Bojer, Bouton, Ayres in Herb. Kew. Rodriguez, Balfour f. (var. Balfouri), fide S. Moore, l.c. The flowers are given as white in Thouars’s figure, yellowish green in that in the ‘ Botanical Magazine,’ and dull bluish green in Deans Cowan’s figure. It is altogether a smaller plant than A. grandiflorus, Lindl. JEONIA. A small genus of about five species, all natives of the Mas- carene Islands. Zonta Macrostacuya, Lindley, Gen. and Sp. Orch. Pl. p. 245. —Epidendrum macrostachys, Thouars, Orch. lles Afr. t. S3.— Beclardia macrostachya, Ach. Rich. Orch. Maur. p.19, t. 11. fig. 2. —Aerides macrostachyon, Spreng. Syst. Veget. p. 719; Boj. Hort. Maur. p. 313.— Aeranthus macrostachys, Reichenbach fil. in Walp. Ann. vi. p. 980; S. Moore in Flor. Maur. et Seych. p. 352. Madagascar, Ankafana, Deans Cowan in Herb. Brit. Mus.! Mauritius, La Savanne, Bojer. Bourbon, Balfour in Herb. Kew. This plant has a rather stout stem covered with imbricated linear oblong green leaves arranged distichously and slightly re- curved, 5 or 6 inches in length by 3 inch in breadth, the apices unequally bilobed. The roots are rather numerous, stout and long. The racemes, one or two to a plant, are erect, about 9 inches high, with numerous large white sweet-scented flowers. The lip is obovate and crenulate, the apex bilobed, with or with- out a small cusp between the two lobes. The column is rather Short; the lobes of the rostellum are linear acute and thrust for- ward over the somewhat large stigma. The anther is helmet- shaped with an acuminate peak in front: inside are two flaps making it almost biloeular. The pollinia are circular in outline, rather flat, and with a groove. There is but one gland, which is ovate-acuminate, the broad apex being notched. Richard (/. c.) 496 MR. H. N. RIDLEY ON THE figures two, whence the plant has been dissociated from its near allies ZE. polystachys and Aphrodite. Fonts Avzznri, Lindley, Bot. Heg. t. 817 ; Gen. and Sp. Orch. Pl. p. 244.—Epidendrum voluere, Thouars, Orch. Iles Afriq. t. 81; folio edition pl. iv.—Aeranthus volucris, Reichenbach fil. in Walp. Ann. vi. p. 900; S. Moore in Flor. Maurit. & Seych. p. 352. Æ. caule longiusculo ; foliis dissitis lanceolatis; racemis longis gracilibus paueifloris; floribus magnis albis; bracteis brevibus ovatis obtusis ; sepalis obovatis, petalis subsimilibus parum latio- ribus et acutioribus; labello quam petala longiore, lobis laterali- bus circa columnam convolutis, lamina biloba obovata, lobis ro- tundatis ; calcare brevi cylindrico recto obtuso; columna brevi; lobis rostelli ovatis foliaceis. Madagascar, Thouars! in Herb. Brit. Mus. Tanala, Deans Cowan (icon.). Mauritius, fide Bojer. Bourbon. The flowers of this plant are as large as those of Æ. macrosta- chya, one inch across, to which species it is most nearly allied. They are white with a yellow centre, according to Deans Cowan. The stem and leaves much recall those of Cryptopus elatus, the latter being lanceolate and shorter, and more distant than in most of the other species of the genus. The peduncle is very long in proportion to the stem, in Deans Cowan’s sketch nearly 2 feet long, while the stem is only 6 inches. The lower part bears a few distant dark-brown sheaths. The flowers are few. The petals are 2 an inch long, 5 lines across in the broadest part. The lip is 1 inch long and j inch in diameter. Neither in Thouars's type- specimen in the British Museum nor in Deans Cowan's drawing do I see the cusp between the lobes of the lip figured by Thouars. The leaves are figured and described by him also as acute, but, according to Deans Cowan, they are slightly unequally bilobed at the apex. Æ. ROSEA, n. sp. ; caule semipedali, radices crassiusculas emittente ; foliis dissitis obtuse lanceolatis uncialibus ; pedunculis paucifloris patentibus; floribus expansis mediocribus ; bracteis parvis ovatis ; sepalis spathulatis obtusis viridibus ; labello roseo, quam sepala multo longiore, lobis lateralibus convolutis rotun- datis obtusis, lobo medio multo majore, obeuneato bilobo, laciniis obtusis ; caleare brevi cylindrica recta. Madagascar, Ankafana, Deans Cowan! in Herb. Brit. Mus. This species is most nearly allied to ZE. Auberti, of which it ORCHIDS OF MADAGASCAR. 497 has exactly the habit, but the raceme is shorter and stiffer, the flowers smaller and differently coloured. The leaves are 1 inch long by $ in diameter. The peduncle 12 inch long, the flower 4 an inch in diameter. Monta POLYSTACHYA, Benth. § Hook. f. Gen. Plant. p. 584. —Epidendrum polystachys, Thouars, Orch. Iles Afrig. t. 82.— Listrostachys polystachys, Reichb. fil. in Walp. Ann. vi. p. 909; S. Moore, in Flor. Maur. and Seych. p. 354.—Angrecum poly- stachyum, Rich. Orch. Maur. p. 74, t. 10. f. 25 Lindley, Comp. Bot. Mag. ii. p. 205. 4E. caule brevi, foliis caulem amplectentibus oblongis loratis, ineequaliter bilobis; racemis sepe longis erectis laxis; bracteis brevibus ovatis laxis; floribus mediocribus; pedicellis brevibus; sepalis lanceolatis acuminatis acutis viridibus patentibus; petalis subsimilibus, angustioribus brevioribus ; labello albo, sepalis sub- squilongo, lobis lateralibus circa columnam convolutis, parum crenulatis, lamina rotundata, cuspide acuminata longa; calcare brevi clavato saccato; lobis rostelli porrectis; caudiculis pol- liniorum linearibus, glandula ovata emarginata. Madagascar, Thouars! in Herb. Brit. Mus. Bourbon, Thouars. Mauritius, Bojer. The leaves are about 8 inches long by 5lines in diameter. The raceme in the type-specimen is 6 inches long, the flowers rather distant. The sepals are 7 lines long, the lip including the cusp is of the same length, the latter is 4 inch. In Thouars's figure the flowers are given as entirely white; but in plants flowered by Mr. Christy at Sydenham, March 1885, the sepals and petals were light green and the lip white. CRYPTOPUS. This monotypic genus is confined to the Mascarene Islands. The only evidence I have of its being a Madagascar plant is a flowerless specimen collected by Lyull, which necessarily is rather doubtful. It seems, however, common in Mauritius, and will probably be found in Madagascar when the eastern coast is more fully explored. C. en ATUS, Lindley, Bot. Reg. sub t. 817; Hook. Journ. Bot.1. p. 45, t. 115; S. Moore, in Flora Maurit. and Seych. p. 349.— Angrecum elatum, Thouars, Orch. Iles Afriq. t. 79.—Beclardia 498 MR. H. N. RIDLEY ON THE elata, Richard, Orch. Maur. p. 78, t. 11. fig. 3.—Epidendrum dipterum, Sieber, Herb. Maur. no. 210. Madagasear? Lyall! in Herb. Kew. Mauritius, Sieber, Herb. Maurit. 210! in Herb. Brit. Mus. Bourbon, Richard! in. Herb. Lindley. Sine loc., Thouars! in Herb. Brit. Mus. Aerides coriaceum, Swartz, Schrad. Diar. Bot. 1799, p. 234; Willd. Spec. Plant. iv. p. 131.—Saccolabium coriaceum, Lindley, Gen. and Sp. Orch. Pl. p. 224, from Madagascar, is quite inde- terminable. NEOTTIER. UConrunis, Thouars. Corymbis, Thouars, Orch. Iles Afr. table i.—Corcurborchis, tbid.—Corymborchis, Thouars, fide Blume.—Hysteria, Reinwardt, Blume Cat. Hort. Buitenz. p. 99.—HRhynchanthera, Blume, Bijdr. fig. Ixxviii.—Macrostylis, Kuhl & Hasselt. ed. Breda, fasc. 1, t. 2. A small genus of six or seven species scattered over the tropics of both Worlds. C. corymbosa, Thouars, Orch. Iles Afr. tt. 37, 38.—C. Thouarsil, Reichb. f.in Bot. Zeit. 1849, p. 868.—C. disticha, Lindley, Folia Orchidacea (pars).—Corymborchis Thouarsii, Blume, Orch. Ind. Archipel. p. 126, pl. 44. fig. 1, A—c. Madagascar, Nossibé, Pervillé no. 142. Bourbon, Thouars. West Africa, Ins. St. Thomas, Don in Herb. Brit. Mus. This species is nearly allied to C. veratrifolia, Blume, with which it was confused by Lindley under the name of C. disticha, from which it is distinguished by its smaller flowers and always persistent column and robuster habit. Blume quotes Corymborchis as the original name given by Thouars to the genus; but in the copy of the ‘Orchides d'Iles d'Afrique’ in the British-Museum Library, the name is Corcur- borchis in the synoptical table and Corymbis on the plate. Pogonta. The only species of this genus that occur in the tropics of the Old World belong to the section Nervilia. Two are found in the Mascarene Islands, P. Thouarsii, Blume (Arethusa simplex, Thouars, . 4239. » = T. umbellatus, Kunth, d. 4240. » = Cannamois simplex, Kunth. 4243. Piquenier’s Kloof, flum. Olifant, = Dovea macrocarpa, Kunth, 9. 4430. Prope Cape Town, = Thamnochortus imbricatus, Mast. 4431. Cape Flats, = T. dichotomus, R. Br. 4432. » = Elegia parviflora, Kunth. 4433. » = Restio Eleocharis, Mast., d. 4434. » = Thamnochortus distichus, Mast., d. 4435. » = Willdenovia striata, Thunb., d. 4436. = Thamnochortus dichotomus, A. Br., d. ” 4437. Table Mountain, = Restio compressus, Rottboll, 9. 4439, 4440. Table Mountain, = R. bifidus, Thunb., 9. 4441. Prope Cape Town, = Lamprocaulos Neesii, Mast., d. 4442. ,, » = Restio multiflorus, Sprengel. 4444. ,, » = Elegia deusta, Kunth. 4445. Table Mountain, = Dovea ebracteata, Kunth, 9. 4446. Muisenberg, = Hypolena Eckloniana, Mast. 4446 (2). Cape Flats, = Thamnochortus spicigerus, R. Br. 4446 (3). » = T. erectus, Mast. 4447. In Mont. Diabol., = Hypodiscus albo-aristatus, Mast. 4448. Cape Flats, — Restio bigeminus, JVees, 9. 4449. » = Thamnochortus Burchelli, Mast., d. 4450. » = T. spicigerus, F. Br. 4451. » = T. imbricatus, Mast., 9. 4452. » = Restio furcatus, Nees, 2. 4453. » = R. cuspidatus, Thunb., g. 4455. Table Mt., = Thamnochortus umbellatus, Kunth, var. di- stachya. 4456 d,4457 9. Table Mt., = Elegia juncea, Linn. 4458 9,4459 3. Muisenberg, = Willdenovia striata, Thunb., 9. 4460, 4461, 4462. Mont. Diabol., = Restio tetragonus, Thunb. 4463, 4464. Mont. Diabol., = R. triticeus, Rottb. 4465. Muisenberg, = Hypodiscus Willdenovia, Mast., d. 4467. Cape Flats, = Willdenovia striata, Thunb., monstros. 4168. " = Dovea mucronata, Mast., 9. 4469. = Restio multiflorus, Sprengel. » 594 SUPPLEMENTARY NOTES ON RESTIACEÆ. 4640. In Mont Tab., = Elegia acuminata, Mast. 4722. In Mont Diab., = Thamnochortus fruticosus, Berg., 2. 4730. Table Mt., = Restio compressus, Rottb. 4732. Cape Town, = R. quinquefarius, Wees. 4758. In Mont Diab., = Thamnochortus dichotomus, R. Br., 9. 4760. Restio quinquefarius. SPECIERUM A CL. REHMANN COLLATARUM RELATIO. 623. = Restio compressus, Rottb. 627. =R. bifidus, Thunb. 631, 632, 635. Ad Montem Tabular., = Elegia deusta, Kunth. 922 et 980. In Mont. Diabol., = Restio filiformis, Poiret. 924. pro parte, = Elegia parviflora, Kunth. 925, 926. = Restio cuspidatus, Thunb. 929. = R. triticeus, Rottb. 1184. = Hypodiscus albo-aristatus. 1454. = Restio triticeus, Rottb. 1712. = R. multiflorus, Sprengel. 1802............. R. Eleocharis. 1805, 1806, 1807. Cape Flats, = Elegia parviflora, Kunth. 1808. = Restio bifurcus, JVees. 1810............. = Elegia deusta, d ? 2555. In mont. prope Worcester, = Hypodiscus albo-aristatus, Mast. 2556. In mont. prope Worcester, = Willdenovia striata, Thunb. 2562. » » » = Restio intermedius, Kunth. 2564. » » » = Elegia glauca, Mast. 2565. » » » = E. parviflora, Kunth. 2567. ” » » = Restio curviramis, Kunth. 2568. » » » = Elegia verticillaris, Kunth. 2572. » » » = Restio Gaudichaudianus, Kunth. 2674. In mont. prope Worcester, = Cannamois scirpoides, Mast. MR. 8. LE M. MOORE'8 STUDIES IN VEGETABLE BIOLOGY. 595 STUDIES IN VEGETABLE BroLoay.—l. Observations on the Con- tinuity of Protoplasm. By Spencer Le M. Móónz, F.L.S. [Read 2nd April, 1885.) (Prates XIX.-XXI1.) Continuity of protoplasm, other than through the sieves of sieve-tubes, was first announced in Phanerogamia by Dr. Tang] *, in his well-known study of the endosperm of Strychnos Nux- vomica and certain palms, It is not my intention to quote the bibliography of a subject which has so fascinatingly influenced authors that a large number of memoirs have already been pub- lished upon it; it will suffice here to refer to Messrs. Schaar- schmidt’st and Gardiner’st recent papers, wherein ample bibliographical details will be found. It will be remembered that Tangl discovered the thick un- pitted walls of the large cells forming the main mass of the endosperm of Strychnos Nux-vomica to be penetrated by threads of protoplasm, which join similar threads running through con- tiguous cell-walls, whereby the protoplasm of each cell is placed in communication with that of its neighbours. This structure is not to be seen in sections treated with water, a medium which causes strong swelling of the walls, accompanied by their stria- tion, and the differentiation of the layers of thickening into an outer and an inner zone. After the action of dilute alcohol upon dry sections, or, better, of iodine in slightly diluted alcohol, or in potassium iodide, with or without subsequent treatment with ehloriodide of zinc, the groups of connecting threads are easily to be seen with moderate powers, the outer threads of each group bowed, the middle ones straight, somewhat after the fashion of the strings of a mandolin. A bird’s-eye view of the wall shows that it is uniformly perforated by crowds of small round openings. Minute examination led Tangl to conclude that these openings are entirely occupied by protoplasm, whose inability to take up colouring-matter shows that it is referable to the limiting layer (Hautschieht), a strueture well developed in these cells. The walls of the epidermis and immediately underlying thin-walled * Pringsheim's Jahrb. f. Wiss. Bot., Band 12 (1879). t ‘Nature’ 1885, Jan. 29th. t ‘Nature, 1885, Feb. 26th. 596 MR. S. LE M. MOORE’S STUDIES tissue are not penetrated by threads *, which are also absent from the proximal wall of the layer bounding the chamber in which the embryo lies; except for this they may be traced with ease throughout the endosperm. Had Tangl worked with the seeds of Strychnos Ignatia he would have found the examination of the threads a much easier task ; to see them it is sufficient to place sections of the hard dark endosperm in water—a cell from the central mass of the endo- sperm so treated is shown at figure 1. The area of thickening is much swollen, and in the case represented shows a couple of strongly marked stris: close to each other and to the primary wall, but this is by no means the only method of striation ; some- times the outer half only of the swollen area t is marked with a number of fine, closely-placed lines, or it may be uniformly so marked throughout. The lumen of the cell is polyhedral in out- line, with convex sides and acute angles, the latter usually point- ing towards the corners of the cell; its granular protoplasm contains oily globules. The threads can easily be seen with a 3-inch objective, but higher powers are necessary for their satis- factory examination, which shows them to be, except for their larger size, similar to the threads of S. Nux-vomica ; there is the same grouping, the same outward bulging of the flank-threads, the same inequality in calibre, and the same continuity through the primary wall with the threads of neighbouring cells. Many of the threads of cells treated with water alone appear to stop short at the primary wall, but if sections be placed for a few minutes in a solution of iodine in alcohol, to which is added a small quantity of water, the nature of the intercellular commu- nication is brought more clearly into view (fig. 2). The primary wall can now be distinguished only as a fine bright line traversed by the stained threads, the course of many of which, not being in the same plane as that of the section, appears to be inter- rupted; the protoplasm within the angular lumina is more deeply stained than that of the threads; it is invested by the more lightly stained limiting layer (represented by the dark outline of the cell-contents shown in figs. 4 & 17). Great difference exists bi In ordinary circumstances, for in sections placed for twenty-four hours in chloriodide of zinc continuity can be traced right up to the epidermis. * By "swollen area" I do not mean the primary wall, which itself swells up somewhat in water. Cf. figs, 1 & 2. IN VEGETABLE BIOLOGY. 597 in the calibre of the threads: exceedingly thin at times, they are occasionally so wide in parts as almost to suggest the idea of rifts in the wall, but in this event it is seldom that the rift-like appearance extends very far along the thread. A modification of this rifting here and there occurs in the form of nodular swellings fairly evenly distributed down a thread, or confined to some part of its course (fig. 3). A bird's-eye view of the wall (fig. 5) shows the rounded openings of the thread-embracing canals; they are larger than those of S. Nuz-vomica, as will immediately be seen on reference to Tangl's figure*. As already mentioned Tangl finds the threads of S. Nuz-vomica to consist only of the limiting layer of the protoplasm; this con- clusion he founds upon their reaction to colouring media— solutions of carmine in ammonia and in aleohol—neither of which is capable of staining either the limiting layer or the threads. Mr. Gardinert was, I believe, the first to show that the threads can be made to exhibit the ordinary dye-reactions of protoplasm ; the best medium for this purpose he found to be a dye which he cails picric Hoffman's bluet. In his just-quoted memoir Gardiner enters elaborately into the nature of aniline blues under the impression that Hoffman’s is the only, or at all events the best, dye for this purpose; his care is, however, quite need- less, for I have succeeded in staining the threads of S. Ignatia in the ordinary way, first placing sections in picric acid, washing and then leaving them in Judson's Oxford-blue for a few minutes, and mounting in water or in strong or dilute glycerine. Ex- cellent results have also been obtained with a dye prepared by Gardiner's method, substituting Sands’ for Hoffman's blue. Gardiner's experiments tended to show that the threads do not consist of the limiting layer of protoplasm alone, even if that structure enters at all into their composition, Tangl’s failure to dye them having apparently been caused by the very faint stain imparted to minute pieces of protoplasm by carmine. For per- manent preparation the best mounting medium is water or caleium chloride ; glycerine either pure or dilute soon causes some slight change in the wall, whereby the threads become invisible ; * L. c. tab. v. fig. 12. t Phil. Trans. vol. 174 (1884). t To 100 cub. centim. of strong alcohol is added the same bulk of distilled water ; the solution is then saturated with picric acid, and Hoffman's blue added to it until it has assumed a dark blue-green colour; it is then filtered. 598 MR. S. LE M. MOORE'S STUDIES this is especially the case with sections which have lain before staining in chloriodide of zinc. Section of the endosperm in a plane parallel to the axis of growth shows that it is composed of an epidermal layer greatly thickened upon its outer side, beneath which lie several rows of small cells with but slightly thickened walls. These walls react with water very much as do the primary walls of the deeper- lying large thickened cells, that is to say they swell up, but not nearly to the same extent. With a low power (fig. 6) it is not possible to see the primary wall of these cells, but it can be made out with more or less ease right up to the epidermis with a }-in. objective. These small cells are followed by layers of cells inter- mediate in size between them and the underlying large cells. In sections which have lain for twenty-four hours in chloriodide of zinc, well-stained threads piercing the walls can be traced as far as the epidermis (fig.7); the protoplasm is therefore continuous throughout the endosperm. In his memoir in the ‘ Philosophical Transactions,’ Gardiner contents himself with noticing the fact of continuity in the endosperm of S. Ignatia without entering into any details. His failure to observe the interesting differences between this species and S. Nux-vomica was probably due to the use of chloride of zinc alone, an excellent reagent for tissues which do not swell up in it to any great extent, but much inferior to solution of iodine iu aleohol in the case under notice. To this author belongs the credit of first discovering an instance in which continuity through the walls of endosperm-cells can be seen without special reagents (Bentinckia Conda-panna) *. Here, however, comparatively high powers are required, and as the walls are said to show merely an “appearance of striation,” it is clear that the fact cannot be verified with the ease that it admits of when S. Ignatia is examined f. Tangl ¢ expressly says that the endosperm-cells of S. potatorum do not show continuity. The chief difference between these cells and those of S. Nu«-vomica is that their lumen is not round, but produced towards one or more of the faces of the primary wall, * Phil. Trans. memoir. t For this reason, and because of the low powers sufficient for the demon- stration, the use of S. Ignatia seeds is strongly to be recommended for class purposes. t Le. IN VEGETABLE BIOLOGY. 599 often to such an extent as to touch the latter. Professor Stras- burger* doubts the accuracy of Tangl's statement; he believes that communieation is maintained by groups of threads running through the wall from pit to pit, but he is not quite satisfied about the point; and Gardiner leaves matters in the same state. The seeds sent me from Kew were, I suspect, from the same sample as those which Gardinerf examined; they were so old and dry as to afford little chanee of satisfactory examination ; however I have sometimes seen the method of continuity shown at fig. 8, and occasionally caught sight of fine threads running concentric with and occupying the interval between the two intramural threads there figured. The only doubt which I have with regard to this species is whether the cell-walls themselves be not penetrated by threads. "Treated with water both primary wall and secondary layers swell up greatly, the former especially at the angles, the latter to differentiate into an inner many- striated zone, and an outer one marked by a great number of minute dots; the striation of the inner zone is continued along- side the canals where they are present (fig. 9). If a section is placed in a drop of solution of iodine in alcohol upon a slide for a minute or two, and a cover-slip being put on, is then carefully irrigated with water, in some part of the section the dots just referred to may appear in the form of interrupted lines which seem to radiate on all sides, from the lumen to the circumference of the cell. "Whether, however, these be cords of intercellular communication or small rifts occurring only in old seeds, or, as they sometimes seem to be concentrically disposed, merely inter- rupted strie of the layers of thickening, I am quite unable to decide. The cells of the horny endosperm of S. spinosa swell up so much in water that the lumen is frequently all but obliterated ; the layers of thickening differentiate in water into two sharply- defined zones, a narrow inner strongly, and a wider outer faintly, striated one. Through the walls of cells treated with iodine in alcohol and partially swollen up in water continuity can be made out if care and patience be brought to the examination ; it is, however, a matter of difficulty to see the threads, the remarkable striation of the wall greatly impeding the view of them (fig. 10). * « Ueber den Bau und das Wachsthum der Zellhaute.' t Phil. Trans. memoir. 600 MR. S. LE M. MOORE’S STUDIES The chloriodide of zine method is quite inapplicable to this case, from the violent swellings whieh that reagent causes; for the same reason it is impossible to stain the threads with aniline dyes. The subepidermal cell-layers have thicker walls than have the other species of the genus. As is the case with S. Ignatia, so here prolonged use of chloriodide of zine serves to demonstrate continuity up to the epidermis. Mr. Gardiner* has discovered continuity in the endosperm of a large number of seeds referable to many of the natural orders ; but with the exception of Tamus and Dioscorea, which have un- pitted endosperm, communication is established through the wall at the bottom of pits. The endosperm of at least two species of Diospyros (embryopteris and melanoxylon) greatly resembles that of Strychnos Nux-vomica; dry sections near the surface show a number of round lumina lying apparently in a homogeneous matrix; but if sections be cut in planes near and parallel to either of the three planes mutually at right angles which roughly divide the seed of D. embryopteris into symmetrical halves, the cells composing the core of the endosperm will be seen to be greatly elongated. Placed in water the contour of the cells becomes more or less evident, but the walls swell up only to a slight extent, and that slowly. In this case the method of demonstra- tion invented by Gardiner, with the slight modification already mentioned (p. 597), was resorted to—indeed, it is the only one which will satisfactorily bring the threads into view. Sections were placed in chloriodide of zine for several hours (usually about a day); they were then removed and examined in this condition, or washed until disappearance of the brown stain, and dyed with picric blue. Figure 11 shows one of the cells with round lumen and its connections with its neighbours as seen before, and fig. 12 another as seen after, the action of the dye. One remarkable point about this is the difference in the number of the connecting threads; close to such a state of things as is drawn at fig. 12 one may find cells intercommunicating by means of a much larger number of threads. As this cannot be due to incomplete stain- ing, since precautions were taken to ensure saturation with the dye, one cannot but conclude that the threads are apt to become obliterated. The protoplasm of the long cells is also connected with that of the neighbouring cells on all sides (fig. 13). Con- * Phil, Trans, memoir. IN VEGETABLE BIOLOGY. 601 tinuous rifts in the wall may sometimes be observed in thin sections—one of these is shown at fig. 14; whether such result from splitting of the tissue during section or are congenital or consequent on absorption I do not know. Figure 15 shows a small portion of the wall penetrated by holes of varying calibre. Investing the seed are several layers of thin-walled cells forming a somewhat spongy testa, immediately beneath which the outer- most endosperm-cells are smaller and thinner than the under- lying ones ; nevertheless continuity can be traced across their walls. In all essentials D. melanoxylon resembles the species just described. One point of great interest remains. Hitherto, as far as I am aware, no one has succeeded in tracing the connection between the intramural threads and the cell-protoplasm. Tangl* says that employment of ordinary reagents causes total plasmolysis, and subsequent authors seem, by their silence, to acquiesce in the truth of the statement. So much attention has recently been directed to this matter + that I have made a special study of the plasmolysis of the cells of Strychnos, and find that Tangl is incorrect. It is true that in the case of many cells the ex- hibition of strong reagents is followed by a plasmolysis which is total, but this is by no means always the case. The best way of observing plasmolytie threads is to place sections in a drop of solution of iodine in alcohol on a slide, a minute or so afterwards placing a cover-slip upon them and examining either in this state, or after addition of a small quantity of water. Figures 16, 17, and 18 were drawn from sections treated in the latter way, the first of them from S. Nuz-vomica, the second from S. Ignatia, the last from S. potatorum. Although easily overlooked, the threads can readily be made out with care; in many instances they may be seen to run into the intramural threads. Figure 18 repre- sents an unusual condition; although plasmolysis has been violent, yet some of the threads extend unbroken across nearly half of the cell’s lumen. It will be observed that in the cases of partial plasmolysis here figured the protoplasm remains for a part of its course in close apposition to the wall of the cell, and it would seem at least possible that the persistence of the threads is due to this fact, the close relation between wall and plasma acting as * L.c. , t H. de Vries, “ Untersuch. üb. d. mechanisch. Ursachen d. Zellstreckung ; Gardiner, Phil. Trans. memoir, and Proc. Roy. Soc. 1882; Bower, Quart. Journ, Mier. Sc. 1883. 602 MR. 8. LE M. MOORE' 8 STUDIES a steadying force whereby the shock of plasmolysis is lessened, especially at the opposite side of the wall, which is usually where the threads occur. The demonstration of the threads is a matter of cautious treatment with water, for, if the iodine tincture be made too dilute, the cell-wall swells up violently, and the threads are broken; however, it often happens that they can still be distinguished in the form of frays upon the edge of the contracted pellet of protoplasm (fig. 4). S. potatorum differs from its con- geners in the minute pitting of the cell-wall: the pits in ordinary section are seen to be very narrow slits ; in oblique section they present a pectinate appearance upon the wall (figs. 18 & 19). I have also, on very rare occasions, seen plasmolytic threads in sections of the endosperm of Diospyros embryopteris treated for several hours with chloriodide of zinc, and stained with picric blue. Years ago Pringsheim * showed that such threads could be seen in cells of fern-prothallia, Riccia, &c. acted upon by weak chloriodide of zinc, but I was unprepared to find them after pro- longed action of the strong reagent. Figure 13 represents the best example obtained ; two only of the threads could be traced with distinctness into intramural threads, the connection of the rest not being patent. The threads, whether of Strychnos or of Disopyros, stained well, those of the former with iodine, those of the latter with picric blue. Continuity in the Floridec. It has long been known that the cells of certain Floridez are placed in communication with each other by means of plasma- containing apertures in their walls. A cursory examination of Nügeli's works t is sufficient to show that he had very clear con- ceptions upon the subject. In the case of several genera (e.g. Callithamnion, Laurencia, Peyssonellia) he figures pores which he describes as having each a small elliptical opening in the centre ; and it is plain from the context that he understood them to function as organs of intercellular communication. Kützing t, too, occasionally figures continuity ; and Zanardini $, in the text * * Bau und Bildung der Pflanzenzelle,’ t ‘Die neuern Algensysteme;' On the Nuclei, Formation and Growth of Vegetable Cells, Ray Society's volume for 1845 ; Sitzb. d. Konigl. Bayer. Acad. 1861. 1 ‘Phycologia Generalis. See tabb. 63, 65, 79, and others. $ ‘Iconographia Phycologica Adriatica, Chondrus? adriaticus (tab. 38), and Schizymenia minor (tab. 62). IN VEGETABLE BIOLOGY. 603 accompanying two of his figures, speaks of cells * anastomosing ” and “uniting to form a network.” At the same time it must be remembered that continuity is only casually figured in the great illustrated works on Phycology, and has often been overlooked even by observers of the first order*. Indeed, it is not until we come down to recent times that we find the subject to have been specially attended to. Mr. Archer f seems to have broken the first ground by his study of Ballia callitricha and allied species, the young cells of which he concluded to be in open communication by means of a pore, which is immediately sealed up by the development, at both its ends, of a plug or “stopper” formed of a substance giving none of the reactions of starch or cellulose. Professor Wright f, of Dublin, observed similar phenomena in Polysiphonia and Griffithsia setacea. A propos of the growth of Polysiphonia, he says (p. 516):—* In its early condition the tips of the rays and of the two ends are quite distinctly open, and the protoplasmic con- tents freely communicate with the tube-like cells above and below and the siphon-cells around it; but as it becomes mature, and as a cell-wall is definitely formed around it, these pore- openings are closed up, and these piugs will prevent sometimes even the *stopper'-like form described by Archer; so that here, as in Griffithsia, it seems to me that these plugs or stoppers are, as it were, only the result of the cell finally closing itself up.” Merely mentioning that Agardh § figures some cases of undoubted continuity, wecome to the poorly illustrated memoir of Sehmitz ||, whose views are quite different from those of his predecessors. He holds that, with the exception of the Corallines, whose cells intercommunicate by wide open pits, the pits are closed by means of an exceedingly thin membrane, on each side of which lies a plate of very thick substance, dyed intensely with hematoxylin &c. "These plates, he says, are connected together by means of numerous strands of protoplasm, which penetrate the closing membrane chiefly or exclusively at its circumference, and unite laterally to form a hollow cylindrical band of intercellular com- * For example, Pringsheim, in his “Beitr. z. Morph. d. Meeres Algen," Abhandl. d. Kónigl. Acad. d. Wissensch. z. Berlin, 1862. t Trans. Linn, Soc., new series, Botany, vol. i. 1 Trans. Roy. Irish Acad. vol. xxvi. 1879. $ ' Florideernes Morphologi.' | Sitzb. d. Kónigl. Acad. d. Wissensch. z. Berlin, 1883. 604 MR. S. LE M. MOORE’S STUDIES munieation. He notes that the substance of the closing membrane greatly resembles the slime of sieve-tubes, and thinks that the plates may perhaps result from local differentiation of the parietal protoplasm. Soon after the appearance of Schmitz’s paper we find Mr. Hick* returning to the old view of Nägeli, but obser- ving that in some cases a “ diaphragm,” traversed by a strand of protoplasm, is developed within the * collar" through which the connecting-thread passes. Mr. Massee t has recently studied the development of Polysiphonia urceolata and fastigiata. He figures the cells as, at an early stage of their history, opening into one another by wide canals, unprovided with ring, plate, or membrane, but as afterwards shut off by the formation of plates of cellulose perforated for the passage of threads of protoplasm. Lastly, the subject has been studied by Mr. Gardiner, whose preliminary notice f I have failed in obtaining a sight of, after many endea- vours. However, from a short paper in * Nature' (Feb. 26th, 1885), it appears that Gardiner's results are, at least in the main, confirmatory of Schmitz's, as he says that, by the aid of chloriodide of zinc, he has been enabled to make out the existence of a pit- closing membrane in every case examined. It may then be said that four views of protoplasmic continuity in the Floride: have been enunciated :— I. Continuity is direct, by means of a pore functioning as a simple canal of communication.—Nageli (and also Kützing, Zanardini, &c.). IL. Continuity is direct, by means of a wide or narrow strand of protoplasm passing through a ring of harder protoplasmie substance, within the circumference of which a closing membrane may be formed.—Hick. IIT. Continuity direct, but soon interrupted by the formation of “stoppers " over the mouths of the pore.—Archer, Wright. IV. Except in the Corallines, continuity is indirect always, the connecting-threads passing through a pit-closing membrane.—Schmitz, Gardiner. CHONDRUS MAMILLOSUS.— The common species of this genus (C. crispus) was very faithfully figured by Kützing §, the figure * ‘Journal of Botany,’ Feb. and March 1884. t Journ. of Roy, Microsc. Soc. 1884. 1 Proc. Camb, Phil. Soc. Feb. 11th, 1884. $ L.c. tab. 73. — IN VEGETABLE BIOLOGY. 605 being by no means improved upon by Hick *. Its rarer congener, C. mamillosus, shows continuity in a still more striking form. The thallus of both may be said to be formed of two orders of cell—outer small elliptical ones placed in longitudinal rows per- pendieular to the surface, and inner larger ones provided with radiating arms, aptly compared by Mr. Hick with multipolar nerve-cells. Both authors figure continuity throughout the thallus, in the outer cells by means of fine threads, on which the rows appear like strung beads, in the inner ones by the coalescence of their arms, “ uniting to form a network,” as Zanardini says of his C.? adriaticus. The arms of the inner cells of C. mamillosus are much longer than those of C. crispus, and the lumen is some- times reduced to a mere junction-point of the arms (fig. 20). In many cases the arms of neighbouring cells are not continuous, and the cells are placed in communication by means of a single fine thread of protoplasm, at or near the centre of which is placed à shining bead-like nodule, through which the thread passes without undergoing division. In other cases a nodule is not seen, and the arms of neighbour cells appear to simply coalesce. The small outer cells are more numerous than is the case with C. crispus ; they unite with their under- and overlying as also with their lateral fellows by single fine filaments, which may or may not be provided with a nodule. In no instance was it found possible, with sections well stained with Judson's Oxford blue, to resolve the connecting filament into two or more, nor was any evidence obtained of its division in passing the nodule. PorvrpEs RorUNDATUS.—A section of the thallus in a plane at right angles to the axis of growth shows that it may roughly be said to consist of three tissue-areas:—1st, the epi- and hypodermis, composed of small cells more or less elongated perpendicularly to the surface; 2nd, a middle zone of larger cells filled with starch- granules; 3rd, fibre-like cells forming the central mass. A view of this is given at fig. 22. Continuity seems to hold throughout the thallus of this type, although it is impossible to see the con- nections of all the cells in any one section. Also, the continuity is direct, by means of a slender thread upon which asmall nodule staining well may frequently be seen. By means of these threads the epidermal cells unite, either singly or in pairs, with a cell of the subjacent layer, which itself communicates, alone or with its lateral fellow, with the longer cell immediately beneath it; and * [, c. tab. 244. fig. 12. È) LINN. JOURN.—BOTANY, VOL. XXI. -Y 606 MR. S. LE M. MOORE’S STUDIES the same relation holds with the two succeeding layers, and with the outermost layer of granular cells, but deeper down I have failed to trace any dual connection. The cells of the middle zone are more or less ovoid in outline, with the exception of the inner- most ones, which are lengthened out more or less perpendieularly to the surface, and often correspondingly narrowed. From the extremity of each of these latter cells a filament of protoplasm runs into one of the cells of the central zone, union occurring, in all observed cases, near an end of this cell (fig. 23). The central- zone cells course irregularly, but with a general up-and-down direction, through the frond, accompanied by rows of smaller cells with granular contents ; they are greatly elongated and somewhat clavate at the ends, and united each by a slender thread of proto- plasm to its supra- and subjacent fellow, but never with their cir- cumjacent neighbours. Examination with high powers failed to resolve the threads, nor could their division in theneighbourhood of the nodule be ascertained (fig. 24). Similar structure is pos- sessed by Furcellaria, from which Polyides differs so slightly as scarcely to be distinguishable, except when fruiting. CERAMIUM RUBRUM *,.— The branches of this type are com- posed of an axial range of large round cells, covered over with much smaller ones, and with others intermediate in size packed in the intervals left between the central cells. Fig. 25 a shows two of the large cells, and the way in which they run towards each other by means of wide pits. The cells have, as is usual in the Floride:, a dense layer of protoplasm investing the more fluid contents, which are granular and contain erystalloids. The pro- toplasm of the pits is either apparently homogeneous or distinetly granular; over it can usually be traced a delicate limiting layer. The pits do not meet, owing to the interposition of a plate of membrane riddled with fine pores, through each of which passes a fine strand of protoplasm into the pit of the adjoining cell (figs. 26,a & b, 256). The membrane is circumscribed by two annular pieces of highly refractive substance, and presenting, a8 it itself does, all the reactions of protoplasm, one belonging to and surrounding the distal end of each pit, but so closely placed are they as to be easily mistaken for a single piece. These pieces, together with the membrane, will be spoken of as forming à "ring" In longitudinal section the central cells appear as 8° * I suspect that this may be referable to some other species; for the axial cells of other specimens (typical and varietal) of C, rubrum which I have had the opportunity of examining all have much shallower pits. IN VEGETABLE BIOLOGY. 607 many large beads threaded upon a string made up of conjoined canals and rings. Similar but smaller rings are interposed be- tween the central and intermediate cells. The protoplasm of the pit comes flush up with the contour of the ring, and remains in close connection with it, and this makes discovery of the mem- brane-piereing threads difficult. The small outer cells present a type of continuity already dwelt upon. Each cell, polyhedral in outline, and with a dense layer of protoplasm placed, with but a few small intervals, against the whole wall, is connected with its circumjacent fellows by a single fine filament, at or near the centre of which is placed a bright bead-like particle, similar to those on the connecting- threads of Chondrus and Polyides (fig. 27). Examination with a high power (900 diameters) failed to indicate any division of the thread in passing the bead. By studying the development of the beads it is shown that they are of the same nature as the rings, but they differ from them in not being provided with a membrane. CERAMIUM RUBRUM, var.—In longitudinal section are seen a central row of long, nearly rectangular, surrounded by a number of smallcortical, cells ; others, intermediate in size, occupying the interval between the pairs of large cells, as well as forming a con- tinuous layer immediately investing them. In the floor of the central cells is a small ring, the membrane of which is perforated at several points of its cireumference, and occasionally elsewhere. Threads of protoplasm pass through the perforations, and con- nect the protoplasts. The cortical cells resemble those of the form last mentioned. The thallus of LAURENCIA HYBRIDA is made up of external, nearly isodiametral, and a central core of cells lengthened in the direction of the axis of growth. The walls of either type are greatly thickened. The protoplasts are connected with their neighbours by strands which run up the pits, each to a single ring, undivided through the centre of the membrane of which, or, as is usually the case, divided at its circumference or whole face, it passes into the contiguous pit. CHYLOCLADIA ARTICULATA.—The branches, resembling those of an Opuntia in miniature, consist, in fully grown parts, of a series of chambers, of which the roof and floor, situated at the narrowed portions, are made up of several layers of short cells. Cells many times longer than broad occupy the centre of a young 2y2 608 MR. S. LE M. MOORE’S STUDIES chamber, but by its expansion during growth gaps occur between them. Fig. 28 shows the method of connection between the short and long cells, and between those of either kind and their fellows. The long cells are frequently clavate, and the ends are joined to their supra- and subjacent fellows by saddle-shaped surfaces. Their protoplasm is continuous through circumferential or scat- tered pores in the closing membrane of often broadly contoured rings (fig. 29a, b,c). Similar rings serve as media of communi- cation between the sides of contiguous long cells. The walls of the chambers are built up of small cells, externally underlain by several layers of larger ones resembling those of their roof and floor. These latter cells communicate by means of rings, through whose membrane a single or several strands may pass (fig. 30 a, b). The connection between the external cells is by means of fine nodule-bearing filaments, of whose division in passing the nodule proof was not obtained. Of the genus CALLITHAMNION, a small fragment only, belong- ing to an undetermined species, has come under observation. In one case the achromatin fibres and commencing formation of a cell-plate were well seen (fig. 31a). Atthe next stage examined (fig. 81) the cell-wall has closed up except where it yields passage to a fine thread, at either end of which is a gleaming particle recognizable from the ordinary protoplasm with which the young cells are gorged. Fig. 32a shows the ring defi- nitely formed ; and 32 6 division of its intercellular protoplasm in passing through the membrane: in this case the ring has been ruptured ; otherwise demonstration of division at this stage is very unsatisfactory. GRIFFITHSIA SETACEA.—The branches are composed of a single range of cells with broad ends, between which, at the blind end of somewhat wide pits often filled with granular protoplasm, is a small thin ring provided with a membrane pierced with holes over all its surface. PriLoTa SERICEA.—A figure showing continuity at and near the growing-points is given by Hick*. The frond consists of an indefinite main axis bearing a number of opposite secondary axes, themselves bearing axes of the third degree, andsoon. The axis of a branch is formed of scutcheon-shaped cells, each of which gives origin on each side to a cell which is the basal cell of a branchlet, and with which, as with its preceding and succeed- * L. c. tab. 244, fig. 9. IN VEGETABLE BIOLOGY. 609 ing axial cell, communication is ensured by the agency of a fila- ment, slender at first, but rapidly inereasing in calibre. In the earliest condition observed by me a bright point of plasma has made its appearance at the top of the protoplasm upon the contiguous faces of the ultimate and penultimate cell; these points rapidly grow, and soon form a small ring through which the eonnecting filament passes. After the formation of several axial cells the branch becomes corticated : the first step in this is the shutting-off, except for a fine filament, of a small cell from both the upper and under face of the axial cell at its distal end. A couple of small wing-like cells then make their appearance, one on each side of the proximal wall of the new cell, with which latter they remain in filamentary connection. Meanwhile the basal cell of the branchlet on each side has lengthened down- wards, and, dividing transversely, it occludes from the external world the side of the axial cell. At the proximal end of the- latter small cells are now budded off, the whole of the cells thus formed repeatedly dividing and growing into a firm cortex round the axial cell. Since every cell remains in permanent filamentary connection with its parent, it follows that there is continuity in one way or another throughout the thallus. Before cortication has advanced to such a stage as to shut out the axial cell from sight, the continuity of the axial cells has plainly become indirect. Fig. 33 shows a surface view of a ring in this condition, its membrane having a couple of pores. Further down the evidence of indirectness is much clearer. Fig. 344 is a view in longitudinal, and fig. 34 b one in transverse section, ofa well-grown axial cell. In consequence of the tenacity of the protoplasm (a fact of such constant occurrence in the Floridew) demonstration of continuity is very difficult. The longitudinal section shows a broad ring, from either side of whose circumter- ence stretch threads and films of protoplasm to join the proto- plasm of the cell. In the transverse section a number of threads are seen passing into fine pores upon the circumference and just within the eontour ofthering. The ring is stopped with mem- brane upon which filmy cell-protoplasm can in some parts be seen ; it is probably this which prevents a sight of the pores all round the ring, for the pores can be traced with tolerable ease where there is no overlying protoplasm. N ear the centre is a bright point which may also mark the position of a pore. The axial cells communicate with their daughter cortical cells by means 610 MR. S. LE M. MOORE'8 STUDIES of rings similar to, but smaller than, those between the axial cells themselves. DELESSERIA ALATA.—The midrib of the frond is composed of several series of long wide cells with broad ends arranged in longitudinal rows separated by smaller cells, of which some have a wide, others a narrow lumen (fig. 85). Examination of the larger cells by means of longitudinal and transverse sections of the midrib of old parts of the frond, reveals a method of continuity different from anything hitherto remarked upon. Reference to figs. 85 and 36 will show that a large part of the floor and roof of neighbour cells is closed up by membrane, round which at certain points are placed small rings, to each of which runs a stout cord of protoplasm. The rings are themselves covered in with membrane similar to that which isolates the cells; this membrane is pierced either by several holes, in which case the *strand of protoplasm is divided in passing it, or in some cases where the ring is small, there is a single hole at the centre of the membrane (figs. 37 and 38). All round the large membrane can usually be traced with more or less ease a slender ring of highly refractive substance, similar to that of which we have seen the contour of the rings to be composed. Fig. 35 shows these matters in transverse section; in this case there are four small rings visible, and apparently a fifth veiled behind a cloud of protoplasm ; their number, however, varies between three and five. The only interpretation that can be put upon these facts is that the large membrane surrounded by its fine ring of gleaming protoplasm is homologous with the ring of Ptilota or Ceramium ; and that the small rings are accessory structures. Itis therefore proposed to call the former the “primary " ring, and the small rings "secondary " ones. The rings will be spoken of as “ com- pound ” or “simple,” according as they have or have not secondary rings. l In some cases it would perhaps seem that a simple ring is interposed between two of the large cells. I am inclined to think that fig. 40 represents such a ring, although it was not possible to make out satisfactorily that there were not small secondary rings at the circumference of the larger one. The primary ring is closely attached to the inner boundary of the incomplete roof-and-floor wall of the cells, which is frequently wavy in outline. Even when best developed, it is very slender ; in profile view it can be easily recognized as a continuation of the IN VEGETABLE BIOLOGY. 611 secondary rings (fig. 37,a). The secondary rings, although usually cireumferential in position, are not always so. Fig. 41 shows a small membrane, near the centre of which are placed what appear to be smali secondary rings; a string of proto- plasm is attached to one of them. The larger of the cells surrounding the rows of wide ones communicate with the latter, and with their fellows, by means of rings which in some cases at least are compound (fig. 41). The rings between the cells with narrow lumen are small and simple. The small epidermal and immediately underlying cells are con- nected by fine nodule-bearing threads similarly to those of Cera- mium rubrum (fig. 27). Working with dried material, I have laboured under a distinct disadvantage in studying the development of these rings. All that I have learnt about the matter is this :—Below the almost hemispherical apical cell of a growing branch are two demilune cells, one beneath the other. Beneath thisis a similar but larger demilune cell which divides into five cells, two on either side ofa central scutcheon-shaped one. The outermost cell on each side takes part in the formation of the thin lamina of the frond; the central cells rapidly elongate in the direction of the axis of growth, and after a time a transverse septum divides each of them in turn (fig. 42). Although the cells for some time after their formation are in very close apposition, it is possible, with the aid of chloriodide of zinc and by careful focusing, to see nume- rous small dots of protoplasm between the newly formed cells, These ultimately disappear with the exception of a single fine thread, at the centre of which is swung a small bright nodule of protoplasm (fig. 42). The intercellular threads appear to persist until division of the cells is definitely completed, whereupon, except for the one thread, they rapidly vanish ; by this means con- tinuity is maintained throughout the whole thallus. Until now cell-division has been at right angles to the plane of the thallus ; but by a series of divisions parallel to that plane the formation of a midrib is announced. In the earliest condition of distinct continuity between the largest as also between the other cells, I have never seen more than one strand of protoplasm. Some little way below the growing-point the connection between contiguous large cells is as shown at fig. 43, where at least three fine strands of protoplasm pass each of them through a small nodule appa- rently without undergoing division. At this stage the ring is 612 MR. $. LE M. MOORE'S STUDIES still in the simple condition; a view from above of the cell- closing membrane with three nodules in the stage represented at fig. 43 will be found at fig. 44. POLYSIPHONIA NIGRESCENS *.— Each branch of the frond has a single longitudinal series of large axial cells, every one of which is surrounded by a number of smaller siphon-cells equal in length to it. The axial cell communicates with every siphon-cell by means of a well-defined ring, the membrane of which is every- where pierced with holes (figs. 45 and 46). "The rings are usually placed halfway up the axial cell; but it often happens that they are not all upon the same level. The upper and under surfaces of the latter cell are largely composed of membrane resembling in this respect the large cells of Delesseria alata. In fig. 47 an en- deavour has been made to represent this, as seen in a longitudinal section through the roof and floor of two axial cells. Upon the right a small portion of the very slender primary ring is visible, and several secondary rings also appear in it; although these rings are small, the protoplasm distributed to them is divided in passing them (as is shown, but not satisfactorily, by the ring near the median line of the section). Fig. 48 represents a bird's- eye view of a membrane ; several small secondary rings can be distinguished at its circumference ; while upon a large part of it lies a quantity of tenacious protoplasm. The siphon-cells are connected with their neighbours by small rings with perforated membranes (fig. 50). Upon the circum- ference of the axial cells at the nodes, and alternating in position with the rings connecting an axial cell and its siphon-cell, are placed a number of small rings; from each of these a delicate strand of protoplasm, sometimes here and there thickened, emerges to pass into the small ring of a cell much less in size than the siphon-cells between and beyond the outer portion of which it is situated (fig. 49). Similar cells intereommunicating by small rings are situated in the neighbourhood of the cell just mentioned. PoLYsIPHONIA FASTIGIATA differs greatly from the foregoing. In the oldest condition in which I have seen it, the axial cells communicate with each other by means of a small ring with per- forated membrane. The protoplasts—ameeboid in appearance from the presence round their equator of a number of stout pro- * I have not been able to get a sight of Kolderup-Rosenvinge's ‘ Bidrag til Polysiphonia s Morphologi,’ quoted by Schaarschmidt (7. ¢.). IN VEGETABLE BIOLOGY. 613 cesses, the agents of axial-siphonal continuity—are very dark in colour, but, when boiled in caustic potash and stained with chlor- iodide of zine, indirect continuity can be made out without much difficulty (fig. 51). HALURUS EQUISETIFOLIUS.— But if the large cells of Delesseria alata and Polysiphonia nigrescens are in great measure composed of membrane, the row of larger cells, of which the axes of Halurus equisetifolius are made up, show the opposite condition of things. The protoplasm in these cells is placed in continuity by means of a relatively small broadly-contoured ring placed at or near the centre of the wide flat wall of partition (fig. 52). Upon the inembrane lies protoplasm in such quantity that the perforations are masked; the semi-profile view of the figure, however, shows that some at least of the threads are circum- ferential. Each of the axial cells of a branch bears near its top a whorl of frequently ramifying branches composed of a single row of small cells several times longer than broad. With the basal cell of every branch the axial cell communicates through a small perforated ring; and a similar but smaller ring is placed between every pair of cells of the branch. The origin of these rings seems to be the same as that of those hitherto described; that a membrane is developed in them very early is shown by fig. 53, which repre- sents the membrane between the ultimate and penultimate cell of a branch from a preparation boiled in caustic potash for a minute or two, and then stained with picric blue. From the branchlet’s basal cell ultimately descend corticating fibres re- sembling in structure the branches themselves. The connections between the cells of these are similar to those between the cells of the branch. BarriA CALLITRICHA.—-If one of the main branches of this species be examined a short way above the point where cortica- tion begins, the junctions of its single series of axial cells will present the appearance shown at fig. 54. It would seem at first sight that, in accordance with Harvey’s * description, the upper cell protrudes into the lumen of the lower; but, as Mr. Archer t showed, the reverse is the fact: this can be ascertained without much difficulty by cutting off the branchlets of an uncorticated branch and turning the flattened axis through an are of 90°. In * In Hooker's ‘ Journal of Botany,’ 1840. + Trans. Linn. Soc. 2nd ser. Bot. i. p. 211 ef seq. 614 MR. S. LE M. MOORE’S STUDIES the centre of the upper cell’s floor, as also of its underlying neighbour's roof, is a highly refractive knob-like structure fixed to the wall of partition. These knobs are what Archer calls the “stoppers.” The curious articulation gives origin to what he terms “ bottles.” An idea of the method of continuity in this type wil be gained from inspection of figs. 55 and 56. Each stopper rests upon a thin bed of membrane similar to the mem- brane closing the rings of other Floride; ; and the wide inter- cellular pit is apparently filled with the same substance. The cireumference of the stopper's base is continuous with the thin parietallayer of cell-protoplasm, and can frequently be easily di- stinguished as a dark line from the brighter mass of the stopper: the base itself is filled up with membrane upon which the mark of the pit is very clear (fig. 55). From the circumference of the base of a stopper fine threads run up the pit to be similarly inserted on the contiguous stopper. It will thus be seen that Ballia resembles other Floridee in the method of continuity, its chief points of difference being the extent to which the mem- brane is developed, and the remarkable form assumed by the highly refractive portions of the ring. The stoppers are endowed with an extraordinary power of growth. Originally making their appearance, like the halves of an ordinary ring, as two gleaming pieces of protoplasm upon the contiguous faces of just-formed cells, between which a narrow channel of protoplasm extends (fig. 58), they rapidly enlarge and continue to do so, it would seem, almost without limit (fig. 57)*. The view here adopted is that continuity is always direct in the early history of the cells, and in some cases (Chondrus, Polyides, Furcellaria) persistently so; while in others direct continuity may persist in one part of the thallus, and be supplanted by the indirect form in another (Ceramium rubrum, &c.). The young cells are placed in communication by means of a fine filament upon which is in most cases placed a small nodule, just as a bead is strung upon a thread. The ground for this statement is that in surface views of the nodule only a single small central pore can be seen, and that the thread itself, as slender as the single threads piercing the membrane of rings, cannot be seen to undergo divi- sion in passing the nodule. The next fact to notice is the rapid growth of the thread accom- * Several enlarged stoppers are figured by Archer (/. c. pl. xxviii.). IN VEGETABLE BIOLOGY. 615 panied by concurrent growth of the nodule to form a ring. Meanwhile the protoplasm between tbe two pieces of highly refractive substance which form the contour of the upper and under faces of the ring has commenced to differentiate a pit- closing membrane, which either occupies the whole of the space bounded by the ring’s contour, with the exception of a single narrow canal at or near the centre, through which the con- necting thread passes, or partially fills up the space, leaving circumferential or scattered pores. Whether the original con- necting filament is always really single from the first is doubtful in such cases as the wide axial cells of Halurus equisetifolius and of the main branches of Ballia callitricha. A typical ring with upper and under highly refractive contour-lines and perforated membrane being formed and increasing in size concurrently with the growth of the cell, the strands of protoplasm passing through the membrane may themselves become surrounded by a small ring, and growing rapidly in calibre just as did the original fila- ment, like it may themselves be divided in passing this secondary ring by the development of a membrane in the latter. I am not sure whether some of the large simple rings before mentioned do not eventually become compound: in Ptilota sericea, for instance, on one occasion a small ring was observed, to which convergence of two strands of protoplasm was plainly made out. This is a matter requiring further elucidation. It has frequently been the subject of remark that the parietal protoplasm of the floridean cell is of a denser consistence than that which it surrounds. A glance, for instance, through the plates of Thuret's and Bornet’s ‘Etudes physiologiques ' will be sufficient to establish this point. Figs. 25 « and 27 of the present memoir may also be quoted in proof of the same. A modification of it consists in the distribution of dense highly refractive proto- plasm in the form of a network upon the wall. This reticulation, originally observed, I believe, by Nägeli, seems to be of common occurrence. In most of the types examined by me it is more or less easily to be made out; but the most beautiful example I have seen is in Polysiphonia nigrescens, from which a drawing of a portion of a siphon-cell’s contents will be found at fig. 59. In another modification the protoplasm is arranged in broad lines down the wall of the cell; good examples of this occur in Ballia callitricha and Halurus equisetifolius. Now the tenour of my observations—inconclusive, it is true, from the want of living 616 MR. S. LE M. MOORE'S STUDIES material—is to the effect that in the newly formed cell a ring of dense parietal protoplasm arranges itself round the debouching point of the filament connecting it with its neighbour, just as it would arrange itself anywhere else upon the wall. The pit con- tinuing to deepen, two closely apposed pieces of protoplasm are strung upon the pit-traversing thread. Where the pit does not appreciably deepen (Ballia callitricha, Callithamnion sp.), the ring has a somewhat different appearance. I cannot therefore acquiesce in Schmitz’s statement, that a plate of highly refractive substance is placed at either side of the pit-closing membrane, for surface views almost always show the bright contour of a ring; the highly refractive substance upon the membrane some- times met with (e. g. Chylocladia articulata, secondary rings of Delesseria alata) is, I venture to think, brought there subse- quently to the formation of the ring. One point worthy of mention d propos of the comparison of these cells with sieve-tubes is this :—The cell-protoplasm is fre- quently found to be densely aggregated on one side or on both sides of a ring. Good examples of one-sided aggregation are found in the long cells from old chambers of Chylocladia articulata (fig. 60), of two-sided in Halurus equisetifolius (fig. 61) : in the latter type cords of protoplasm may frequently be seen passing from a plate-like aggregation at the top of the cell to the more fluid protoplasm lying beneath it. Running down the cell from stopper to stopper is sometimes found a stout band formed of dense, completely investing more fluid, protoplasm. Ballia callitricha shows this better than any other type studied by me (fig. 57): such a band may sometimes traverse the lumen of the axial cells of Ceramium rubrum; and as its calibre is equal to that of the pit, the impression is produced of a cord coursing uninterruptedly through several of the cells (fig. 62)*. I would suggest that this may be a means whereby more rapid molecular transference is effected. With reference to the question whether the substance of which the bright contour of the rings is composed be really protoplasm I would merely remark that there seems no reason why it should not be so regarded. Archer+ showed that the stoppers of Ballia are soluble in boiling caustic potash; and this is also the case * A somewhat similar disposition of the protoplasm is figured by Hick (l. ^ tab. 243, f. 6) in the axial cells of Ceramium acanthonotum. t L.e. p. 214, IN VEGETABLE BIOLOGY. 617 with the contour of ordinary rings ; but upon this ground there is as much reason for regarding erystalloids not to be of proto- plasmie nature as these ring-contours. tis only in the wide sense which modern research has given to the conception of pro- toplasm that I wish to be understood when referring the compo- sition of the rings to that substance. Postscripr.—As tending to confirm the conclusions arrived at from the necessarily incomplete study of development contained in the foregoing pages, it would have been advisable to quote in more detail Mr. Massee's before-mentioned paper on * Polysi- phonia ( fastigiata and urceolata)” in the * Journal of the Royal Microscopical Society’ for 1884. Massee’s method was to re- move the wall from cells near the growing-point by soaking for several hours in a solution of nitro-picric acid. By this means it was found that the young protoplasts are connected by fine marginal threads, as well as by a thicker central band of proto- plasm, and that, with the exception of an aperture for the passage of the latter, the protoplasts become isolated by the gradual de- velopment of a cell-wall between them. The aperture increasing in size with the growth of the cell, is after a time, spanned by a sieve-plate, which, as already stated, this author considers to be composed of cellulose. Reference should also be made to Mr. Hick’s recent paper (‘Journal of Botany,’ April 1885) on Continuity in several species of Fucus. By the employment of a number of refined processes, Hick has satisfied himself that continuity throughout the tissues of these types may obtain in either of four ways :— (i.) Through a comparatively wide and open pore ; (ii.) by the inter- position of a sieve-like diaphragm; (iii.) by means ofa single ribbon of protoplasm passing through a longand narrowslitinan otherwise impervious diaphragm; (iv.), as (iii.), but the ribbon reduced to a mere filament transmitted through a minute pore in the diaphragm. DESCRIPTION OF THE PLATES. Puare XIX. Fig. 1. An endosperm-cell of Strychnos Ignatia swollen up in water. (Ain this and the following figures marks the primary wall of the cells.) x 450. Du. 2. Piece of endosperm of the same treated with solution of iodine in alcohol and then with a small quantity of water. l x 600. , 3. The same, but after addition cf water, showing nodules upon the intra- mural threads. x 600. 618 MR. S. LE M. MOORE'8 STUDIES Fig. 4. The same, after exhibition of solution of iodine in alcohol: the cell has 11. 12. 13. 14. 15. 16. Fig. 17. 21. 22. 25. been swollen up in water; plasmolytic threads remain as a fray upon the edge of the protoplasm. x 600. . The same. View of wall, from above, showing the debouching points of the intramural threads. Xx 600. . The same. Epidermis and underlying cells treated with water. The section was taken in a plane parallel to the axis of growth. Xx 140. . The same. A subepidermal cell after lying twenty-four hours in chlor- iodide of zine. x 600. . Strychnos potatorum. Continuity of the protoplasm of two endosperm- cells. Chloriodide-of-zine preparation. x 600. . The same. Cells swollen up in water. x 600. . Strychnos spinosa. Small portion of section of endosperm treated with solution of iodine in alcohol. x 600. Section through the short cells of the endosperm of Diospyros embryo- pteris, showing continuity of their protoplasm through the walls ; after lying in chloriodide of zinc for twenty hours. x 600. The same, after twenty-four hours’ action of chloriodide of zinc; the threads stained with picric blue. x 600. A long cell of the same, treated similarly to the last : plasmolytic threads visible at one point. x 600. Rift in the wall of the same. x 600. A small portion of the wall, from above. x 600. Strychnos Nux-vomica. The lumen of the two endosperm-cells showing plasmolytic threads passing into intramural ones, thus demonstrating continuity. After action of solution of iodine and alcohol and sub- sequent treatment with a small quantity of water. x 600. Prats XX. Plasmolytic threads from endosperm of Strychnos Ignatia; and fig. 18, from that of S. potatorum. Same treatment as the last. x 600. . Pectinate appearance produced by the pits in the wall of the endosperm- cells of S. potatorum. x 600. . A stellate cell from the tetraspore-bearing process of Chondrus mamil- losus, its lumen reduced to a mere point of junction. X 600. Larger cells from the same, showing direct continuity. X 600. Transverse section of a portion of the frond of Polyides rotundatuss showing direct continuity. Not drawn to scale. . The same. Connection between a granuliferous cell and one of the long cells of which the centre of the frond is largely made up. X 600. . The same. Direct continuity between two of the small cells at the surface of the frond. x 900. Ceramium rubrum. a. Two of the axial cells, showing continuity of their protoplasm through small perforations in the membrane of à ring: one of the cells has two crystalloids, the other a single one. 6. A ring, from above; it shows a highly refractive edge and perfo- rated membrane. Glycerine preparation. X 600. . The same. a. A ring showing division of the threads in anticipation Fig. 29. 3l. 32. 35. IN VEGETABLE BIOLOGY. 619 of their passage through it (glycerine preparation), b. A ring from the front, showing threads passing through it. Such a view as the last can only be gained by a great deal of focusing (chloriodide-of-zine preparation). x 600. . Thesame, Apparent direct continuity of the small outer cells. Tho parietal protoplasm lies as a dense highly refractive layer upon almost the whole of the wall. x 600. - Connection between the short cells, isolating and the long cells lying in the chambers of Chylocladia articulata. Not drawn to scale. Rings between the long cells of the same, from above. From sections dyed with Judson's Oxford blue. x 600. . Continuity through the rings between the chamber-isolating cells of the same. From a section treated similarly to the last. x 600. Callithamnion sp. (a) A tip of a branch showing divided nuclei, achromatin fibres, and a commencing cell-plate, x 600. (5) A smaller tip than the above, the cell-wall fully formed, the protoplasm in continuity by means of a fine thread, at either end of which is placed a highly refractive piece of protoplasm. Chloriodide-of-zinc preparation. x 450. Cell-junctions of the same, further developed. The rupture of the ring (fig. b) has brought into view several fine threads issuing from its membrane. Fig. a x 450, fig. b x 600. . Ptilota sericea. A small ring froma branch shortly before cortication ; perforated membrane already developed. Chloriodide-of-zinc prepa- ration. x 600. . Connection between axial cells near the bottom of the frond of the same: a, in longitudinal, b, in transverse section. x 450. Delesseria alata. View of the membrane-closed bottom of a large cell, from above, showing a large primary ring and circumferential secondary rings. X 450. Pare XXI. . The same. Cell-junction in longitudinal section. x 450. In this and the last: p, primary ring; s, secondary ring; m, membrane. . Division of protoplasm in passing the rings of the same. The dark line on either side of the ring in fig. a is the primary ring. X 450. . The same. Rings, from above. x 600. . A small, apparently simple ring, between two of the large cells of the same. X 450. . A membrane from a ring, showing two small rings near its centre. x 450. . À primary ring with small secondary rings interposed between two of the cells which surround the large cells of Delesseria alata. From a preparation placed for eighteen hours in chloriodide of zinc. Xx 450. . Row of long cells from which the midrib of Delesseria alata is built up. The uppermost came fourth in order behind the scutcheon-shaped cell formed by division of the third demilune cell. The second cell is preparing for division ; it is connected with its fellows by a fine 620 Fig. 43. 44. 49. 58. 59. MR. S. LE M. MOORE’S STUDIES nodule-bearing filament. The third cell has just divided, and the nodule-bearing filament has not yet putin an appearance. The lower- most cell has already divided. Chloriodide-of-zine preparation. x 600. Wide cells of the same in an early condition, some distance below the growing-point; at least three nodule-bearing threads are visible through the incomplete wali of partition. x 600. A ring at the stage of fig. 43, showing the small slimy membrane and three apparently open secondary rings. X 450. . Polysiphonia nigrescens. A ring interposed between the axial cell and a cortical siphon, showing division of the protoplasm in passing it. x 600. . An axial-siphonal ring, from above. — x 600. . Longitudinal section through the junction of two axial cells of the saine, showing a large primary ring with secondary rings. x 300. . The same in transverse section, showing a large part of the bottom of a cell covered in with membrane, upon the circumference of which secondary rings are here and there to be seen. Xx 450. (The wall of this has been drawn rather too thick, and at the expense of the membrane.) Conneetion between the axial cell and a small cell in the neighbourhood of the siphons. "There were twenty siphon cells in this section. x 450. . Ring between two siphons of the same, showing division of the proto- plasm before reaching it. x 450. 5l. Protoplast from an axial cell of Polysiphonia fastigiata, showing a ring wth perforated membrane. From a teased preparation pre- viously boiled in caustic potash, and subsequently stained with chlor- iodide of zine. x 450. . Half-profile view of a ring of Halurus equisetifolius, obtained in longi- tudinal section of the corticated part of a branch. Xx 450. . Membrane of a small ring between the ultimate and penultimate cell of a branch of the same. The preparation was boiled in caustic potash and stained with picric blue. x 600. - Ballia callitricha. View of the junction of two axial cells of a branch before cortication has advanced to them. x 450. - The same, showing fine threads of protoplasm attached to the circum- ference of a stopper, the intercellular pit filled with membrane &c. Drawn with Hartnack’s No. 8 objective, ocular No. 4, but not to scale. . Two stoppers with threads of protoplasm piercing the pit-closing membrane. (Chloriodide-of-zine preparation.) x 450. . Longitudinal section of a corticated part of a branch of the same, showing a greatly enlarged stopper; from it the protoplasm extended as a wide band to the stopper in the floor of the cell. Xx 450. Apex of a slender branch. The stoppers are represented by a piece of highly refractive protoplasm at either end of the unelosed protoplasm- filled pit. x 600. Part of a siphon-cell of Polysiphonia nigrescens, showing reticulate protoplasm on its wall. x 600, —M— MR. PA, S, rM a IN VEGETABLE BIOLOGY. 621 Fig. 60. Junction of three long cells of Chylocladia articulata. The protoplasm has collected at the top of one of them. x 600. 61. Halurus equisetifolius. Meeting-point of two young cells. A dense layer of protoplasm lines the intercellular wall at each side. x 140, 62. Two axial cells of Ceramium rubrum. The protoplasm runs in the form of a hollow tube from stopper to stopper. x 140. STUDIES IN VEGETABLE Broroay.—IlI. On Rosanoff's Crystals in the Endosperm-Cells of Manihot Glaziovii, Müll. Arg. By SerNcER Le M. Mobs, F.L.S. [Read 2nd April, 1885.] Tu existence of crystals fixed to the wall of their containing-cell and themselves enclosed in an investment of cellulose appears to have been first noticed by Schacht *, who described in some detail the large single crystals found in the leaves of Citrus vulgaris. A few years afterwards Rosanoff t wrote his well-known memoir on the Crystal-masses in the pith of Kerria japonica aud Ricinus communis, and in the petiole of several Aroidex and the floral organs of Encephalartos and Nelumbium ; he showed that the crystals are either directly fixed to the wall of the cell and surrounded by a cellulose investment, or, still retaining the in- vestment, are pedicellate upon the wall. Similar crystals were found by De La Rue? in the leaves of Hoya carnosa. Pfitzer$, working over Schacht’s ground, very elaborately monographed the crystals of Citrus which he found not in the leaves alone, but in their stalks and in the stem: he further proved that they are not, as Schacht had supposed, of sulphate, but of oxalate of lime ; and he made a careful microchemical investigation of the sup- porting pedicels. He also discovered that at an early stage the crystal can be seen lying freely in its ceil surrounded by a nimbus of protoplasm. Besides this, Pfitzer showed that the crystals met with in the cortical parenchyma of Salir aurita, Populus italica, and other trees are secured to their cell-wall by thick depo- sits of cellulose. Poulsen || subsequently wrote a paper on the pedicellate sphere-crystals surrounded by cellulose of Rose-berries ; * Abhand. Senck. Gesell. z. Frankfurt, i. t Bot. Zeit. 1865 and 1867 t Dot. Zeit. 1869. § ‘Flora,’ 1872. A short account of the erystals is given by De Bary in his ' Vergleichende Anatomie’ (p. 147). | Vidensk. Medd. Naturhist. For. Kidbenhayn, 1874. LINN. JOURN.—BOTANY, VOL. XXI. - 622 MR. S. LE M. MOORE’S STUDIES and again * on the single similarly surrounded crystals, which he was the first to observe in the fundamental tissue, bast-paren- chyma, petioles, &e., of various Phaseolee. It is scarcely necessary to say that crystals merely attached to the cell-wall are not included in the category of Rosanoff's crystals, the essen- tial feature of which consists in the relation between them and a support of cellulose developed from the cell’s wall. It will be observed that up to the present time organs in the full tide of vegetation have been the home of these crystals ; the announcement of their discovery in a resting tissue has now to be made. On freeing sections of the endosperm of Manihot Glaziovii, Müll. Arg., of their oil with ether or alcohol, and washing away, as far as is possible, the closely packed aleurone- grains, the walls of the cells are seen to be everywhere studded with single, twin, and sphere-crystals of small size. Fig. 1 General view of oil- and aleurone-freed endosperm tissue of Manihot Glaziovit. x 450. gives a general view of the tissue under a moderate power. The chief forms assumed by the crystals are as follow :— (1) Clinorhombie crystals, sessile, or shortly pedicellate upon wall (fig. 1); truncation is sometimes carried further than is here represented. The size varies considerably. (2) Sphere-erystals fixed directly to the wall or on short pedi- cells, or occasionally supported on narrow bars running across the lumen of the cell. These spheres differ greatly in size ; they * ‘Flora, 1877. IN VEGETABLE BIOLOGY. 623 are frequently placed at the meeting-angles of the cells. They may be composed of small pyramids with blunt apices, or of a greater or less number of flat plates, or of clinorhombie crystals, or of a medley of the above. The masses are sometimes greatly reduced, as is shown at fig. 2, which represents a sphere-mass made up of a clinorhombie crystal capped by two small pyramids. Fig. 2. Fig. 3. Ys e Fig. 2. Small sphere-crystal, composed of a clinorhombic crystal. Fig. 3. Cellulose skeleton investing a five-sided prism. From a preparation treated with hydrochloric acid. x 600. (3) Five- (occasionally six-) sided short prisms with plane faces. "These are best studied in sections treated with hydro- chlorie acid, whieh dissolves out the crystal, leaving a delicate skeleton of cellulose (fig. 3). It often happens that the angles of these crystals are not well developed. (4) Twin crystals are of frequent occurrence (fig. 1), and occa- sionaly mulberry-masses (the crystals composing which are exceedingly small) are to be found. . These crystals are in various ways surrounded and attached to the wall by cellulose. Examples of this attachment are drawn Fig. 5. Fig. 6. Fig. 4. A single cell, showing a shortly pedicellate sphere-crystal, the cellulose bands in association with it stretching some way across the lumen of the cell. x600. . Fig. 5. A clinorhombie crystal in a well-developed investment of cellulose X600. And two pyramids, X 600. Figs. 6, 7, and 8. Cellulose investment of crystals dissolved in hydrochloric acid. x 600. 624 REV. G. HENSLOW ON VERNATION AND FOLIAGE in figs. 4 and 5 and 6-8; the last three after dissolution of the crystals in hydrochloric acid. Fig. 6 represents the invest- ment of a large sphere-mass at the junction of four cells; the support is thicker in fig. 7; in fig. 8 it is exceedingly delicate. The crystals are composed of oxalate of lime; they are also doubly refractive. The investing support, if it swells up at all in chloriodide of zinc, does not do so to any appreciable extent : from a mixture of aniline rose and violet, it selects the former, showing that it has somewhat of the character of collenchyma. Similar erystals are found in the tissues of the embryo; but whether they occur in the adult tissue, I cannot say. Mr. Scott, in his memoir on the latex canals of this species*, speaks of clino- rhombic and sphere-crystals, but is silent as to their attachment to the wall and investment with cellulose. My best thanks are due for aid in prosecuting the above re- searches to the Kew authorities, especially Mr. John R. Jackson, Curator of the Museums, who has been good enough to supply me with seeds; to the Royal Microscopical Society, whose Assistant Secretary, Mr. West, has been very helpful; to Mr. E. M. Holmes, for kindly assisting in the determination of species - and communicating specimens of Floridez; lastly, to Messrs. Fletcher and Davies, of the Mineralogical Department, British Museum, for obligingly looking at my preparations of the endo- sperm of Manihot Glaziovii. On Vernation and the Methods of Development of Foliage, g4 pro- tective against Radiation. By the Rev. George HENSLOW, M.A., F.L.S,, F.G.S. [Read 4th June, 1885.] Iv his work on ‘The Movements of Plants,’ Mr. Darwin says:— ‘The fact that the leaves of many plants place themselves at night in widely different positions from what they hold during the day, but with the one point in common, that their upper sur- faces avoid facing the zenith, often with the additional fact that they come into close contact with opposite leaves or leaflets, ciearly indicates, as it seems to us, that the object gained is the protection of the upper surfaces from being chilled at night by radiation." + * Quart. Journ. Micros. Se, 1884. t P. 284. AS PROTECTIVE AGAINST RADIATION, 625 He further adds* :—* We exposed on two occasions during the summer to a clear sky several pinned-open leaflets of Trifolium pratense, which naturally rise at night, and of Oxalis purpurea, which naturally sink at night (the plants growing out of doors), and looked at them early on several successive mornings after they had assumed their diurnal positions. The difference in the amount of dew on the pinned-open leaflets and on those which had gone to sleep was generally conspicuous; the latter were sometimes absolutely dry, whilst the leaflets which had been hori- zontal were coated with large beads of dew. This shows how much cooler the leaflets freely exposed to the zenith must have become than those which stood almost vertically, either upwards or downwards, during the night. “ From the several cases above given, there can be no doubt that the position of the leaves at night affects their temperature through radiation to such a degree, that when exposed to a clear sky during a frost it isa question of life and death." It is the object of the present communication to show that many, perhaps the majority of, leaves as they develop on the expansion of buds in spring assume the same positions as leaves of those partieular plants whieh are remarkable for sleeping at night, and appear to exhibit a similar care, but in a more perfect way,in protecting the upper surfaces; and the inference to be drawn from all the phenomena presented by young growing leaves, as it seems to me, is the same, namely, to shield their upper sur- faces especially from injury caused by radiation, viz. chill and desiccation. Moreover, it is not an unusual thing for the young leaves only to be subject to hypnotism, the older ones ceasing to rise and fall by day and night: thus Mr. Darwin mentions Me- lilotus as an instance; and I find it is the same with the French Bean. Lastly, experiments made to test this theory are decidedly corroborative. Vernation, as explained and illustrated in text. books on Botany, is solely concerned with the various methods by which the rudimentary leaves are folded up in the buds of different plants; and the internal wooliness, as well as external resinous matters &c., are usually regarded as being safeguards against injury to the buds within from the severity of winter weather. Beyond that surmise, I am not aware that any attempt has ever been made to offer any rationale of the processes of verna- * pp. 293, 294. 626 REV. G. HENSLOW ON VERNATION AND FOLIAGE tion; nor has any one noticed the special behaviour of deve- loping leaves and stipules, as well as of the axes and petioles which support them. If, however, we keep in view the two assumed objects—first, the securing an erect or dependent position of the blades so as to place their surfaces in a plane at right angles to the surface of the earth ; and, secondly, the pro- tection of the upper surfaces,—it will be seen how various are the efforts of nature to secure these two ends during the period of development, and while the young leaves and shoots are succulent and delicate, such being only too readily and often severely injured by the cool nights, sharp frosts, and cutting winds of spring, at the very time when the buds are expanding. The various methods of protection are more perfect than in sleep- ing plants, inasmuch as the young leaves are more delicate than when adult. Testing the effects of desiccation by the heat of the sun, I found that when young clover-leaves naturally conduplicate were forcibly spread out with the upper surfaces exposed to the sun for twenty minutes, the ed ges soon curled in wards, and they lost 37:2 per cent. of weight. Those spread out with their under surfaces exposed remained flat, but lost 43:2 per cent. of weight in the same time, 7. e. rather more than the preceding. In all experiments with clover-leaves exposed at night, I found the dif- ferences were not so pronounced as in many other plants. In observing the effects of frost upon delicate leaves one sees that they are especially injured along the margins; and, more- over, where so affected, they are more or less dried up. The first fact gives a significance to the revolute and involute kinds of vernation, in which the margins are rolled outwards and inwards respectively. This led me to suspect that whenever leaves suffered from radiation, not only was there a reduction of temperature, but a loss of moisture would seem to accom- pany the loss of heat. A similar loss of moisture would occur by desiecation, in consequence of cool dry winds, aud must be equally guarded against. If this were so, then the balance would indicate the loss. With this object in view, I weighed a number of undeveloped leaves together of several plants, and made two groups of each, selecting leaves as nearly like as possible for each group of the same plant. I then, following Mr. Darwin's method, fixed all of one group with their upper surfaces exposed upwards with card-clips, such as are used by entomologists for spreading out the wings of butterflies; the other group of leaves AS PROTECTIVE AGAINST RADIATION. 627 werelaid on cork with their leaflets naturally conduplieate or otherwise protected as in nature. In the morning I weighed each group again, and reduced their losses to percentagesof the original weights of each group respectively. I found that there was in every case a generally much greater loss in the case of the leaves artificially exposed than with the others, as will be seen from the following examples :— Loss p. c. Loss p. c. Laurustinus, exposed naturally ............... 877. Spread out, 187 Portugal Laurel ,, none 84 » 12:6 Lime » Qo eene 45 » 214 Laburnum » J) eH 14:8 » 33:0 Ash » fy eee 23 » 118 Rose » »»o Henne 10:6 » 176 Periwinkle » "E 50 » 12:05 Clover » »o Hee 244 » 29:3 Walnut » 187 » 246 Very few nights of the last month (May) were frosty, so that several attempts to carry out experiments were unsatisfactory, as 80 many nights were cloudy, windy, and stormy, and ill-suited for radiation ; yet on one occasion, when about three degrees of frost occurred, the Walnut, Lime, and Laburnum leaves when spread out showed to the eye unmistakable injury ; whereas these three kinds of leaves, placed as they occur on the trees, did not appear to be hurt. Other leaves treated in the same way and subjected to the same slight frost did not appear to have suffered; but I had not at that time thought of weighing them. I will now add a selected series to illustrate the various methods adopted whereby the young and developing leaves are protected from injury by radiation. E I. SımPLE Leaves. (1) Opposite: (i) erect.— This position is assumed in all cases of opposite leaves which I have had an opportunity of examining. The leaves face one another with their upper surfaces more or less in close contact, concealing the bud between them. In shrubby Veronicas and Hypericums and Periwinkle the leaves are almost, as it were, glued together, so firmly do they cohere all round the margins. In others, such as Aucuba and Laurustinus, they are more slightly adjusted. The only exceptions that I have met with were Snowberry, Weigela, Honeysuckle, and the leaflets of the Elder, all being of the same family, Caprifoliacee *. The pairs of leaves had their edges in- * I first discovered this difference in Weigela, and that led me to examine the others, which revealed the fact that this peculiar form of vernation is an ordinal character. 628 REV. G. HENSLOW ON VERNATION AND FOLIAGE volute, and one of each pair slightly embraced the opposite leaf, and so wrapped up the bud within it. (ii) pendulous. Y have not met with any iustance of opposite leaves being pendulous in the young state. (2) Alternate: (i) erect.— Good examples may be seen in the Common and Portugal Laurels. As the upper surfaces of different leaves cannot be in contact, issuing from separate nodes and at differ- ent elevations, each leaf is conduplicate, 7. e. the two halves of the blade are folded together and thereby press their upper surfaces of the halves against one another ; indeed so tightly is this done, that it is not at all easy, in any instances, to sepa- rate them. It may be noted here that the condu- plicate vernation is an extremely common one both for simple and compound leaves, and the significance of it will be now very apparent. It is sometimes further complicated by having the surfaces plicate, as in the Vine, Beech, Maple, Currant, Rasp- berry, &c., conditions which are probably additional safeguards against radiation. As other instances of the erect position may be mentioned Rhododendron, the leaves of which, as of the Dock and Primrose, are revolute; but those of the Violet, Pear, and Michaelmas Daisy are also erect but involute, while the Cherry has them convolute. In all these the undeveloped leaf is cylindrical or an elongated cone, erect, and offers no extent of surface, while the margins, the most sensitive parts, are specially protected. (ii) pendulous. The Lime, Hazel, and Ampelopsis Veitchii are good examples. In the case of the Lime, as soon as the bud ex- pands and escapes from the winter (stipular) scales, the inner stipules develop considerably: those on the upper side are COU- cave and ovoid and cover the upturned edges of the conduplicate leaves, whieh at once take a position in a vertical plane; the stipules at the sides elongate much more than the former, fur- nishing some lateral protection to the whole bud, which now curves strongly downwards, and somewhat resembles a mussel in shape (fig. 2). As the bud continues to develop, the branch becomes more and more strongly curved downwards, so that the leaves are held vertically (figs.3 and 4) ; and as the lower and older ones increase in size, they assume a horizontal position and undertake Portugal Laurel. NENNEN RB T AS PROTECTIVE AGAINST RADIATION. 629 to protect the younger ones, which are concealed beneath them. Thus the protecting care is handed on to each leaf as it arrives at maturity, until the whole series are developed and the branch and leaves become horizontal. The tile-like arrangement of the uppermost stipules and subse- quently of the leaves themselves, thus protecting the edges of the Fig. 3. The Lime in different stages of development. vertically placed leaves beneath them, reminds one of a very similar method of protection in Trifolium repens when asleep. In this plant the two basal leaflets rotate so as to bring their upper surfaces in contact and their blades vertical, while the ter- minal leaflet revolves through 180? and comes down upon the upper edges like an arched roof above them * (figs. 5 and 6). Fig. 5. Fig. 6. Trifolium repens. . Fig. 5. Leaf during day time. Fig. 6. Asleep, during night-time. After Darwin. In the case of the Hazel the process is much the same, but with Ampelopsis Veitchii the blade spreads out at once, however young, and is not conduplicate; but as it hangs vertically from its very birth to its fall, it does not require any further protection beyond what the leaves above it happen to supply by overhanging it. The branches of this species cling so tightly to the wall that very possibly a good deal of heat is radiated from the wall itself on to * Darwin, Movem. of Plants, p. 349, fig. 141. LINN. JOURN.—BOTANTY, VOT. XXI. 3A 630 REY. G. HENSLOW ON VERNATION AND FOLIAGE the back, i. e. the underside of the leaf. The Beech differs from the Lime in having its young leaves dependent only and not pro- tected. They are feebly conduplicate, but with a plicate surface, and the stipules do not form any protection; and, contrary to what takes place in the Lime, the older, basal leaves, instead of covering the younger ones, hang below them, the slender shoot curving at the apex so that the terminal and younger leaves have a tendency to be vertieal, but not to so pronounced an extent as in the Lime. It may be noticed that the Beech does not open its buds till a later period than the Lime, and at a time when spring frosis are nearly over. The Spanish Chestnut agrees closely with the Beech, the leaves being at first subconduplicate, and similarly plicate, while each leaf on expansion is spread out below the terminal shoot. Both trees belong to the same section of the Cupulifere. . II. Compounp Leaves. (1) Opposite: (i) erect.—The Ash and the Elder are illustrations. As with simple leaves, the leaf- lets of compound leaves are almost invariably conduplicate ; the Elder, however, as stated above, having the margins of the leaflets involute as well. They all stand erect at first and only gradually assume the horizontal position on their complete development. The leaflets of the Ash are at first similarly clustered together, but simply conduplicate. With Weeping Ashes, in order to place the young leaves erect, the petiole is obliged to make a very strong curve upwards, not necessitated in the ordinary form of the tree. (ii) pendulous. I have not met with an instance of opposite compound leaves being pendulous in the young state. (2) Alternate: (i) erect.—Of compound leaves there are the two types, digitate and pinnate. Of the former, Lupin will illustrate this position. A specimen with white flowers had the leaflets all conduplicate at first and erect, forming a small but dense mass. Of the pinnate type with erect leaflets many instances might be mentioned, as, e. g., Goutweed, Chervil, Sumac, Rasp- berry , and Rose. As a rule, besides being conduplicate and, 1n some cases, plicate in addition, the leaflets are all pressed together laterally, thereby affording a certain amount of mutual protection. This is well seen in the Rose. In the case of the Rose the developing bud is protected by the adnate stipules (fig. 7), and the young leaf, when it merges, is at first erect (as in the Pea, figs. 8 and 9). The leaflets are condu- plicate and all five pressed together laterally ; for at this stage AS PROTECTIVE AGAINST RADIATION. 631 the petiole has not elongated sufficiently to allow of the pairs of leaflets being separated. Asthe petiole grows the whole collection of leaflets becomes more pendulous until they expand, and the Fig. 7. Rose. Younger leaf erect; older with petiole curved and half-developed leaflets. Fig. 8. Pea. Leaf emerging from between the stipules; the leaflets conduplicate. Fig. 9. One stipule removed to show the erect bud within and proteeted by the stipules. leaf ultimately assumes the horizontal position. In the Garden Bean the leaves are conduplicate ; but the margins are involute as well, so that the leaflets resemble so many quills. (ii) pendulous. Of the twotypes the following may be selected:— The French Bean, Wood-Sorrel (Oxalis), Clover, and Laburnum as being ternate, and the Horse Chestnut and Virginian Creeper as being digitate ; while the Walnut will exemplify the pinnate type. The French Bean (Phaseolus vulgaris) bears at first a pair of unifoliate leaves; but those which sueceed them are ternate. While very minute in size (one half to three quarters of an inch in length) the leaf is horizontal, but the leaflets are conduplicate with their edges uppermost. When a little larger (say, one and a quarter inches) the petiole bends down angularly, and the leaflets are in vertical planes, holding identically the same positions as when adult and asleep. As they increase in size, the leaflets rise up and take a horizontal position. They now become subject to hypnotism, falling at night and rising again by day. The Wood-Sorrel (fig. 12), Clover (fig. 11), and Laburnum (fig, 19) all agree in having their three leaflets conduplicate, 632 REV. d. HENSLOW ON VERNATION AND FOLIAGE pressed closely together laterally and suspended vertically. The petiole is suberect in the Laburnum and Clover, but strongly Fig. 10. Fig. 11. Fig. 12. Fig. 10. Laburnum. Fig. 11. Clover. Fig. 12. Woodsorrel: a, nat. size; b, magnified; c, vertical aspect, in diagram as seen asleep. curved at the apex in the Wood-Sorrel, so that the minute leaf is under the concave end (fig. 12, a and b). When the leaf of the Oxalis is asleep, the leaflets fall vertically and bring their undersides in contact (fig. 2, c), but never resume the conduplicate condition again. The sleeping condition of Clover has been described above ; but here, as in allother instances, conduplication once lost is never resumed *. This seems to indicate that the danger from exposure in the very young state being much greater than in the adult, the pro- tection is correspondingly more perfect. In the Horse Chestnut (fig. 13) the digitate leaf has at first all the leaflets dependent and more or less covering one another. This is exactly similar to the condition of the leaflets of Lupinus pilosus when asleep, as described by Mr. Darwin t (fig. 14). It Lupinus pilosus. “Leaf, seen la- terally, asleep at night-time. Walnut. After Darwin. * This is due to the fact, at least in Clover, that a thick layer of ehlorophyl- laceous tissue is developed over the fibro-vascular bundle of the midrib, thereby preventing the two halves closing again. t L. c. p. 941, fig. 137. Horse Chestnut. AS PROTECTIVE AGAINST RADIATION. 633 may be added that the leaflets of Lupin are at first clustered together and conduplicate. The Virginian Creeper has its five leaflets at first conduplicate and suberect; they gradually curve over and spread themselves vertically like a star, not unlike the method adopted by Lupinus pubescens, as described by Mr. Darwin. The petiole of the Walnut-leaf (fig. 15 *), on emerging from the bud (the scales of which, like those of the Ash, are petiolar), eurves strongly downwards, so that the leaflets, which are conduplicate, stand in a vertical plane. As the basal ones expand they still remain with their surfaces vertical, and it is not until they are approaching maturity that the petiole rises up and the leaflets spread themselves out horizontally. Besides vernation, conduplication, and the subsequent vertical position of leaves and leaflets, as caleulated to protect them from the evil effects of radiation, hairs and tomentum &c. must not be forgotten as being bad conductors of heat, and therefore very important aids to protect the organs clothed with them. Foliar organs and axes, when young, are often very hairy, silky, or woolly, as the case may be, which in older states become more or less glabrous, either by the hairs becoming more sparsely scattered by epidermal growth, or by vanishing altogether. Similarly the stellate pubescence or woolly clothing, which is not an unfrequent character of the young condition, often disappears as soon as the surfaces thus protected are sufficiently advanced to require such additional aids no longer. As examples may be mentioned the young shoots of Poplar, Apple, Ivy, &c., while the leaves of Colts- foot are at first densely villous, but soon lose the cottony webs from the upper surface as they become adult. CowcLusrow.—The examples given in this paper could, of course, be multiplied indefinitely ; but enough seems to have been stated to justify a belief in the general accuracy of the deduction that vernation, conduplication, the various positions taken up by developing leaves, &c., all conspire to protect them from the evil effects of radiation. * In the printing of p. 632 the woodcuts representing Walnut and Labur- num petiole respectively have accidentally been transposed. LINN. JOURN.—BOTANY, VOL. XXI. 3B 634 MR. W. JOSHUA ON BURMESE DESMIDIEZ. Burmese Desmidiex, with Descriptions of new Spe occurring in the neighbourhood of Rangoon. By W. Josuva, F.LS. [Read 5th February, 1885.] (Prarzs XXII.-XXV.) Dortna the winter of 1883-84 I received, through the kindness of Dr. Robert Romanis, F.L.S., of Rangoon, some collections of freshwater Alge, taken from the leaves of Pistia Stratiotes. On preparing these for microscopic use, and on frequent subsequent examinations, they appeared to contain so many Desmidiew entirely new to science that I thought a detailed enumeration would be necessary. The best collection was made after the rainy season from a pond, or old tank, in the vicinity of Rangoon, situated on the glacis of the fort surrounding the Pagoda, to the right of the south approach. This tank is only of small dimensions, being not more than 30 ft. square and about 20 deep. It appears to be of considerable antiquity, having been placed there in Burmese times, and is said never to be disturbed by cleaning out, 8 condition conducive to the multiplication of these interesting forms. The surrounding soil is a friable sandstone in beds of shale, producing alum. The locality lies at the eastern estuary of the Irrawady, about 26 miles from the sea, and on elevated ground, It may be observed that many of the species are considerably smaller than their northern prototypes; still some of the Cos- marie: and Docidieæ are of extraordinary size and beauty. I am not aware that any special list of Burman Desmids has hitherto been recorded, save a few odd species mentioned by Dr. G. Zeller, in Journ. Asiat. Soc. of Bengal, vol. xlii. (pt. ii. Nat. Hist. i., iv. 1873), the Desmid portion of the Algæ collected in Arracan and British Burma by Mr. S. Kurz (J. c.) being there stated to be in the hands of Mr. W. Archer of Dublin. The only lengthened account of Eastern Desmids that I am acquainted with, is that by Dr. G. C. Wallich*, from a district of Lowe? Bengal, situated about 120 miles north-west of Calcutta, 1855; and a few notes of species collected by Herr Grunow, of Berndorf., Vienna, from the island of Banka, Singapore, 1855. * Ann. & Mag. Nat. Hist. (3) v. pp. 184 and 273, pls. 7, 8, 13, and 14. MR. W. JOSHUA ON BURMESE DESMIDIER. 635 As might be expected, some of Wallich's and Grunow's species exist here, as also others of Tropical America; while some bear a strikingresemblance toforms found by Dr. Schweinfurth in Central Africa, and described by Prof. Cohn in * Desmidiew Bongoenses,' one especially noticeable, Cosmarium Euastron, having the re- markable cuneiform-stellate markings seen in his C. tholiforme. Cosmarium Pardalis, Closterium Isidis, and Pleurotenium nodu- losum, De Bary, f. tenuior, also closely resemble some of the Rangoon forms. Docidium, Euastrum, and Micrasterias are well represented, as may be expected in an exotic list. Of the 186 species and varieties, in 16 genera, collected by Dr. Romanis, 100 have their representatives in Europe; and of the 7 Docidies, 4 are new. The drawings have been made by a vertical camera, and en- larged from 200 to 900 diam. The list, though a long one, is, I feel sure, only tentative; for if a further search be made in other parts of the empire, doubtless many more interesting novelties will yet be found. DESMIDIE®, Kütz. 1. Desmipium, 4g. D. (AProaoNvuM) BAILEYII, f. TETRAGONA, Delponte. D. Scuwarrzi, dg. Conjugated: zygospores. D. CYLINDRICUM, Grev. D. quapratum, Nordst. Conjugated zygospores. (Pl. XXII. fig. 3.) D. QUADRANGULATUM, Ralfs. 2. ONYCHONEMA. O. LÆVE, Nordst. Conjugated: zygospores. 3. BAMBUSINA, Kitz. B. BrREBISSONII, Kütz. 4. SPHÆROZOSMA, Corda. S. PULCHRUM, Bailey, P. TRILOBUM, n. var. figs. 1 & 2.) S. cellulis triplo latis quam longis, breviter utrinque convexis forma irregularibus, a vertice conspectis 3-lobatis, lobis basi 322 (Plate XXII. 636 MR. W. JOSHUA ON BURMESE DESMIDIER. valde elongatis: isthmo nonnihil elongato. Longit. 28 p, latit. 40 p, crassit. 45 p: isthmi latit. 10 p. Cells three times as wide as long, on each aide slightly convex, connecting isthmus rather elongated. Cells rather irregular in shape, in end view three-lobed ; base of lobes much elongated. Length 28 u; width 40 p; thickness 45 p; width of isthmus 10 p. S. FILIFORME, Ehrenb. SPrmROozOsMA sp.? Probably new. S. EXCAVATUM, Ralfs. S. excavatum, B. WALLICHI, Jacobs. 5. GonatozyGon, De Bary. G. AsPERUM, Breéb. 6. MICRASTERIAS, Ág. M. MAHABULESHWARENSIS, Hobson. Very fine; abundant. M. truncata, Corda, v. CRENATA. M. rotata, Grev. M. PINNATIFIDA, Kitz. M. rrorica, Nordst. M. APICULATA, Menegh. n. f. (Plate XXII. fig. 13.) AM. prominentiis centralibus in quavis semicellula spinis robustis circumdatis. A variety with central protuberances in each semicell circled with strong spines. M. cnUXx-MELITENSIS (Khrenb.), Ralfs, var.: see Delponte, pl. iv. 7, 12. (Plate XXII. figs. 10 & 11.) M. Arama, Wallich (l. c. p. 279, pl. xiii. fig. 11). Entirely identical with the Bengal species, and abundant. M. 1ncisa, Kütz., var D. Wallich. M. rorracka, Bailey. M.rux,n.sp. (Plate XXII. fig. 12.) . M. magna, orbicularis, levis: semicellulis distincte 4-lobatis, lobis subdivisis, valde attenuatis, basi leviter inflatis, apicibus profunde furcatis, radios rectos sistentibus: lobis terminalibus anguste cuneatis, divergentibus, attenuatis. Longit. (sine MR. W. JOSHUA ON BURMESE DESMIDIER. 637 brachio) 180 p, latit. (cum spinis) 217 p: isthmi latit. 25 p: expansio lobi polaris 52 y. Large, orbicular, smooth; semicells distinctly 4-lobed; lobes subdivided ; slightly inflated at base; much attenuated ; deeply furcate at ends, forming straight rays; end-lobes narrow- cuneate, divergent, and attenuated. Differing from M. radiosa, Ralfs, by its larger size and linear form of subdivisions. Length, without arm, 180; breadth, with sp., 217 u; width of isthmus 25 u; spread of polar lobes 52 y. M. cERATOFERA, nob., Journal of Botany, vol. xxiii. p. 34, tab. 254. figs. 4a, 4b. M. EvasrROIDES, n. sp. (Plate XXII. fig. 14.) M. mediocris, subquadrata; semicellulis 5-lobatis, lobo termi- nali truncato, profunde constrictis, marginibus denticulatis superficie spinis brevibus ornatis. Longit. 103 y, latit. 90 y. Medium; subquadrate ; semicells 5-lobed, deeply constricted, denticulate at edges, and furnished with short spines on surface; end-lobe truncate; length 103 pu, breadth 904. In appearance resembling some of the forms of Euastrum verrucosum; but pro- bably a new Micrasterias allied to M. mahabuleshwarensis, Hobson, and M. Wallichi, Grun. 7. EvasTRUM, Ehrenb. . AUSALUM, Ralfs. . AUSALUM, Var. AMPLIATUM. - SINUOSUM, Lenorm. . DIDELTA, Ralfs. . BINALE, Ralfs, n. f. crassum. (Plate XXIII. figs. 11, 12, ) . TURGIDUM, Grun. . SUBSTELLATUM, JVordst. GEMMATUM, Bréb., MONONCYOLUM, Wordst. . DECEDENS, Reinsch. . CUNEATUM, Jenner. sPINULOSUM. Delponte, f. 1nERMiUS, Nordst. . OBLONGUM, Grev., f. SCROBICULATUM, Nordst. . ELEGANS, Bréb. . SUBLOBATUM, Bréb. . HYPOCHONDRUM, Wordst. SHEE pd Bd bd bd bx Bd bd C2 Dd Ed bd [d bd 638 MR. W. JOSHUA ON BURMESE DESMIDIEJE. E. RETRORSUM, n. sp. (Plate XXIII. figs. 14 & 15.) E. mediocre, diametro plus quam duplo longius, prominentiis marginalibus instructum ; semicellulis obscure 3-lobatis, angulis lateralibus produetis, versus apicem recurvis, centro inflatione magna; cytiodermate crasso: isthmo vix attingente j partem latitudinis: constrictione lineari. Longit. 75 p, latit. apud punctum constrictionis 35 y. Medium size, rather more than twice as long as wide ; semi- cells obscurely 3-lobed; lateral angles produced, and pointing backwards towards apex, with large central inflation ; cytioderm thick ; isthmus scarcely one fourth of width ; constriction linear; cells ornate with marginal protuberances. Length 75 p; breadth at point of constriction 35 p. E. opEsuM, n. sp. (Plate XXIII. figs. 19 & 20.) E. parvum, nonnihil longius quam latum: semicellulis 8- lobatis, centro non inflatis ; lobis globose rotundatis, lobo polari inciso, a latere conspectis ovatis, apicibus truncatis: membrano pallide carneo, aliquando indistincte granulato. Longit. 50-80 p, latit. 30-35 : isthmi latit. 9-15 p. Small, rather larger than broad; semicells 3-lobed; each lobe globularly rounded; polar lobe with incision; side view oval, with truncated ends ; no central inflation of semicells ; membrane a faint flesh-colour, sometimes indistinctly granulated. Length 50-80 u; breadth 35-30 p; width of isthmus 9:15 p. E. FLAMMEUM, n. sp. (Plate XXIII. figs. 3, 4, & 5.) E. mediocre, nonnihil longius quam latum: semicellulis 4- lobatis, profunde constrictis, prominentia magna prope basin cujusve lateris, lobis terminalibus truncatis, dilatatis, spinum robustum magnum apud angulos singulos ferentibus; sinu apicali profundo, magno, versus centrum nonnihil latiore, lobis lateralibus tres spinos magnos rectangulares ostendentibus. Longit. (sine spinis) 48 p, latit. (sine spinis) 28 p : constrictionis latit. 7 p: sinus polaris longit. 12 u, basi latit. 4 p. Medium size, rather longer than broad; semicells 4-lobed ; constriction deep, a large protuberance on each side near base ; end-lobes truncate, dilated, bearing at each angle a large stout spine; the apical notch deep and large, somewhat widened to- wards the centre, with three large rectangular spines at each side of lateral lobes. Length 48, without spines; width, without ae we l MR. W. JOSHUA ON BURMESE DESMIDIE Æ. 639 spines, 28u; width of constriction 7; length of polar notch 12 u, and width at base 4 p. E. conarrorpEs, n. sp. (Plate XXIII. fig. 10.) E. parvum nonnihil longius quam latum: semicellulis 4- lobatis, anguste constrietis, prominentiis quatuor truncatis in quavis semicellula; angulis latis hirsutis, truncato-elongatis : membrano pallide roseo. Longit. 40 p, latit. 30 p: isthmi latit. 13 y. Small, rather longer than broad; semicells 4-lobed; con- striction narrow, with wide angle; four truncate protuberances in each semicell; angles rough and truncate-elongate, giving the whole plant a rough and coral-like appearance; membrane pale pink. Length 40u; breadth 30 p; width of isthmus 13 y. E. rRUNCATUM, n. sp. (Plate XXIII. figs. 6 & 7.) E. mediocre, subquadrangulare, fere duplo longius quam latum : semicellulis 3-lobatis, marginibus levibus crenatis, lobis lineari- divisis, basi globose inflatis, lobo polari attenuato, leviter de- presso: prominentiis centralibus magnis, duabus in quavis semi- cellula: isthmo lato, sinu lineari stricto. Longit 65 p, latit. ad basin 37 u: lobi polaris latit. 24: isthmi latit. 16 x. Medium size, subquadrangular, about twice as long as broad ; semicells 3-lobed, with smooth crenate edges; polar lobe atte- nuate, with slight depression ; central protuberances large, two in each semicell; isthmus wide, with close linear notch ; divisions of lobes linear; basal portion a circular inflation. Length 65 p; breadth at base 37 u; breadth of polar lobe 24 p ; width of isthmus 16 p. E. arrenvatum, Wolle. E. sERRATUM, n. sp. (Plate XXIII. figs. 1 & 2.) E. vix duplo longius quam latum, profunde constrictum ; semicellulis 6-lobatis, prominentiis duabus magnis in quavis cellula, late a puncto divisionis remotis, lobis lineari-divisis, ad basin globosis, angulis acutis, interne conspectis distincte ser- ratis, lobis terminalibus incisis, angulis exterioribus divergen- tibus, attenuatis. Longit. 58 y, latit. 35 p: isthmi latit. 8 p. Scarcely twice as long as broad ; constriction deep ; semicells 6-lobed ; angles sharp, having on inside view a decided serrate appearance, with two large protuberances in each semicell, widely removed from poiut of division ; isthmus nearly quarter of width ; eud-lobes incised, with exterior angles divergent and attenuated ; 640 MR. W. JOSHUA ON BURMESE DESMIDIEE. divisions of lobes linear, circular at base. Length 58, width 35 p; width of isthmus 8 y. E. EXILE, p. sp. (Plate XXIII. figs. 16, 17, & 18.) E. parvum, quadratum, diametro plus duplo longius: semi- cellulis 1-lobatis, constrictionibus tribus lateralibus, et promi- nentia uniea conica, a vertice conspectis ovatis: apicibus trun- eatis; angulis polaribus spinis longis robustioribus projectis: isthmo divisione lineari. Longit. (sine spinis) 30 p, latit. ad punctum divisionis 18 u: isthmo vix 3 latitudinis. Small, quadrate, rather more than twice as long as broad; semicells 1-lobed, having three lateral constrictions and one conical protuberance; apex truncate; polar angles with long, rather stout spinous projections; end view ovate; isthmus scarcely one third of width; division linear, Length without spines 30; width at point of division 13 p. E. DIVERGENS, n. sp. (Plate XXIII. figs. 8 & 9.) E. fere tam longum quam latum, centro profunde constrictum, angulo lineari valde dilatato: semicellulis profunde 3-lobatis, granulis magnis, supra centri inflationem concentrice dispositis; lobis lateralibus semifusiformibus, apicibus truncatis, parte supe- riori granulis margaritaceis paulum elongatis, seriebus yertica- libus dispositis, minute ornatis; lobo polari quadrato, truncato (latitudine equanti tertiam partem diametri transversi cellule). Longit. 48 p, latit. 48 p: lobi polaris latit. 15 p: isthmi latit. 10 p. About as long as wide, in the middle deeply constricted ; linear angle much widened; semicells deeply 3-lobed ; granules large above central inflation, in concentrie arrangement ; lateral lobes semifusiform; ends truncate; the upper part of lobes minutely ornate with pearly granules a little elongated, placed in vertical rows ; polar lobe quadrate, end truncate. Some- what similar at first sight to Æ. breviceps, Nord., and E. bellum, Nord., also .E. platycerum, Reinsch ; but differing in shape of basal lobes. Length=breadth 48,4; width of polar lobe 154; breadth of isthmus 10 p. E. onBicULAnE, Wallich (l. c. p. 282, plate xiv. figs. 8-11). 8. STAURASTRUM, Meyen. S. BIFIDUM, Ralfs, var. (Plate XXIV. figs. 8& 9.) S. PROBOSCIDEUM, Breéb., f. JAVANICA, Nordst. N. MARGARITACEUM, Ehrenb., var. nirtum, Nordst. MR. W. JOSHUA ON BURMESE DESMIDIEX. 641 S. LEPTOCLADUM, JVordst., var. 3. cuisrATUM, Wolle. S. GRACILE, Ralfs. Several forms. S. GRACILE, B. cURTUM, JVordst. S. VESTITUM, Ralfs. S. Avicuna, Bréb. S. BRASILIENSE, JVordst. S. LEPTOCLADUM, JVordst. (9-gona.) S. SALTANS, n. sp. (Plate XXIII. fig. 21.) S. majus; semicellulis quadrangularibus dorso late curvatis, lateribus apicis spino magno expanso utrinque instructis, basi truncatis, profunde incisis, apud marginem granulis globosis minutis instructis: angulis superioribus brachii valde incurvis, apicibus bifurcatis; spinis valde robustis, superiori duplo et ultra majore; brachio granulis rotundatis serie regulari dispo- sitis, instructo. Longit. 45 p, latit. 93 p, cum spina 150 p. Rather large ; semicells quadrangular ; the back widely curved, furnished at each side of apex with a large spreading spine; base truncate, with deep incision, furnished at edge with small globular granules; upper angles of arm much incurved, aud with bifureate apices; spines very strong, that on the upper side more than double the length of the other; arm having a regular series of : circular granules. It differs from S. leptocladum, Nordst., in the absence of the lateral crenation and subglobose inflation of basal lobe. Length 45; width to end of spine 93 p, with spine 150 p. M S. HORRESCENS,n. sp. (Plate XXIV. fig. 3.) S. magnum, fere tam longum quam latum, subovatum, pro- funde constrictum, isthmo latitudine, spinis exclusis, equanti 4 partem diametris transversi: membrano toto spinis robustis, truncatis, in serie regulari dispositis, obtecto. Longit. sine spinis 73 p, cum spinis 85 p, latit. sine spinis 70 p, cum spinis 80 u: isthmi latit. 14 p. Large, about as long as broad, subovate, deeply constricted ; breadth of isthmus one fifth part of transverse diameter, exclusive of spines; the whole membrane covered in regular disposition with strong truncated spines. Differing from S. Pringsheimii, Reinsch, the semicell of which forms nearly a perfect ellipse, and has no inflation. of angle at point of juncture. Length without spines 73 u, with spines 85; width without spines 70 y, with spines 80 p; width of isthmus 14 p. 642 MR. W. JOSHUA ON BURMESE DESMIDTEZ. S. sexANGUIARE, Bulnk., n. var. (Plate XXIII. fig. 24.) S. BIFURCUM, n. sp. (Plate XXIII. figs. 25-28.) l S. parvum, forma variabili, brachiis duobus aut tribus in- structum; cytiodermate granulato, marginibus serratis: semi- cellulis a fronte conspectis cuneatis; basi lata ; ad latera singula lobi apicalis spino majore instructis; lobo basali inflatione sub- globosa. Longit. 23-48 y, latit. 35-73 p. Small, variable in outline, sometimes with two, sometimes three arms; cytioderm granulate, margins serrate ; semicells in front view cuneate, with broad base, furnished at each side of apical lobe with a larger spine; basal lobe with subglobose inflation. Its nearest relation is with S. incisum, Wolle. Length 23-48 p; breadth 35-73 u. S. CUSPIDATUM, f). DIVERGENS, JVordst. S. LONGISPINUM, Arch. S. cYATHODES, n. sp. (Plate XXIII. figs. 22 & 23.) S. mediocre: semicellulis cyathiformibus, triangularibus, brachiis longe productis, profunde fissis et trifurcatis, medio profunde constrictis, puncto junctionis minuto, a vertice con- spectis, brachiis minute granulatis, granulis seriebus tribus dispositis: membrano haud omnino lævi. Longit. sine spinis 30 p, latit. sine brachiis 22 p: longit. ab apicibus 64-74 p, latit. ab apicibus 47-50 u: ? connexionis isthmi latit. 3 p. Medium size: semicells cyathiform, triangular, with long pro- duced arms, deeply cleft and trifurcate ; deeply constricted at middle, point of juncture very small; in end view with minute granulations on arms arranged in three rows; membrane not perfectly smooth. Length without spines 30; breadth without arms 22 u; length from apices 645, 745; breadth from apices 47 u, 504; width of connection 3 p. S. BACILLARE, Bréb. . DEJECTUM, Bréb., B. connatum, Lund. . INCONSPICUUM, Nordst. S. PUNCTULATUM, Bréb. S. cYRTOCERUM, Bréb. S. DILATATUM, Ehrenb. S. STRIOLATUM, Naeg. S. rELIFERUM, Ralfs. S FURCULOSTELLATUM, Reinsch. MR. W. JOSHUA ON BURMESE DESMIDIEZ. 643 S. GRANULATUM, Reinsch. (Plate XXIV. figs. 4 & 5.) S. oRBICULARE, Ralfs. S. PoLYMORPHUM, Bréb. S. FurcatuM, .Ehrb., Bréb., B. sENARIUM, Ehrenb., Nordst. (Plate XXIV. figs. 6 & 7.) S. LEPTACANTHA, f. 9-gona, Nordst. S. LEPTODERMUM, Lundell. S. MINUSCULUM. S. PLATYCERUM, n. sp. (Plate XXIV. figs. 1 & 2.) S. eytiodermate levi, |pellucido: cellulis a fronte conspectis ovatis, brachiis triplo longioribus quam latit. corporis ipsius ; apicibus divergentibus, tricuspidatis, marginibus hirsutis, promi- nentiis spinosis regulariter dispositis; radiis 4, 5, raro 6. Cel- lul; latit. ad basin 18 u, brachii longit. cum spinis 33-504: ex- pansio brachiorum:83-160 p. Cytioderm smooth and transparent ; cells in front view ovate ; arms 3 times as long as breadth of body ; diverging apices tricus- pidate; margins rough with regularly placed spinous protu- berances; rays four, five, rarely six. It comes nearest S. aspi- nosum, Wolle, but has not the slender rectangular spines, and is a larger species. Width of cell at base 18; length of arm, including spines, 33 u, 50 u ; spread of arms 83 p, 160 p. 9. XaANTHIDIUM, Ehrenb. . ARMATUM, Bréb. - ACANTHOPHORUM, JVordst. . ANTILOPEUM, Kutz. - ANTILOPEUM, f. JAVANICA, JVordst. . ANTILOPEUM, fJ. TRIQUETRUM, Lund, f. BRASILIENSE, Vordst. . ANTILOPEUM, f. ANGULATUM, n. var. (Plate XXIV. fig. 16.) X. a fronte conspectum quadrangulare, minute punctatum, apicibus convexis ; cellulis plus duplo latis quam longis, profunde constrictis, angulo lineari, nec inflatis nec granulatis; apicibus? angularibus, apud punctos singulos spinam longam subulatam ferentibus. Latit. sine spinis 484, cum spinis 98 p: isthmi latit. 10 p. In front view quadrangular, with convex ends, minutely punc- tate; cells rather more than twice as wide as long, with deep con- striction, angle linear ; no inflation or granulation ; apices angular, P pd pd pd pd pd 644 MR. W. JOSHUA ON BURMESE DESMIDIER. bearing a long subulate spine at each point. Breadth without spines 48 » ; breadth with spines 98 u; width of isthmus 10 p. 10. ArtHropEesmus, Ehrenb. A. GIBBERULUS. A. Incus, Hass, Forms. (Plate XXIV. figs. 10, 11, & 12.) A. ocrocornis, Ehrenb. A. sUBULATUs, Nordst., n. var. GnaciLIS. (Plate XXIV. fig. 13.) A. mediocris, profunde constrictus, angulo lineari ; semicellulis utriusque anguli polaris spinis longis convergentibus ornatis, nonnihil longioribus quam latis, apicibus spinarum fere conni- ventibus: dorso leviter curvato, eum spinis uniformiter arcuato. Longit. 27 u, latit. sine spinis 35 u, cum spinis 65 p: constrictio 6-7 y. " Medium size; deeply constricted, with linear angle; semi- cells at each polar angle furnished with long converging spines ; the two apices of opposite spines nearly meeting ; rather longer than wide; back slightly bent, forming a uniform curve with the spines. Length 27; breadth without spines 35 p, with spines 65 u; constriction 6°7 p. A. ARCUATUS, n. sp. (Plate XXIV. fig. 14.) A, mediocris, profunde constrictus, angulo late hianti, non spinosus, fere tam longus quam latus: membrano levi aut leviter punetato : cellulis subovatis, dorso late rotundato : angulo utrius- que poli spinoso, spina valida, divergenti. Longit. 40 p, latit. sine spina 36 p, cum spina 70: constrictio 10 u. Medium size; deeply constrieted, with wide gaping angle; without spines; about as long as broad; membrane smooth or slightly punctate; cells subovate; back widely rounded, with à strong divergent spine at each polar angle. Length 40 p; breadth without spine 36 u, with spine 70; constriction 10 p. A. APICULATUS, n. sp. (Plate XXIV. fig. 15.) A. parvus, subtrapezoideus: profunde constrictus, angulo latiore: semicellulis lateribus rotundatis, dorso curvato, inflato: angulis polaribus spinis apiculatis verticalibus ornatis. Small, subtrapezoidal, rather longer than wide; semicells with rounded sides and curved inflated back ; polar angles furnished with apiculate spines, set vertically; constriction deep, with rather wide angle. Not common. No measurements made. MR. W. JOSHUA ON BURMESE DESMIDIER. 645 11. CosuanrUM, Corda. C. Evastron, n. sp. (Plate XXIV. figs. 30-34.) C. mediocre, ovatum, 1j-13-plo longius quam latum, apicibus non truncatis, profunde constrictum: strictura lineari aut an- guste euneata: semicellulis semirotundatis: angulis inferioribus fere rectis, apicibus ovatis: membrano crasso, dilute pallido- flavescente (colore intensiore in centro zonali cellule), papillis magnis, basi expansis, stellato-dispositis. Longit. 65-95 p, latit. 47—50 u: crassit. 27-33 p: isthmi latit. 15 p. Medium size, oval, 14 to 14 times longer than wide; ends not truncate, deeply constricted; stricture linear, or narrow wedge-shaped; semicells semicircular; lower angles nearly straight ; apices oval; membrane thick, very pale yellow, deepen- ing in zonal centre of cell; papille large, with spreading base, and showing a clearly defined quincunx arrangement with stellato- cuneiform rays in oblique light. Smaller and more globose than C. tholiforme, Cohn, to which it comes nearest. Length 65 p-90 p ; breadth 47-50 p; thickness 27-83; width of isthmus 15 p. . TURGIDUM, Bréb. . GRANATUM, Bréb. . DEBARYIT, Arch. . PULCHERRIMUM, Nordst. A small var. . CUCURBITA, Bréb. , CRENATUM, Ralfs. . CAPAX, n. sp. (Plate XXV. fig. 8.) C. magnum, cylindricum, duplo longum quam latum, medio vix constrictum, annulo marginali ad punctum junctionis : lamellis chlorophyllaceis multis, curvatis; granulis amylaceis passim di- spersis: membrano pallide fusco, regulariter minute punctato. Longit. 165 p, latit. 90 p. Large, cylindrical, twice as long as wide; middle scarcely con- stricted, with a marginal ring at point of juncture. The chloro- phyll-plates of contents many and curved, with starch-granules scattered throughout; membrane pale brown, regularly and minutely punctate. Length 165; width 90 p. C. OBSOLETUM, Hantzsch. C. SMOLANDICUM, Lund. C. Cucumis, Corda. C. Borrrtis, Menegh., var. INDICUM, n. var. (Plate XXIV. fig. 19.) C. semicellulis profunde constrictis, centro non inflatis, granulis Qaaananaaa 646 MR. W. JOSHUA ON BURMESE DESMIDIEZ. delicatis regulariter et symmetrice dispositis, spatio lucido ovato ad centrum cujusve cellule: isthmo latitudine inter j et } partem diametris transversi. Longit. 88 p, latit. 65 u: isthmi latit. 16 p. Semicells with fine granules regularly and symmetrically arranged, with a clear oval space in centre of each cell 20 u by 13 p; without central inflation, deeply constricted (width of isthmus between the fourth and fifth part of transverse diameter). Differing from C. Botrytis, B. mesoleium, Nordst., in the narrow isthmus and arrangement of granules. Length 884; breadth 65 u; width of isthmus 16 y. C. TETRAOPHTHALMUM, Kfz. C. amanum, Breb. C. Menecuinit, Bréb., and forms. C. PYRAMIDATUM, Bréb. A. variety with slight inflation in semicell. C. LATUM, Zréb. C. Pataneuta, Bred. C. coxsPERSUM, Ralfs. C. DriapEMA, n. sp. (Plate XXV. fig. 7.) C. magnum, fere 13-plo longius quam latum, cylindricum: semicellulis leviter constrictis, late marginatis, granulis magnis in seriebus equidistantibus dispositis, intersticibus minute punctatis. Longit. 75 p, latit. 50 p. Large, nearly 1j times longer than wide, cylindrical; semi- cells slightly constricted, with wide marginal band; granules large, symmetrically placed in equidistant rows, minutely punctate in interstices. Length 75 u; breadth 50 y. C. RECTANGULARE, Grun. . C. ezoBosuM, Bulnh. C. NORIMBERGENSE, Reinsch. C. ARMATUM, n. sp. (Plate XXIV. figs. 21-25.) C. parvum, subhexagonale, nonnihil longius quam latum, con- strictione lineari, recta: semicellulis trapezoideis, basi latis, apice truncatis, prominentiis obtusis 7-13, in serie interiore dispositis (ad basin poli multo majoribus); marginibus prominentiis validis conicis, 4-5 utroque latere fimbriatis, dorso nudo: cellulis multum inflatis, a latere conspectis sphericis. Longit. 35 p, latit. 25 p: isthmi lat. 83 y. Small, subhexagonal, rather longer than wide; median division a straight line; half-cells trapezoidal, with wide base furnished MR. W. JOSHUA ON BURMESE DESMIDIER. 647 with an inner row of blunt protuberances 7-13, much enlarged at base of polar end; end truncate, fringed at margin with strong conical projections 4-5 in each side, none on the back; cells considerably inflated, appearing spherical in side view. This species has some resemblance, especially fig. 4c and 4 d, with C. vosgesiacum, Lemaire; it stands also near C. Reinschii, Arch. Length 354; breadth 25 p; width of isthmus 83 p. C. puNcTULATUM, Bréb. C. cuNEATUM, n.sp. (Plate XXIV. figs. 17 & 18.) C. mediocre, constrictione profunda: cellulis ovato-cuneatis, fere tam longis quam latis; semicellulis acute cuneatis, dorso nudo, centro granulis 4 utroque latere dorsi, nec ad apices pro- vectis, marginibus extremis granulis 14-17 majoribus, rotundatis, seriatim dispositis ornatis ; angulis lateralibus a latere conspectis, semiacutis : isthmo latitudine circa 1 partem diametris transversi. Longit. 46 p, latit. 46 u : isthmi latit. 11 p. Medium size; cell ovato-cuneate, about as broad as long; semicells sharply euneate, with broad back, furnished at ex- treme edge with a series of 14-17 large circular granules, diminishing in size on each side as they approach the linear median line, and in many specimens showing a double row of granules above the centre; there is also a central series of granules, 4 on each side of back, not carried to the apex ; seen in side view, lateral angles semiacute ; constriction deep. It stands nearest to C. monomazum, Nordst. Breadth length 46 p ; breadth of isthmus 11 p. C. sPINosuM, n. sp. (Plate XXV. figs. 3 & 4.) C. mediocre, paulo longius quam latum, semiellipsoideum ; membrano interiore minute et regulariter punetato, serie unica aut duplici punctorum majorum nonnihil proeul a constrictione addita; semicellulis seriem spinarum circa 6 utroque latere (non ad apicem provectarum) ferentibus: cellulis a latere conspectis globosis: isthmo latitudine fere 4 diametri transversi: angulo sinu lato apud punctum junctionis. Longit. 51 p, latit. (sine spinis) 40 y. Medium size; rather longer than broad, semielliptical; inner membrane minutely and regularly punctate, with a single or double row of rather larger punctations, encircling at some distance the constriction ; furnished with a row of spines about 6 on each side of semicell, which are not extended to apex; seen in side view, they form two divergent rays, approaching nearer at 648 MR. W. JOSHUA ON BURMESE DESMIDIER. base and widening upwards; in side view, cells globose (width of isthmus about half the transverse diameter); angle at point of juncture forming a wide notch. Length 51 p; breadth without spines 40 u. C. sunpTUMIDUM, JVordst., P. PLATYDESMIUM, JVordst. C. IncIsuM, n. sp. (Plate XXIV. figs. 28 & 29.) C. mediocre : cellulis plus 13-plo longis quam latis: constric- tione angustata, lineari: semicellulis subquadratis; basi recta : angulis apicalibus rotundatis, apice truncato, late depresso; membrano interiore minute granulato: isthmo angusto. Longit. 38 p, latit. 26 p, crassit. 19 p: isthmi latit. 18 u. Medium size; cells rather more than 13 times as long as wide ; constriction narrow, linear ; semicells subquadrate ; base straight; apical angles rounded; apex truncate, with wide depression ; inner membrane minutely granulate; isthmus narrow. Nearest some of the forms of C. Meneghinii, Bréb. Length 38 p; breadth 26 u; thickness 19 u; width of isthmus 18 p. C. ANNULATUM, Naeg.,f. Except in shape of granules, identical with C. elegantissimum, Lundell. . PACHYDERMUM, Lund. . PACHYDERMUM, f. MINUS, Nordst. . GEMINATUM, Lund. . Barr rrr, Wolle. . PogTIANUM, Arch. . PYGMEUM, Arch. . TETRAGONUM, Naeg., Arch. . CONTRACTUM, Kirschner. . SUBROTUNDUM, Delponte. . BrcARDIA, Reinsch. . UNDULATUM, Corda, y. MINUTUM, Wittr., var. ORNATUM, Schaer. C. DEPRESSUM, Naeg. C. QUADRATUM, Ralfs, f. C. PARDALIS, Cohn, f. MINOR, n. var. (Plate XXIV. fig. 20.) C. INORNATUM, n. sp. (Plate XXIV. figs. 26 & 27.) C. mediocre, nonnihil duplo longius quam latum, fere cylindri- cum, apice leviter truncato: lateribus late rotundatis; centro leviter constricto ; angulo latiore: semicellulis a latere conspectis ellipsoideis; membrano crenato; verrucis indistinetis, nullis in eO00000700000n MR. W. JOSHUA ON BURMESE DESMIDIEX. 649 centro, sepe in serie externa modo visis, Longit. 44 p, latit. l8 u: isthmi latit. 15 p. Medium size; rather more than twice as long as wide; nearly cylindrical ; apex slightly truncate ; widely rounded at sides, in middle slightly constricted, with rather wide angle; side view of semicells elliptical; membrane crenate; verruce very faintly visible, absent in centre of cells, the outer row only often the only ones visible. Length 44 p; breadth 18 p; width of isthmus 15 pg. C. EXASPERATUM, n. sp. (Plate XXV. figs. 1 & 2.) C. parvum, fere tam longum quam latum: cellulis anguste ellipsoideis, profunde constrictis, angulo late hiante, prominentiis validis, truncatis, irregularibus obtectis, in seriebus 2-3 horizon- talibus in utraque semicellula dispositis: a vertice conspectis angustis, ellipsoideis; cytiodermate crasso. Longit. cum promi- nentiis 40 u, sine prominentiis 35 p : latit. cum prominentiis 42 p, sine prominentiis 38 p: isthmi latit. 11 p: crassit. sine prom. 16 u, cum prom. 23 y. Small; about as long as wide; cells narrowly elliptical, co- vered with rows of strong, truncated, rather irregular protu- berances, arranged in 2 or 3 horizontal rows in each semicell ; deeply constricted, with wide gaping angle; end view narrow, elliptical; eytioderm thick. Length with protuberances 40 p, without protuberances 35 u; breadth with protuberances 42 p, without protuberances 38 u; width of isthmus 11 p; thickness without protuberances 16 u, with protuberances 23 p. C. BIFARIUM, n. sp. (Plate XXV. figs. 5 & 6.) C. parvum, paulo longius quam latum ; profunde constrictum, angulo anguste lineari, semicellulis subovatis; angulis rotundatis, minute regulariter granulatis, prominentiis 5-6 magnis spher icis in seriebus duabus horizontalibus dispositis ; a vertice conspectis subsphericis, utroque latere polari prominentiam rotundatam, elongatam ferentibus. Crassit. sine papillis 15 p, longit. 25 p, latit. 23 u: isthmi latit. 6 p. Small; rather longer than wide; deeplv constricted, with narrow linear angle; semicells subovate; angles rounded, mi- nutely and regularly granulate, with 5 or 6 large spherical protu- beranees in two horizontal rows; end view subspherieal, show- ing on each side of polar end a rounded elongate protuberance. LINN. JOURN.— BOTANY, VOL. XXL 3c 650 MR. W. JOSHUA ON BURMESE DESMIDIEJE. Thickness without papille 15,4; length 25u; breadth 234; breadth of isthmus 6 p. C. Hammert, Reinsch. C. nETUSUM, Perty. C. venustum, Bréb. 12. PLEUROTÆNIUM, Naeq. . Noposum, Bailey. . CLAVATUM, Kitz. . INDICUM, Grun. . INDICUM, f. MINOR. "UU et 13. Docitpium, Bréb. D. Bacutum, Breb. D. GRANULIFERUM, n. sp. (Plate XXV. figs. 11 & 12.) D. magnum, diametro 14-15-plo longius, ad basin semicellule nflatione subhemispherica indistincte undulatum, apicibus trun- eatis angustioribus: membrano crasso, granulis magnis regula- riter obtecto, granulis singulis fere sphericis, centro nodulosis: zygosporis ignotis. Longit. 680-885 y, latit. apud apicem 35- 40 u: inflationis latit. 45-60 u: latit. in centro semicellule 45-60 u: granulationes 3 in 1 p. Large; 14-15 times longer than wide, with a subhemispherical inflation at base of semicell ; indistinctly undulated ; apices trun- cate, rather narrowed; membrane thick, regularly covered with large granulations; each granule nearly spherical, having 4 central nodule ; zygospore unknown. In size and shape resem- bling Pleurotenium maximum, Reinsch. Length 680 p-885 p; width at apex 35 4-40 u; width of inflation 45 u-60 p; width in middle of semicell 45 u-60 p ; granulations 3 in 1 p. D. rEssELLATUM, n. sp. (Plate XXV. fig. 15.) D. medioere, diametro 8-9-plo longius, a medio ad apicem, vix angustatum, preter apices, projectionibus magnis irregularibus quadratis, sursum deminutis, obtectum: membrano non punctato: apice leviter angustato, spinis 8-9 truncatis exsertis instructo. Longit. 200-330 p, latit. 25-35 u : apicis latit. 16-24 p. Medium size; 8-9 times longer than broad from middle to apex, scareely narrowed, covered with large irregular quadrate projections not extending to apex, but diminishing iv size upwards; membrane not punctate; apex slightly narrowed, MR. W. JOSHUA ON BURMESE DESMIDIEZ. 651 furnished with 8 or 9 truncate spines set outward. It appears sufficiently distinct from D. verrucosum, Bailey, as shown in the figure, and those of Rev. F. Wolle, Desm. U. States. With slight inflation at point of juncture. Length 200-330; breadth 25- 35 p; width of apex 16-24 u. D. ANNULATUM, n. sp. (Plate XXV. fig. 13.) D. mediocre, diametro 10-11-plo longius, basi paulo majore: semicellula crenulationibus 12-14 ornata seriem regularem pro- jectionum annularum sistentibus: membrano incolore, distincte punetato: apice truncato, leviter angustato, sepe dentibus bre- vibus obtusis instructo. Longit. 235 p, latit. suture 25 p, latit. apicis 16 u. Medium size; 10-11 times longer than wide, with slight en- largement at base; semicell with 12-14 crenulations, forming a regular series of annular projections; membrane colourless and distinctly punctate ; end truncate, slightly narrowed, frequently furnished with short blunt teeth. Length 235,4; width at suture 25 u; width at apex 16 y. D. BURMENSE, n. sp. (Plate XXV. fig. 14.) D. magnum, diametro 15-20-plo longius : semicellula profunde constricta, constrictionibus 11-12: nulla extra basin inflatione : puncto junctionis haud valide notato: apicibus truncatis, leviter angustatis, dentibus 8-10 brevibus obtusis instructis: membrano tenui, indistincte punctato. Longit. 660-850 p, latit. ad basin 35 p ; apicis latit. 20-27 p. Large; 15-20 times longer than broad, having in semicell ll or 12 deep constrictions; no extra basal inflation; point of Junction not strongly marked ; apices truncate, slightly narrowed, and furnished with 8-10 short blunt teeth; membrane thin, in- distinctly punctate. Its nearest resemblance is to D. constrictum, Bailey, from which it differs in its greater length, undulate habit, and attenuated apices. Length 660-850 p; width at base 35 p; width at apex 20-27 p. D. coronutatum, Grun. Abundant. Often joined in long filaments. Showing points of junction and covering of apex (Pl. XXV. figs. 16 & 17). D. minurum, Naeg. 14. TRrPLOcERAS, Bailey. T. aRACILE, Bailey. 802 652 MR. W. JOSHUA ON BURMESE DESMIDIEX. 15. CrosTERIUM, Nitzsch. C. Dranz, Ehrenb. C. costatum, Corda. C. Exsis, Delponte. C. BacrrrzuM, n. sp. (Plate XXII. figs. 4, 5, & 6.) C. robustum, fere rectum, diametro 6-10-plo longius, cylindri- cum: apicibus plus minus attenuatis, conos obtusos sistentibus : membrano crasso, levi, in adulto pulchre pallido-ochraceo: lineis mediis 2-3 apud punctum junctionis. Longit. 200—235 p, latit. 24-35 u. Stout, nearly straight; 6-10 times longer than wide, cylin- drical; apices more or less attenuated, forming obtuse cones; membrane thick, smooth, in maturity of a fine delicate ochre; 2 or 3 median lines at point of juncture. Nearest to C. didymotocum, Corda, and C. antiacerosum, De Notaris. Length 200-235 p ; width 24-35 p. . LAGOENSE, Nordst. . STRIOLATUM, Ehrenb. . LINEATUM, Ehrenb., f. SANDVICENSIS, Nordst. . RaArrsir, Bréb. - ACEROSUM, Schrank. . SETACEUM, Ehrenb. . JUNCIDUM, Ralfs. . VENUS, Kutz. INTERMEDIUM, Ralfs. . INCURVUM, Bréb. . ACUTUM, Bréb. . LErBLEINU, Küfz. . PORRECTUM, Nordst. - SUBTILE, Bréb. Conjugated with quadrate zygospores. . NEMATODES, n. sp. (Plate XXII. figs. 7, 8, & 9.) C. lanceolatum, diametro 10-11-plo longius, sensim curvatum, centro haud inflatum, striis tribus transversis, apicibus paulo utrinque dilatatis, inflationem prominentem ovatam ferentibus, ultra abrupte attenuatis: membrano crasso, flavido-fusco, lon- gitudinaliter striato, striis 18-920. Longit. 187 p, latit. 20 p, latit. ad inflationem 13 p. Lanceolate ; 10-11 times longer than wide, gradually curved, cO ocOooo0on000000000020 MR. W. JOSHUA ON BURMESE DESMIDIEX. 653 no central inflation ; apices slightly dilated on both sides ; beyond suddenly attenuated ; central part with 3 transverse strie; mem- brane thick, yellowish brown, marked longitudinally with 18-20 strie. It comes near to C. lagoense, Nordst., but is of a more slender habit, and has the incrassate apicallobes. Length 187 p; width 20; width at inflation 13 u. 16. Pentum, Bréb. P. DELICATULUM, n. sp. (Plate XXV. figs. 9 & 10.) P. parvum, subcylindricum, duplo et ultra longius quam latum, utroque latere medii leviter emarginatum: semicellulis apice conieis, membrano interiore valde attenuato, minute regulariter punetato. Longit. 63 p, latit. 26 p. Small, subcylindrical, rather more than twice as long as wide ; on both sides of middle slightly emarginate ; semicells conical at end; much attenuated inner membrane minutely and regularly punctate. Length 63; breadth 26 p. P. MABGARITIFERUM, Ehrenb. P. sPIBosTRIOLATUM, Barker. P. piaitus, Bréb. P. LAMELLOSUM, Bréb. P. (Cyzinprocystis) BREBISSONII, Menegh. P. cLosrerioivEs, Ralfs. P. MINUTISSIMUM, Wordst. P. NAvICULA, Bréb. EXPLANATION OF THE PLATES. Pare XXII. Figs. 1 & 2. Spherozosma pulchrum, Bailey, B. trilobum, n. var.: 1, filament, X400 ; 2, end of filament seen crossways. 3. Desmidium quadratum, Nordst.; conjugated zygospores, x 400. 4, 5, & 6. Closterium Bacillum, n. sp., x 200. 7, 8, & 9. Closterium nematodes, n. sp.: fig. 7, X300; and 8 & 9, x 400. 10 & 11. Micrasterias Crux-melitensis (Ehrenb.), Ralfs, var. n. f., x 300. 12. Micrasterias Luz, n. sp., X 300. 13. Micrasterias apiculata, Menegh., n. f. 14. Micrasterias euastroides, n. sp., x 300. Pare XXIII. Figs. 1 & 2. Euastrum serratum, n. sp.: 1, cell, X600; 2, side view of same, x 300. 654 MR. W. JOSHUA ON BURMESE DESMIDIEZ. Figs. 3, 4, & 5. Euastrum flammeum, v. sp. : 3, cell, X600 ; 4, end view, x 300; 5, side view, x 300. 6 & 7. Euastrum truncatum, n. sp. : 6, cell, x 600; 7, side view. 8 & 9. Euastrum divergens, n. sp. : 8, cell, x 450 ; 9, cell, x 600. 10. Euastrum coralloides, n. sp., x 600. 11, 12, & 15. Euastrum binale, Turp., n. f. crassum : 11, cell, «500; 12 & 13, x 2000. 14 & 15. Euastrum retrorsum, n. sp. : 14, cell, x 400 ; 15, side view. 16, 17, & 18. Euastrum exile, n. sp.: 16, cell, X900; 17, side view, and 18, end view, x 900. 19 & 20. Euastrum obesum, n. sp., X 330 and 300. 21. Staurastrum saltans, n. sp., X 500. 22 & 93. Staurastrum cyathoides, n. sp.: 22, front view, x 300; 23, end view, x 400. 24. Staurastrum sexangulare, Bulnh., n. var., x 400. 25, 26, 27, & 28. Staurastrum bifurcum, n. sp.: 25 & 26, x 600; 27 & 28, x 300. Prate XXIV. Figs. 1 & 2. Staurastrum platycerum, n. sp., x 400. 3. Staurastrum horrescens, n. sp., x 400. 4 & B. Staurastrum granulatum, Reinsch, x 400. 6 & 7. Staurastrum furcatum, Ehrenb., 8. senarium, Ehrenb. : 6, front view, X600; 7, upper markings. 8 & 9. Staurastrum bifidum, Ralfs, var.: 8, end view, x 000 ; 9, side view, x450. o 10, 11, & 12. Arthrodesmus Incus, Menegh., vars.: 10 & 11, X400; 12, x 300. 18. Arthrodesmus subulatus, Kurz, n. var. gracilis, x 400. 14. Arthrodesmus arcuatus, n. sp., x 300. 15. Arthrodesmus apiculatus, n. sp. 16. Xanthidium antilopeum, Kütz., f. angulatum, n. var., X 400. 17 & 18. Cosmarium cuneatum, n. sp.: 17, cell, X600; 18, oblique view of semicell, x 600. 19. Cosmarium Botrytis, Menegh., f. indicum, n. var., x 400. 20. Cosmarium pardalis, Cohn, f. minor, n. var., X400. 21-25. Cosmarium armatum, n. sp.: 21, cell, x 500; 22, cell, x 900; and 23, x 600. 26 & 27. Cosmarium inornatum, n. sp., X600. 28 & 29. Cosmarium incisum, n. sp., x 600. 30-34. Cosmarium Euastron, n. sp. : 30, cell, x 450; 33, stellate arrangement of granules; 94, side view of granules. Prate XXV. Figs. 1 & 2. Cosmarium exasperatum, n. sp. : 1, front view, x 600; 2, end view, x 600. 3 & 4, Cosmarium spinosum, n. sp.: 3, cell, x 450 ; 4, end view. EOCENE FERNS FROM IRELAND AND SCOTLAND. 655 Figs. 5 & 6. Cosmarium bifarium, n. sp. : 5, cell, x 900; 6, end view. 7. Cosmarium Diadema, n. sp., x 450. 8. Cosmariwm capaz, n. sp., x 300. 9 & 10. Penium delicatulum, n. sp., x 450. 11 & 12. Docidium granuliferum, n. sp.: 11, cell, X300; 12, granules, x 1000. 13. Docidium annulatum, n. sp., x 300. 14. Docidium burmense, n. sp., x 300. 15. Docidium tessellatum, n. sp., X 300. 16 & 17. Docidiwm coronulatum, Grun.: 16, junction of cells, x 300; 17, apex of cell, x 400. Eocene Ferns from the Basalts of Ireland and Scotland. By J. SrARKIE GARDNÉR, F.L.S., F.G.S. [Read 7th May, 1885.] (Prati XXVI.) I FERL hardly justified in bringing a communication before you based on such very meagre material. I have only been able to get together, from all sources, five varieties of fern, and these consist of poorly preserved and almost unique fragments ; they nevertheless represent, so far as the ferns, a great number of expeditions to County Antrim and to Mull, and much diligent collecting there, and contrast favourably with any brought from the other extremities of the formation in which they occur. Two, those from Glenarm and Ballypalady, are from the lowest stage but one, palxontologically, of the basalts, and belong, I be- lieve, to almost the oldest known Eocene. One is new to science and the other recorded for the first time from Great Britain. One, from Mull, is from, palzontologically, the lowest stage of the basalts. The others, from Lough Neagh, are from a vast series of deposits, synchronous, I believe, with those of Bovey Tracey, and representing the latest stage of the Eocene preserved in the north of our Isles. Attention has already been called to these latter by Mr. W. H. Baily in the ‘ Reports ' of the British Association. Except the two from Lough Neagh (figs. 8 & 9), which belong respectively to the Geological Survey of Ireland and to Canon Grainger, F.G.S., the specimens were collected by means of the Government Grant in aid of research, and will be deposited in our National Museum. 656 MR. J. STARKIE GARDNER ON EOCENE FERNS FROM I am afraid, from a botanical aspect, that their interest is not great ; but, geologically, they supply data of considerable value towards fixing the relative age of one of the most remarkable formations in the world, and not less important because entirely destitute of fossil remains save and except plants. I allude to the enormous basaltic formation of the North-east Atlantic, which once stretched, there is every reason to believe, continuously from Antrim through the Faroés and Iceland to Greenland. GLEICHENIA HIBERNICA, sp.nov. (Plate XXVI. figs. 5-7.) Eocene basalts, Glenarm and Ballypalady, County Antrim. Pinpe narrow oblong, acuminate, with probably entire and even caudate apices, length 10 centim., breadth 4 centim., cut down within a short distance of the rhachis into numerous linear lobes, 30 millim. in extreme length and 5 to 6 millim. broad. The lobes are at angles of 50° to 60° with the rhachis, obtuse or subacute, opposite near the base and alternate higher up the pinna. Rba- chis prominent, slender, glabrous; primary veins of the segments undulating, directed alternately towards each bundle of veinlets. The lowest veinlets fork once and extend to the margin of the sinus. The veinlets are fine and grouped in fasciculi, simple at first and starting at an angle of 55^, they divide into 3, the outer simple and the central one forked ; though sometimes two, at times all three of the veinlets fork. "The venation becomes more simple towards the apices. The margins of the segments seem faintly undulate. The species seems to have been rigid, and the texture coriaceous. Below the last segment is a small ear-like expansion in one of the specimens, destitute of midrib but traversed by forked veinlets. The sori were removed before the pinnæ became imbedded, but their position is defined by a small narrow elliptical scar, situated near the base of the most forward—that farthest from the midrib —of the secondary veinlets of each fasciculus (fig. 7), though Where these are more complex, near to the bases of the pinnules, the scars occur on both the outside veinlets instead of on one. Their position leaves no doubt as to the genus in which the fossil should be placed, and the whole habit agrees with the Mertensia section of Gleichenia, and especially with G. dichotoma, a native, according to Sir W. Hooker, “ of tropical and subtropical regions, almost universal in the New and in the Old Worlds, Pacific THE BASALTS OF IRELAND AND SCOTLAND. 657 Islands, and as far north as Japan." The frond seems to have been dichotomous, for fig. 5 appears to represent parts of two basal pinne, while 6 may be a terminal one. A species belonging to the same section of Gleichenia, with very similar pinne, occurs in the Middle Eocene of Bournemouth, but is distinguished by the conversion of some of its pinnz into hook-like tendrils, differing from those of any existing fern. No other Tertiary fossil species at all resembles it except one from tle very ancient Eocene of Sezanne; but similar species existed in the late Cretaceous of Aix-la-Chapelle. The Bournemouth species is almost limited to a small patch, only a few yards in extent; but the Irish species seems to have been spread rather widely, having already been found at two localities many miles apart. These seem, however, to have been near its northern limits, for it has never been found among the numerous fossil floras met with in this basaltic formation nearer to Arctic regions. The genus Gleichenia seems to have reached its culminating development in Europe in Cretaceous times, and thenceforward to have rapidly disappeared. The likeness between the fossil and its living representative falls short of specific identity. So vast an interval has elapsed that the marvel is that the growth and venation of any fern should have continued, while all else around it has progressed, without modification. We have examples of such, however, in Chrysodium vulgare, Osmunda javanica, Onoclea sensibilis. Other Eocene species are only separated from their living allies by trifling cha- racters: among these is the Gleichenia like G. dichotoma, but with a climbing habit, and close allies of Pteris cretica and Lygo- dium palmatum, a Phymatodes, and an Adiantum. ` In the latter category must the Irish G. hibernica be placed ; though at any moment specimens gathered from a fresh locality may show à closer agreement, amounting, perhaps, to identity, absolute and complete. No other Tertiary fossil can readily be confounded with it. The few specimens I possess from Glenarm were found close to- gether during my last day's work there. Mr. Stewart obtained à specimen from Ballypalady in which only the rhachis and mid- ribs remained distinct, and I possess another in the same matrix in which some of the veinlets are faintly discernible. 658 MR. J. STARKIE GARDNER ON EOCENE FERNS FROM GLEICHENIA, species indeterminable. (Plate XXVI. figs. 10 & 11.) Eocene basalts, Ballypalady, Antrim. The pinnules are narrow oblong, 5 millim. wide, and cleft in segments down to the midrib. These segments are somewhat faleate, at an angle of 55° with the midrib and barely 2 millim. across. The midrib is slender and prominent, but the rest of the venation is obscure. A simple veinlet can be traced on each segment, and here and there secondary veinlets diverging from it; though the matrix is too coarse to retain very delicate impressions. The remains of this fern are of the scantiest description, no more than two fragments of pinnules, barely 15 millim. long, being known to me. These are traced out in yellow on the deep brown matrix, the outline being comparatively sharp. This colouring is rarely met with ; I have only seen it in Nelumbium Buchii and one or two other leaves, and I am inclined to interpret its presence as an indication that the leaves possessed a thick, but not coriaceous texture. It would be useless to attempt any generalizations upon such material. There is, however, one fact too significant to be entirely passed by. This is, that while innumerable examples of this type are met with in the so-called Cretaceous floras with Dicotyledons of America, Europe, and Greenland—such as Gleichenia Kurriana, Heer, which is indistinguishable from it, and in the Heersien flora Benitzia minima, Sap., and even in Eocenes of so late a date as the Woolwich beds—not the remotest trace of any fern resembling it, so far as I know, has yet been brought to light from any of the later Eocene beds of Europe*. Trifling as it may be, its occur- rence in the Antrim basalts accords with the views I have ad- vanced, and supported by, I believe, irresistible evidence, as to the great antiquity of this vast basaltic formation. ONOCLEA HEBRIDICA, J. S. Gardn. (Plate XXVI. figs. 1-4.) Filicites hebridicus, Ed. Forbes, Quart. Journ. Geol. Soc. vol. vii. p. 103, pl. 2. fig. 2 (1851). Eocene basalts, Isle of Mull. There are no known characters by which this fern can be defi- nitely distinguished from the living Onoclea sensibilis. It is a * I have since ascertained that it reappears, possibly after an enormous interval, in the Oligocene of Gurnet Bay, Isle of Wight. THE BASALTS OF IRELAND AND SCOTLAND. 659 question whether the fossil should be allowed to retain the specific name given by Edward Forbes, who described it without being aware of its close relation to an existing species: or whether we should follow Prof. Newberry, who has united a later and more robust species, from North-American Tertiaries, with the living one. The former course appears preferable, at least provisionally. The pinne are exceedingly variable as in the existing species, both as regards their cutting and venation—fig. 1, from near the base, and fig. 2, from the apex of a frond,representing two of the extremes. The slender stipes, from 1 to 2 feet and more in length, which are seen crossing some of the slabs recently obtained from Mull, appear to be grooved and glabrous. The fertile frond is identical with those of O. sensibilis (fig. 4), except that the sori are somewhat smaller (fig. 3). A soft texture is indicated in the fragmentary state in which the fossil is usually found, and the character of the impressions in the shale points in the same direction. The discovery of this Onoclea at Mull is particularly interesting, since it has as yet not been met with elsewhere in Europe. It is limited to the upper layer, about a foot thick, of the black shaly leaf-bed at Ardtun Head, where much macerated fragments occur in some abundance, in company with a large palmate leaf, com- pressed reeds, and Equisetum. The bed is fluviatile, and seems to have been formed alongside a river, in a swamp liable to inun- dation, similar to the Tilbury flats before their level was raised. The age of these beds has not been ascertained with any certainty; but after a careful review of both the stratigraphical and palzon- tological evidence, I can no longer hesitate to regard their horizon as at least as old as the Lower Eocene, and probably below that of the Thanet Beds. The fossil Onoclea is not exclusively confined to Mull; it has also been brought from Atanekerdluk, in Greenland, by Mr. Whymper, in an equally macerated condition. It was at first named Woodwardites arcticus by Heer; but Dr. N ewberry subse- quently pointed out its true affinity, and it is satisfactory to know that the identification of the Mull and Greenland specimens is supported by one of the highest living authorities on fossil plants, the Marquis de Saporta. Along with fragments with copiously anastomosing venation, named Woodwardites, are others of more robust aspect and much simpler venation, originally named Peco- pteris Torellii, Heer, but afterwards corrected by Heer, first to 660 MR. J. STARKIE GARDNER ON EOCENE FERNS FROM Hemitelites, and finally to Osmunda Torellii, Heer. It seems probable that these may belong to a variety of the same fern, such as the one found in the Fort Union Group of Dakota, and still found living in the United States, The Fort Union Group is a freshwater series considered by American geologists to belong to the Eocene Formation, containing numerous plants, among them being O. sensibilis in considerable abundance. The type of this fern met with in it is more robust than any living, or than the fossils met with in the older localities ; and it is perhaps worth mentioning, without entering into the merits of the ques- tion here, that I am inclined, if not to doubt the Eocene age of the deposit, to regard it as a very late member of the formation. 'These specimens were collected by Dr. Hayden and are finer than any obtained elsewhere. In speaking of them Prof. Newberry remarked, * there is little room for doubt, therefore, that during the Miocene age a species of Onoclea (Euonoclea) flourished in the interior of our continent, of stronger habit than either of the living varieties, and holding a middle position between them." He seems to regard its horizon in the United States as Miocene, and living it is quite common at the present day. Like so many other species and genera of Ferns and Coniferz, whose characters are so sharply defined that they appear to have no affinities with other existing species, Onoclea sensibilis is a survival from a flora belonging to a past so remote that nearly all its companion species have become extinct or modified beyond specific recognition. We are far from able to trace its pedigree; but we at least know that in Mull it once occupied a station separated by thousands of miles from its nearest exist- ing habitat. It either once ranged synchronously from America through Europe to its present habitats on the temperate coasts of Asia, having since died out in the intermediate areas, or it emigrated from some original home to the Amur and Japan, and, vid Europe and Greenland, to America. Its limited palæ- ontological range lends little support to the former theory, but is not opposed to the latter. Prof. Newberry thinks that it had an American origin; but it is clearly found in older rocks in Scotland and Greenland than in the United States. With such slender data speculations are scarcely warranted ; but a satisfactory interpretation of the known facts would be, that we had a well-developed ancestral form of Onoclea sensibilis, relatively small and fragile, established in Mull during the cooler THE BASALTS OF IRELAND AND SCOTLAND. 661 Lower Eocene period; that the increasing temperature drove it north ; that once established in the latitude of Greenland, it followed the then existing coast-line across the Atlantic; and, lastly, descended into America when the decreasing temperature of the Miocene rendered a southerly move imperative. It seems to have increased in size and strength, and, as a recent fern, is described by Newberry as varying “ in the size, outline, and ner- vation of the sterile frond, from 6 inches to 3 feet in height, from a finely reticulated to an open dichotomous nervation; from a bipinnate frond with remote obovate pinnules to a pinnate form with wave-margined pinne and broadly alate rhachis.” He has named the American fossils Onoclea sensibilis fossilis, finding that the recent species “ plainly includes all the characters of the fossils before us"*. It adapts itself to a considerable range of temperature, growing in America from Florida to Canada, and in Asia from the Amur to Japan and Manchuria. Further descrip- tions and illustrations of the American fossils by Prof. Newberry are to be published. The Ferns from the Lough-Neagh deposits belong to an altogether different and much newer horizon. The paleonto- logical evidence, so far as it is available, points conclusively to the Middle or Upper Eocene as likely to be about their true age, while the stratigraphical evidence warrants us, I believe, in including them in the Eocene basaltic formation. It is very significant that these, occurring in a later formation, should present us with the first instance of the occurrence of species of the English Eocene in that of North Britain; and the induction is almost irresistible that the increased temperature enabled them to occupy stations previously beyond their range. The two species are Goniopteris Bunburii, Heer, and G. stiriaca, Unger, both known in Ireland from unique specimens. The latter is one of the commonest plants at Bovey Tracey; and the former is also met with there, as well as more frequently towards the upper part of the series at Bournemouth. The Lough-Neagh deposits exhibit quit: a striking analogy with those at Bovey Tracey; and were we dealing with invertebrate or vertebrate fossils instead of with fossil plants, the community of two such well-marked species as these would be regarded as conclusive evidence of the synchronism of the deposits. * Ann, Lyc. Nat. Hist. N. Y. vol. ix. p. 89 (April 1865). 662 MR. J. STARKIE GARDNER ON EOCENE FERNS FROM Gontoprerts Bunsurtt, Heer. (Plate XXVI. fig. 9.) Eocene basalts, Lough Neagh. This should be credited with a far wider range in the European Tertiaries than has previously been extended to it; but it isa type apparently rare, if not unknown prior to the Middle Eocene period. Though it comes very near in venation and outline to Goniopteris diversifolium and G. tetragonum from Brazil, the per- sistent absence of any traces of sori in all of the numerous spe- cimens that I, as well as others, have come across, suggests that it would be better placed in some genus in which the fertile fronds are separate. One fulfilling these conditions and bearing an other- wise strong resemblance to it is Onoclea orientalis ; but this differs in possessing free veinlets. The genus Onoclea is practically re- presented only by three existing species, forming respectively the subgenera Ewonoclea and Struthiopteris, the former with copi- ously anastomosing, and the latter with free veinlets. In Onoclea germanica the segments are cut down, and the veinlets look as if they must unite but for this; but in some forms of the closely related O. orientalis the pinnules have merely waved margins, yet the veinlets never anastomose. Some principle or repulsion keeps them from contact, however nearly they may approach; yet it is quite certain that a common ancestor must at some period have united the two very distinct existing types; and the vena- tion we are dealing with must almost of necessity have been one ofthelinks, Genera have their rise and fall, and such a large proportion of the fossil plants I have.already examined belong to genera poorly represented now, that I am predisposed to search among such for survivals of the floras of Eocene age. The case stands thus :—As far as outline and venation go, our fern might be placed in Goniopteris, Nephrodium, Diplazium, Polybotrya, or other genera. These in a living state are sepa- rable by their sori. The fossils are destitute of any trace of sori. True, they are fragmentary, and have been floated down by water, and the sori might therefore have been so completely rubbed off as to leave no trace. The balance of probability is against this ; for it is not uncommon for sori to be preserved on ferns asso- ciated with them in the very same deposits. The determination must thus be based on a nice appreciation of probabilities, and may rest on an accumulation of data which separately would be trivial and valueless. It was described as Lastre@a by Heer. THE BASALTS OF IRELAND AND SCOTLAND. 663 GoNIOPTERIS STIRIACA, Unger, sp. (Plate XXVI. fig. 8.) Eocene basalts, Lough Neagh. The only British locality for this species prior to its discovery at Lough Neagh was Bovey Tracey. It seems to have been local, but not uncommon there. All the fronds found were destitute of sori; but Prof. Heer considered it to be identical with a large species from the Aquitanian of Monod in Switzerland, whose fronds he estimated to have been 3 feet in length and a foot wide. From these he supplemented his diagnosis, and fixed the size, form, and position of the sori. If the identification is correct, several other supposed species should also be united with it, such as Lastrea helvetica and L. dalmatica. It is also given an Arctic range by Heer; but there seems, from the de- scription, some little doubt as to the identity of the Greenland species with ours. A very considerable difference in the vena- tion of pinnules from different parts of the same frond might be looked for in a species of such large dimensions, and the minute apex of a pinnule from Lough Neagh, when magnified, has a quite different aspect, and is almost exactly like Œ. Bunburii in its venation. While, however, considerable latitude would be permissible as to the extremes of venation and cutting that might safely be associated in one species, when the specimens are all from the same locality and bed, the large number of existing genera (e. g. Acrostichum and Nephrodiwn) in which pinne of this type occur should make us anxious to note even the most inconsiderable persistent differences. Thus a fern found by the Marquis de Saporta in the gypsum of Aix, Provence, appears at first glance completely identical with it; but it differs in reality in an important particular, the anastomosis ofthe inferior veinlets of the segment not being continued up to contact with the pair above. This peculiarity removes it from the group G. prolifera, Mett., of Tropical America, to which our British specimens seem very closely allied, and associates it with G. crenata, Mett., also of Central America and the Antilles. T may here mention a slight peculiarity in the Lough-Neagh specimen, which is, that instead of the veinlets of contiguous segments anastomosing regularly in pairs, the lower veinlet on one side, in at least some instances, anastomoses with three in succession on the other. Want of recognition of even so slight a character as that on which Saporta insists might lead to very erroneous views as to the former distribution of a species. G. stiriaca seems to have been 661 EOCENE FERNS FROM IRELAND AND SCOTLAND. a fern requiring a relatively moderate temperature in our Eocene period; and I think temperature must have excluded it from Bournemouth, where otherwise every variety of condition favour- able to fern-growth seems to have existed. Bovey was evidently a much higher station and far from the sea. This species is found in many of the relatively temperate Middle and Upper Eocene floras of Europe, and, as before remarked, possibly extended to circum- polar regions, just as Aspidium Lonchitis extends from Naples and Greece to Disco, the subtropical Trichomanes radicans to Ireland, and Hymenophyllum to Norway; it is also met with iu the American Tertiaries. It must have been a robust fern and of somewhat coriaceous texture, the pinnules being with difficulty detached from the rhachis, and, judging from its presence in the Bovey beds, addicted to marshy stations. Attention had already been called to this specimen by Mr. W. H. Baily. in the * Reports of the British Association for 1883. The stone is in the possession of the Rev. Canon Grainger, F.G.5., of Broughshane, and is quite full of fragments of pinne. It was originally deseribed as Polypodites stiriacus by Unger, and the generic name was altered successively to Lastrea, Heer, Goniopteris, A. Braun, and Phegopteris, Ettingsh. DESCRIPTION OF PLATE XXVI. Figs. 1-4. Onoclea hebridica, J. S. Gardn., Ardtun Head, Isle of Mull. l. Portions of pinnz from a large frond, showing copiously anastomosing venation. l 2. Upper part of a frond with less anastomosing venation, and denticulate to entire margins. 3. Fragment from a fertile frond with sori. 4. Similar fragment from living O. sensibilis. 5-7. Gleichenia hibernica, J. S. Gardn., Glenarm, Antrim. 5. Parts of a pair of probably basal pinne. 6. Parts of a pair of probably apical pinnze. 7. Veinlets magnified, with scar of sorus. 8. Goniopteris stirjaca, Unger, Lough Neagh. 9. Goniopteris Bunburii, Heer, Lough Neagh. 10 & 11. Gleichenia, sp., Ballypalady, Antrim. Figs. 1-7. Collected by the author, and deposited in the British Museum by direction of he Royal Society. Fig. 8. In the cabinet of the Rev. Canon Grainger, Broughshane, Fig. 9. In the collection of the Geological Survey, Dublin. INDEX. Abies Webbiana, on Sundukphoo, 385. Abutilon indicum, 373, 378. A myriophylla, 217; splendens, Acalypha hologyna, 441. Acampe, 456, 475; Renschiana, 475. Acanthacez (Philippines), 291. Acanthocladus microphyllus, 213. Acanthus spinosissimus, 531. Acer Campbellii, 388 ; levigatum, 388. Achyranthes argentea, 403. Achyrocline flaccida, 222 ; satureioides, 222; Schimperi, 401. Acokanthera, 402. Acrocephalus villosus, 434. Actinidia strigosa, 387. Adenoplusia axillaris, 424. Adesmia grisea incana, 215; longipes, 216; pendula, 216. Adiantum, 657 ; cuneatum, 239, 240. Adventitious budding in Ferns, 364. JEcidiospores, non-intervention in case of certain Heterzeeious Uredines, 368. JEcidium, 369. JEgilops biuncialis, var. biaristata, 537 ; triunciale, 537, var. breviaristata, 5387; uniaristata, 532. ZEluropus litorali:, 537. JEonia, 494, 495 ; Auberti, 496 ; macro- stachya, 495, 496; polystachys, 496, 497 ; rosea, 496. Aeranthus, 493, 494; Aphrodite, 496 ; Arachnites, 494, 495; Curnowianus, 486; gladiifolius, 476, 494, 495; macrostachys, 499; pectinatus, 487 ; polystachys, 496 ; rectus, var. ?, 477; recurvus, 477; sesquipedalis, 475; Volucris, 496. LINN. JOURN.— BOTANY, VOL. XXI. Aerides, 493; coriaceum, 498; macro- stachyon, 495. Aerobion crassum, 486; implicatum, 484. Aerostichum, 663. JErua lanata, 403; scandens, 374; velutina, 374. JEschynanthus, 253 ; 301. Affinities of East-African Plants coll. by J. Thomson, Sir J. D. Hooker on, 395. Afghanistan Alge, Dr. J. Schaar- schmidt on, 241. African Plants coll, by J. Thomson, Prof. Oliver on, 392. Aglaia macrobotrys, 298. Ainslizea pteropoda, 389. Aitchison, Dr., Afghan. Algæ coll. by, 241. Ajuga lobata, 390. Alectoria nidulifera, 272. Alectra asperrima, 402. Alepidea amatymbica, 394, 400. Alge of Afghanistan, 241. Algæonema simplex, 299. Algo-Lichen Hypothesis, Rev. J. M. Crombie on, 259. Alisma ranunculoides, 526. Alismaceæ (Philippines), 292. Alleanthus (Philippines), 295. Allium tenuifolium, 525. Allomorpha, 287. Alopecurus, 237 ; aristulatus, var. geni- culatus, 237. Alpinia gigantea, 316; gracilis, 316; arviflora, 316. Alsophila Baroni, 455 ; tomentosa, 300. 3D philippinensis, 666 Alternanthera echinata, 232. Alyssum gemonense, 532; leucadeum, 532, 533; montanum, 525. Amamirta Coceulus, 373. Amarantacee of Patagonia, 232; of Philippines, 291. Amarantus chlorostachys, 232. Amaryllidee of Patagonia, 235; of Philippines, 292. Amblystegium serpens, 366. Ambrosia, 222; artemisizfolia, 222. Ammannia, 241, 345; cryptantha, 945; pentandra, 241 ; rotundifolia, 345. Amomum, 253, 254. Ampelidez (Philippines), 290. Amphorchis, 512, 517 ; calearata, 517, 518, 519; inermis, 517; laxiflora, 517; lilacina, 517, 518; occidentalis, 517. Anabeena, 241 ; cylindrospermum, 242, 243, 250. Anacardiacez of Patagonia, 215; of Philippines, 290. Andrachne australis, 299. Andropogon, 209, 236 ; argenteus, 236 ; avenaceus, 236; virginicus, 209. Anemone, 241, 393, 394 ; capensis, 394 ; obtusiloba, 387; tetrasepala, 241, 245; Thomsoni, 397. Angelica, 350. Angrecum, 475; aphyllum, 492; api- eulatum, 483; articulatum, 483; Brongniartianum, 480; caudatum, 481 ; caulescens, 488 ; citratum, 481; clavigerum, 485; Cowanii, 484; crassum, 486 ; cryptodon, 483; distichum, 487; eburneum, 480; elatum, 497 ; Ellisii, 483; fastuosum, 483; filicornu, 483; fragrans, 477 ; fuscatum, 483 ; Gilpinz, 491 ; gladii- folium, 476 ; globulosum, 491; hya- loides, 483; implicatum, 484; in- apertum, 489; maxillarioides, 479 ; modestum, 482; multiflorum, 488; pectinatum, 487 ; polystachyum, 497 ; rectum, 477; recurvum, 477; ros- tratum, 485 ; Rutenbergianum, 484 ; sesquipedale, 475, 478; sp., 405; spathulatum, 478; superbum, 480; teretifolium, 484; virens, 480. Anonacem, 26; of Philippines, 290, 293, 294. Anosporum, subgen. of Cyperus, 25, 27, 33, 34, 72. cephalotes, 34; collocata, 117; monocephalum, 34; pallidum, 27, 118 ; vulgaris, 35. Anplectrum, 287. Anthemis brachycentros, 531; Chia, 531; Cota, 531; graveolens, 533; INDEX. hydruntina, 533; montana, 533; peregrina, 530. Anthistiria, 253. Anthoceros, 268. Anthospermum, 394. Anthyllis Hermannie, 529. Antidesmus montanum, 299. Antimonius, 298. Antirrhzea pbilippinensis, 312; sp. 312. Aphananthe, 296. Apium australe, 219; graveolens, 219; ranunculifolium, 219. Aplogonum Baileyii, f. 635. Aplostellis ambigua, 498. Apoeynacem of Patagonia, 226; of Philippines, 291, 313. Apodocepbala pauciflora, 417. Apogamy in species of Ferns, 365. Apospory in Ferns, F. O. Bower on, 60. tetragona, in Polystichum angulare, var. pulcherrimum, 362. Aracez (Tonglo), 390. . Aralia bipinnatifida, 388; pseudo-gm- seng, 388. Araliacez of Philippines, 291, 310; of Tonglo, 388. Ardisia longipes, 421; ?macroscypha, 490 ; myriantha, 419 ; oligantha, 419 ; umbellata, 420. Areca (Timor-Laut), 372. Arethusa simplex, 498. Argemone mexicana, 303. Arisema Griffithii, 386, 390; Hooker- ianum, 386 ; Jacquemontii, 390 ; nepenthoides, 390; speciosum, 390; utile, 386. Aristea, 894 ; alata, 405. Aristida, 297 ; setifolia, 237. Aristolochia altissima, 530. Aristolochiacese (Philippines), 291, Arjona, 234 ; patagonica tuberosa, VAT. patagonica, 234. Arnoldia minutula, 263. Arnottia, 517. Aroidex, 297 ; of Philippines, 292. Artemisia, 392, 393, 394. Arthonia cinnabarina, 262. . Arthrodesmus apiculatus, 644, 654; arcuatus, 644, 654; gibberulus, 644; incus, 644, 654; octocornis, 644; sub- ulatus, var. gracilis, 644, 654. Arthrolobium repandum, 530. . Articulated laticiferous vessels Wm Hevea, D. H. Scott on, 506. Artocarpus incisifolia, 371 ; integrifolia, 371. INDEX. 667 Arundina speciosa, 301, 541, 543, 544, 548, 549. Asclepiadex of Patagonia, 226 ; of Phi- lippines, 291. Ascomycetes, 261. Aspidium capense, 240, var. cristatum, 365; falcatum, 363; Filix-mas, 365 ; Lonchitis, 644. Asplenium (Athyrium), 254; aniso- dontum, 299; Brackenridgei, 297 ; faleatum, 374 ; (Drepanophyllum) Hohenackerianum, 254; macrocar- pum, 254; persicifolium, 297 ; scandens, 297; sorsogonense, 299; jyightianum, 299 ; Woodwardioides, Asterolinon serpyllifolium, 226. Asteropeia, 328. Astilbe rivularis, 388. Asystasia chelonioides ?, 373. Athamanta multiflora, 531. Aria, three forms of proliferation in, Athyrium, singular mode of develop. in Filix-femina, 354, 358, 359, 360, 361; Filix-feemina, var. clarissima, 354, 358, 360, 361, var. plumosum divaricatum, 354, var. plumosum elegans, 357. Atriplex lampa, 233,.var. angustifolia, oe i pamparum, 233; patagonica, Aucuba japonica, 389. Avena condensata, 530. Azorella, 219, 268; glebaria, 219. Baccharis artemisioides, 221 ; coridifolia, 221; Gilliesii, 221; glutinosa, 221; salicifolia, 221; tenella, 221. Bacillariacex (Afghan), 244. Beckea, 287. Balanophora (Timor-Laut), 372. Balanophoree (Philippines), 292. Ball, J., Contributions to Flora of N. Patagonia and adjoining territory, Ballia, continuity - protoplasm in, 616; callitricha, 603, 612, 615, 616, | 620. Bambusacex (Philippines), 293. Bambusina Brebissonii, 635. Barleria, aff. repenti, 403. Baronia, 337. Bartsia, 392, 393, 394. - Batrachium confusum, var. hetero- phylla limosa, 531. Bauhinia, 254 ; Blancoi, 373. Beciardia macrostachya, 495. Begonia, 253, 347; Cathcarti, 388; | heteropoda, 347 ; Lyallii, 347 ; nossi- beea, 347 ; quercifolia, 300. Begoniacem, of Philippines, 291; of Tonglo, 388. Bellevallia comosa, 526. Benitzia minima, 658. Benthamia spiralis, 502. Bentinckia Conda-panna, continuity protoplasm in, 598. Berberidee of, Patagonia, 211; of Tonglo, 387. Berberis heterophylla, 211; insignis, 387 Berkheya, 394; Spekeana, 401. Berteroa obliqua, 529. Bicoronella, (Madagascar) 456, 500; gracilis, 500, 501, 502, 522; longi- folia, 500, 501; parviflora, 500, 522. Bidens chrysanthemoides, 223 ; heleni- oides, 223; leucantha, 401; pilosa, 401. Bignoniacez (Philippines), 291, 313. Biswarea tonglensis, 388. Bixinez (Philippines), 290. Blaeria, 394. Blechnum, 239; hastatum, 239. Bletia sylvatica, 466. Blumea tetraptera, 313. Blumenbachia m ultifida, 218. Bobartia, 105, 111. Bohol (Philippines), plants of, 301. Bolus, H., Restiaces, S. African, coll. by, 591. Bonatea cirrhata, 511; incarnata, 510. Boopis anthemoides, 220; laciniata, 220 ; scapigera, 220. . Boraginex of Patagonia, 227 ; of Phi- lippines, 291 ; of Tonglo, 389. Borassus (Timor Laut), 372. Bornet, Recherches Gonidies des Lichens, 262, 263. Botanic notes from Darjeeling to Ton- glo and Sundukphoo, by C. B. Clarke, 384. Botany of Timor Laut, 370. Botryanthus commutatus, 529. Bower, F. O., on Apospory in Ferns, 360. Brachylepis Candolleanus, 226. Bracts of Cyperus, nature of, and as a character, 9, 10. Brassica campestris, 212; incana, Bridelia coccolobefolia, 441. Briza Lamarckiana, 228. Brizopyrum spicatum, 238. Brodiza aurea, 235. Bromeliaces ot Patagonia, 234. Bromus unioloides, 239. 668 Bruguiera caryophylloides, 373. Bryophagus gloeocapsa, 279. Bryum cespiticium, peculiar growth in, 366; viridulum, 551, 558. Buchanania nitida, 300. Buchnera urtiezefolia, 296. Buddleia axillaris, 424 ; Colvillei, 389 ; sphzrocephala, 425. Bulbochsete pygmæa, 249. Buibophyllum, 462, 464, 465 ; Baronii, 463; clavatum, 464; erectum, 464; Hildebrandtii, 464 ; longiflorum, 465 ; minutum, 463; multiflora, 463; nu- tans, 463, 464; occlusum, 464, 465 ; Thompsonii, 464. Burmanniacez (Philippines), 292. Burmese Desmidiese, with Description of New Species from Rangoon, by W. Joshua, 634. Burseracem (Philippines), 290, 294. Buxbaumiacee, 552. Byssocaulon niveum, 262. Cadia pedicellata, 338 ; pubescens, 338. Cesalpinia pulcherrima, 373. Calamintha suaveolens, 531. Calanthe, 465 ; conspicua, 901; sylva- tica, 466; veratrifolia, often barren, 539; versicolor, 466. Calceolaria mexicana, 390. Calendula fulgida, 530. Callicarpa americana, 315; bicolor, 315 ; Blancoi, 815; Cumingiana, 315. Callistemma brachiatum, var. Sibthorpi- ana, 581. Callithamnion, continuity protoplasm in, 602, 608, 615, 619. Calodendron, 394, 395; capense, 394, 395, 399. Calothrix, 263. : Calycerese of Patagonia, 220. Calyciflorse of Madagascar, 336. Calycophyllum grandiflorum, 311. Calysaccion obovale, 373. Campanula Rosani, 536; Tenorii, 536 ; versicolor, 536. Campanulacee, of Philippines, 201 ; of Tonglo, 389. Campteria biaurita, 657. Canavalia obtusifolia ?, 373. Cannamois, 591 ; scirpoides, 594; sim- plex, 593; virgata, 591, 592. Capnodium Footii, 268. Capparidez (Philippines), 290. Caprifoliacese (Philippines), 291, 310. Carapa moluecensis, 373. Cardamine trifoliolata, 387. Cardopatium corymbosum, 531. Carex, 12, 24, 236, 253, 297; Baroni, 451; bengalensis, 391 ; Cumingiana, INDEX. 297, 301 ; Cumingii, 297 ; Daltoni, 391; decora, 891; fibrata, 297,301 ; fili- cina, 991; hispida, 528 ; incrassata, 236 ; nepalensis, 391 ; nubigena, 391; pellucida, 391 ; phacota, 391; phala- roides, 236 ; polycephala, 391 ; remota, . 391 ; riparia, 236 ; serrulata, 529 ; sp. (Kilimanjaro), 406 ; vesiculosa, 391. Carica Papaya, 303. Carionia elegans, 310; 310. . Carissa edulis, var., 431; Schimperi, triplinervia, Carruthersia, 297 ; pilosa, 301. Carum multiflorum, 531. Caryophyllacex (Tonglo), 387. Caryophyllee, of Patagonia, 213°; of Philippines, 290. Cassia cæspitosa, 209; didymobotrya, 400 ; javanica ?, 373. Cassibeera triphylla, 239. Casuarinee (Philippines), 292. Caucalis, 393 ; infesta, 393, 400; mela- nantha, 395, 400. Cebu (Philippines), plants of, 301. Celastracese (Tonglo), 387. . Celastrinex, of Patagonia, 215; of Phi- lippines, 290, 294; of Tonglo, 387. Celastrus stylosa, 387. - Census, genera and sp. plants of Philip- pines, 290. ] Centaurea alba, var., 533, 534, 539; amara, 534; angustifolia, 530; Auberti, 466; deusta, 528, 534, var- nobilis, 535, var. tenacissima, 534 ; divaricata, 536 ; hybrid, 534 ; Lippi, 535. Centrosis sylvatica, 466. Cephalanthera grandiflora, 539. Cephalostachyum Chapellieri, 454. Ceramium, continuity protoplasm 1D, 600 ; rubrum, 606,607, 611, 614, 616, 618, 621. . Cerastium, 393 ; arvense, 213; chilense, 213; Commersonianum, 213; semi- decandrum, 213, var. unilaterale, 398 ; vulgatum, 393, 398. Ceratodon purpureus, 366. Ceratophyllez (Philippines), 292. Cestrum Parqui, 229. Cheetacme madagascariensis, 443. Chailletiacez (Philippines), 290, 294. Chamabaina squamigera, 390 Chara, 249. Characee (Afghan), 249. Chascolytrum subaristatum, 238. Cheilanthes micropteris, 239. - Chelidonium, laticiferous vessels of, 573. Chenopodiacem, of Patagonia, 232 ; of Philippines, 291. INDEX. Chenopodium album, 232; ambrosi- oides, 233; murale, 232. Chiodecton myrticola, 262; nigrocinc- tum, 262. Chirita Kurzii, 390. Chlorophyllacez, 261, 262. Chondrus, continuity protoplasm in, 607, 614; ?adriatieus, 602, 604; crispus, 604, 605; mamillosus, 604, 605, 618. Choranthacez (Philippines), 291. Chroococcus, 241, 242, 243, 250. Chroolepidez, 261. Chroolepus, 271. Chrysodia vulgare, 657. Chrysoglossum, self-fertilization in, 547, 548, 549. Chrysomyxa ledi, 370 ; development of uredospores in, 370. Chrysopogon, 236 ; nutans, 237; sti- poides, 236. Chrysosplenium lanuginosum, 388 ; ne- palense, 388. Chuquiraga erinacea, 224 ; hystrix, 224; Kingii, 224, 225 ; spinosa, 225. Chylocladia articulata, continuity pro- toplasm in, 607, 616, 619, 621. Cichorium spinosum, 528. Cinchona, hybridization in, 378. Calisaya, 376, 377, 378, 379, var. Ledgeriana, 374,375, var. microcarpa, 375, 379, var. pallida, 37; Ledgeriana, 974. 375, 376; micrantha, 377, var. calisayoides, 375; officinalis, 376, 378, var. Condaminea, 378; succirubra, 376, 378. Cinchona-bark, chemical analysis of, 376. Cineraria abyssinica, 401. Cirrhopetalum, 465, of Madagascar, 456 ; Thouarsii, 457, 465. Citrus vulgaris, crystals on, 621. Cladium Melleri, 451; pantopodum, 451; xipholepis, 451. Cladonia, 272; coccifera, 273, 282; pyxidata, 273. Claraz, G., List of Plants in N. Pata- gonia and S. Province of Buenos Ayres, J. Ball on, 211. Clarke, C. B., Botanic Notes from Dar- Jeeling to Tonglo and Sundukphoo, 384; Notes on Flora of Parasnath, à mountain of N.W. Bengal, 252; on Indian Species of Cyperus, with re- marks on some others that specially illustrate the Subdivisions of the Genus, 1. Clathrospermum, 319. Cleistanthus Blancoi, 315; ferrugineus, Clematis, 393, 394; Bojeri, 317; cir- 669 rhosa, var. balearica, 527; edentata, 318; grata, 318 ; laxiflora, 317; mau- ritiana, 317; microcuspis, 317; oli- gophylla, 317; simensis, 318, 397; Thunbergiana, 394 ; Thunbergii, var. sericea, 397 ; trifida, 317. Clerodendron, 298 ; brunsvigioides, 435 ; intermedium, 391. Closterium, 247, 652; acerosum, 652; acutum, 652; antiacerosum, 652; bacil- lum, 652, 653; Cosme, 247 ; costatum, 652; Diane, 652; didymotocum, 652; Ensis, 652; incurvum, 652; intermedium, 652; juncidum, 652; Lagoense, 652, 653; Leibleinii, 652; lineatum, f. Sandvicensis, 652 ; Nematodes, 652, 653; porrectum, 652; Ralfsii, 652; setaceum, 652; strio- latum, 652; subtile, 652; Venus, 652. Cluytia lanceolata, 404; Richardiana, 404. Cnicus lanceolatus, 224. Coast Flora of Japygia, S. Italy, H Groves on, 523. Cochlostyla, 285. Cologyne corymbosa, 390; (Pleione) Hookeriana, 390. Ocnogonium confervoides, 262; Linkii, 262. Colchicum Bertolonii, var. Cupani, 530. Colea parviflora, 428. Coleochsetacese (Afghan), 249. Coleochete scutata, 249. Coleosporium Senecionis, 370 ; develop. uredospores in, 370. Collema, 260, 263, 279 ; evilescens, 279 ; furvum, 283; glaucescens, 265. Columbia Blancoi, 308 ; floribunda, 308. Combretaces (Philippines), 290, 294, 310. Commelynacez, of Philippines, 292; of Timor Laut, 371. Compositze, 312; of Patagonia, 220 ; of Philippines, 291, 312; of Tonglo, 389. Conferva, 249. Confervacez, 261, (Afghan) 249, (Phi- lippines) 292. Connaracez (Philippines), 290, 294. Conocephalus ovatus, 299. Conomitrium, 552, 554. Continuity of Protoplasm, S. le M. Moore's obs. on, 595; in Floridesm, views of diff. authors thereon, 603, 604. . Contrib. to Flora of N. Patagonia, 202. Contrivances for Self-fertilization in Tropical Orchids, H. O. Forbes on, 538. Convolvulacez (Philippines), 291. 670 Convolvulus dissectus, var. angustifolia, 209 ; tenuissimus, 530. Conyza tetraptera, 313. Cora, 271. Cordia subcordata, 373. Coreopsis abyssinica, 395,401 ; sp. nov. ?, 401 Cornacese of Philippines, 291; of Ton- glo, 389. Corymbis, 498; corymbosa, 457, 498 ; disticha, 498 ; of Madagascar, 456; Thouarsii, 498 ; veratrifolia, 498. Corymborchis(Corcurborchis)Thouarsii, 498. Cosmarieze, 634. Cosmarium, 245, 645; abruptum, 247 ; Aitchisonii, 246, 250; amenum, 646; annulatum, 648; armatum, 646, 654 ; Baileyii, 648; capax, 645, 655; bi- cardia, 648; bifarium, 649, 655; Botrytis, 245, var. indicum, 645, 654, B. mesoleium, 646 ; consper- sum, 646; contractum, 648 ; cre- natum, 645; Cucumis, 645; cu- curbita, 615 ; cuneatum, 647,654; de- pressum, 648; De Baryii, 645; Dia- dema, 646, 655 ; elegantissimum, 648 ; exasperatum, 649, 654; geminatum, 648; globosum, 646; granatum, 247, 645; Hammeri, 650; Hookeri, 246, 250; incisum, 648, 654; inornatum, 648, 654; latum, 646; Meneghenii, 247, 646, 648; microsphincta, 246; minutum, 247; monomazum, 647; nitidula, 246; norimbergense, 646; obsoletum, 645; Oliveri, 246, 250; pachydermum, 648, 6. minus, 648 ; Pardalis, 635, 654, f. minor, 648, 654; parvula, 247; phalangula, 646 ; phaseolum, 246 ; platydesmium, 648; Portianum, 648; pulcherri- mum, 246, 645; punctulatum, 647 ; pygmeum, 648; pyramidatum, 247, 646, var. simplex, 247, 250; quadratum, 648; rectangulare, 646 ; Reinschii, 647 ; retusum, 650; smo- landieum, 645; spinosum, 647, 655; subrotundum, 648; subtumidum, 648; tetragonum, 648; tetroph- thalmum, 646; tholiforme, 635, 645; turgidum, 645; undulatum, 246, y. minutum, var. ornatum, 246, 250, 648; venustum, 650; vosgesia- cum, 647. Crassulacez, of Patagonia, 218; of Philippines, 290. Crepis rubra, 531 ; sp. ?, 401. Crocus Thomasii, 537. Crombie, J. M., on the Algo-Lichen hypothesis, 259. INDEX. Crossandra, aff. undulefolix, 403. Crotalaria Dilloniana, 399; Thomsoni, 399. Croton Cumingii, 297. Crotonoide:s, laticiferous 311. Cruciferx, of Patagonia, 212; of Philip- pines, 290 ; of Tonglo, 387. Crudia Blancoi, 909 ; spicata, 309. Cryptogamia (Philippines), 292. Cryptopus elatus, 496, 497. . Cryptostylis Arachnites, self-fertilization in, 547. Cucurbitaceze of Philippines, 291; of Tonglo, 388. Culm of Cyperus, variations in, 6. Cuming’s Malacca plants described as from Philippines, (footnote) 287. Cuphea, 209, 218; glutinosa, 209 ; spicata, 218, var. racemosa, 218, _ Cuphocarpus aculeatus, 350; inermis, tubes of, Cupuliferz (Philippines), 292, 294. Cyanophycez (Afghan), 241. Cyanotis hirsuta, 406. Cyathea integra, 299. Cyathula, an globulifera ?, 403. Cycadacex (Philippines), 292. Cyclostemon Cumingii, 800. — Cylindrocystis, 242; Brebissonii, 653. Cylindrospermum, 241, 250. Cymbidium, 472 ; equitans, 457 ; flabel- latum, 472; Huttonii, 472; Stape- lioides, self-fertilization 1n, ; umbellatum, 465. Cymbopogon, 253. Cynoglossum lanceolatum, 402. — Cynometra bijuga, 373; ramiflora, 373. Cynorchis, 513, (gEucynorchis) 512, (§Parviflore) 513. —— of Madagascar, 456. angustipetala, 512, 514; Arnot- tioides, 512, 513, 516; Boryana, 512, 513, 516 ; brevicornis, 518, 516; calanthoides, 512, 514, 515; fastigiata, 512, 515; flexuosa, 513, 514; gibbosa, 512, 514; grandiflora, 512, 513, 514; hispidula, 513, 516, 517; lilaeina, 513, 515; parviflora, 512, 513; purpurascens, 513, 515; uniflora, 512, 5138. . Cyperacer, of India, 23; of Patagonia, 236; of Philippines, 292; of Tonglo, 391. Cyperus (Indian), bracts of, 9 ; culm of, 6; glumes of, 18; leaves of, 8; nut of, 2; rhachilla of spikelet, 16; rhizome of, 2; stamens of, 19; etyle of, 20; umbel of, 11. as, Indian species of, Mr e. B compressus, „ ô, 16, 18, 96, 99 103. yke 01, : 154; concolor, B73 6, Tertus, 41 1057 conspec t subgenera» . congestus, ot, 182; conglomerati osporum: d pielidium: 3&5 | 5, AT, 91, Mb 113, var: g ME 112, icy peru 933; Juncellus, a9; Ma |o var Y 6, var. ? " gocotrensits {13s 3cu3, pycreus 33 contrac ‘ r mandelints 39, alopecuro d dives 104; © us, correctus: 21 ompar : + jnunt mà ©. corymbos , M mbosus, 7, 9, yrocerus € Y ,* € ygmaeus 10, Ins 9, yacros chy and Jsolept M naco pared, 159; Qossyrensi» qi; erue us, Ob» pysinien ; cuminatus, i73 633 curvatu 7, curve’: 5, 12, adustus, 9! * ptiacus, 22, : 119; cuspid tus, 8 pgus equals, 29 ; NUUS 62, eh gifolia, 97 M Burchell, var en yifola, 25 albidus, 573 „erinus, , var B panice 0 albomargt ; \bostriatus, cyrtole pi ; vrtostachy?» 59; 119, 131; \ ecuroide ; 15, damietten® , 2g.” dehiscens aA, 22, 74, 187) , 2 2, alterni densifolius, ensus, j. dente olius, 19> 7, 130, tus, 1803 datus, M 169; dep us, ; mabilis, 95 auropu? diandrus, 91, 69; B castaneus, 55 3, 1825 piguus, 62; 8" (enus, iaphantr d, ; diaphanum. 72 4, 166, 1873 dersontb 196 ; diaphanWw 72, 99 gi, var B ngulatus, ^^ 60, 615 gusti- Nicarj* difform; 19, 18b folius, hyllus, 120; 133; asus, 13s 1217 digitatus» uatilis, 5. arenarl’” by, 22, 9h 125, 188; qilutus, ©? 93, 194; 106, 173 areolatus, T5 entinus, jphyllus T , 15383 dissitiflorus, 04; ; aristatus, Pi" ,02, 93, $ i gissolutus: 5, 145; distacby 99 i95 B- Maingasi, 94; articulatus 4T, 8 distans, 9, 1493, 144, 5,4 sicho 10, 155, ipo; sten 49, 955 Atkin- ph \ius, 9 W, 615 divaricatis 415 soni, ; atratus, DT 3 troferru divergen? 62; dive? 5, 2l 187, ineus, 9; at onervatus, 67; atro- 23 atesti, 188; di- nitens, , A9, ; 5i 66, Ẹ Drur mondii, 16; atropur pure" Aucherl, 22,113; 44, We» 197 effusus, 109, aurantiacus 85; aureus, ap, Vi elatus, 5, 189, 190, var. B- auricomus 6, ` 88, 89, onux, elegans, 9: 14, 7A, australi 1; 8 tumnalis: 195; jega tis, 125, 190; ele- Babakan, 151; pakensis, 6, Y ntulus, 66; eleusm 85 131, badius, . 165, 1 3 9. Y 143, ; j|iotianus 62, 198 ; artert, 6 ; be hal : ngatus, , 169; ema gina- 147, 151}, ,. v olor, 167; tus, 199? eminens 190; enodis, 1 Q8 lovi 5 pighum's ; , 151; Eragrostis, o, 425 48, bh Boivin, 63; vorneensiss 45 ; pi, 59: 63, 65 var. P cyrtosta che chiatus, ; vracbyphyllo^r 85; go, vers V. neurotropi go, var. € prachys achyus 165 196 ; micront 59, var minor, p9, var- zeus, 9b ; romoides, eros hys, Í y. ô ytolepi?: Browne, Vio Jpiferus. Y 59; er thræus, 104; esculentus, pillosus, eo; V lboso-8tO oniferus 1o, 167, 9, Yi8 181, 2 2, var g 113; Jposus, 175, 1825 helodes, igi, var xta, igi; ew jnerassa , 110; caducum. morphus, 33; € \tatus, 4, 19 23, qucus, 118; pito us, 9b 143, 185, 186, 187 a nalicula 1 esce amoena gi, var dives, 187, var 195; pillaris, ri Oatesi, 23 ; exiguus, ;, e pansus 152; castanen”> , 88, 89; ru falcatus, 1 Fajcieulosus, 9 nensis, 120; ee hal nthus, fascicular ; 41,91 fasciculo is, 02 cephalotes, 5, 9, ferax, i; fer neus, É yx H 117, 4 2, tructurall festivus, o; Ki jdingt, 122; fli 26 rysanth s, 61 nus, ke 202, 5515 flieulmi* $ 138 cimicinus eT; 198, 199 fim piat us. 149; flaccid 49, 1415 Cleaver 56 ; 85; flavesce?® 10, 19 93, 24, 30, 40 ; comosus, 67 69; , 42, 4% 9i 3, 65, 69, 72 INDEX. 65, var. 6. Fontanesii, 38, var. y. paraensis, 38, 8. rubromarginatus, 6 ; flavescentis, 37, 38, 39; flavicomus, 69, 70, 71; flavidus, 3, 4, 8, 19, 47, 122, 202, var. africana, 121; flavis- simus, 110; flavus, 15, 33, 196; flexuosus, 16; fluitans, 28, 118; foliosus, 54; Fontanesi, 39; For- skalei, 135, var. £. virescens, 187; fugax, 55; fulvescens, 182; fuscus, 119, 135, var. protracta, 137, var. virescens, 104; gangeticus, 147; Gatesii, 55; geminatus, 175; glaber, 104; glareosus, 86 ; globosus, 17, 18, 20, 24, 47, 50, 51, 54, var. B. nilagirica, 49, var. y. stricta, 49; glomeratus, 84, 96, 140, 141, 184; gracilescens, 178; gracilinux, 162, 169; gracilis, 120, 173; gra- mineus, 64; graminicolus, 145; Grayii, 198; Griflithii, 6, 95, 96, 101, 102; grossarius, 57; Gula- Metthi, 157; Gussoni, 37; gymno- leptus, 44; gymnos, 155; hzmato- stachys, 116; Haspan, 3, 4, 7, 8, 9, 10, 19, 20, 22, 23, 47, 119, 122, 133, 202; hebes 149; Heldreichianus, 163; Helferi, 128, 160 ; helodes, 171; helvus, 62; herbivagus, 177; Her- manni, 178, 179 ; hexastachyus, 167, 170, 171; Heynei, 152; Hochstetteri, 69, 70, 71, var. B. pinguior, 70, var. y. russa, 71; holosericeus, 55; honestus, 149 ; Hookeri, 189; Hook- erianus, 34, 54; hyalina, 180; hya- linolepis, 82; hyalinus, 19, 46, 94; Hydra, 167,172, 178 ; hyemalis, 120 ; imerinus, 450; incompletus, 194 ; in- crassatus, 55; incurvatus, 147, 148; indeterminata, 99, 120, 125, 133, 142, 145, 148; inflexus, 91; infra- apicalis, 142; intactus, 54; inter- medius, 42, 166, intermedius, var. indica, 40; intricatus, 91; inunda- tus, 14, 15, 19, 20, 21, 22, 31, 32, 68, 73, 74; inundatus and C. procerus compared, 31; involutus, 113; Iria, 23, 131, 132, 137, 138; Iria, var. flavescens, 138 ; iridifolius, 127; isch- nocornus, 66 ; ischnos, 197 ; Jacquini, 143; jeminicus, 4, 10, 15, 59, 84, 95, 169, 175, 177, 202; junceus, 120; junciformis, 79; juncoides, 120; jungendus, 47 ; Junghuhnii, 49; Khasiana, 48; Koenigii, 158 ; Kur- zii, 129; Kyllingimoides, 34, 197; Vetus, 5, 181, var. y, 184; levigatus, 3, 10, 18, 19, 20, 21, 77, 79, 106, 195, 202, var. (3. junciformis, 79; La- marckianus, 47, 164; lamprocarpus, 173; lagorensis, 127 ; Lagunetto, 64: lanceolatus, 42, 47; lanceus, 50, 60, 60, var. B. Grantii, 66, var. 8. macrostachyum, 66; lateralis, 77; lateriflorus, 166; laterifolius, 152; laticulmis, 120; latispicatus, 6, 8, 40, 42, var. B. acaulis, 41, var. y. aphylla, 41; laxus, 125, 175; Lech- leri, 9, 116, 117; lepidus, 85, 124; leptophyllus, 182; leptos, 120 ; lepto- stachyus, 122, 167; leucocephalus, 34, 85, 105, 107, 108, 111, 140; leu- colepis, 61; leucostachys, 77, 450; levis, 59; ligularis, 196; limosus, 81 ; littoralis, 168, 173; lividus, 49; lon- gifolius, 130; longus, 4, 5, 140, 160, 161, 162, 163, 167, 169, 170, 174, 202, var. e. adoensis, 166; var. e. elon- gata, 166, var. 7. gracilis, 166, var. C. maculatus, 166, var. 9. pallescens, 161, var. pallida, 165, var. tenuiflora, 174; Louisiane, 56; lucida, 24; lucidulus, 5, 15, 16, 18, 84, 99, 183, 202; lucidus, 9, 13, 48, 194; luteolus, 60 ; lutescens, 181, 182, var. B. fulvescens, 182; Luzule, 13, 19, 116, 117; macer, 125, 154, 170; macranthus, 42, 66 ; macropus, 101; macrostachyus, 69; maderaspa- tanus, 94; malaecensis, 7, 147; mar- garitaceus, 96, 110; marginellus, 149, 150; maritimus, 167, 184; Martia- num, 118; Maximiliani, 62; maxt mus, 190; megapotamicus, 68; me- lanocephalus, 57, 66 ;. melanopus, 66; melanorrhizus, 178 ; melanostachyus, 18, 65, 67; membranaceus, ; Meyenii, 97; Michauxianus, 192; Michelianus, 82; microcarpus, 4, 122; microdontus, 55, var. 8. texensis, 55 ; Microiria, 138; microlepis, 137, 138; microstachyus, 85, 125; minimus, 56; minor, 54; mitis, 165; moestus, 125; monocephaloides, 34; monoce- phalus, 34, 82 ; monogynus, 34; mo- nophyllus, 147, 158; monostachyus, 71,195 ; Monti, 16, 17, 19, 20, 21, 31, 32, 72, 202, var. B. stylosa, 79 ; mucro- natus, 47, 77, 79, 80; Mulenpulla, 82; multifolius, 194; multiplicatus, 125, 129, var. Kurzii, 129; multistriatus, 14; Mundtii, 6, 43, 58, 63, 64, var. B. distichophylla, 64; mutica, 18; my- riostachyus, 165; natans, 28, 34; Neesii, 188, 189; nemacaulos, 193; nervosus, 178; neurotropis, 59; Dt gricans, 42; nigroviridis, 125; nila giricus, 50; niloticus, 155; nitens, 12, 18, 43, 45, 104; nitidulus, 194; nitidus, 57,66; niveum, 110; niveus, OPERUM, INDEX. 673 96, 108, 109, 197 ; nodosus, 7 ; nudi- eulmis, 161; nudus, 120, 155, 157; nutans, 140, 143, 145; Nuttallii, 13, 55; obstinatus, 44; obtusiflorus, 96, 110, var. 8. flavissima, 110; ochre- oides, 164 ; odoratus, 170 ; officinalis, 167 ; oleraceus, 175 ; Olfersianus, 24, 62, 63, 65 ; olidus, 55 ; oligostachyus, 85; olivaceus, 54; olivaris, 167; ornatissimus, 69; ornatus, 152, 153; ovularis, 15, 199, 200, 201; oxylepis, 114, 115; pachyrrhizus, 111, var. (3. minor, 111 ; pallescens, 169; pallidus, 27, 115, 118; Pangorei, 147, 160, 161, 172; paniceus, 201; panicoides, 138; paniculatus, 51, 55; pannoni- cus, 19, 20, 80; Papyrus, 11, 123, 191; paraensis, 38; paraguayensis, 90; parciflorus, 125, 138; parvulus, 64; patens, 18, 44, 45; patente, 45; patuliflorus, 69, 70; patulus, 104; pauciflorus, 149; ^ paucispiculatus, 450 ; pauper, 10, 41; pectinata, 40 ; pectinatus, 90, 104; pectiniformis, 90, 97, 98; peduneulosus, 131; pen- natus, 151, 194; permutatus, 42, 67 ; persicus, 106; pertenuis, 159, 169, 174; petrzus, 104; phymatodes, 178, 179; piceus, 43; picreus, 43; pictus, 32; pilosus, 12, 16, 20, 21, 32, 54, 96, 147,148, 150, var. 6. babakensis, 151, var. B. obliqua, 151, var. y. poly- antha, 151; piptolepis, 149; platy- culmis, 120; platyphyllus, 7, 9, 20, 202; platystachys, 170; platystylis, 5, 12, 19, 20, 22, 27, 28, 115, 117, 202; plenus, 54; pleuranthus, 77; pozformis, 37; polystachyus, 5, 12, 13, 18, 51, 154, 184, 202, var. «. Cleayerii, 56, var. ferruginea, 12, 51, 54, var. filicina, 55, var. č. holosericea, 55, var. 8. laxiflora, 53, var. leptostachya, 55, var. macrostachya, 56, var. e. mi- cans, 54, var. y. minor, 54, var. y. paniculata, 55; Preslii, 166; proce- rulum, 172; procerus, 4, 5, 9,14, 15, 16, 20, 31, 32, 68, 138, 147, 150, 152, 153; — andC.inundatuscom pared, 31; propinquus, 62; proteinolepis, 5, 96, 113, 167; protractus, 133; pseudo- bromoides, 40, 61 ; pubisquama, 127 ; pugioniformis, 82; pulcherrimus, 115, 119, 132; pulvinatus, 44 ; pumilus, 2, 12, 14, 18, 43, 44, 46, 47, 57, var. B. mutica, 45, var. punctata, 25, var. y. punctata, 46; punctatus, 46; puncti- culatus, 14, 16,68, 202, var.8. quinqua- gintitlora, 69; pungens, 113; Purshii, 1; pusillus, 87, 90 ; pygmeus, 14, 21, 22, 23, 28, 29, 30, 72, 83, 91, 103, LINN. JOURN.— BOTANY, VOL. XXI. 202, var. 3. Aztecorum, 83; quinque- florus, 151; quitensis, 85 ; racemosus, 125, 142, 188, 190, var. B. khasiana, 189; radians, 6, 100, 101, 102, 103; ra- diatus, 185 ; radicans, 101 ; radicosus, 167, 169; Ranko, 16, 64; reflexus, 116; repens, 178; reptans, 3, 63; resinosus, 138; retrofractus, 7, 15, 199; retusus, 69, 71, 178; Retzii, 172; riparius, 120; rivularis, 21, 37, 65, 77, 78, var. ò. acutata, 67, var. y. depauperata, 65, var. B. eluta, 65; Roestelli, 174, var. y. cyprica, 165, var. €. badia, 165; rotundus, 4, 5, 10, 11, 13, 21, 51, 99, 111, 140, 165, 166, 167, 173, 178, 181, 202, var. acuta, 172, var. carissalis, 99, var. e. laxata, 172, var. pallida, 178, 8. pen- dulus, 175, var. procerula, 174, var. a. Salsola, 171, var. 6. 100-flora, 171, var. y. Amalis, 171, var. 6. procerula, 172; Roxburghii, 190; rubicundus, 18, 40, 95, 96, 104, 105, f. rubiginosus (=C. viscosus), 115; Salzmanni, 88 ; sanguinolentus, 57 ; Santonici, 138, 142; scariosus, 4, 8, 159, 202 ; Schimperianus, 8, 160, 162; Schom- burgkianus, 107 ; Schweinfurthianus, 99; Schweinitzii, 103; scirpoides, 127 ; scopellatus, 55 ; Seemannianus, 16, 195; seminudus, 138, 158, var. Pangorei, 159; semitrifidus, 177; Serra, 104; seslerioides, 85; seta- ceus, 37; setifolius, 198; setigerus, 183; Sieberi, 142; Sieberianus, 178; silletensis, 115, 119, 132, 133; so- lidus, 177; solutus, 89; Sonderi, 51; sorostachys, 107, 140; spa- diceus, 152, 172; spaniophyllus, 147; spectabilis, 140, 142; spha- celatus, 5, 96, 154, 183; spheero- cephalus, 110, B. leucocephalus, 110, spherospermus, 124; spinuliferus, 60; squamulatus, 145; squarrosus, 45, 82, 89, 91, 94, var. B. lancinus, 95; var. stenocarpa, 87; Stauntoni, 59; stellatus, 120; stoloniferus, 21, 22, 72, 111, 154, 168, 172, $. pallida, 51, 167, 173; stramineus, 10, 39; strictus, 46, 50, 54; strigosus, 96, 154, 183; subulatus, 51, 53, 149; suleinux, 36, 56; surinamensis, 116 ; tegetiformis, 7, 8, 147, 157, 158, 159; tegetum, 8, 9, 17, 84, 140, 158, 160, 161, 162, 166, 167, 169, 202, var. B. ambigua, 162, 166, var.? y. pro- tracta, 162; tenellus, 56, 65; Tene- riffz, 104, 105; Tenorianus, 178; Tenorii, 178; tenuiculmis, 9, 15, 99, 100, 163; tenuiflorus, 166, 169, 170, 3r 674 174, 175; tenuispicus, 122; tereti- fructus, 47 ; tetrastachys, 167 ; texen- sis, 55; thermalis, 165; Thomasi, 183; Thomsoni, 177 ; Thouarsii, 54; toluccensis, 125; torosus, 152, 153; Torreyanus, 55, torsus, 174; tortuosus, 44 ; tremulus, 71; trinervis, 85; tri- stachys, 43 ; truncatula, 18 ; truncatu- lus, 45 ; tuberosus, 15, 172,173, 178; tunicatus, 120 ; turfosus, 63 ; turgidu- lus, 7, 130; turtuosus, 49; umbellatus, 3, 5, 13, 15, 16, 32, 120, 186, 200, 201, var. e. cylindrostachys, 201, var. 6. laxata, 201, var. y. picta, 201; umbrosus, 125; uncinatus, 90, 91; unioloides, 24, 60; var. angulata, 41, var. B. angulata, 60, var. 6. bromoides, 61, var. capensis, 61, var. y. capensis, 61; usitatus, 4, 15, 84, 176; vagina- tus, 10, 119, 120, 124; variegatus, 62, 67 ; vegetus, 19, 115, 236; venustus, 186, 188 ; versicolor, 91; verticillatus, 185; vestitus, 177; Vicaryi, 81; virens, 9, 19, 116; viridis, 137, 172; viridulus, 3, 77; viseosus, 7, 114; vulgaris, 47; Wallichii, 149; Wightii, 40, 99, 104, 105; xan- thinus, 37 ; xanthopus, 142; Zollin- geri, 15, 99. Cypripedium Argus, 301. Oyrtandracez of Philippines, 291, 294 ; of Tonglo, 390. Oyztopera plantaginea, 471; squalida, Cyrtopodium, 471; bieolor, 472; plan- tagineum, 471. Gystoeoccus humicola, 266. Cytinaceze (Philippines), 291. Dactylococcus infusionum, 248. Dalbergia Baroni, 337. Dalea alopecuroides, 309; nigra, 303, Danais vestita, 408, Daphne papyracea, 390. Darjeeling, Botanical notes on, by C. B. Clarke, 384. Dascycoleum, 295, 298. Datiscacez (Philippines), 291. Davallia, 297 ; decurrens, 299 ; repens, Deeringia celosioides, 373. Delarbrea of Timor Laut, 374; sp., 373. Delesseria alata, continuity of proto- plasm in, 610, 612, 613, 616, 619, Delphinium, 393, 394; macrocentron, Dendrobium chrysanthum, 546; cru- INDEX. menatum, self-fertiliz. in, 538; cultri- forme, 473; fusiforme, 474. Desmides, 634, 635. , Burmese, W. Joshua on, 634. Desmidiacez (Afghan), 245. Desmidium, 247, 635; Baileyii f. tetra- gonum, 635 ; cylindricum, 635 ; qua- drangulatum, 635; quadratum, 247, 635, 653, var. excavatum, 247, 250; Schwartzii, 635. Desmodum umbellatum, 373. Desmostachys, 332; acuminata, 332; deltoidea, 332 ; Planchonianus, 332; Renschii, 332. Diagramsillustrating Apospory in Ferns, 363-367. Dianthus rupicola, 530. Dichonema, 271. Dichostylis pygmea, 82. Dichrocephala chrysanthemifolia, 400. Dichronema candida, 450. Diclidium, 17, 18, 34, 83, 191; as sub- gen. Oyperus, 34, 191. Dicliptera, near maculata, 403. Dicotyledones (Philippines), 292. Dicranum, 552; incurvum, 551, 557; tamarindifolium, 552, 557 ; viridulum, 551; viridulus, 555. Didymocarpus vestita, 427. Dilivaria ilicifolia, 373. . Dillenia indica, 307; philippinensis, 307 ; speciosa, 507. Dilleniacex (Philippines), 290, 307. Dilobeia Thouarsii, 437. Dioscorea acuminata, 449. Dioscoreacex (Philippines), 272. — . Diospyros, continuity of protoplasm m, 602; embryopteris, 600, 602, 618; melanoxylon, 600, 601. fusco-velutina, 422; gonoclada, 424; maritima, 373; megasepala, 423 ; spherosepala, 423. Diplazium, 662. Diplolophium abyssinicum, 395, 400. Dipsacez, Patagonia, 220. Dipsacus, 220 ; sylvestris, 220. Dipferocarpese (Philippines) 290, 294. Dipterocarpus grandiflora, 298. Dirina repanda, 262. Disa incarnata, 521; macrantha, 539; Rutenbergiana, 521. uu Discaria febrifuga, 215; longispina, 215. Disperis Hildebrandtii, 522; of Mada- gascar, 456 ; oppositifolia, 522. Disporum, sp., 391. Distichiz, 552. Distichlis maritima, 238. | ,Distichophylla ($), 552. — ee INDEX. Docidium, 634, 650; annulatum, 651, 655; Baeulum, 650 ; Burmense, 651, 655; constrictum, 651 ; coronulatum, 651, 655; granuliferum, 650, 655; minutum, 651 ; tessellatum, 650, 655 ; verrucosum, 651. Dodonza madagascariense, 335. Dolichos Lablab, 373. Dombeya floribunda, 325, 326; ma- crantha, 325; repanda, 326. Dovea, 574, 575, 576, 579; Bolusi, 576, 591; ebracteata, 577, 593; macro- carpa, 593; mucronata, 590, 591, 593; paniculata, 577; racemosa, 518; tectorum ?, 578, 590. Draba gracillima, 387. Dracena xiphophylla, 449. Dracontomelum Cumingianum, 300. Dracunculus vulgaris, 529. Drepanophyllex, 552. Droseracez (Philippines), 290. Druery, C. T., Further Notes on Repro- duetion of Athyrium, 358; Observa- tions on a Singular Mode of Develop- ment in Lady Fern (Athyrium Filix- fcemina), 354. Dryopeia oppositifolia, 522. Duvaua fasciculata, 215 ; preecox, 215. Dyer, W. T., Report on Botany of H. O. Forbes's Exped. to Timor Laut, 370. East-African plants coll. by J. Thom- son, described by Prof. Oliver, 392. Ebenacez (Philippines), 291. Echinolena madagascariensis, 452. Echinops, 393, 394 ; amplexicaulis, 401. Egyptian tomb, pollen from, 251. Ehretia an var. abyssinice ?, 402. Elxagnacee (Philippines), 292. Elxodendron, 333; griseum, 334 ; ni- tidulum, 333; trachycladum, 338 ; vaccinioides, 333. læoselinum Asclepium, 529. Elatineæ (Philippines), 290. Elatostema diversifolia, 390. Elegia, 574, 575, 579, 590; acuminata, 580, 583, 586, 594 ; asperiflora, 592 ; coleura, 586; deusta, 590, 593, 594; equisetacea, 583; filacea, 589; fistu- losa, 591; glauca, 579, 592, 594; Juncea, 581, 582, 583, 584, 591, 593; membranacea, 581, 582, 583; parvi- flora, 586, 588, 589, 593, 591; pro- piqua, 582, 583, 584; rigida, 587, 92; spathacea, 588 ; stipularis, 587 ; thyrsifera, 583,585 ; vaginulata, 586 ; Verreauxii, 589; verticillaris, 590, 592, 594. Eleocharis, 7 ; ovata, 527. Elettaria, 254. 675 Elynanthus, 590. Embelia Gamblei, 389. Emex spinosa, 529. Emilia angustifolia, var., 401. Encephalartos, crystals in, 621. Encyonema cespitosum, 244; Lunula, 244; ventricosum, 244. Endocarpon, 266 ; miniatum, 274; pu- sillum, 266. Enkianthus himalaicus, 389. Eocene Ferns from Basalts of Ireland and Scotland, J. S. Gardner on, 655. Epacridese (Philippines), 291. Ephebe, 260 ; pubescens, 263. Ephebella Hegetschweileri, 263. Epidendrum, 465; dipterum, 498; ma- erostachys, 495; polystachys, 497 ; umbellatum, 465 ; voluere, 496 Epilobium, 393; oliganthum, 345; palustre, 345; stenophyllum, 395, 400. Epipactis viridiflora, 539. Epithema Benthami, 297. Equisetum, 69. . Eragrostis, 238 ; delicatula, 238 ; maxi- ma, 454; megastachya, 238. Eranthemum, sp., 373. Eria, 301 ; albido-tomentosa, 545, 548; javensis, 546, 548, 549; retroflexa, 301; vulpina, 301. Eria, self-fertilization in, 546. . Erica arborea, 393, 395, 401 ; javensis 546 ; Scoparia, 529 ; verticillata, 529 Ericaceæ of Philippines, 291, 294; of Tonglo, 389. Erigeron bonariense, 221. Eriocaulese (Philippines), 292. — Eriocaulon aquaticum, 450 ; fluitans, 450 ; melanocephalum, 450. Erioglossum edule, 373. zs. Eritrichium albiflorum, 227 ; californi- eum, 227; tenellum, 227. . Erodium cicutarium, 214; Gussonl, 538. . Ervum, 216; agrigentinum, 530; hir- sutum, 216. oe . Eryngium, 219 ; Barrelieri, 530 ; humi- fusum, 219; paniculatum, 219. Erythrina, 385. Erythroxylon, 328. su Erythroxylum firmum, 327; laurifolium, 327. Euangrecum, sect. of Angrecum (Mada- gascar), 480. Euastron, 635, 645, 654. Euastrum, 245, 635, 637 ; ansatum, 637, var. ampliatum, 637; attenuatum, 639 ; bellum, 640; binale, n. f. cras- sum, 637, 654 ; breviceps, 640 ; coral- 676 INDEX. loides, 639, 654; cuneatum, 637; | Felicia, 394; muricata, var. ?, 400. _ decedens, 637; didelta, 637; diver- | Ferns, apospory in, 300; normal life- gens, 640, 654 ; elegans, 637 ; exilis, 640, 654 ; flammeum, 638, 654 ; gem- matum, 637; hypochondrum, 637; cycle and modifications in, 363. , Eocene, from Basalts of Ireland and Scotland, J. S. Gardner on, 655. inermis, 250; monocyclum, 637; | Festuca bromoides, 239; Fenas, 537. obesum, 638, 654; oblongum, | Ficoideæ of Patagonia, 219 ; of Philip- f. scrobiculatum, 637; orbiculare, pines, 291. 640; platycerum, 640; retrorsum, | Ficus, 443; aff. acanthophyllæ, 374; 638, 654; serratum, 639, 653; sinuosum, 637; spinulosum, 250, f. intermius, 245, 637; subloba- tum, 637; substellatum, 637; apodocephala, 445; megapoda, 444 ; podophylla, 444 ; sphærophylla, 443 ; spp. 374; tiliæfolia, 443; tricho- phlebia, 445. truncatum, 639, 654; turgidum, 687; | Filaments in stamens of Cyperus taken verucosum, 637. as subgen. charac., 20. Eucalyptus (Timor Laut), 372. Filices of Madagascar, 454; of Pata- Eucynorchis (§ Cynorchis), 512, 513, 515. gonia, 239; of Philippines, 292; of Tonglo, 391. Eucyperus, 15, 19, 20, 22, 33, 83, 84. Fimbristylis, 22, 27, 28, 117, 128 ; nu- (Indian) as subgenus of Cyperus, tans, 288. 33, 83; 8 A. — Aristati, 85; § B. | Fissidens, 550; adiantoides, 554, 559, Compressi, 95; § C. — Platystachi, 105; $ D. — Bobartia, 110; § E. — Viscosi, 114; $ F. — Luzuloidei, 115; $ G. — Pseudanosporum, 117; $ H. — Haspani, 119 ; § I. — Diffusi, 124 ; § K. — Fusci, 131 ; § L. — Glomerati, 140; $ M. — Marginati, 146; $ N. 1g rymbosi, 153; $ O. — Papyri, Eugenia, 289; javanica, 373; loise- leuroides, 341. Euhabenaria, 504. Euhibiscus, 325. Eulophia, 467, 472; beravensis, 467, 470; bicolor, 472 ; ensata, 470; gal- bana, 469; madagascariensis, 469 ; pileata, 468 ; ramosa, 470 ; reticulata, 470; Rutenbergiana, 460; scripta, 468, 469; spinulosum, 245, var. Oli- vieri, 245 ; vaginata, 467. Eumystacidium, 489. Euonoclea, 662, 666. Euonymus frigidus, 387. Eupatorium, 220, 417; erodiifolium, 220. Euphorbia aleppica, 531; tetraptera, 44 Euphorbiaces (Philippines), 292. ———, Dr. Pax, obs. on anat. and classif. of, 570. Exacum, 2538. Exaltati without stolons, 5. Exocarpus, 336. Experiments, culture of sori of Athy- phm Filix-feemina, by F. O. Bower, Faba, 316. Fatonia lanceolata, 374; pilosa, 374. 560; algarvicus, 555; Arnoldi, 553, 558 ; Bambergeri, 557 ; Bloxami, 553, 558; bryoides, 553, 554, 555, 556, 557, var. cespitans, 556; Closter, 553, 558; collinus, 554, 559; cras- sipes, var. rufipes, 556; cristatus, 559; Curnovii, 553, 556; decipiens, 559, 560; Donnellii, 553, 557; ex- iguus, 553, 557 ; exilis, 551, 553, 559, 556, 557, 558; floridanus, 556; fon- tanus, 553, 556; Garberi, 553, 558; gracile, 557 ; grandifrons, 554, 559 ; Hallianus, 554, 560; Hallii, 563, 558; holomitrius, 555; hyalinus, 550, 554, 558; impar, 553, 554; m- constans, 551, 554; incurvus, 553, 556, 557, 558; insignis, 559 ; intro- limbatus, 553, 555; Julianus, 554, 560; Langei, 554, 559 ; limbatus, 553, 556; Loscosianus, 556; majus, 994, 559, 560; Mildeanus, 556; minutu- lus, 558, 556 ; obtusifolius, 553, 558 ; Orrii, 553, 556; osmundoides, 558, 558; Panizzii, 556; polyphyllus, 554, 559; polypodioides, 554, 559; pu- sillus, 555, var. madidus, 557; Ba- venellii, 553, 557; rivularis, 509, 554; rufulus, 553, 556; rupestris, 559 ; Sardagnai, 555 ; Sardous, 557; sepincola, 555 ; serrulatus, 559 ; stric- tus, 559; subbasilaris, 554, 560 ; sub- grandifrons, 559; subimmarginatus, 555; synanthus, 553, 554; syniscus, 554 ; tamarindifolius, 553, 557 ; taxı- folius, 554, 557, 558, 559; texanus, 553, 556; ventricosus, 553, 556 ; virl- dulus, 558, 555, 556, var. crassipes, 556, var. incurvus, 557, var. pusil- lus, 554, 555, a w 7 INDEX, Fissidens, Notes on European and N. American sp. of, by W. Mitten, 550; list of sp., 553; literature thereon, 551. Fissidentaceæ, 552. Flagellarieæ (Philippines), 292. Flaveria Contrayerva, 223. Flemingia strobilifera, 373. Flora of East Africa, observations on, by Sir J. D. Hooker : affinities with that . of Abyssinian highlands, 395 ; African | mountain vegetation compared, 394 ; European element in, 393; origin of the flora, 395; S.-African element in, 394. Flora of Japygia, S. Italy, H. Groves on, —— of Madagascar, H. N. Ridley on, 456 ; J. G. Baker on, 317, 407. of Mountains E. Equatorial Africa, 392. — — of N. Patagonia, J. Ball on, 202. of Parasnath, N.W. Bengal, C. B. Clarke on, 252. of Philippines, and probable deri- vation, R. A. Rolfe on, 283, 303. of Philippines, census of gen. and spec. plants, 240 ; endemic vegetation, 293, 294; literature thereon, 285; Malayan features of Flora preponde- rate, 205. of Timor Laut, W. T. Thiselton Dyer and Prof. Oliver on, 370, 373. Floridex, continuity of protoplasm in, eoz; views of diff. authors thereon, Feniculum, 350. Foliage, develop. of, as protective against radiation, Rev. G. Henslow on, 624. Forbes, H. O., Botany of Exped. to Timor Laut, 370. , on Contrivances for ensuring Self- fertilization in Tropical Orchids, 538. Fragaria chilensis, 228; Daltoniana, Fragilaria, 245. raxinus Ornus grafted on excelsior, 16. Fritillaria cirrhosa, 391. Frost, effects of, on plants, 627. Fucus, continuity of protoplasm in, 617. Fumaria, 211; agraria, 527; capreo- lata, 211. Furcellaria, continuity of protoplasm in, 606, 614. Further notes on a singular mode of reproduction in Athyrium Filix- femina, var. clarissima, by C. T. Druery, 358. | | | | 677 Gaertnera phanerophlebia, 425; phyl- lostachya, 425. Gagea foliosa, 530 ; Granatelli, 530. Gaillardia Doniana, 228; megapota- mica, var. scabiosoides, 228. Galilea mucronata, 114. Galium, 219, 393; Aparine, 393, 400; corymbosum, 219; pusillum, 219; triflorum, 389. Ganophyllum falcatum, 296. Gardenia, 311. Gardner, J. S., on Eocene Ferns from Basalts of Ireland and Scotland, 655. Garuga mollis, 297. Gastonia, 350; cutispongia, 351 ; emir- nensis, 350. Gastrorchis, 466. Geissorhiza, 394. General Mediterranean Groves on, 524. Gentiana capitata, 389. . Gentianex, of Philippines, 291; of Tonglo, 389. Geophila Gerrardi, 413. Geraniacem, of Patagonia, 213; of Philippines, 290 ; of Tonglo, 387. Geranium, 393, 398; intermedium, 213; patagonicum, 213; simense, 395, 398, var. glabrius, 398. Gerardia genisteefolia, 229, Geunsia Cumingiana, 315. Gimbernatia Calamansanay, 310. Gladiolus, 394, 405; Garnierii, 406 ; ignescens, 406; Newii, 406; Quar- tinianus, 405; Watsonioides, 405. Gleichenia dichotoma, 657; Kurziana, Flora, H. 658. . Gleocapsa, 242, 263, 271; Itzigsohnii, 9 Glæocapsæ, 279, Gloeocystis vesiculosa, 248. Gloothece, 241, 243, 250. Gleotrichia, 244. Glumacez of Parasnath, 253. Glumes of Cyperus, 18; colour as a character, 18; mucronation of, 18. Glyceria festuczformis, 597; fluitans, 239; tonglensis, 391. - . Glycosmis pentaphylla, 373; sapin- doides, 373. lina 215 Glycyrrhiza astragalina, . Gnaphalium americanum, 222; auricu- latum, var., 401; luteo-album, 222; spicatum, 222; unionis, 401. Gnetacex (Philippines), 292. Gomphia ep? 590, dependens, 331 ; lanceolata, 330; obtusifolia, 330 ; perseæfolia, 330. — Gomphocarpus abyssinicus, 402 ; physo- carpus, 402. 678 Gomphocentrum, sect. of Mystacidium, 487, 488. Gomphonema dichotomum, 245 ; gra- cile, var. major, 245 ; parvulum, 245. Gomphospheeria aponina, 241. - Gomphrena rosea, 232; villosa, 232. Gonatozygon asperum, 636. Gongrosira, 271. Goniopteris Bunburii, 661, 662, 663; crenata, 663; prolifera, 663; stiriaca, 661, 663; tetragonum, 662. Goodenoviex (Philippines), 291. Goodyera nuda, 499; procera, self- fertilization in, 547, 550. Gottschea, relation to Fissidens, 550. Graminesz of Patagonia, 236; of Philip- pines, 292; of Tonglo, 391. Grammangis Ellisii, 472. Grammatocarpus volubilis, 218. Grammatophyllum Ellisii, 472. Graphidei, 262. Graphis contexta, 262; elegans, 262; heterospora, 262. Grewia asiatica, 255; cuneifolia, 326; eriopoda, 300; ferruginea, 326; ma- cropbylla, 326; occidentalis, 398; trinervata, 327. Griffithsia, continuity of protoplasm in, 603 ; setacea, 603, 608. Groves, H., on Coast Flora of Japygia, S. Italy, 523. Guettardella, 298 ; philippinensis, 301, 312. Guilleminea lanuginosa, 232. Guizotia Schultzii, var., 401. Gunnera, 268 ; macrophylla, 303 ; per- pensa, 341. Gussonia, 487 ; aphylla, 491, 492, 493; exilis, 491, 492,493; Gilpine, 491; globulosa, 491 ; physophora, 491, 492. Gutenbergia cordifolia, 400. Gutierrezia linearifolia, 221. Guttiferz (Philippines), 290. Gymnadenia, 515; Boryana, 516; fas- tigiata, 515 ; purpurascens, 515. Gymnema? vel Sarcolobus ? sp., 373. Gymnochilus nudus, 499; recurvus, 499. Gymnochilus (Madagascar), 456. Gymnogramme leptophylla, 364. Gymnosperine (Philippines), 292, 293. Gynerium argenteum, 238. Gynura vitellina, 401. Gyrinopsis Cumingiana, 300. Habenaria, 502, 516, 519; Arach- noides, 508, 509, 510; bimaculata, 506; cirrhata, 511; citrina, 507; de- pauperata, 505; dissoides, 511; fili- formis, 504; graminea, 504; Hilsen- diversifolium, 662; - INDEX. bergii, 509; Hildebrandtii, 503 ; hyperborea, 539; imerinensis, 505; incarnata, 510; intacta, 539; aff. kili- manjari, 405; malacophylla, 509; minutiflora, 503; multiflora, 503; misera, 503; nutans, 507; aff. nyi- kanz, 405; papillosa, 504; pleista- denia, 404; purpurea, 506, 507; Ru- tenbergianum, 510; simplex, 509; spirale, 504; spiralis, 502, 508 ; stricta, 510; tenerrima, 505; Thom- soni, 404; tridentata, 539; truncata, 509, 510. Heemadoracex (Philippines), 292. Halopeplis amplexicaulis, 529. Haloragese (Patagonia), 218. Halurus equisetifolius, continuity of protoplasm in, 613, 615, 616, 620, 621. Hantzschia Amphioxys, 241, 245, var. vivax, 245. Haplostellis truncata, 498. . Hebenstreitia dentata, 402; integrifolia, 402. Hedeoma multiflora, 231. Hedychium, 254. Hedyotis trichoglossa, 409. Heleocharis ezspitosissima, 450. Helianthemum angustifolium, 528; glaucum, 527; leptophyllum, 527; sessiliflorum, 530 ; spilosepalum, 528 ; vulgare, 528. Helicia, 295, 296 ; castaneæfolia (ftnote), 295; philippinensis (ftnote), 205. — Helichrysum adenocarpum, var. alpi- num, 401; foctidum, 401; leuco- spherum, 417 ; xylocladum, 418. Helinus mystacinus, 309. Helionites Zollingeri, 299. Heliotropium anchusszfolium, var. an- gustifolium, 227 ; curassavicum, var. parviflorum, 227. Helosciadum nodiflorum, 526. Hemicarex Hookeri, 391. Hemichlzna bulbosa, 175. Hemiphragma heterophyllum, 390. Hemitelites, 659. Henslovia philippinensis, 287. Henslow, Rev. G., on Vernation and Methods of Development of Foliage as protective against Radiation, 624. Heptapleuron, 255. Herminium, 504. Herpestis radicata, 229. . Heterxcious Uredines, reproduction in, without szecidiospores, 368. Heterostachys Rittneriana, 233. . Hevea, articulated laticiferous tissue 1n, 566, 568, 570. Hevea brasiliensis, 566, 568; Spruceana, 566, 570. INDEX. 679 Hewittia bicolor, 373. Hibiscus, 324; campylosiphon, 308; crassinervis, 2308; gossypinus, 398; grewizfolius, 308; palmatifidus, 324 ; surattensis, 973 ; Vidalianus, 308. Hieracium, 394. Hippomarathrum Bocconi, 528. Holarrhena? madagascariensis, 424. Holboellia latifolia, 387. Holmes, E. M., Remarks on Cinchona Ledgeriana as a species, 374. Holocarpa veronicoides, 414. Holorages (Philippines), 290. Homalium confertum, 341. Hooker, Sir J. D., Distribution of East- African Plants, with reference to J. Thomson's coll, 392; Introductory Note on Flora of Parasnath, N.W. Bengal, 252. Hordeum jubatum, 239; leporinum, 537; pratense, 239. Hormosiphon, 271. Hoya, 259 ; carnosa, crystals in, 621. Hyalis argentea, 224. Hyalocalyx, n. gen. of Turneracez, 256 ; setiferus, 258. Hydrocharides (Philippines), 292. Hydrocotyle asiatica, 348; filicaulis, 348 ; superposita, 348. Hydrophyllacez of Patagonia, 227 ; of Philippines, 291. Hydrostachys stolonifera, 435. Hymenophyllum, 664. Hyoscyamus, 27. Hyperieinee of Patagonia, 213; of Philippines, 290. Hypericum, 393, 398 ; connatum, 213 ; inceolatum, 398; peplidifolium, Hypnum, 551, 560; adiantoides, 551, 560 ; bryoides, 551; cupressiforme, 366 ; taxifolius, 551. Hypochæris, 226 ; neapolitana, 528. Hypodiscus albo-aristatus, 591, 593, 594 ; Oliverianus, 592. Hypoestes, 373; antennifera, 403; flo- ribunda, var., 373 ; jasminoides, 432 ; Rothii, var. pubescens?, 403; sta- chyoides, 431; trichochlamys, 432; unilateralis, 431 ; verticillaris, 403. Hypolæna Eckloniana, 590, 593; sp., Hypolytreæ, 25. Hypothesis, Algo-Lichens, J. M. Crom- bie on, 259. Hypoxis, 394. Hysterionica jasonioides, 221. Ichnocarpus frutescens ?, 313; Navesii, 313; ovatifolius, 313, Ilex, 394; Cumingiana, 308; cymosa, 308 ; Lobbiana, 309; luzonica, 309 ; philippinensis, 309. Ilicineæ (Philippines), 290, 293, 308. Illecebraceæ (Patagonia), 232. Impatiens kilimanjari, 398; sp. nov., 399 ; Thomsoni, 398. Incompletæ (Madagascar). 435. Indian species of Cyperus, Mr. C. B. Clarke on, 1. Inflorescence of Cyperus, flexible yet chief charac., 14. Iodina rhombifolia, 210, 234. Ipomæa phylloneura, 426; sagittata, 528. Irideæ, 235. Isanthera discolor, 297. Isoëtes echinospora, 366 ; lacustris, 366. echinospora and lacustris, vege- tative budding in, 366. Tsoglossa angusta, 439 ; gracillima, 430 ; Melleri, 431. Isolepis, 14, 23, 83, 202; echinulata, 91; hamulosa, 93; Micheliana, 14, 16, 21, 28, 29, 30, 202; pygmza, 28. —— Micheliana as related to Cyperus pygmeeus, 28. . Isopyrum adiantifolium, 387. Italo-Grecian Flora, H. Groves on, 529. Ixora emirnensis, 412; sp., 973; timo- rensis, 373. Jasminum abyssinicum, 401 ; auricula- tum, 401; settense, 401. u Joshua, W., on Burmese Desmidiez, 634. Juglandez (Philippines), 292. J uncacenm 4 Patagonia, 236; of Phi- lippines, 292. Juncellus, 19, 20, 22, 27, 32, 33, 36, 71; as subgenus of Cyperus, 33, 71. Juncus acutus, 236, 527; multiflorus, 528. Juniperus communis, 216; nana, 216; procera, 394, 395, 404. . Justicia neglecta, 402 ; sp., 402 ; tricho- phylla, 429 ; triticea, 429. Kayea philippinensis, 301. Kerria T ponica, crystals in, 621. Kibessa, 287. Kickxia arborea, 313; Blancoi, 313. Kidston, R., on Lycopodites (S.) Vanu- xemi in Britain, and affinities, 560. Kitchingia schizophylla, 340. Kniphofia, 394 ; Thomsoni, 406. Knorria imbricata, 556. Knoxia brachycarpa, 253. 680 Kobresia, 12. Keeleria cristata, 238. Kolowratia elegans, 316. Kundmannia Sicula, 528. Kyllinga, 32, 33, 34; panicea, 201. , as a genus, 32; Clarke prefers it as a subgen. of Cyperus, 32. Labiatz, of Patagonia, 281; of Philip- pines, 291, 315; of Tonglo, 390. Lactuca brevirostris, 298. Lady Fern, singular mode of develop. in, 354. Lagecia cuminoides, 528. Lagoseris bifida, 531. Lagostomus trichodactylus, 217 (foot- note). Lamprocaulos Neesii, 593. Lantana Clarazii, 229; Kisi?, 403; Sellowiana, 230; sp., 403. Larrea nitida, 213. Lasianthus Biermanni, 389. Lasiosiphon, 394. Laticiferous (Articulated) Vessels in Hevea, R. H. Scott on, 566. 5 tubes of Euphorbia, Dr. Pax on, Tl. Lauracez (Tonglo), 390. Laurencia, continuity of protoplasm in, 602; hybrida, continuity of proto- plasm in, 607. Laurinez (Philippines), 292. Leaves of Cyperus, nature and charac- ters of, in diff. sp., 8. . Lecanora, 262, 264, 266 ; cinerea, 273; galactina, 275 ; gibbosa, 275, 282. Lecidea alboatra, 275, 282 ; geographica, 275, 282; lutea, 262; microsperma, 262; muscorum, 266. Ledum palustre, 370. Leguminose of Patagonia, 915; of Philippines, 290, 309; of Tonglo, 388. Lemna, 268. Lemnacee (Philippines), 292. Lentibulariess (Philippines), 291. Leonotis, 393, 394; rugosa, 403. Leontodon asper, 531; fasciculatus, 529; saxatilis, 531. Lepidagathis laxa, 300. Lepidium bipinnatifidum, 212 ; bona- riense, 212; Chichicara, 212; pube- scens, 212. Lepidodendron, 564. Lepidospermum thermale, 590. Leptocarpus paniculatus, 500. Leptogium, 263, 279. Leptospermum flavescens, 296, 300; Seoparium, 296. Leptostachyus, 17, 20, 32. INDEX. Leucas decemdentata, 373 ; masaiensis, 403; stachydiforme, 403. Leucobryaceze, 552. Leucopogon suaveolens, 296, 298, 300. Leycesteria formosa, 389. Leyte island, plants of, 300. Lichen. See under Algo-Lichen hypothe- sis, 259. Lichenospheria Lenormandi, 263. Lichina, 263. Lightfootia, 394; abyssinica, 395, 401. Ligusticum greecum, 531. Ligustrum Cumingianum, 297. Liliacez of Patagonia, 235; of Philip- pines, 292 ; of Tonglo, 391. Lilium Wallisii, 298. Limodorum, 466 ; concolor, 469; ebur- neum, 480; flabellatum, 472; plan- tagineum, 472: scriptum, 468 ; tuber- eulosum, 466. Linaria canadensis, 229; reflexa, 530; Sieberi, 530. Lindsaya concinna, 298; 299. Linnean Herbarium, Restiacee pre- served in, 590. uu Liparis, 457, 460, 462; angustifolia, 458; bicornis, 458; csspitosa, 408; connata, 462; flavescens, 460; longi- caulis, 461; longipetala, 459; lutea, 458 ; nepalensis, 390 ; ochracea, 461; ornithorrhynchos, 460, 461; parva, 462 scandens, Lipocarpha, 107. _ Lippia, 230; asperifolia, 403 ; lycioides, 230 ; oligophylla, 434; rubiginosa, 230; seriphioides, 230. Lissochilus (Madagascar), 456; mada- gascariensis, 471; Rutenbergianus, 471. List of Plants collected by Mr. J. Thom- son on Mountains of E. Equatorial Africa, by Prof. D. Oliver, with Ob- servations on Distribution, by Sir J. D. Hooker, 392. Listrostachys polystachys, 497. Litobrochia spinulifera, 657. Litsea cinnamomea, 299. Loasa prostrata, 218. Loasaceæ (Patagonia), 218. Lobelia coronipifoliæ, var., 401. f Loganiaceæ of Philippines, 291; 9 Tonglo, 389. Loiseleuria procumbens, 342. Lolium perenne, 239; subulatum, 537, var. brasilianum, 239. Lophatherum geminatum, 454. 3 Loranthaceæ (Philippines), 292, 293, 294. . Loranthus, sp., aff. L. rigido, 374. C——— Mo Ó— INDEX. Lotus, 393; tigrensis, 395, 400. Luhea, divaricata, 210. Luzon (Philippines), plants of, 301. Luzuloidesze, 19. Lycium elongatum, 228; fiüifolium, var. minutifolium, 228; Wilkesii, 229, Lycopodiacez (Philippines), 292. Lycopodites, gen. emended, 560, 561, 562, 564; arborescens, 561; flexi- folius, 561; Lacoei, 561; simplex, 56l; Vanuxemi, 560, 561, 562, 564, 565, 566. Vanuxemi, in Britain, with remarks on affinities, R. Kidston on, 560. Lycopodium, 374; carinatum, 374; hastatum, 299 ; megastachyum, 454 ; obtusifolium, 455; pachyphyllum, a palmatum, 657; Phlegmaria, 214. Lyngbya, 263. Lysimachia prolifera, 889; serpyllifolia, 226. Lythrariex (Philippines), 290. Lythrum rotundifolium, 395, 400. Macaranga mnyriolepida, 442; ribesioi- des, 442. Machilus odoratissimus, 390. Maerostylis, 498. Maerura, (Madagascar) section of An- graecum, 475; sp. of, 475. Madagascar, flora of, 317, 407. ——— new gen. of Turneracee from (Hyaloealyx), 256 ; Orchids of, 450. æsa, sp., 373. - Magnoliaceæ, of Philippines, 290, 307 ; of Tonglo, 387. Malanthera Brownei, 401. falax angustifolia, 458; brevifolia, 457 ; esspitosa, 458 ; flavescens, 460. Malpighiaceze (Philippines), 290, 294. Malvacer (Philippines), 290, 308. andragora autumnalis, 523 ; caules- cens, 300. Manihot, laticiferous tissue in, 566; Systemat. position of, Dr. Pax versus Bentham, 572. — Glaziovii, Rosanoff's Crystals a 1 Endosperm-cells of, 621, 622, 524. Margyricarpus alatus, 217; Clarazii, 217; microphyllus, 217; setosus, - 216, 217. Mariscus, as subgen. of Cyperus, 34, 193. ` —, 9, 15, 17, 20, 32, 33, 34, 83, 114, 193, 199; biglumis, 199; brachy- stachyus, 196; flavus, 196; Kraussii, LINN, JOURN.—-BOTANY, VOL, XXI. 681 197; mucronatus, 114; ovularis, 199; paniceus, 199, 201 ; retrofractus, 199; umbellatus, 200. Marsdenia, fruit of, 373. Marsileaceæ (Philippines), 292. Marumia, 257. Masters, Dr. M. T., Suppl. Notes on Restiaceæ, 574. Mathurina, 256. Maytenus chilensis, 215. Medicago Echinus, 530; lupulina, 215. Medinilla lanceolata, 344; leptophylla, 243 ; lophoclada, 344. Melandryum macrocarpum, 528. Melanophylla alnifolia, 352; aucubse- folia, 353. Melanornia, 272. Melanthera madagascariensis, 418. Melaspilea arthonioides, 262. Melastoma pencillatum, 300. Melastomacex, 287 (Philippines), 290, 310. Melhania laurifolia, 326. Melia Candollei, 297. Meliacex (Philippines), 290, 294. Melica macra, 238; papilionacea, 238; violacea, 238. Melicocea triptera, 309. Melilotus, hypnotism in, 625. Melochia velutina, 373, var. glabrata, 373. Melodinus Cumingii, 301. Melothria emirnensis, 346; Rutenber- giana, 346. . Memecylon Eleagni, 343; olezfolium, 343. Menispermacese (Philippines), 290. Menodora decemfida, 226. Mentzelia albescens, 218. Metabasis ceretensis, 530. . Methods adopted in various plants against effects of radiation in young leaves, 627-633. Michelia, 268. Micrasterias, 635, 636, 637; alata, 636 ; apiculata, 636, 653 ; ceratofera, 637; erux-melitensis, 635, 653; euastroides, 637, 653; foliacea, 636; incisa, var. B. Wallich, 636; lux, 636, 653; Mahabuleshwarensis, 636, 637; pin- natifida, 636; radiosa, 637 ; rotata, 636; tropica, 636; truncata, var. crenata, 636 ; Wallichi, 637. Microccelia, 487, 490 ; exilis, 493. Microcoleus Aitehisonii, 241, 250. Microglossa volubilis, 400. Micromelum pubescens, 373. Micromeria approximata, 530; canes- cens, 530; nervosa, 528; punctata, 403. 3 F 682 Micropterygium, relation to Fissidens, | 550. Mikania scandens, 221. Milla aurea, 235. Millingtonia pinnata, 314 ; quadripin- nata, 813, 314, 315. Mimosa dasyphylla, 338; hamata, 339; myriacantha, 330 ; nissobiensis, 338. Mitten, W., notes on European and N.- American species of Mosses of Genus Fissidens, 550. Modecca peltata, 345. Momordica Charantia, 973 ; ovata, 298. Monochilus (Madagascar), 456. 9 Boryi, 499; gymnochiloides, 499. Monocotyledons of Madagascar, 447 ; of Philippines, 292. Monopetalæ (Madagascar), 407. Moore, S. le M., Observ. on Continuity of Protoplasm, 595, summary of views, 614; on Rosanoffs Crystals in the Endosperm-cells of Manihot Glaziovii, 621 ; Studies in Veg. Biol., 595, 621. Mosses of Genus Fissidens, W. Mitten on, 550. Mougeotia, 247, 248, 250. Mucidines, 265. Mucuna (Stizolobium) sp., 373 ; (Timor Laut), 374. Muehlenbeckia chilensis, var. injucunda, Muntingia Calabura, 303. Murraya exotica, var., 373. Mussenda frondosa, 311; fuscopilosa, 410; grandiflora, 311; macropoda, Mycoderms, 265. Myosotis albiflora, 227. Myrica luzonica, 294, 316 ; Vidaliana, Myricace: (Philippines), 292, 316. Myriophyllum, 218, 340; axilliflorum, 340; proserpinacoides, 218; verti- cillatum, 341. Myristica guatteriæfolia, 298 ; insipida, 374. Myristicaceæ (Philippines), 291. Myrsinacez (Tonglo), 389. Myrsinez (Philippines), 291, 294. Myrtacex (Philippines), 287, 290, 293. Mystacidium, 487; caulescens, 488; graminifolium, 490; inapertum, 489; ochraceum, 488 ; tenellum, 489. Neetrocymbe, 272. Naiadacez (Philippines), 292. Narcissus serotinus, 529. INDEX. Nassauvia glomerulosa, 225; rosulata, 225. Nastus capitatus, 454. Navicula ambigua, 245 ; appendiculata, 244; Brebissonii, 244; limosa, 245; oblonga, 245; Tabellaria, 244; viri- dis, var. commutata, 244, var. longior, 244. Nectandra amara, var. australis, 209. Negros (Philippines), plants of, 301. _ Nelumbium, crystals in, 621; Buchu, 658. Nematonostoe, 280. Neobaronia, 336; phyllanthoides, 337. Neottia Nidus-avis, 539. Neottiex, 498. Nepenthacese (Philippines), 291. Nephrodium, 662, 663; crassifolium, 209; giganteum, 299; obscurum, 299 ; Otaria, 299 ; recedens, 299. Nephromium, 263. Nervilia, 408. Nicotiana acutifolia, 209. . Nierembergia filicaulis, 229; hippo- maniea, 229; rigida, 229. Nitzschia vermicularis, 245. Nostoc, 244, 263, 268, 271, 279; liche- noides, 265. Nostoeacese, 260, 261. Notes on Afghanistan Algæ, by Dr. J. Schaarschmidt, 241. Notes on Flora of Parasnath, N.W. Bengal, 252. Notes on Ranunculus lingua, by F. C. S. Roper, 380. Notes on Restiacem, Dr. Masters on, 574. Notholæna Marantæ, 529; vellea, 529. Notobasis syriaca, 528. Nut of Cyperus, 22; its charac. of primary importance in classif., 44; structural peculiarities of, 23-25. Nuts of Cyperi sink in water, ; Nuts and continuity of protoplasm in Strychnos Nux-vomica, 595, 596, 597, 598, 600, 601, 618 ; Ignatia, 596, 597, 598, 600, 601, 617 ; potatorum, 598, 601, 602, 618 ; spinosa, 599, 618. Nyetagines (Philippines), 291. f Nylander, On Germination Spores 9 Vaucillaria, 265. Nympheacez (Philippines), 290. Oberonia, 456 ; brevifolia, 457. Obryzum, 263. Observations on the Continuity of Pro- toplasm, by S. le M. Moore, 595. Observations on a singular Mode © Development in the Lady Fern (Athyrium Filix-foemina), 354. INDEX. Occurrence of Articulated Laticiferous Vessels in Hevea, D. H. Scott on, 5606. Occurrence of Lycopodites (Sigillaria) | Vanuxemi in Britain, with remarks on its 560. Ochna serratifolia, 329 ; vaccinioides, 329; Wightiana, 330. Ochnacesz, 290, 294. Ochrocarpus ovalifolius ?, 373. Octodiceras, 552, 554. Octomeles, 295; sumatrana, 299. Ocymum sp., 403. Odina speciosa, 297. Œdogoniaceæ (Afghan), 249. CEdogonium longicolle, 249, var. af- ghanicum, 249, 250, var. senegalense, 249, 250; Pringsheimii, 249. CEnothera mendocinensis, 218 ; mollis- sima, 218 ; odorata, 218. Olacines (Philippines), 290. Olax emirnensis, 331; 331. Oldenlandia latifolia, 409. Oleaceze (Philippines), 291. affinities, by R. Kidston, | psittacorum, Oliver, Prof. D., List of Plants of Timor | Laut coll. by H. O. Forbes, 373; on Plants collected by J. Thomson in E. Equatorial Africa, 392. Omphalaria, 263. Onagrariez of Patagonia, 218; of Phi- | lippines, 291. Oncoba capreæfolia, 320; Petersiana, 320 ; tettensis, 320. Oncostemum neriifolium, 421; platy- cladum, 421 ; venulosum, 422. ` Onobrychis alba, 531; echinata, 531. Onoclea, 659, 660; germanica, 662; hebraidica, 659; orientalis, sensibilis, 657, 659, 660, var. fossilis, Ononis Sieberi, 529. Onychonema læve, 635. Oocystis Nagelii, 248. Oophore generation (Ferns), 363. Opegrapha filieina, 262; 262; varia, 262. Ophioglossum vulgatum, 254. 662; | herbarum, | Ophiopogon spicatus, 297; Wallichi- | anus, 391. Ophrydex, 500. Ophrys, 529; apifera, 539, 546; exal- | tata, 530; Speculum, 529. Oplismenus bromoides, 452. Orchidacez, of Kilimanjaro, 405; of Madagascar, 456; of Philippines, 292, 293 ; of Tonglo, 390, Orchids of Madagascar, H. N. Ridley on, 456. 683 Orchids, Self-Fertilization in some tro- pical, H. O. Forbes on, 533. Orchis longicruris, 529; mauritiana, 476. Origin of East Equatorial African flora, Sir J. D. Hooker on, 395. Origin of Lichen gonidia, and relations to thallus, 259; theories connected therewith, 260. Ornithogalum collinum, 530: diver- gens, 532; excapum, 532; refractum, 532, var. Adalgisa, 532. Orobanchaceæ (Philippines). 291. Orthosiphon brevicaulis, 433; emirnen- sis, 433 ; secundiflorus, 453. Oryza coarctata, 255. Osbornia octodonta, 296. Oscillaria, 241, 242. Osmelia conferta, 300; philippinensis, 300. Osmunda, 364; javanica, 657; sensi- bilis, 660; Torellii, 609. Osteospermum, 394. Othonna, sp. nov., 401. Owenia cerasifera, 374; 373. Oxalis, movements of leaflets in pur- purea, 625, 631. amara, 214; Darwinii, 2141; floribunda, 214; Griffithii, 387; ma- cropoda, 328; Martiana, 214; myrio- phylla, 328. . Oxypetalum solanoides, 227. cerasifolia, Paico hembra, 233; macho, 233. Palmæ (Philippines), 292, 293. Palinellacez, 261; (Afghan) 240. Panax amplifolium, 351 ; Cumingiana, 310; confertifolium, 351; multi- bracteatum, 351; pentamerum, 352 ; zanthoxyloides, 351. ! Pandanee (Philippines), 292, 293. Pandanus, 447; ceratophorus, 418; concretus, 448 ; microcephalus, 447 ; oligocephalus, 418: seyehellarum, 449 ; (Timor-laut), 372. Pandorina morum, 249. Pangium edule, 293. Pannarize, 263. Papaver, reticulate laticiferous vessels of, 573; Rheeas, 251. Papaveraceæ (Patagonia) 211. — Pappophorum alopecuroides, 293. Papyrus, 84; elatus, 189; Pangorei, 158, 161. Paracaryum glochidiatum, m Parasnath, N.W. Bengal, Flora of, 252. Paratrophia Cumingiana, 310. Paratrophis philippinensis, 297. Paris polyphylla, 391. > 389. 684 Paritium, 325. Parviflorz ($ Cynorchis), 512. Passifloreze ( Philippines), 291. Patagonia, Flora of, 202; poverty of, opinions on (Ball), 205; table proport. sp. nat. orders in N. Pat. and Argentine Repub., 208. Pavonia glechoinoides, 200; periana, 398. Pecopteris Torellii, 659. Pectinaria, sect. of Angrecum (Mada- gascar), 487. Peddiea, 394. Pellea ternifolia, 239. Pelletiera serpyllifolia, 226; 226. Peltophorum ferrugineum, 373. Penium Brebissonii, 653; Closterioides, 653; delicatulum, 653; digitus, 653 ; lamellosum, 653; margaritiferum, 653; minutissimum, 655; navicula, 653; spibostriolatum, 653. Pennisetum triticoides, 453. Pentacena polyenemoides, 232. Pentapterygium serpens, 389. Pentas carnea, 400; micrantha, 408; purpurea, 400 ; Schimperiana, 400. Pentopyxis stipulata, 389. Peperomia reflexa, 890; trichophylla, 436. Peracarpa carnosa, 389. Perigynium of Cyperus cephalotes, whether present or absent, 26. Peristylus gramineus, 504; purpureus, 506 ; spiralis, 502. Petunia nyctaginiflora, 229, Peyssonellia, continuity of protoplasm in, 602. Pfaffia lanata, 232. Pfeifferia, 285. Phacelia glandulosa, var. patagonica, 227. Phaius, 466; self-fertilization in sp. of, 539, 541, 545; albescens, 540, 542 ; amboinensis, 540, 542; Blumei, 539, 540, 541, 542, 544, 545, 548; Humblotii, 466; pulchellus, 466, 467; stuppeus, 466; tetragonus, 466 ; tuberculosus, 466 ; villosus, 466, Phellolophium madagascariense, 349. Philibertia solanoides, 226. Philippines, Flora of. and probable deri- vation, R. A. Rolfe on, 283, 303. Philydraces (Philippines), 292. Phleum arenarium, 532; grecum, 532. Phlomis, 528; ferruginea, 530; fruti- cosa, 528, 530. Phenix acaulis, 253. Phornothamnius thymoides, 342. Schim- verna, INDEX. Phycochromacez, 261, 263. Phyllaetidium, 262, 271. Phyllagonium, 552. Phylauthoidez, no laticiferous tubes, 571. Phyllanthus buxifolius, 299. Phylliseum, 263, 272. Phymatodes, 657. Physalis viscosa, 228. . Physia, 264, 266 ; lithotea, 277; parie- tina, 263, 265, 266; pulverulenta, 277, 282. Physma, 263. Phyteuma limoniifolium, 551. Pilea, 300 ; capitata. 446; longipes, 446 scripta, 390; sp., 390 ; ternifolia, 390 umbellata, 447 ; umbrosa, 590. Pimpinella laxiflora, 349. Pinardia Coronaria, 528. . Pinus insularis, 207, 802; Murkusi, 297, 299, 309. . Piper, aff. canino, 274; capense, 496 ; paehyphyllum, 456. . Piperacez, of Philippines, 291, of Ton- glo, 390. Piptanthus nepalensis, 388. Pipturus asper, 298. Piriqueta, 256. Pistia, 28, 118; Salvinia, 118. Pittosporex (Philippines), 290. Pittosporum stenopetalum, 320. Plantaginem, of Patagonia, 291; of Philippines, 291. . Plantago, 231; albicans, 525 ; Bismare- kii, 221; gnaphaloides, 281; major, 231; patagonica, 231. - Plantathera, 511; graminea, 504. Plants, E. Equatorial African, coll. by J. Thomson, described by Prof. Oli- ver, 392. . . Platycoryne (Madagascar), 456 ; Pervil- lei, 521. Plectranthus cymosus, 434 ; sp. ?, 403. Plectronia Boiviniana, 411; buxifolia, 411. Pleurococcus mucosus, 248. —— Pleurotznium clavatum, 650; indicum, f. minor, 650; maximum, 650 ; no- dosum, 650; nodulosum, 635 ; tenuior, 635. Pleurothallis disticha, 457. . of Plowright, ©. B., on reproduction O Heteræcious Uredines, 968. — , Plumbaginem, of Patagonia, 221; 9 Philippines, 201. . Poa alopeeurus, 233; alpina, 891; attica, 537 ; denudata, 238, var. mr nor, 238; lanigera, 228. . Podocarpus, 394; elongata, 395, ion madagascariensis, 447 ; Mann, 391. ————— G INDEX. Podospermum Tenorii, 556. Podostemacez (Philippines), 291. Pecilostachys geminatum, 453; Hilde- | brandtii, 454. Peederia Tacpo, 312. Pogonia (Madagascar), 456; Rensch- iana, 498 ; Thouarsii, 498. Pollen from Egyptian Tomb, C. F. White on, 251. Pollinia (Parasnath), 258. Polyalthia, 319; Chapellieri, 318; lu- cens, 318. Polybotrya, 662. Polyedrium minimum, 248. Polygala, 212; amara, 321; pilosa, 371; spinescens, 212, 213, var. Aspa- lathoides, 212. Polygalex, of Patagonia, 212; of Philip- pines, 290. Polygonaceze, of Patagonia, 233; of Phi- lippines, 291. Polygonatum oppositifolium, 891; punctatum, 391; verticillatum, 391. Polygonium brachypodum, 435; Meiss- nerianum, 435. Polygonum camporum, var. australe, 233. Polyides, continuity of protoplasm in, 606, 607, 6l4; rotundatus, 605, 618. Polypetalx, 317. Polypodium flaceigerum, 299 ; irioides, 914; lamarioides, 800; longifolium, 299; palmatum, 299; papillosum, 299; sessilifolium, 299; simplicifo- Num, 297 ; splendens, 299; steno- phyllum, 299. Polypogon monspeliensis, 237. Polypothrix amphibia, 243. Polysiphonia, continuity of protoplasm in, 603 ; fastigiata, 604, 612, 617, 620 ; nigrescens, 612, 613, 615, 620 ; urceo- lata, 604 617. Polystachya of Madagascar, 456, 473 ; anceps, 473 ; cultrata, 473; Jussieu- lana, 475; luteola, 474, 475; minuti- flora, 475 ; rosea, 474 ; rosellata, 475; virescens, 474. Polystichum angulare, 362, var. pul- cherrimum, 862, 366, 368 ; apospory, and prothalloid growth in this var., 363. Polytrichaces, 552. Polyxias Cumingiana, 310. Pontederiacex (Philippines), 292. Popomia, 319, Populus italica, crystals in, 621. Portulaca oleracea, 373; (Philippines) Posidonia caulinia, 526. | ] 685 Pothos ineequilaterus, 299. Premna obtusifolia, 873. Primula rotundifolia, 380 ; petiolaris, 389 ; var. (? sp.), 389. Primulacem, of Patagonia, 266; of Tonglo, 589. Procris grandis, 299. Prosopis juliflora, 303. Proteacez (Philippines), 292. Prothallus, reproduction of (Ferns), 9362. Protococcacese (Afghan), 248. Protococeus, 242, 262, 266, 267, 275; (Afghan), 248 ; viridis, 262, 266. Protoplasm, continuity in, S. le M. P , ) , Moore on, 595, 621; Hicks’s and Masse's methods, 617. Prunus nepalensis, 358. Pseudopyereus, 72. Psilocarya eandida, 450. Psoralea badocana, 296; foliosa, 399; hypnorum, 277, 283. Psorospermum cerasifolium, 324 ; disco- lor, 323; leptophyllum, 323 ; tricho- phyllum, 323. Psychotria lucidula, 413; malayana, 312; mesentericarpa, 412; sp., 373; Tacpo, 312. 0. Pteris, 200 ; atrovirens, 657; biaurita, 657 ; cretica, 864, 657 ; Fraseri, 656 ; hibernica, 656, 657; ludens, 299; quadriaurita, 657 ; spinulifera, 657 ; tripartita, 374. Pternandra, 287. m Ptervcarpus erinaceus, 309; Vidalianus, 309. i Ptilophyton Vanuxemi, 564, 565. , Ptilota, continuity of protoplasm in, 610; sericea, 608, 615, 619. . Puccinia, develop. of uredospores in P. gramineis, 369. graminis, 869; rubigo-vera, 369. a we Pycreus, 16, 19, 20, 21, 22, 23, 33, 35, 36; angulatus, 60; albo-marginatus, 69; capillaris, 48; diffusus, 8&2; Elliottianus, 62; flavescens, 37 ; in- firmus, 62; lævigatus, 77 ; Maximi- liana, 62 ; megapotamicus, 68; mu- cronatus, 77 ; Olfersianus, 55 ; pan- nonicus, 80; paraensis, 38 ; poly- stachyus, 51; propinquus, 62; pul- vinatus, 44; pumilus, 46; retusus, 71; squarrosus, 94. | , subgen. of Cyperus, 33, 55. Pyrenacantha chlorantha, 331. Pyrenomycetes, 267. Pyrus, 217; foliolosa, 988 ; lanata ?, "388; malus, 217; microphylla, 388; rhamnoides, 388 ; vestita, 38s. obscura, 369 ; 686 Quercus, 530; 7Egilops, 531; pseudo- coccifera, 530. Radermachera Banaibanai, 314; pin- nata, 314 ; quadripinnata, 313, 314. Radiation, protective against, in verna- tion and development of foliage, Rev. G. Henslow on, 624. Ramalina, 264, 266. Randia, sp., 373. Rangoon Desmids, 634. Ranunculacee, of Philippines, 290; of Tonglo, 387. Ranunculus asiaticus, 531; Cymba- larie, 387 ; diffusus, 387; Flammula, 382, var. floribus minimis, 387; Lin- gua, 380, 381, 383, 384; oreophytes, 395, 297 ; palustris flammeus major, 382. Lingua, F. C. S. Roper on, 380; submerged leaves of, 382. Ravisara, Madagascar name for Iso- glossa Melleri, 431. Rehmann, Restiacez coll. by, 594. Remarks on Cinchona Ledgeriana as a species, E. M. Holmes on, 374. Remarks on Reproduction of Heterzci- ous Uredines, by C. B. Plowright, 368. Renealmia exaltata, 316; gracilis, 316. Report on Botany of H. O. Forbess ex- ped. to Timor-Laut, by W. T. Dyer, 370. Reproduction, in Ferns, 360; in Hete- recious Uredines, 368. Restiacee, 574; coll. by Bolus, 591; coll. by Rehmann, 594; in Linnean Herbarium, 590; Supplt. Notes on, by Dr. Masters, 574. Restio, 575, 591; bifidus, 590, 591, 593, 594; bifurcus, 594; bigeminus, 576, 593 ; compressus, 593, 594; curvira- mis, 594; cuspidatus, 590, 591, 593, 594; dichotomus, 590; elatus, 592; Elegia, 584; Eleocharis, 593, 595; filiformis, 591, 594; fraternus, 591; furcatus, 593; Gaudichaudianus, 592, 594; glaucus, 580; intermedius, 594; multiflorus, 575, 593, 594; pa- chystachyus, 592; paniculatus, 590; perplexus, 592; quinquefarius, 591, 594; racemosus, 578, 579 ; scoparius, 592; Sieberi, 591, 592; subfalcatus, 592; subverticillatus, 592; thyrsifer, 574, 534, 585; thyrsoides, 590 ; tri- florus, 590, 591; triticeus, 594; vagi- natus, 590, 591; verticillaris, 590; vimineus, 590, 591. Rhachilla, spikelet of Cyperus, 16 ; great var. of, 16; its form not of sec- INDEX. tional value, 17; solubility of wing, a useful character, 17. Rhamnew, of Patagonia, 215; of Philip- pines, 290. Rhamphicarpa ?, 402. Rhipsalis horrida, 347. Rhizome of Cyperus, structure and peculiarities of, in diff. species, 2; how far to be used as a charac. of groups, 5. Rhizophoree, of Philippines, 290; of Timor-Laut, 371. Rhodamnia, 287. Rhodoelada, 327 ; rhopaloides, 328. Rhododendron, 385; arboreum, 389; barbatum, 389; campanulatum, 385, 389, various colours as grown at Sundukphoo, 385; cinnabarinum, various colours of, on Sundukphoo, 386, 389; Dalhousiz, 389; of Dar- jeeling, 385 ; Falconeri, 389; javani- eum, 299, of Sundukphoo, 385. Rhodolzna acutifolia, 322. Rhopala, 328. Rhus, 337 ; succedanea, 388. Rhynchanthera, 498. Rhynchospora candida, 450. . Rhyneosia senna, var. foliolis lanceolatis, 216; var. foliolis late ovatis, 216; texana, 216. Ribes, 217, 885; glaciale, 358; magel- lanicum, 217. . à Riccia, continuity of protoplasm in, 602. Ricinus, crystals in, 621. Rivularize, 261. Roccella, 262. Rolfe, R. A., on Hyalocalyx, a new genus of Turneraces, from Madagas- car, 256. , Romanes, Dr. R., Burmese Desmids coll. by, 634. Romulea, 393; camerooniana, 406. Roots of Cyperus, how alter, 3. Rosa sericea, 388. " Rosaceæ. of Patagonia, 216 ; of Philip- pines, 290; of ‘Tonglo, 388. ils Rosanoffs crystals in endosperm-ce of Manihot Glaziovii, &e., S. le M- Moore on, 621. Rottboelliz, 18. Rourea platysepala, 336; santaloides, 336. Rousseauxia, 343. Rubia cordifolia, 389. 219; Rubiacem, 311; of Patagonia, ot Philippines, 291, 293, 291, of Tonglo, 389. . 988: Rubus calycinus, 388 ; Hookeri, 380 ; lasiocarpus, 988 ; macrocarpus, 388. Ruellia brevicaulis, 428. 311; INDEX. 687 Rumex. 393; conglomeratus, 233; cu- neifolius, 233; magellanicus, 233; obtusifolius, 393, 403 ; pratensis, 233; sanguineus, 233; thyrsoides, 529; tuberosus, 533. Rutacom, of Philippines, 290; of Tonglo, 87. Ryssopteris dealbata, 301; microstemma, 301. Saccolabium aphyllum, 492 ; ceum, 498. Salaeia Calypso, 335; dentata, 334; oleoides, 334. alicinem (Philippines), 292. Salicornia fruticosa, var. peruviana, 233. Salix aurita, crystals in, 621; Hum- boldtiana, 210. Salpichroma rhomboideum, 228. Salsola, 11. Salvadoraceæ (Philippines), 291. Salvia hematodes, 536; triloba, 530. Salvinia, 28. Samar (Philippines), plants of, 300. Samydaces (Philippines), 291. Sanguinaria, laticiferous sacs in, 573. Santalacez, of Patagonia, 234; of Phi- lippines, 292. Sapindaces, of Philippines, 290, 293, 209; of Tonglo, 388. Sapium, 316, sp., (footnote) 294. Sapotacese (Philippines), 291. Satureja cuneifolia, 531. Satyrium, 519 ; calceatum, 520 ; gracile, 220; gramineum, 506; rostratum, 520 ; spirale, 502; trinerve, 519. —— of Kilimanjaro, 405 ; of Mada- gascar, 4506. Saxifraga purpurascens, 318. Saxifragese, ot Patagonia, 217 ; of Phi- lippines, 290 ; of Tonglo, 388. Scabiosa, 393, 530; Columbaria, 393, 400 ; grandiflora, var. canescens, 530 ; maritima, var. villosa, 530 ; ucranica, var. eburnea, 530. Scenedesmus quadricauda, 249, var. a, ecornis, 249, Schaarschmidt, Dr. Julius, Notes on Afghanistan Alg: coll. by Dr. Aitchi- son, 241. ` Schismatoclada concinna, 407 ; vibur- noides, 407. Schistostegeæ, 552. Schizandra elongata, 387 ; grandiflora, 37. Schizocasia, 297. Schizophyceæ, 242. Schenus capensis, 590; mucronatus, 114; nigricans, 527. coria- Schwendener’s theory of Algo-Lichens, 260, 261. Scirpus, 7, 108, 117, 236; coronarius, 108; hamulosus, 93; lappaceus, 91 ; maritimus, 236; trialatus, 130. Scitaminex, 316, (Philippines) 292, 316. Sclerochatium, 590 ; thermale, 589. Scoparia flava, 229. Scorzonera Colunine, 530; Tenorii, 536. Scrophularia bicolor, 530; filicifolia, 531 ; lucida, 531. Scerophularines, of Patagonia, 229; of Philippines, 29]. Seutellaria indica, 297; luzonica, 297, 315. Scytonema, 263, 271, 279. Seytonemeze, 261. Selago, 394; cephalophora, 402; Thom- soni, 402. Self-fertilization in Tropical Orchids, H. O. Forbes on, 538. Seligeriacez, 552. Semecarpus Perottetii, 298. Senecio, 223; laricifolius, 224; oligo- leucos, 224; pinnatus, 223, var. glan- dulosus, 223; punctatus, 223; pur- pureo-viridis, 419 ; sp. nov. ?, 401. Serrafaleus Lloydiana, 537. Seseli, 350. Sesuvium portulacastrum, 219. Setaria caudata, 236. Sexual organs in life-cycle of Ferns, 303. Sibthorpia pinnata, 390. Siculo-Japygian Flora, H. Graves on, 530 Sida humilis, 373, var. repens, 373; Schimperiana, 398. Sideritis approximata, 525; romana, 525. Sigillaria (see Lycopodites) Vanuxemi, 561, 562, 564, 565 Silene antirrhina, 213, var. pteroneura, 213. Simarubes (Philippines), 290. Sinapis dissecta, 527; pubescens, 530. Sirosiphon, 242, 271, 279; Bornetii, 242. Sirosiphonez, 261. . Sisyrinchium bogotense, 235; chilense, 235; Clarazii, 235; pusillum, 235; tinctorium, 235. Skimmia Laureola, 387. Skitophyllum tamarindifolium, 557. Smilacina divaricata, 391; oleracea, 386, 391; purpurea, 391; sp., 391. Smilax ferox, 391; menispermoidea, 391. Solanacee, of Patagonia, 227 ; of Philip- pines, 291 ; of Tonglo, 390. 688 Solanum, 227 ; eleagnifolium, 227 ; my- oxotrichum, 426; nigrum (forma), 402; sodomseum, 528; sp. 402; Tweedianum, 227; verbascifolium, 13. Solidago linearifolia, 221 ; microglossa, 221. Sonchus, 393 ; asper, 226, 398, 401 ; ole- raceus, 226. Sonerila, 253, 287. Sopubia Dregeane, var., 402; stricta, 427. Sorghum nutans, 236. Sorostachys Kyllingioides, 107. South-Mediterranean Flora, H. Groves on, 524. Sparmannia, 394; abyssinica, 396, 398. Spartina coarctata, 286. Spathoglottis plicata, self-fertilization in, 542, 548. Spheranthus gracilis, 400; suaveolens, 400. Spherosepalum alternifolium, 321. Spherozosma, 635; excavatum, 686; excasatum, (9. Wallichi, 636; fili- forme, 636; pulchrum, f. trilobum, 635, 653 ; sp.?, 636. Spike of Cyperus of differential im- portance, 14. Spilanthes, 222; Acmella, 404 ; Heleni- odes, 222. Spilonema paradoxum, 263. Spireea belia, 388. Spiranthes africana, 502 ; australis, 539. Spirogyra Lutetiana, 248; mirabilis, 248; porticalis, 248; punctata, 248. Spirostachys patagonica, 233. Sporangium in life-cycle of Ferns, 363. Spore-generation of Ferns, 363. Sporogonium of Mosses, stalk produces .. normal plants, 366. Sporophore of Ferns, 363. Staberoha, sp., 590. Stachymacris indica, 315. Stackhousia muricata, 296. Stackhousiacee (Philippines), 290. Stamens of Cyperus, 19; their numbers &c. as a charac., 19. Statice brasiliensis, 226 ; cancellata, var. glabrata, 590, var. minutiflora, 537 ; virgata, 537. Staurastrum, 640; aspinosum, 643; Avi- cula, 641; bacillare, 642; bifidum, var., 640, 654; bifurcum, 642, 654; | brasiliense, 641; eyrtocerum, 642; cuspidatum, j3. divergens, 642; cya- thodes, 642; cejectum, B. connatum, 642; dilatatum, 642; furculostellatum, 642; furcatum, B. senarium, 643, 654; gracile, 641, B. curtum, 641; horre- INDEX. scens, 641, 654; granulatum, 643, 654; incisum, 642; inconspicuum, 642; leptacantha, 643; leptocladum, 641, var. B. cristatum, 641; leptoder- mum, 643; longispinum, 642; mar- garitaceum, var. hirtum, 640 ; minus- culum, 643; orbiculare, 643; platy- cerum, 643, 654 ; polymorphum, 643; Pringsheimii, 641; proboscidium, f. javanica, 640; punctulatum, 642 ; saltans, 641. 654; sexangulare, 642; striolatum, 642; teliferum, 642; ves- titum, O41. Stauroneis acuta, 244, 250; anceps, var. intermedia, 244, var. tenuicollis, 244, 250; dilatata, 244; Phoenicen- teron, 244. Stellaria bulbosa, 387 ; longissima, 387 ; sikkimensis, 887. Stenodia lanceolata, 227, 229. Sterculia alata, 308; Blancoi, 308; fætida, 373. Sterculiaceæ (Philippines), 290, 294, 308. Stereocaula, 263. Stereocaulei, 279. Stereocaulon furcatum, 263; ramulosum, 263. , Stereospermum Banaibanai, 314; pm- natum, 314; quadripinnatuw, 913, 314; Seemannii, 314. uu Stevia satureimfolia, var. angustifolia, 220, var. patagonica, 220. Stietina, 263. Stigmatidium erassum, 262. Stigonema, 263, 271, 279. " Stipa bicolor, 237; caudata, 237 ; Cla- razii, 237 ; intermedia, 237 ; pogona- thera, 237; pulchella, 287. Streptopetalum, 256. Striga ?, 402. a Strigula Babingtonii, 268 ; complanata, 268. - Strombosia (Timor-Laut), 373, 374. Struthiola, 394 ; Thomsoni, 404. Struthiopteris, 662. . Strychnos abyssinica, 402 ; Nux-vomica, continuity of protoplasm in, 597. le Studies in Vegetable Biology, by S. M. Moore (I.) 595, (IL) 621. a Style of Cyperus, 20 ; primary subgene founded on, whether 3-fid or 2-fid, 2]. Styracese, of Philippines, 291; of Tonglo, 389. Styrax Hooke:i, 389. ; Submerged leaves of Ran unculus Lingua, 382. , Supplementary Notes on Restiace®, by Dr. M. T. Masters, 574. INDEX. Suriraya angusta, 245. Sussea conoidea, 447. Swertia, 393, 391; pumila, 395, 402; Schimperi, 395, 402. Symbiosis, 265, 267. Symphonia acuminata, 322; eugeni- oides, 322; pauciflora, 322. Symphorema, 295. Symphyosiphon Thelephoroides, 242. Synalissa. meladermia (footnote), 279; picina, 279. Synechococcus, 241, 243, 250. Synedra, 245. Synethis, 265. Tabernsemontana affinis, 209 ; orientalis, 373; parviflora, 373. Taccacez (Philippines), 292. Tagetes, 223. Talanma mutabilis, 307; Villariana, Tangl, obs. endosperm of Strychnos, continuity of protoplasm in, 595. Tarenna macrochlamys, 411. Telephium madagascariensis, 347. Terminalia bialata, 310 ; Calamansanay, Ternstræmiaceæ, of Philippines, 290, 294; of Tonglo, 387. Tencrium Polium, 525. Thalamifloræ (Madagascar), 317. Thamnochortus, 589, 591; argenteus, 591; Burchellii, 593; cernuus, 591, 592; dichotomus, 591, 592, 593, 591 ; distichus, 590, 593; elongatus, 591, 592 ; erectus, 593 ; fruticosus, 594; imbricatus, 591, 593 ; spicigerus, 591, 593 ; umbellatus, 593. Thapsia garganica, 530. Thelymitras, self-fertilization in, 539. Thespesia campylosiphon, 308; po- pulnea, 373. Thomson, J., Plants coll. in E. Equa- torial Africa, 392. Thunbergia convolvulifolia, 428; fus- cata, 403. Thylachium laburnoides, 319 ; lauri- folium, 319. Thyielacea, near Synaptolepis, 403. hymelæaceæ, of Philippines, 292; of Tonglo, 390. hymus capitatus, 528; striatus, 530. Thysanotus, 296 ; chrysantherus, 304. Tiarella polyphylla, 388. Tiliacese (Philippines), 290, 294, 308. Tillwa minima, 218 ; muscosa, 218, Tillandsia coarctata, 234. Timor-Laut, Botany of, 370. Tina polyphylla, 335. Toddalia pilosa, 329. LINN. JOURN.—BOTANY, VOL. XXI. 689 Todea africana, 365. Tonglo, Botanic notes on, by C. B. Clarke, 384. Torreya cespitosa, 56. Tournefortia sarmentosa, 373. Trentepohlia, 25, 262, 271; bifoliata, 34 Treub, Natuur d. Lichen, 264. Trichocline heterophylla, 225. Trichomanes radicans, 664. Trichostomum, 552. Trifolium, 393; dalmaticum, 529; movements of leaflets in pratense, 625 ; repens, 629; simense, 395, var. an- gustifolia, 399 ; sp., 399. Triglochin laxiflorum, 529. Trigonella corniculata, 525; monspe- liaca, 525. Trillium Govanianum, 391. Triploceras gracile, 651. Tripteris, 394; sp. nov. ?, 401; Vail- lantii, 401. Tristellateia australasica, 373. Triuridex (Philippines), 292. Tropæolum pentaphyllum, 214. . Tropical Orchids, self-fertilization in, 538. Tsileondroaha, 350. Tupistra amarantoides, 391. Turnera, 256. Turneracex, n. gen. Hyalooalyx, from Madagascar, 256. Turrza mombasana, 399. Typha angustifolia, 236. . Typhacez, of Patagonia, 236 ; of Philip- pines, 292. Uapaca clusioides, 441; myricefolia, 440. Ueberlinia, 395; rotundifolia, 397. Umbel of Cyperus, nature, develop., and analysis of, 11; value as character doubtful, 15. Umbelliferze, of Patagonia, 219; of Phi- lippines, 291. Umbilicaria pustulata, 277, 282. Umbilicus horizontalis, 525, 528. Ungeria monocephala, 34. Unona, 319. . Uredines, Heterzecious, reproduction in, 368. Urera, 445; acuminata, 446 ; sphæro- phylla, 445. - Urostigma (Timor-Laut), 371. Urtica neglecta, 530; spathulata, 234. Urticacer, of Patagonia, 234; of Philip- pines, 292; of Tonglo, 390. Usnea ceratina, var. scabrosa, 272. Utricularia ibarensis, 427; rosulata, 301; spartea, 427. 3G 690 Vacciniacez (Philippines), 291, 294. Vaccinium Nummularia, 389. Valerianella Vesicaria, 529. Vandez (Madagascar), 467. Varicellaria, 265. Vegetable Biology, Mr. Moore's Studies in, 595, 621. Veprecella hispia, 342, 343. Verbascum garganicum, 536 ; macru- rum, 528; viminale, 536. Verbena bonariensis, 230; erinoides, 231; incisa, 231; litoralis, 230; teucrioides, 231. Verbenaces, of Patagonia, 229; of Philippines, 291, 293, 294, 315. Verbesina australis, 222. Vernation and Methods of develop- ment of Foliage as protective against radiation, Rev. G. Henslow on, Vernonia Andersoni, 312, 313; cinerea, 373; Cumingiana, 297, 312, 313; Melleri, var., 400 ; philippinensis, 297, 312, 813 ; polytricholepis, 415 ; strep- toclada, 416 ; trichantha, 416 ; voluta, 415. Veronica cana, 390. Veronicacee (Tonglo), 390. Verrucaria Garovaglii, 266; 262. Vesicaria andicola, 212. Viburnum cordifolium, 389 ; erubescens, 389 ; luzonicum, 298, 310. Vicia sativa, var. Cosentini, 530. Vigna lutea, 373 ; sp., 400. Villaria philippinensis, 311, 316. Vinsonia humilis, 449 Viola, 387; abyssinica, 395, 397; bi- flora, 387, var. canescens, 387 ; dis- pm 387; Hookeri, 387; serpens, nitida, Violariee, of Philippines, 290; of Tonglo, 387. Viscum apodum, 439 ; cuneifolium, 438 ; granulosum, 438 ; lophiocladum, 437 ; Radula, 439; rhytidocarpum, 438; trachycarpum, 439. Vitacese (Tonglo), 388. INDEX. Vitex littoralis, 297 ; Negundo, 373; trichantha, 434 ; trifolia, 373. Vitis himalayensis, 388. Vittaria elongata, 374. Vivaona, Madagascar name for Dilobeia Thouarsii, 437. Voacanga Cumingiana, 301, 313. Volvocacez (Afghan), 249. Warea tonglensis, 388. 20. | Weddelia biflora, 373; mossambicensis, 401. Weinmannia fraxinifolia, 339 ; minuti- flora, 339. White, C. F., on Pollen from Funereal Garlands found in Egyptian Tomb, | 251. Wightia, 255. . Willdenovia Luceeana, 590, 592 ; striata, 598, 594. . Wollaston, G. B., obs. on devel. bulbils of Ferns, 358. Woodwardites arcticus, 659. | Wornaskioldia, 256. | Xanthidium acanthophorum, 643; anti- | lopeum, 643, f. angulatum, 643, 654, | f. javanica, 643, g8. triquetrum, f. brasiliense, 643 ; armatum, 643. Xanthium ambrosioides, 222. Xanthostemon Verdugonianus, 296, Xylophylla ensifolia, 336. Xyridez (Philippines), 292. Zanthoxylon acanthopodium, 387; oxyphyllum, 387. Zehneria aff. mucronate, 373. Zephyranthes Adersoni, 235. Zingiber, 254. . Zollingeria macrocarpa, 309; tripters, 309 Zoochlorella, 269. Zooxanthella, 269. Zygnemaces (Afghan), 247. _ Zygoglossum umbellatum, 465 hi Zygophyllez, of Patagonia, 213; of Phi- lippines, 290. END OF THE TWENTY-FIRST VOLUME. PRINIED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET. Dum e Eg oe unn a td M \ SS PS Li G I mri mel Q 43 e : _ ~ = C zx "2 D ————— 088 me £ | 7 m 5 Qd. S. ( E du t } TSS Tl > v uc ÁN —— - PEN i ~ a a N j WI = . SS EN p UL, XM Wwrevmmmmmmm mmm T yp [ STI uL—— — — MUI] : "uum Uia i EIE Rin. m m"— m A EL cO egg au uL EHI CH (t) Lug nM UM z RARI NT NL i Eo eaa 2X5 ME R AIDS YPERACEA . unten Brosi i Z 2 z 2 S Ha 4 |. C.F. White Linn. Soc.Journn. Bor. VoL. XX. Pr..6 ©, xi Fute. del 1.2.5. ANCENT EGYPTIAN. 9, eqeau hth + NE + RECENT POLLEN AND ANTHER Oi PAPAVER RH(EAS. Linn.Soc. Journ. Bor Vou. XXL Pr, 7. o mm m RUMP EN S "Ww > A Sx v HYALOCALYX SETIFERUS, Rolfe. IN.Fitch ime Q o ounn. Bor. Vor. XXL. Pr..8 e ® Mintern Bros .unp. LI / ` — M BÀN LL C — I — "d \ p SN / / V T Jj Soc. LICHENS. LINN. T HYPOTHALLINE CONDITION OF E: E E- i an z ut 8 5 5 4 1 - P = 2 co We 2, 7o, b a Bit J s v. * ae y HH HY 3 . Li / esas E + "n HYPOTHALLINE ^ iv. INN. Soc.JourRN.Bor Vou. XXI. Pr. d T L Rolfe. qe. VILLARIA PHILIPPINENSIS, £c 1 n. RA Rolfe del. N Fitch lit {v — fe 2 A : N : NN. Soc. Journ. Bor Vo ORY IN FERNS. . APOSP l-7 ATHYRIUM FILIX FCEMINA ver Cl Bow er. N ) / ~ A ` N APOSPORY IN FERNS Hm 3. ATHYRIUM FILIX FOZMINA var CLARISSIMA. COIA TRAT t+ ewes a ani ` an ratt cr i aj . a : >x s ~~ ~ d ~ - > . is ^ [te] E ~ Q ^ EH E E - oO 3 c zZ, E Z = c ~ — "n ) - ) z ) < 2 - 3 J z 5 A D v 3 9 9 2 75 —- 043 Lo 1 \ ) N (Y icaves | eUbmerno hth leaf 7 | Submerged IN 1 | Linn. Soc. Journ Bor. Vor, XXI Pr HN Ridley imp. Fitch ~ A 14 BICORNELLA GRACILIS. B.5-9 B PARVIFLORA JN Fitch lith LINN. Soc. Journ. Bor. VoL . XXI. Pr..16. Fitch. hth Forbes d . . ee ad. SELF-FERTILIZATION IN ORCHIDS. p? zu > A Se >x< S e Es] ai & 2 O Ss [2] c A z S r eaP | £D. Sp.; £.Javensis m SP. SOGLOSSU Chry. Vood yera Fitch lith FERTILIZATION IN ORCHIDS. SELF- FR Kidston . Linn. Soc. Journ. Bor Vou . XXI. PL.18. Kidston del. Hammond lth H Hanhart imp LYCOPODITES VANUXEMI, Goeppert, sp. / \ aa No7 CONTINUITY Linn. Soc. Journ. Bor Vor XXI. PLI9. i 1 j 1 IOPLASM fint | Linn Soc Journ. Bor Von. XXL. F12 / / \ H N N \ i h i | j | ww NTINUITY OF PROTOPLASM Mintern imp - ff ior Vou. XXI Fi 2l Linn. Soc. Journ Cr FAO ive VI da. YR CONTINUITY OF PRC Linn. Soc. Journ Bor Vor. XXI P1.23 W. Joshua dei. V p X v 7 zT ( wt W.Joshua del a RN N BN a BURMESE DESMIDIEA E Carter sc. 25 Vor XL PI LINN Soc. Journ. Bor ] i 1 Joshua de te UD tie tla ste? REUS LN a Se © "E La N "n T at a ee Se es Ste es yum sirami MS BE L DAC #0949008 60: Aero ct j IN op X 300 J Starkie Gardner. Linn. Soe. Journ. Bor VoL. XXI P1. 26 NC WZ Mibi temere ren repito x Morgan hth. l _ » A )IC 7 GLE NIA HIBERNICA. SON ,EA HitBRIDICA.5 [GLE ICHE? l NOME URII. 10-11, GLÉEICHENIA, Sp 8,GONIOPTIZ RIS STIRIACA 9,G.BUN BUI West Newm anım