ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA Vol. XI. ANNALS m P ROYAL BOTANIC GARDEN, CALCUTTA. iliam. < > Vol. XI. - —————— 00 ASIATIC PALMS-LEPIDOCARYEAE By DR. ODOARDO BECCARI. DUET. I. THE SPECIES OF CALAMUS WITH 238 PLATES. OA COU TTA Printed at the Bengal Secretariat Press 1908. i ASIATIC PALMS-LEPIDOCARYEAE By Dr. ODOARDO BECCARI. THE SPECIES OF CALAMUS WITH 288 PLATES. CALCUTTA: | Printed at the Bengal Secretariat Press. 1908. Published at the BENGAL SECRETARIAT Book DErôrT, Writers’ Buildings, Calcutta. IRSE NERIS QIHE!ICI.JAAIL AGENTS. In India— Messrs. THACKER, SPINK & Co., Calcutta and Simla. Masses. Newman & Co., Calcutta. Messrs. HIGGINBOTHAM & Co., Madras. Mzssns. TRACKER & Co., Lo., Bombay. Messrs, A. J. ComBrrper & Co., Bombay. Tux SUPERINTENDENT, AMERICAN Baprist Misston Peress, Rangoon. Mrs. RApHABAI ATMARAM Sacoon, Bombay. Messrs. R. Campray & Co., Calcutta. Rar Sans M, Guras Sineu & Soss, Proprietors of the Mufid-i-am Press, Lahore, Punjab. Messrs. Tuowrsox & Co., Madras. Mzssas. S. Murtuy & Co., Madras. Mzssns. Goran Narayen & Co., Bomba Messes. 8. K. Lanier & Co., Printers and Book-sellers, College Street, Calcutta. Messrs. V. Katyanarama IYER & Co., Book-sellers, &., Madras. Messrs. D. B. TangaroREVALA, Sons & Co., Book-sellers, Bombay. Masses. G. A. Nargsonw å Co., Madras. Mz. N.B. Maruvr, Superintendent, Nazir Kanum Hind Press, Allahabad. Tug CarncurrA ScHoor-Boox SociETY. Mz. Sunpge Panpurane, Bombay. Messrs. A. M. & J. Fencuson, Ceylon. Messrs. TEMPLE & Co., Madras. Mzssns. Compripes & Co:, Madras. Messrs. A. R. Prttar & Co, Trivandrum. Messrs. A. CnaxpD & Co., Punjab. In England— Mz. E. A. ARNOLD, 41 & 49 Maddox Street, Bond Street, London, W. Messrs. CoxsrABLE & Co., 10 Orange Street, Leicester. Square, London, W. C. Messrs. GrRINDLAY & Co., 54 Parliament Street, London, S. W. Mzssns. KzoaN PauL, TRENCH, TRÜBNER & Co., 43 Gerrard Street, Soho, London, W Mr. B. QvanrTCR, 11, Grafton Street, New Bond Street, London, W. Mzssus. W. TnackgR & Co., 2, Creed Lane, London, E. C. Messes, P. S. Kine &jSox, 2 & 4 Great Smith Street, Mzsszs. H. S. Kino & Co., 65 Cornhill, London, E. C. Mz. B. H. BrackwzLL, 50-51 Broad Street, Oxford. Messrs. Detcurow BELL & Co., Cambridge. Mz. T. Fisoer Unwin, 1 Adelphi Terrace, London, W. C. Westminster, London, S. W. On the Continent — Mzssns. R. FRIEDLÄNDER & Sonn, Berlin, N. W. Carlstrasse, 11. Mz. Orro HanRASSOWITZ, Leipzig. Mz. Rovotr Havrr, Halle.a-S, Germany. Mz. Kart Hrersemany, Lei ig. Mz. Ernest Lerovx, 28 Rue Bonaparte, Paris. Mz. Martinus Nisnorr, The Hague, * In palmis semperparens juventus ; in palmis resurgo." Mart. DEDICATED TO THE ILLUSTRIOUS MEMORY OF William Griffith 1810—1845. FELLOW OF THB LINNEAN SOCIETY OF LONDON; MEMBER OF THE IMPERIAL ACADEMY OF NATURAL SCIENCES OF BONN; CORRESPONDING MEMBER OF THB IMPERIAL y BOTANICAL SOCIETY OF RATISBON, OF THE ROYAL ACADEMIES OF SCIENCES OF TURIN AND CHRISTIANIA, AND OF VARIOUS OTHER LEARNED SOCIETIES. SURGEON IN THE MADRAS ESTABLISHMENT THE HONOURABLE FAST INDIA COMPANY AND SUPERINTENDENT OF THEIR BO ANIC GARDEN AT CALCUTTA FROM 1842 TO 1844. AUTHOR OF PALMS OF BRITISH EAST INDIA. Caleutta, March 1908. PREFACE. ——— : A monograph of the genera Calamus and Daemonorops was prepared by me many years ago, as part of a general work on “ Asiatic Palms” which it had been my intention to publish in my “Malesia.” The “Istituto di Studi Buperiori" of Florence having, however, withdrawn its subsidy towards the continuation of the work intended to illustrate the Malayan botanical collections gathered by me—collections which had become the property of that Institute— my studies of the Asiatic Palms were interrupted, and only a partial use of my notes could be made by Sir Joseph Hooker in the sixth volume of the “Flora of British India.” | The present monograph would probably never have seen the light had it not been my good fortune, in 1899, to meet in Florence Sir George King, who most courteously and generously offered to arrange for the publication of the work in the “Annals of the Royal Botanic Garden of Calcutta,’ of which publication he had been the founder. This unexpected and welcome aid enabled me, after ten years’ interruption, to resume the stucy of the two genera mentioned, and now, thanks to the most valuable help given me by Lieut.-Col. D. Prain, the worthy successor to Sir George King in the Superintendentship of the Calcutta Botanic C arden and in the Editorship of these “Annals,” this monograph of the genus Calamus is issued, with the hope that monographs of Daemonorops and of the remaining Lepidocaryeae, if mot of all the Asiatic Palms, may follow. In the course of my study of Calamus and Daemonorops I have had the advantage of dealing with very extensive material. No pains have been spared. in bringing together specimens of these plants from their native countries, The greatest liberality has, moreover, been shown by the Directors of the leading botanical establishments, who have accorded me full opportunities of making use of the collections under their charge. As a result I have been able, except in the case of five species described and figured only by Rumph, but not met with again by modern botanists, to provide a description of my own and to take a photograph of every one of the some 200 known species of Calamus. ‘The list of those who. have in various ways aided me in my work. is a oe one, and I cannot too heartily and gratefully acknowledge the valuable assistance I have received from so many friendly helpers; but I feel above all deeply indebted to Sir George King and to Lieut.-Col Prain, since these gentlemen have provided me with an opportunity of publishing this work. accompanied by natural-size phototype reproductions of my negatives. I am also under much obligation to them for sets of the Calami, and other — Palms ‘from the Calcutta Herbarium, placed freely at my nem whenever required. I have likewise to express my thanks to Captain A. T. Gage, Ann. Ror. Bor. Garp. CarcurrA Vor, XI. ii Curator of the Calcutta Herbarium, to whom I fear I have on many occasions given much trouble during the preparation and publication of the present volume. | I wish also to tender my warmest thanks to Sir Joseph Hooker, who has encouraged me to undertake the task of preparing a general work on the Asiatic Palms, and to Sir William Thiselton-Dyer, his successor in the Directorship of the Botanical Museum of Kew, which has been my main source of information regarding the Palms that form the subject of the present study. I have likewise to express my thanks to Professor A. Engler for the loan of the sets of Calami and Daemonorops of the Berlin Herbarium, carefully arranged for me by that enthusiastic Phoenicologist Dr. Udo Dammer. The Berlin collection has proved very rich in new species, chiefly the fruits of the explorations of Dr. Merrill in the Philippines, of the German botanists in New Guinea, and specially of Dr. Warburg in various parts of Malesia and the adjacent countries. My cordial thanks are also due to Professor L. Radlkofer who has granted me the use of some of the type specimens of Martius, preserved in the Herbarium at Munich; to the late Professor Crepin and to Professor Durand for still other types of Martius that exist at Brussels; to the late Professor Suringar and to my lamented friend Dr. Boerlage of the Leiden Herbarium who selected, on my behalf and sent to Florence, an instructive specimen of every one of the species of Blume; to Dr. J. W. ©. Goethart, who has more recently sent me valuable contributions from the same s Rijks Herbarium of Leiden; to the late Professors Regel and Maximowicz, also lost friends; and to their successor Professor Fischer von Waldheim, for the loan of the entire collection of Palms belonging to the St. Petersburg Herbarium. I have also to thank the following friends for their kindness in supplying me with specimens of Palms of which I was in need:—Mr. Casimir de Candolle of Geneva; Dr. John Briquet, Conservator of the Herbarium Delessert; Mr. G. Beauverd, Conservator of the Herbarium Barbey-Boissier : Dr. A. Zahlbzuckner of the Vienna Herbarium; Professors E. Bureau 5 Poisson and E. Bonnet of the “Muséum d'Histoire Naturelle” of Paris. To Mr. H. N. Ridley of the Botanic Garden of Singapore I am indebted for many of the Palms that grow in that island and in J ohore, regions which he has thoroughly explored, momenta But for the largest contribution to my work I am under a das | T NE io the Rev. Father.Scortechini, so untimely lost to Science, who most po pa m < bg the whole of his Malayan Palms, with his nores and drawings, altho this group of plants was ore of th d intended to eta M | | | ose he dM intended 111 I have derived valuable help from a most splendid set of the Calami and Daemonorops cultivated in the Botanic Garden at Buitenzorg, beautifully represented by extraordinarily large and complete specimens. For this collection I am indebted to Dr. Melchior Treub, the eminent Director of that great establishment: to him and to his assistants I wish gratefully to express my warm thanks. It gives me much pleasure also to acknowledge the kind help of many friends who have most generously supplied me with invaluable specimens from their collections. Thus I have to thank my late friend Baron Ferdinand von Mueller for many Australian and Papuan Palms; Mr. Louis Pierre, of whose monumental Forest Flora of Cochin China the botanical world greatly regrets the discontinuance, for many Indo-Chinese specimens; Sir D. Brandis, Mr. C. B. Olarke and Mr. J. Sykes Gamble, for Palms from various parts of India; my late dear friend Signor Leonardo Fea, for Palms from Central Burma; the late Dr. K. Schumann for some from New Guinea; Dr. Schweinfurth for the few species growing in the Niam-Niam country in Central Africa; Mr. Gustav Mann, formerly Conservator of Forests in Assam, for an almost complete collection of the Palms of the various districts of that Province; and Mr. E. H. Man, for an equally important collection from the Andamans and Nicobars, of whose Civil Commission he was for so many years a member. In conclusion, it may be remarked that from the commencement of my own explorations I gave special attention to the collection of Palms. The material brought together by myself to represent these Princes of the Vegetable Kingdom is, therefore, as regards the tropical Asiatic Archipelagos, probably more important than that existing in any other Museum. ‘This material is now the property of the “Istituto di Studi Superiori" of Florence; and I feel certain that the authorities who superintend it must be very pleased to see an important part of their collection now magnificently illustrated through the enlightened munificence of the Government of Bengal. O. BECCARI, FronENcE, 1905. CONTENTS. I Introductory Essay ep iv The Leaves Al up EIN . . Ro 4^ UM v The Leaf-Sheaths . ‘ ‘ . , . e . . vi The Leaf-Sheath foem Lo 12 4 14. 0*4 9 vii The Ochrea . " ‘ ^ SU . e viii The Petiole ee as : V o s ko 4*1 ix The Rachis . . š : ; * : ^ i š b e INUNMD 5 4 (Wc e eas ee A xi TheSpadices . .. - - . Ko WW oe . xii The Spathes : " ‘ o rg : s e RUE xii The Partial Inflorescences uni Jae oe (74. o xiv The Spikelets .- . Eo o4 9 ey xv The Flowers . ‘ . e ‘ . . e e . xvi The Fruiting Perianth . (080023 ^ E UL RUN xvi The Fruit. š : - e . : " . . i xviii TheSeed . . : . e ; ‘ . . xix Extra-floral Nectaries í ‘ ‘ . . . i R xx Calami under Cultivation . ^ ; . xxi Uses of, Trade in, and native boi of e P ` s xxii Note on the species of Calamus in the Philippine Islands . . xxiii Identification of Loureiro’s Cochin-China Calami e e xxiv Note on Roxburgh's species of Calamus (o ox o xxv Note on Rumph's species of Calamus and Daemonorops . : xxvi Note on the date of publication of the species of Calamus Daemonorops named by Martius, Griffith and Blume * xxvii On the classification of the species of Calamus xxviii Diagnostic characters of the General Calamus and Daraionorope ‘ xxix Calamus or Palmijuncus? bo a á e : xxx Geographical Distribution . : : : : 2 j Definition of the genus . Eo 4 Poe ^ v — « Anomalous or infrequent chatacters ‘ ‘ 4 : ^ š ‘ Systematic conspectus e o” lt lt - -* -* n Eua Oe MEME . 2 o0 o 5 * * ^! Description of species . . CC e Cw Description of Plates I—II accompanying wt woo» No Biological and died Notes Ge LU ow ee Spinosity . . , . > - . s ° The Stem ` . . e A . . . * Nomina nuda " Eoo oe es DOCU a S o « e t. Se . — — * | * t -—* CALAMUS. INTRODUCTORY ESSAY. 1. Biological and General Notes. Tue species included in the genus Calamus are usually slender elegant Palms. which have, during the evolutionary period, acquired the power of raising their leafy crown above the heads of the loftiest trees in primeval tropical forests of the Old World. We may suppose that originally the species of Calamus were delicate, standing Palms endowed with a very active and rapid growth, and of such a structure as to render the increase of their stem in length more easy than its increase in diameter. This peculiarity, coupled with their tendency to overgrow other plants in search of light and of eonditions more suitable for the fertilization of their flowers may have been acquired, according to an hypothesis of my own,* during the period—very remote in the evolution of organised beings— which I have termed the plasmatical era," by means of the hooked spines with which these palms are furnished, which endow them with the faculty of suspending themselves from neighbouring plants and even of rising above them. A Calamus in order to change its erect habit into a climbing one had, according to this hypothesis, necessarily first to acquire the organs needed to ensure this essen- tial condition of. its existence. These organs are, with hardly an exception, common spines which have assumed a hooked shape. The means whereby Palms have been enabled to acquire spines of this kind is ee one of those morphological problems which, like the metamorphosis of any other organ of living beings, we are unable to explain scientifically, but as to which we are obliged to remain satisfied with some more or less plausible hypothesis. I suppose therefore that the spinosity of Palms, especially that which besets the leaf-sheaths, was originated by the stimulus induced in the very sensitive peripheral tissues by animals in search of nutriment in the youngest and most tender parts of the plant.T I suppose therefore that'the young central parts of every spinous Palm: must have been coveted for nourishment by numerous animals, had they not been defended by spines. It is quite impossible for me to explain now, even superficially, how the stimulus produced by the action of certain animals on the irritable vegetable tissues may have given rise (in very remote times) to hereditary epidermal outgrowths or hyperplasia; organs of such a nature as we may consider the spines to be. Nor can I explain how it is possible that the stimulus which exercised its action at a definite point may have induced the production of spinous organs in almost every other part of the plant. As we have already seen, the spines which enable the Calami to climb are hooked, or are of the kind that, in the descriptions of the species, we have agreed to term “claws.” These claws are almost exclusively met with on the axial parts of the spadices, on the leaf-sheath flagella (abortive spadices), on the leaf-rachis. and, especially, on its prolongation or cirrus. | |* Beccari: Nelle Foreste di Borneo, p. 293. f Beccari: Nelle Foreste di Borneo, p. 179. Axs. Roy, Bor. Garp. CarcurrA Vor. XI. | : 2 INTRODUCTORY ESSAY. According to the view now explained we can imagine that the hooked spines owe their origin to a special sensibility of the protoplasm, which at certain determinate points, where the reaction to the stimuli happened to be more effective, induced the tendency to stretch towards and twine round extraneous and heterogeneous bodies. The causes which have given origin to the hooked spines ought apparently to belong to the class of causes which have produced the numerous other contrivances whereby an erect plant acquires the power to climb. This statement is, to a certain extent, borne out by the fact that all Palms which have hooked spines on the leaf- rachis and on the spadices, or which have clawed cirri at the ends of the leaves, or have the leaf-sheath flagella similarly armed, are climbing species; whereas, when a Palm is only armed with straight spines it is certainly an erect one, or is bushy and short-stemmed, like Mauritia, Eugeissonia, Sagus, Zalacca, | Oncosperma, etc. All scandent Calami, or to speak more generally, all climbing Asiatic Palms owe their fitness for this kind of existence to the transformation into “claws” of the short straight spines which defend certain parts of the plant; whereas species of the American genus Desmoneus and of the African genera. LEremospatha and Ancistrophyllum, which also climb, are indebted for this property to the transforma- tion into rigid and valid hooks of the smaller and apical leaflets of their fronds.* Among the -numerous species of Calamus known to me, only one, the small almost stemless C. pygmaeus, found on the tops of the mountains of Borneo, is able to raise itself a few feet from the ground through the surrounding. shrubs by means of the small deflexed branchlets of its filiform^spadices, which act as hooks. As all Calami have, without doubt, originated in the densest tropical forests, and as the power of attaining the summits of the trees among which they have to struggle for.air and light has been one of their most important needs, contrivances of various kinds and different nature have been evolved for the accomplishment of this purpose; consequently, the numerous modifications induced by this circumstance in their organs, supply most important characters whereby it is possible to distinguish the various species of the genus Calamus, The spinosity, together with slight morphological modifications in certain organs, is the principal contrivance by which Calami climb. These modifications are:— (1) the extension of the summit of the leaf-rachis into a cirrus: (2) the extension of the main-axis of the appendix into a long filiform-clawed appendix : (3) the transformation of the spadices into long whip-like clawed Organs, With the help of these characters it is nof diffieult to bring the greater part of the species of Calamus together into more or less natural groups. The spathes also afford important characters that can be made use of in classifying the species of Calamus, though to a less degree than in the case of Daemonorops, a genus in which the chief funetion of the spathes is to protect the flowers whereas in (Calamus the Spathes also often to a certain extent assist the plant to climb, oo © Owing to. the seoulies anda ce S "Owing to the peculiar arrangement which prevails in the Botanical Museum of Florence, where the library ds been placed some kilometres away from the herbarium, I have not of late had any opportunity of consulting e paper, by F. O. Bower, on the modes of climbing in the genus Calamus (Annals of Botany, vol. I, p. 126). BIOLOGICAL AND GENERAL NOTES. 3 The spinosity, length and scandent nature of the stem, and the unwieldy size of the leaves and spadices of many Calami are causes of much annoyance, labour and loss of time to the botanist who is obliged to collect specimens of such plants. On thé other hand, this study has been greatly hindered by the very incomplete and fragmentary manner in which, on account of these difficulties ın collecting them, the species as a rule are represented in herbaria. Another not uncommon cause of error has been the differences that exist between the spadices of the two sexes, to say nothing of the association, which has frequently taken place in herbaria, of the leaves of one species with the spadices of another, Moreover, not a few species have been based on specimens belonging to only one sex, or on portions of but a single plant, and it has seldom happened that the description of a species has been based on the inspection of a good and large suite of specimens taken from many individuals. If indeed we had been content to describe only those species of which we possessed complete specimens, that is to say entire full-grown leaves, male and female spadix and fruit, the number of thosé secured to science would have been far smaller than it now is. Owing to the fragmentary nature of the material available for the study of these Calami it has been necessary to give a very rigorous, minute and full description of the various parts of the specimens actually at hand. These often have belonged to but one individual ; consequently the descriptions now offered, like those of other authors, are not only in many instances lacking in completeness, but frequently fail to give all the characters essential to the collectivity of individuals that constitutes a species. They only indicate the peculiarities of a single member of such a collectivity, or to be more precise those of only a small portion of some of its organs. I am led to make this remark because in more than one instance it may happen that the specific characters which I have assigned to a species will prove inconstant, or, when larger and more complete specimens become available, even be found to be not altogether exact. It is indeed impossible, when we have only one half or a fragment of a leaf available for study, to form a precise idea of the degree of variability or of the extent and amplitude of the characters of the leaves of a given species. When, for example, the statement is made that the leaf of a particular species has a_ petiole 10 cm. long, with three pairs of leaflets on each side of the rachis, it is not meant by this that in every leaf of the numerous individuals which constitute the species the petiole must always be of that length and the leaflets always be exactly of this number. The phytographer, when he has not sufficient material at his: disposal, is of necessity compelled to describe the individual. On the other hand, the student who tries to recognize a species from a description made under such circumstances will be careful to give no more than their due importance to the characters assigned by the author and will make sufficient allowance for variability. In the case of Calami, as in that of many other Palms of large size, the conditions are not as they are in the case of small plants whereof a considerable number of specimens from different places may be brought together for comparison’ in the herbarium, so that the student is enabled to acquire a complete knowledge of the degree of variability exhibited by the individuals which constitute a species, From what has been said as regards the very fragmentary condition of the material Ayn. Roy. Bor. Garp. Gatcurra Vor. 2 4 INTRODUCTORY ESSAY. now available for the study of Calami, it will be realized that, very often, even the most essential characters of a species have been derived from small portions of a single plant, which have accordingly had to be described with great precision because we cannot predict which of those characters are constant and which are variable. The fragmentary state of herbarium specimens of Calamus is, moreover, often a great hindrance in the grouping of species according to their natural affinities, and this task has more than once to be accomplished by the author as the result of a mental reconstruction of the entire plant from the few fragments that are available to him for study. The incompleteness of the specimens of Calami, the difference between the male and female spadices, or between the flowers and the fruit, and in particular the not infrequent fact that the spadices in herbaria are not from the same plant as the associated leaves, have Jed even the most experienced botanists to propose a few non-existent species: nor am I certain that the present work is entirely free from the same blemish. | II.—Spinosity. The nature of the spinosity on the different organs of the Calami is of great diagnostic importance, We know of no quite spineless Calamus and even C. Meyenianus, which usually has unarmed leaf-sheaths and leaves, now and then bears a few prickles on the spadices. In every case the spinosity of Calimus is due to an hypertrophic growth of the peripheral tissues of the plant. I know of only one exception to this rule, in C. javensis var. acicularis (PLATE 41), which bears on the petiolar part of the leaf a few spinescent organs analogous to the long hooks that occur on the prolonga- tion of the leafrachis in the African palms of the genera Ancistrophyllum and Eremospatha and in the American genus Desmoncus. The spinescent hooks which provide the Palms of the genera mentioned with the means of climbing are obviously quite different as regards origin from the hooked thorns of a Calamus, and are no more than transformed and abortive leaflets; of this nature also I take the spines at the base of the petiole of C. javensis var. acicularis to Ve. The spines of Calamus are to be differentiated into those that serve for defence and those that help the plant to climb. Spines of the former category are long and straight, and it is to indicate these that in the descriptions of species I have restricted the proper name of “spines.” They are essentially organs of defence, to be found especially on the leaf-sheaths, at the bases of the petioles, and sometimes also on the leaf-rachis and on the primary spathes. When these true straight spines are shorter and broad-based they are more particularly termed ‘‘ prickles” Spines of very small size, or “spinules,” are besides produced on the main nerves or “cost” of the leaflets and on their margins ; ( from these we gradually to rigid or criniform ‘‘hairs” and to “ bristles.” = . The spines which subserve the function of fixin and so enabling it to climb, are short with a broad sharp tip so that they resemble the hooked nails of g the plant to adjacent trees, base, and have a curved very a cat; these in the descriptive SPINOSITY. 5 portion are especially termed “claws.” There are, however, numerous transitions between the two principal sorts of spines. In the non-scandent species true claws are very scarce, very imperfect, or wholly absent, Calamus Oxleyanus is the only species known to me which is provided with a long and powerfully clawed cirrus at the extremity of the leaf, but nevertheless ‘appears to be a non-scandent Palm. With this solitary exception, which (see observations on C. Oxleyanus) is perhaps not absolute, the non-scandent species of Calamus are almost exclusively armed with straight spines. C. erectus, for example, is densely be- set with long laminar spines on the leaf.sheaths, on the petiole and leaf.rachis and spathes, and only bears some small rudimentary claws on the tail-like appendix at the summit of the spadices. The spines which defend the leaf-sheaths are usually laminar, flat, elastic, more or less elongated, very sharp, often smooth and shining, light.coloured, brown or almost black or of the colour of the surface of the sheaths, solitary or scattered, or more or less confluent by their bases and seriate, or even disposed in annular horizontal or oblique rings or whorls; the spines near the mouth of the sheaths are often longer and more slender than the others. Frequently with the larger spines are intermingled smaller ones or even small wart-like pungent tubercles. In some species (C. platyacanthus, C. ornatus, C. palustris, etc.) the spines have a very broad base which is concave beneath and swollen above, where it is sharply separated by a definite line or narrow furrow from the laminar acuminate point. Sometimes in place of spines the sheaths bear brittle rigid. eriniform bristles, as in C. Muellerü, or small rigid hairs seated on a bulbous base, as in the species of the group of C. ciliaris; in these the hairs are usually deciduous with age or at a certain time separate from their bulbous permanent bases; these latter render scabrid the surface of the parts on which they are situated. In one variety of C. tenuis the bases of the spines are unusually extended laterally, while the points remain atrophied; and, as a few of these spines are aligned close together, their bases remain in contact right and left so as to form continuous, slightly raised, more or less oblique submembranous ridges across the sheaths. In C. corrugatus the surface of the sheaths is rendered uneven by the presence of merely annular raised wrinkles. Some spines when young have their margins fringed with a furfuraceous scurf. Very rarely the spines of sheaths are curved or hooked, and I recollect only C. javensis VAR. tenuissimus as a form in which they have a tendency to become so. The spines on the petioles, especially those near their base or along their margins, as well as those that sometimes occur on the first or basilar spathe or on the rachis of certain leaves are usually less laminar, thicker and stronger than those of the leaf- sheaths. : Very peculiar are the spines of C. tomentosus, which consist of small black points rising from the centre of small mamillate swellings or tubercles, Very curious too are the spines of the leaf-sheaths of O. radulosus and O. spathulatus, which instead of being, as is usually the casé with the ^ spines of leaf-sheaths, horizontal or deflexed, are ascending, semiconic, short, thick, fiat beneath and have a distinct axillary swelling at the base in their upper part. | ibo | 6 INTRODUCTORY ESSAY. Sometimes the spines are confluent and, by their united bases, form membranous crests which are crowned by permanent or deciduous needle-like spiculae. Not infrequently the spines, especially when rather strong, leave on the surface of the sheaths or even on the underside of the leaf-rachis, a more or less distinct and deep impression of their outline; this is due to the fact that when the leaves are closely packed in the terminal shoot the spines are turned upwards, and are pressed against the surface of the organ from which they originate, while, after the expansion of the leaves, these spines become at first spreading and at length hori- zontal or deflexed. Whenever the rachis is prolonged beyond the ultimate leaflets as a filiform or whip-like appendix, or when a similar appendage crowns the summit of the spadix, or when leaf-sheath flagella are present, the armature of these parts consists of hooked prickles or claws, while at the same time the lower surface of the leuf-rachis and the attenuated portions of the spadices interposed between two partial inflor- escences are usually similarly armed. The claws at times are slender but more frequently they are robust, with a broad and swollen base and a very sharp and short curved point. In the first or lower portion of the rachis they are usually solitary, but they became 2-3-nate and even digitately 5-nate upwards when the rachis is prolonged into a cirrus; in this ease the claws usually form ł-whorls at regular intervals. In very robust species they even form {-whorls, the claws being then 6-7 in number and confluent by their bases. It is, however, very rare, except at the extreme apex, for the circle of claws to be closed and form a complete whorl. The leaf-sheath flagella and prolongations of the spadices, when present, are similarly armed. . The different kinds of spines of Calamus are all, as already stated, outgrowths of the peripheral tissues and consist of their lengthened cells with tapering ends (clostres), which are very closely united and form externally a very resistent sheath of prosenchyma, while the interior is composed of a more or less abundant wall-like or muriform parenchyma. In the spines of ©, Flagellum var. karinensis I have happened to find besides a few slender bundles of spiral vessels. Every kind of spinosity, but in an especial manner that of Calamus, is assumed by me to owe its origin to certain stimulative causes acting on tissues liable to hypertrophic cellular growth, I suppose therefore that a bite, a sting, even simple contact, or any other agency capable of irritating the !protoplasm, when in the full force of its vitality, may have produced, during the plasmatical era—that is to say, in the very early periods of the evolution of organisms, and under special eireumstances—certain local hypertrophic excrescences. These, having in the course of time acquired the power of being transmitted to the progeny, may have been changed into spines, | In the category of spines, rather than in that of rudimentary leaf-blades, I would also include the scales of the loricated fruits of the Lepidocaryee, III.—The Stem. The stem of Calamus is usually slender and very long; for as Calami are Palms of very rapid growth, the annular thickenings or rings of the stem where the leaves THE STEM. 7 are attached are very far apart and consequently the internods are very long and sheathed by the basal tubular part of the leaves. In a very few species the stem is erect ; in this case it never attains a great height. In C. arborescens, which is I believe the largest known erect Species, the stem reaches a height of 5-6 metres, and, when divested of its leaves, & diameter of, at most, 6 cm. The stem of C. bacularis, another non-climbing species, is about the size of a common walking cane. The species of Calamus, which possess erect stems, appear to be of less rapid growth than the climbing ones, as their joints are rather short. I know only one =- Species which is almost stemless; this is C. pygmaeus the small Bornean Palm . already alluded to, certainly the most diminutive species of the whole genus, C. Lobbianus appears also to be an almost stemless plant. ©. tonkinensis and C. salicifolius are bushy species with rather short slender recumbent stems. €. acantho- spathus appears to be at first erect, but ultimately subscandent, never however attain- ing a great length. The stem, in climbing species, with the sheaths on, varies from 4-5 mm. in diameter in some varieties of C. javensis and in C. filiformis, to 10 em, in C. andamanicus; when divested of the sheaths, the range is from 2-3 mm. up to 4 cm. The length of the internodes is usually great and attains the extraordinary length of 90 cm. in €. Scipionum. The total length of the stems of Calami has been greatly exaggerated. Roxburgh (Flora Indica, Vol. III, page 777) assigns to his OC. ezíensus a length of from two to three hundred yards, but I think that even if feet instead of yards be meant, this length is still excessive. Loureiro, too, gives to his C. rudentum the extremely unlikely length of 500 feet. With regard to this subject I have to say that I never measured a Calamus at all approaching these dimensions, and that all those observed by me were certainly under 50 metres (about 150 feet). As Calamt are climbing plants, the entire length of their stems may be approximately estimated by the height of the trees which support them; often, however, the lower part of their stems is procumbent creeping long distances on the soil of the forest, so that a Calamus plant may be considerably longer than the height of the tree from which it is suspended. This is brought about by the fact that the leafy crown of a palm of this kind is, on account of its being too slender, unable to maintain itself freely above the aérial plane of the forest and as new leaves and spadices or flagella are successively produced, those which preceded them and are situated lower down the stem gradually decay or lose their hold and do not succeed in retaining the plant in position by their hooked spines. Thus it happens that while the top of the palm strives to rise above and even to spread its fronds over the crowns of the loftiest trees, it has on account of its weight a contrary tendency to find a lower level. As the result of these two opposing agencies, the crown of the plant remains constantly at about the same aérial level, while the lower portion of the naked stem creeps for a considerable distance on the ground below, : The internodes of the stems of the Calami are sometimes exactly cyl- indric, but not infrequently are slightly enlarged upwards; those of the higher or adult part of the plant usually have a slightly raised longitudinal swelling or we 8 INTRODUCTORY ESSAY. ridge on the side of the insertion of the spadices, As the spadices are onm on the leaf-sheaths, and as the leaves, though really gis PIN A pim d i i t has it on the side opp . i i de has the ridge on one side, the nex ThE ha pe d course of the fibro-vascular bundles which pass into the ai dio into its homologue the leaf-sheath flagellum. The fibro-vascular bundles ^ a internode of the stem pass into the sheath next above, which dea M MR i i slight longitudinal swelling i gth; their course is marked by a g g j Ni ge pe the side which corresponds with the analogous swelling on o the internode; this ridge usually ends near the mouth of the sheath where th e spadix is inserted. The stems of Calami are usually rather spongy in the centre, and are always very hard externally, though they are at the same time flexible and elastic, | The stems of Culami divested of their sheaths have almost always a polished surface, and look as if covered with a thin layer of vitreous varnish; sometimes, however, the surface is opaque, but almost always is of a straw-yellow colour, especially when dry. This vitreous appearance of the naked canes of many Calami is due to the presence of a very thin layer of a silieious substance which entirely covers them. When the stem of a Calamus,” more particularly of one that has a very” polished and shining | surface, is bent, small thin lamellæ may be seen to crack off as the result of the fracture of the thin eoating of silicious material with which the cuticle of the stem is laden. In these lamelle no cellular structure whateve be made out. The cells which constitute the immediately underlie this cuticle, r can epidermal layer of the Stem, and are also more or less laden with silica, frequently: to such an extent that even after incineration they retain their form and preserve their minutest details, so that even the characteristic structure of the stomatic cells. continues unaltered. The histological elements of the inner part of the stem are, however, only very partially silicised. This abundant silicious secretion in the tissues of Palms h which the wood of Pal in a fossil state may not depend entirel y on a molecular substitution of siliea for cellulose in the wall of their cells, but may also be due to the impregnation of that substance which probably already existed during life in the tissues of the Palms of past ages. suggests to my mind. ms is found silicised The more polished and vitreous the surface of a stem is, the more it iş impregnated with silica and the more it is valued for economic purposes, In the category of species whose stems are thus valuable are included €, cæsius, Q, optimus, C. tenuis, C. Rotang, ete, IV.—The Leaves. The leaves or fronds of scandent Calami often appear at first sig owing to their being separated from each other by considerable interval the elongated character of the internodes, When, however, together, as is the case near th e apex of the stem where less dense crown, a spiral tristichous a ht alternate, s because of the leaves are closer they form a more or ‘rangement is quite obvious, THE LEAVES. 9 A complete Calamus leaf is made up of a basilar portion or sheath, which usually takes the form of a. tube completely enclosing each joint or internode, and of the leaf proper. Within the latter we have to distinguish a medium portion or rib known as the rachis, and to right and left of the rachis the leaflets, which vary in number according to the species. The first or lower portion of the rachis which usually is destitute of leaflets may be held to represent the petiole. Then in whole groups of species the rachis is prolonged beyond the distal leaflets as a long and slender whip-like clawed appendage termed the cirrus. The leaves of an adult Calamus plant are always pinnate except in the cases of C. flabellatus, where they are furcate-flabellate, in that of C. radiatus where they are digitate-radiate and in those of C. digitatus and C. pachystemonus where we find only 2—4 leaflets approximated at the apex of the petiole. These three kinds of leaves, however, are, it is to be noted, kinds that first appear after the germination of the seed, even in those species that ultimately have regularly pinnate leaves in the higher part of the plant. On account of the diversity of form assumed by the leaves in Calamus it is therefore most important to know when describing them whether they be leaves from the lower or leaves from the upper part of the plant, because the differences between leaves from these two situations are often very remarkable. I have therefore in these descriptions of Calami, unless it is otherwise stated, always record- ed the characters of the leaves of the adult plant. In the measurements given by me of the leaves of Calamus it is to be noted that I include in the leaf that portion of the rachis which has no leaflets and which is usually spoken of as the petiole. But the sheath is not included nor is the terminal cirrus, when a cirrus is present ; ; 8o that the length of the leaves is measured along the rachis from its junction with the sheath to the point of attachment of the uppermost leaflets. As the first leaves in a very young plant of Calamus are flabellate, furcate, radiate or digitate, we may conclude that the pinnate leaves of the full grown plant are derivates of these simpler forms and consider C. digitatus, C. pachystemonus, C, radiatus and C. flabellatus as survivals of the primitive types of the genus. The cirrus is often very long and is always armed with “claws.” These claws, as we have already seen in discussing them generally, are sometimes solitary, but are more frequently 3-nate or 5-nate-digitate, especially towards the tip of the cirrus and in more robust cirri are even 7-nate and arranged in 4—3 whorls with more or less uniform naked intervals between the groups of claws, In appearance and in function the cirrus corresponds exactly to the leaf-sheath flagel- lum, to be discussed further on: in origin, however, it is altogether different. | In cirriferous leaves the leaflets attached along both sides of the rachis may either cease abruptly at a definite point or may gradually decrease in size upwards. The leaves of the lower part of the stem and the radical leaves are non- cirriferous, and often end in two more or less confluent leaflets or in bifurcate leaflets even in species where the leaves in the higher p of the — normally have a typical cirrus. Axx, Roy, Bor. Gano, Catcutra, Vou, XI, 10 INTRODUCTORY ESSAY. In a few species, such as C. ornatus and . C, conirostris the leaflets gradually diminish in size towards, and become abortive at the apex of the leaf, while the rachis becomes more “clawed” there than it is lower down, without, however, the development of a terminal portion devoid of leatlets. If in the higher part of the stem of an adult Calamus the leaves are destitute of a cirrus, then it is certain that the species is never cirriferous except as an abnormality, such as I have observed in O. heteroideus, in its var. pallens, | and in C. exilis, where occasionally a very slender cirrus appears at the apex of their otherwise constantly paripinnate leaves. Whole groups of species are characterised by the absence of cirri from their leaves, and it never happens that a species with non-cirriferous leaves in the upper part of its stem is cirriferous lower down: the opposite condition is, however, of common occurrence. For example, in a terminal portion of C. asperrimus, one metre in length, with five fully expanded leaves, the lowest of these leaves ended in two leaflets with no rudiment of a cirrus between them, the next above had a scarcely aculeolate cirrus 7 cm. in length, while the sixth had the cirrus 55 cm. long and densely armed with 3-nate claws. In the intermediate leaves the cirrus exhibited a gradual transition between these two extremes. When there is no cirrus the leaves end in two equal or sub-equal leaflets which are usually the smallest on the whole leaf, and may be perfectly free at the base or may be, to a greater or less extent, connate. Sometimes a very short rudimentary cirrus appears normally between the two terminal leaflets, as in C. Blumei and C. rhomboideus ; still more rarely the leaf terminates in a small solitary undivided leaflet as.in C. ramosissimus. In most Calami the leaves are pinnate and the leaflets very regularly disposed. on both sides of the rachis, but in this case I have frequently observed all the leaflets on one side slightly smaller tkan the corresponding leaflets of the opposite side, and at times even more remotely inserted on the rachis, Very frequently the leaflets are inequidistant, geminate or attached in couples, as in ©. latifolius, or aggreguted in three, four, or even larger numbers along the rachis, as in C. gracilis, C. fasciculatus, ete. s In Calami the leaflets are always symmetric at the base except in C. Blumet and to a less degree in C. tomentosus ; they never ar | ms e sigmoid or faleate. í Most frequently they are broadly li more or less ensiform, or narrowly lanceolate and sini) ii the M una of breadth to length varies from 1:10 to 1:90. There is, however, no lack of other forms, such as lanceolate, oblanceolate, ovate, obovate, elliptic, oblong, spathulate, etc. When the leaflets are comparatively broad toy ona -Gieu “solved on cas m ‘ and concave ix the in or are spoon-shaped, specially towards their apices: in a small group, that to which (€. rhomboideus | e 8 aped belongs, they are more or less lozenge The leaflets almost always terminate in a symmetric, more or less acuminate and bristly-brushed tip; very often, however, we may observe here a small notch or indentation on the lower margin not very far from the apex; this notch in many In shape the leaflets vary a good deal. THE LEAVES. 11 cases is inconspicuous or even obliterated, while in; others it is rather conspicuous. as in €. australis, C. pseudo-tenuis and especially in C.: Lobbianus. I suppose that this notch may possibly be the last trace of the derivation of an acuminate leaflet from « truncate one, Among Calami leaflets with a truncate transversely cut or premorse end, analogous to those of a typical Ptychosperma, are only found in €. caryotoides. The leaflets in Calami, without a single exception, are attached to the rachis by a very narrow point, tapering always more or less gradually or even abruptly towards the base where the lamina, commonly doubled downwards, forms just at the junction with the rachis a small cavity on the lower surface, within which in many cases nestles a small swelling which resembles an extra-nuptial nectary. In Calamus the leaflets are never decurrent along the rachis. Very important characters are derived from the main nerves or ‘ coste” of the leaflets, as their varying distribution appears to correspond to a varying group- ing of the fibro-vascular bundles in the axial parts of the plant. Most of the Calami have narrow leaflets with one or three primary nerves or costee, running throughout their entire length, sharply prominent on the upper surface and usually less distinct on the lower. In not a few species there are five main or primary nerves, less often seven to nine, and even, in C. rhomboideus, twelve; but, whatever the number of the costze may be, the central or midrib is almost always the strongest. All the primary nerves in the leaflets of Calamus traverse the entire length of the blade and meet at its apex, except in the species of the C. rhom- boideus group, where the -nerves nearer to the margins usually disappear at different levels and do not reach the apex. Between the primary nerves we find interposed others more numerous and more slender; very frequently one of these secondary nerves runs along or very near the lower. margin so as usually to slightly thicken it; in the leaflets of Calamus a primary nerve is never found running along the margin. - When the leaflets are rather broad, as is the case in (C. palustris, C. latifolius, C. javensis, etc. their lower margin is often bordered on the upper surface by a polished shining band a few millimetres wide; sometimes a few more similar bands are found also in different places along the lamina; these lines correspond with the portions of the leaflet that are outermost in the leaf-bud. As the leaflets before the leaves expand are longitudinally plicate and closely packed together, these polished bands look as if they had been produced by friction as they were being forced out from the central terminal shoot. The primary and secondary nerves are connected by means of ME veinlets which are more or less interrupted and anastomosed. In some species secondary venation forms a very elegant network and is very conspicuous, the component veinlets being close together, slightly sinuous, thin but very well defined and continuous, that is to say crossing transversely the whole width of the lamina; good instances of this style: of venation are seen in C. insignis, C. heteracanthus, ete, Usually these transverse veinlets are more distinct on the upper than on the lower surface. Ann. Roy. Bor. Garp. Carcurra Vor, XI. "(19 INTRODUCTORY ESSAY. Very commonly we find the central costa and some of the lateral primary nerves more or less heset, now on the upper, now on the lower surface and now on both with hairs, bristles or spinules; the presence or absence or varied situation of these supply easy though not always reliable differential characters. In C. ciliaris and in other species of the same group the secondary and tertiary nerves are also covered with hairs, and if these nerves be numerous and very close together, the surface has the appearance of being uniformly hairy, as for example. in C. hispidulus. In a very few species a solitary spine is found, though not constantly, on the upper surface of the central costa near the base of the leaflet (C. tenuis and C. Rotang). In C. spinifolius the leaflets are armed with 3-5 erect and comparatively strong true spines. The leaflets of C. salicifolius are also more or less supplied with a few spines of this character; these, however, are absent from its vam. leiophyllus. In texture the leaflets of Calamus are usually thinish, sometimes sub-herbaceous or membranous, often chartaceous, and occasionally thinly coriaceous; very frequently both surfaces are of the same green colour, but the upper adMaed da most usually shining while the lower is dull. In comparatively few species the lower surface is more or less conspicuously different in colour, being mealy-puly- erulent and subpurpurescent in C. Grifithii, glaucescent in C. cæsius, mealy-ochraceous in C. symphysipus, mealy-white in C. arborescens and C. hypoleucus, or decidedly white and as if coated with a thin chalky layer in C, Lobbianus, C. leucotes and C. discolor. Only in €. deerratus have I observed the lower surface sprinkled with small punctiform scales. | The margins of the leaflets of Calamë are very seldom absolutely smooth; most commonly they are furnished with cilia or very small adpressed or spreading spinules. The leaves of Calami usually retain their light green colour in dried specimens, but certain species, especially those of the C. andamanicus and of the C. platyspathus groups, assume in drying a light brown or tobacco colour, while ©. spathulatus, C. Martianus, C. insignis and a few others are readily recognisable in herbarium specimens by their bright yellow hue, : V.—The Leaf-Sheaths. The leaf-sheaths of the leaves of Calamus are very important organs of the plant. In most Calami, in all those species we may say that are scandent, a leaf- sheath arises from each ring of the stem and forms a complete more or less elongated cylindric tube round the internode immediately above. In the non-scandent species the leaf-sheaths are more or less open along the ventral aspect. | : The leaf-sheaths are always of a firm, tough or more or less coriaceous texture, and are sometimes even woody; but their leading characteristic is that they are always more or less covered with spines, which assume a great variety of forms and furnish one of the most conspicuous and useful diagnostic characters ise this Gee nation of many species. "Eg vimecs THE LEAF.SHEATHS. 13 It would seem that during the period characterised by great morphologieal malleability of organisms the tendency to variation was very active in the direction of acquiring a high degree of spinosity, as being that which secured for the plants its most valuable means of defence. But while differences in the character of the spinosity, of what we way call the * armature" of the leaf-sheaths, afford one of the best characters in distinguishing the species of Calamus, it sometimes happens that two plants, one with powerfully armed the other with smooth leaf-sheaths, cannot be considered specifically distinct; thus C. ornatus var. horridus is formidably beset with very large spines, while its VAR. mitis is almost smooth. The tubular cylindric leaf-sheaths have commonly a kind of pouch in their dorsal upper portion at the base of the petiole, so that they may be termed gibbous; the leaf-sheaths of the non-scandent species which are open on the ventral side are without this peculiarity and are gradually narrowed into the petiole. Leaving the spines out of account the surface of the leaf-sheath is at times glabrous, pulverulent, greenish, glaucous, or more or less clothed with a deciduous or easily removeable furfuraceous whitish-grey, or brown skin of fluffy indumentum, I know _ only C. tomentosus where the sheaths, in common with other parts of the plant, are entirely clothed with a permanent white tomentum. The leaf-sheaths are often longitudinally striate or indented with the impressions stamped upon them by their own spines. In a few species, such as ©. latifolius, C. marmoratus, C. Feanus, C, javensis, the sheaths in the younger parts of the plant are variegated, spotted or marbled with whitish, dark-green or purplish patches, The fact that the most important fibro-vascular bundles, such as those that enter the reproductive organs, pass from the stem through the nodes into the sheaths indicates the complex, almost sympodial character of this part of the leaf. The vascular bundles usually traverse the entire length of the sheaths, and their surface is generally marked externally, as has already been seen, by a more or less longitudi- nally raised ridge which terminates at the insertion of a spadix, or of a leaf-sheath flagellum, laterally near the mouth of the leaf-sheath itself, Owing to this peculiarity of structure the spadices, like the flagella, emerge from near the apex of the sheaths and never arise in the axils of the fronds; only in C. azillaris, where the leaf-sheaths are comparatively short and where one sheath covers a considerable portion of that immediately above, do the spadices, which moreover are inserted far below the mouth of the sheaths, appear axillary. Even when the sheaths are not exactly tubular but are more or less open on the ventral side, as in C, erectus, so that they closely resemble the sheathing base of the fronds of — those Palms that have axillary spadices, the spadices of Calamus retain their usual | Position and emerge laterally from near the apex of their proper sheaths; so that it appears as if the sheath at first formed a closed tube but was split longitudinally | afterwards and the gap kept — by the growing central shoot. RO vna 14 | INTRODUCTORY ESSAY. The sheaths in those species with non-cirriferous leaves, and usually provided with flagella in the upper part of the plant, are without these appendages in the lower portion, as it seems that the flagella appear only when the plant has acquired & certain degree of development, the spadices being produced still later. In many cases in a full grown plant one leaf sheath bears a spadix and the next one a flagellum and so on: in other instances, such as C. radiatus, C. pachystemonus and. C. digitatus, every sheath of a fertile and full-grown plant has its own spadix. VI.—The Leaf-8heath Flagella. The leaf-sheath flagella are sterile or abortive spadices arising from some of the leaf-sheaths at the point where normally fertile spadices are inserted. We have conclusive proof that the flagella are no more than incompletely developed spadices in the fact that we occasionally find them bearing more or less incompletely developed spikelets, as not infrequently happens in C. heteroideus. Martius uses the name “lora” for the flagella, but I have chosen the latter term as being in more common use and more readily intelligible. In very many species the spadix, and especially the female one, is prolonged at its apex into a long slender clawed appendix corresponding exactly to the apical part of a flagellum, but,we must be careful not to mistake this appendix for a “cirrus” which is the corresponding flagelliform prolongation of the leaf rachis. For notwithstanding the very great similarity of the two kinds of appendages, especially in the matter of their function and the identity of their armature of claws, the flagellum has a morphological origin which is quite distinct from that of the cirrus. : The leaf-sheath flagella of some of the.larger species are exceedingly strong and resistent and are sometimes of very great length; I have measured one in C. flagellum which was over 7 metres long. The biological function of the flagella is that of attaching the individual to neighbouring plants by means of the hooked prickles with which they are armed; consequently every Calamus provided with well- developed flageila is undoubtedly scandent. Those Calami that possess flagella are destitute of cirri at the ends of their leaves; and, on the other hand, those species with cirriferous leaves have no flagella, and their spadices are usually comparatively short, panicled and non-flagelliferous at their apices. N As a rule then cirri and flagella, being quite similarly armed with claws, are sub- stitute-organs which take the place of each other in the function of providing the plant with means of climbing, but in some species of Group XII, the sheaths have flagella, the spadices are flagelliferous, and even the leaves are, though usually very imperfectly, cirriferous ; these species seem therefore to have exerted every means in their power, and that to the utmost extent, to attain a climbing habit. The flagella, being morphologically identical with spadices and only differing in the absence of branches and spikelets, consist of an axial portion clothed with cylindric closely “sheathing spathes and, as in the fertile spadices, have the lowest or outermost ‘spathe larger and firmer in texture than the megea ones, and actually armed with straight spines. > E THE LEAF.SHEATH FLAGELLA. 15 When a Calamus is not decidedly scandent, but is a derivate of species which climb and therefore possess weli-formed flagella, we find that rudimentary flagella are present. Flagella are quite absent from species with an erect stem, such as €. erectus; C. arborescens, etc. In the flagelliform species flagella are also usually absent in the earlier stages of the plant and only make their appearance when the plant has reached a certain height and begins to produce spadices. In many species, however, it is found that spadices alternate with flagella. In those species where the leaves are cirriferous and the spadices are short and panicled, we observe now and then a rudimentary flagellum; in O. latifolius, for | example, I have had occasion to note the presence of rudimentary flagella, 8-10 em. in length, filiform, sheathed by quite tubular prickly spathes, while in other cases the place usually occupied by a spadix is indicated by a small protuberance. Such rudimentary flagella have remained abortive because it was not natural for the spadix which they represent morphologically to become flagelliform. Very probably in the non-flagelliferous Calami the young plants may bear such abortive flagella; this I have had an opportunity of noting in ©. erioacanthus. Whether a species be flagelliferous or not may be ascertained from herba- rium specimens even if the flagella are actually missing, because if the spadix is elongated and ends in a weil-developed clawed flagelliform appendix, some of the leafsheaths are almost certainly flagelliferous; on the other hand, if the spadices be short, compact, panicled and not flagelliferous, most probably the sheaths are not flagelliferous and the leaves of the higher part of the plant are cirriferous. VII.—The Ocrea. The ocrea of Calam is a tubular stipule, occasionally split into two parts, bordering the mouth of the leaf-sheaths. In many cases the ocrea is much elong- ated and conspicuous, membranous or chartaceous in texture, glabrous or more or less clothed with hispid hairs or spicules, sometimes even more or less spinous; not infrequently it is very short or is reduced to a short ligule in the axil of the petiole. Sometimes the ocrea is persistent and clothes the base of the sheath immediate- ly above its own; its chief function seems to be the protection of the younger parts of the terminal shoot. Most frequently after the expansion of the leaves the ocrea decays and is lacerated or reduced to fragments or filaments, scarcely retaining any trace of its shape ; ; Sometimes, being deciduous or of a transitory nature, its former presence is revealed only by a narrow scarious rim at the mouth of the sheath. In the African Calami the ocrea is usually rather elongate and tightly clothes the base of the sheath above its own, but its outer side is more elongated or produced than that facing the petiole; in these species, therefore, the ocrea assumes Deed; din i of the mouth of a beaked id and we Pid term it E liguliform." . | -A 16 INTRODUCTORY ESSAY. In C. erectus the ocrea is exceptionally large; it is tubular and entire while still enclosed in the terminal unexpanded shoot, but afterwards it is split along the ventral side into two halves, one on each side of the petiole, resembling two large auricles which are as much as 7 cm. in length, chartaceous and densely clad with hispid hairs. ! A small group of species from New Guinea have a still more striking form of ocrea resembling that of some species of .Kor/halsia ; in fact the ocrea of €. macrochlamys and C. ralumensis attain the extraordinary length of 30-35 cm, and are firm and thin in texture or chartaceous, elongate-conical or almost the shape of asses’ ears, in other cases the ocrea is at first membranous and tightly embraces the sheath above its own and ultimately becomes disintegrated into filaments or fibres. I do not know any Calamus in which the ocrea is transferred into a receptacle for harbouring ants, as is the case in some species of Korthalsia. VIII.—The Petiole. As we have already seen, the basal portion of the rachis of the leaf which does not bear leaflets is treated as the petiole. In Calamus therefore the petiole begins at the mouth of the sheath and terminates at the point of attachment of the first leaflets. The petiole is usually rounded or convex beneath and flat or chan- nelled above; it is frequently armed with prickles, usually longer along the margins than elsewhere. The petiole of the radical leaves and of the young plants is gener- ally a good deal longer and more cylindric than that of the adult plant, and in these the petiole of the upper leaves is shorter than that of the lower ones. IX.—The Rachis. The rachis represents the backbone or axis of the leaves in Calami just as in other Palms, and bears, right and left, a number of leaflets varying according to the species. In the description of the species the term rachis is applied only to that part of the axis of the leaves which bears leaflets, it having been decided to term the basal naked portion of the leaf-axis the petiole and the prolongation : ihe axis beyond the distal leaflets, when this occurs, the Cirrus. The rachis is not uniform throughout its length; being a continuation of the petiole it is more or less convex beneath towards the base and flattish towards the apex, where in all scandent species it is more or less armed with hooked prickles or claws; these are usually solitary lower down and become binate and then ternate towards the apex, and at last in some cases among the cirriferous species even quinate and ,half-whorled and of increased size and strength where the rachis begins to assume the essential features of a cirrus, On the upper surface immediately above the petiole, the rachis is usually very slightly channelled or almost flat, and shows on each side of the central part a longitudinal furrow within which the leaflets are inserted; higher up the two furrows gradually vanish, the central part becomes narrower and is transformed into a salient angle with two converging sides or facets. There, consequently, the rachis is triangular in section, bi-facetted with a salient angle above and flattish beneath. This shape is almost invariably assumed by the rachis from the middle onward to the apox, THE CIRRI AND SPADICES. 17 X.—The Cirri. In speaking of the different kinds of spines and of the leaves I have SR described the appendage of the leaf-rachis termed the cirrus, and have now very little to add. To recapitulate what has already been said; the cirrus is always more or less armed with solitary digitate, aggregate or more or less incompletely and dimi: diately-whorled claws. Every Calamus possessing cirriferous leaves is scandent with, as we have seen, the exception perhaps of C. Ozleyanus. When a Calamus has hooked prickles or claws on the leaf-rachis, on the primary spathes and the axial parts of the spadix, the plant is nevertheless scandent even if there be no cirrus at the end of the leaf, since the clawed leaf-rachis, together with the leaf-sheath flagella with which in this case the plant is provided, take the place of the cirri and perform the functions of climbing organs. On the other hand, when the hooked prickles characteristic of cirri and flagelli are absent from all its organs ihe species is most certainly bushy or has an erect stem. : To avoid confusion I have employed the term “cirrus” for the prolongation of the leaf-rachis, and have restricted the term “flagellum” to the appendage ree from a — of the spadix. XI.—The Spadices. As a general rule the species of Calamus are dioecious, The spadices spring always from the leaf-sheath and most commonly from its mouth laterally to . the petiole. The leaf-sheath in Calamus being almost always considerably elongated and the leaves being far apart from each other, the point of insertion of the spadices is exposed and visible, and it is only in species with a short erect stem and with the leaves grouped at its apex that the spadices at times simulate an axillary insertion. C. azillaris and C. adspersus among the climbing species are the only | ones. known to me where the spadices seem at first sight axillary. As in most Palms the spadices emerge from the axils of leaves or from rings of the stem where once a leaf was situated, it ought to be worth while investi- gating by what morphological modifications and through what intermediate conditions the species of Calamus have arrived at a point of origin for their spadices so unlike that of other Palms and so unusual. The point of insertion of the spadices, as of the flagella, on the leaf-sheaths is frequently marked by a distinct swelling or callosity, and, as we have already seen in discussing the leaf-sheaths, a slightly raised ridge, which marks the course of the fibro-vascular bundles that pass from the stem into the spadices, often also runs downward along the whole length of the sheath. The axial part of the spadices of Calams, usually much elongated and very slender, is armed towards its slender filiform extremity with claws like those of a flagellum ; ; besides therefore fulfilling its main function of bearing flowering branches, it is very often made use of as a subsidiary climbing organ. This is just the reverse of what obtains among the species of Daemonorops, where the spadices are _— deyoid of hooked spines and never are utilized as organs of climbing. ee The spadices of Calamus belong to two principal categories. In the ka y of the species they are much elongated or flagelliform, like those just referred toi Ann. Ror. Bor. Garp. Carcurra Vor. XI. 18 _ INTRODUCTORY ESSAY... Spadices of this kind are more or less armed with claws, have the branches on which the spikelets are borne remote from each other, and have the axial portion interposed between two such branches more or less armed with claws on the outer side. Spadices of this ‘class most usually occur in species that have non-cirriferous leaves and flagelliferous sheaths. The second kind of spadix is peculiar (to those species that have cirriferous leaves, but have leaf-sheaths which are not flagelliferous. In these species the spadices are comparatively short and broad, usually shorter than the leaves, panicled and often pyramidal with numerous approximated and gradually diminishing branches and with a rigid axis, non-flagelliferous at its apex. C. palustris and the other species of the group to which this belongs possess spadices of this character. : Only very few species have short and contracted spadices: C. Lobbianus, ©. conirostris, ©. brachystachys are instances. (C. simpler alone has an undivided spadix with a simple spikelet at each primary spathe. It may be laid down as a general rule that when a Calamus has a long flagelliform spadix the leaves are not cirriferous, aud in this case if the leaf-sheaths do not bear spadices, these are replaced by long clawed flagella. On the other hand, when the leaves are cirriferous the leaf-sheaths are without flagella and the spadices are panicled ‘and comparatively short. In Calamus the spadices are always furnished with a variable number of eylin- dric or very rarely laminar spathes, to be explained presently at greater length; to each spathe there is a corresponding branch or “partial inflorescence.” | As a rule in the species where the spadices are flagelliform there is no very great dissimilarity between the male and the female spadix, though in almost every case the female is less branched than the male one, In the female spadix a simple branch-bearing spikelet springs from each spathe, and thus gives us a ‘simply branched” spadix; in the male spadix the primary branches are divided again and again, so that we have a ‘supra-decom- pound” spadix. In a few species the male spadices, like the female ones, are simply decom- pound, as in O. longisetus, O. leptospadiz, ete. Less often, the female spadix is also more or less partially ultra-decompound ; this I have observed in ‘the case of C. luridus and €. tonkinensis, where however only the basal portions of the lower partial inflorescences are branched again. Moreover, in €. luridus I have occasion- ally found sub-moneecious spadices in the form of inflorescences producing female spikelets in their lower parts and male ones at their apices. | .. A case of monecism is also afforded by €. rudentum, where occasionally in the female spadix near its aper a few spikelets may be composed solely of male flowers; these are apparently fertile, but are more slender than the flowers on exclusively male spadiees. I do not, however, know any truly monecious Calamus with female flowers normally accompanied by fertile male ones on the same spadix, or with distinct male and female spadices on the same plant, though Roxburgh employed the specific name menoicus for the Calamus already characterised by Linnaeus THE SPATHES. 19 as €. Rotang, which certainly is not endowed with the peculiarity, By a super- ficial- observer every Calamus might be supposed to be moneecious, since in the female spadices every female flower is accompanied by another which, though sterile; frequently has all the appearance of being a male. Every Calamus has the true fertile male flowers brought together on a separate plant, and only very exceptionally some male spikelets may be found on a female spadix; all Calami are therefore essentially dicecious and every plant produces only spadices of one sex. -E have never chanced to find on the male spadix any trace of female flowers. The appendages, more or less obviously of foliar nature, which clothe the axial parts of the spadices, are called ‘“‘spathes.” Those spathes that are inserted on the main axis and are visible even before the partial inflorescences emerge from them are the “primary spathes.” The spathes of the inflorescence are termed ^ secondary” or ‘tertiary ” according to the degree of division of the axial part on which they are inserted. XII.— The Spathes. Monk commonly he spathes of Calami are tubular, do not completely envelope the flowers, and form superposed tubular sheaths to the axial part of the spadix. When the spadix is very young the primary spathes are enclosed one within the other, and gradually diminish in size from its base to its apex. In a few species of the group to which C. platyspathus belongs, the primary spathes approach those of a Daemonorops, being at first tubular and completely en- closing the partial inflorescences with their flowers, but at flowering time bursting longitudinally so as to expand their limb into an elongated flat laminar blade with only a small tubular portion at its base; the partial inflorescences are then at liberty to expand their flower—bearing branches. Spathes of this kind somewhat resemble those of Daemonorops longispathus with this difference that those of the Calamus just mentioned give greater evidence of their origin from a tubular pattern. The most aberrant primary spathes among Culami are those of C. hypoleucus, which are cymbiform and resemble those of a Daemonorops of the section Cymbo- spatha. All spathes of typical Calami of the flagelliferous groups are much elongated, while those species that have broad and short panicled spadices have shorter spathes : in both cases, however, they are always tubular and most commonly strictly and closely sheathing ; less frequently the primary spathes, though tubular, are loose, enlarged in their upper part or earshaped, The primary and to a less marked degree the secondary spathes are of a rather firm structure, more or less thinly coriaceous or even almost woody ; very commonly entire and truncate at the mouth and prolonged at the apex into a short limb; not infrequently their upper portion becomes decayed or, as in C. erectus, even lacerated and reduced to filaments. ‘The mouth of the spathes is often provided with cilia, elongated scales or paleolæ which are of a more or less fugacious character. | ' The length of the primary spathes usually corresponds to the distance interven- ing between two partial inflorescences ; similarly the length of the secondary spathes- corresponds to the distance intervening between the insertion of the spikelets in the simply decompound — "—— a russe decompound. | Ann. Roy. Bor. Garp. CarcurrA Vor. XI. 20 INTRODUCTORY ESSAY. .. The mo e distinctly flagelliform the spadices are, the more elongated is the portion of their axis interposed between two partial inflorescences, and in this case the base of the spathe is so gradually connected with the axis as to render it impos- sible to determine from the outside where the spathe begins to be an organ inde- pendent of the axis. When the segments of the axis interposed between two partial inflorescences are very long these are always, at least in their lower parts armed externally, like a flagellum, with simple ternate or even half-whorled claws. As a general rule the spathes are more densely armed with — prickles on the outer or dorsal than on the inner or ventral side. The lowest or outermost primary spathe is almost always somewhat differently shaped from those that succeed it, and provides important diagnostic characters. It is usually thicker in texture and less cylindric than the others; is not infrequently flattened and two edged and is more spinous on the margins and on the back than the upper spathes. In the very earliest stages of the spadix the first spathe includes all the subsequent ones; these emerge one from the other like the tubes of a telescope. ! The upper primary spathes become gradually shorter and narrower, and the ultimate ones help to form the slender filamentous flagellum. XIII.-The Partial Inflorescences. From or near the mouth of every primary spathe, except the outermost, springs, usually distichously, a partial inflorescence. The number of the inflorescences varies both according to the species and to the strength of the individual plant. In several species the partial inflorescences end in a spikelet larger than the lower ones; in others, laterally to the uppermost spikelet springs a more or less distinct caudicle formed of diminutive sheathing spathes and representing the prolong- ation of the axis. The lower inflorescences of a spadix are almost always larger and more branch- ed, and bear more spikelets than the upper ones; there is a gradual diminution in size, degree of branching and number of spikelets of the inflorescences from the base of the spadix towards the apex, where the partial inflorescence is not infre- quently reduced to a few spikelets or even to a single spikelet, "The spathes which sheath the axis of the partial inflorescences follow the same rule as the primary spathes, from which indeed they do not essentially differ, and become gradually shorter and narrower as they approach the apex. XIV.—The Spikelets. The ultimate divisions of the partial inflorescences bear small axes beset with flowers which may be considered spikelets, which name I find preferable to that of * Juli” employed by Martius. The name “Julus” or ‘‘ Amentum ” (Catkin) is usually applied to a cylindric inflorescence where the flowers are attached all round a central axis, whereas in the spikelets of Calamus the flowers are almost always regularly arranged in two series right and left of the central axis so that the spikelets appest flattened exactly as in many Graminee, THE SPIKELETs. 21 Owing to this bifarious arrangement, the flowers when very close together and disposed in the same plane, render the spikelets comb-like or pectinate. Not infrequently, _ however, both series of flowers are more or less assurgent and point upwards so that the spikelets acquire a scorpioid tendency, as for example in 4€. Gambie, and €. Huegelianus and in most of the species of the groups to which C. ezilis and C. heteracanthus belong. In & few instances the disposition of the flowers is not clearly evident, In C. gonospermus, for example, it is not easy to make out whether the spadix is composed of short spikelets with closely packed glomerulate flowers or of contracted partial inflorescences in which what appear to be glomerules of flowers are contracted spikelets. In C. .Lobbianus also the spikelets are short and dense and their original structure, especially when they are laden with fruits, is difficult to recognise. With the single exception of C, polystachys only one spikelet, in all the known species of Calamus, arises from every secondary spathe. In C. polystachys, however, in the lower part of the inflorescences there are 2-3 spikelets in the axil of each secondary spathe, though towards the top the spikelets are solitary. As a general rule the lower spikelets in every partial inflorescence are larger than the upper ones; these gradually diminish in size and number of flowers, the ‘uppermost being the smallest. . Most usually the spikelets are inserted at or near slightly above or inside the mouth of the corresponding spathe; but in some species, for instance in all those belonging to the groups of C. castaneus, C. zeylanicus and €. Zollingerii they are attached to their base by means of a pedicellar part as long as the spathe from which the spikelet emerges (Prate II, fig. 2). If the spikelets be pedicellate even the flowers, as will be seen presently, are borne on pedicellate involucres, The appendicular organs of the spikelets which are spathes of a reduced or diminished nature are termed “spathels.” Even when the primary and secondary spathes are prickly the spathels are always unarmed; the only exception known to me is in the var. insularis of C, paspalanthus, where the spathels are provided with one or two diminutive claws; usually however when the secondary spathes are scabrid, a similar scabridity is present on the spathels and even on the involucres, The various appendicular organs of the spikelets are of much diagnostic impor- tance, but it is not always easy to establish their nomenclature; and to be certain of their exact morphological nature it is essential first to know the sex of the spikelets because the appendicular organs or involucra of the flowers are not the same in the male and female spikelets. Male Spikelets—The male spikelets of Calamé are of simpler structure than the female ones, since the flowers in the first have only one proper involucre (PLATE I, figs. 1-3) while in the second the involucre is double (Prate I, figs. 8-12), moreover, the male spikelets have only one kind of flower, and the flowers with their involucres are solitary in the axil of every spathel (Puare I, figs. 1-3), while the female spikelets have always at least two flowers to each spathel (Prate 1, fig. 10) of which one is female, the other is more male than female, but is „sterile -and speedily deciduous, ig eye 23 © INTRODUCTORY ESSAY. .ss Im the: male spikelets .the partial inflorescences are usually branched two or three times and the division or branchlets which remain in their basal portions may be considered to be compound or branched spikes, while the ultimate divisions of these are the true or simple spikelets. As regards size, the different divisions of the inflorescence and the spikelets follow the general rule; the upper branches and spikelets are gradually smaller than the lower ones and every partial inflorescence assumes a more or less pyramidal form, The terminal spikelet of each division is usually considerably larger than the others. The male spikelets are almost always shorter and more densely furnished with flowers than the female ones. The axis of the male spikelets is clothed with spathels which are commonly broadly and asymmetrically infundibuliform (Prate I, fig. 14, and Prate II, fig. 4h); sometimes, however, their tubular part is so short that they look like bracts or are boat-shaped or spoon-shaped with the axis passing excentrically through them; the spathels are rounded on one side and slightly prolonged on the other, the outer, into a small point. In the axil of each spathel is a flower with its special involucre (PrarE I, fig. 1b, and Prare II, fig. 44). The involucre has usually the form of a small cup (Pate I, figs. 1-36), being more or less concave; in a few instances we find in place of the involucre two small scales or bracteoles united by their bases which clearly explain the origin of the involucre from two coalescing appendicular organs; for example in C., asperrimus, C. siphonospathus and cthers. Authors usually term the involucre a “‘spathellula,” a name that may easily be mistaken for that of a ''spathella," and -one that, as we shall presently see, has not been always properly and uniformly applied. In the simple spikelets it is always easy to, distinguish the spathels from the other appendicular organs as it is from the axils of these spathels that the flowers arise; but in the case of compound or branched spikes it is a spikelet, which has its own spathels, that is situated in the axils of a spathel; in this case therefore we have primary, secondary, and even tertiary spathels, according to the degree of division of the primary spike. But all this is of very slight importance since, as a rule, the secondary and tertiary spathels differ from the primary only in size. It is sufficient if we hold that in a male spikelet we mean by a spathel the appendicular organ which clothes the axis of a spikelet—whether of & primary or a secondary spikelet is of no consequence—and that a spathel of the ultimate bridge eoe: a culis which in turn has its base enveloped by a special involucre. The modifieations, peculiarities and diagnostie characters afforded by the spathels are essentially the same as those of the female spathels to be described further on; but as a rule the spathels of the male spikelets are shorter, broader, more distinetly infundibuliform and more closely packed than those of the other sex. The flowers of the male spikelets are solitary in the axil of each spathel except in a very few species; such as O. viminalis, C. pseudo-tenuis and some few others. In @, viminalis I have counted as many as 8 glomerulate flowers to each spathel; but in this case the glomerules must be considered secondary much reduced diss as. odi flower is provided with its own diminutive bracteole. 3 E. um .SPIKELETS. | | 23 Female Spikelets—The female spikelets are as a rule more elongated. than tlie. male ones and though the flowers are biíarious they are more remotely disposed, "dee! The female spikelets also have their axis constantly clothed with tubular or slightly infundibuliform spathels but these are longer and less distinctly infundibuliform than those of the male spikelets and are not infrequently represented by membranous rings round the axis of the spikelets with which in their lower part they are organically fused (Prate I, figs. 8-124, and Prate II, figs. 1 and 3k). It is therefore often difficult to distinguish externally the point where the spathels are differentiated from the axis. | In some cases the spathels are much reduced in size, are devoid of any tubular portion and resemble bracts. The morphological structure of the female spikelets does not differ essentially from that of the’ male, but -in the female the appendicular organs together with the flowers which arise from the axil of or above each spathe must be considered in their entirety as a secondary much contracted spikelet. This mode of interpreting the female spikelets of Calami is in accord with the general plan of division of the spadices, and assumes that it is supra-decompound as is almost universally the case with the male spadices. According to this interpretation the female spikelets would also have been supra-decompound if the small contracted spikelet existing at each spathel had undergone further development. The flowers of the female spikelets are inserted, like those of the male ones, in ‘or @ little above the axils of their respective spathels, but in the female spikelets the female flower has two involucres in place of one. And from the descriptive point of view it is most important to fix the nomenclature of these two involueres. The organ which remains inside or a little above each spathel and which is immediately in contact with the axis of the spikelet has been termed by me an "involuerophorum?; it corresponds with the “ spathellula" of Martius and with the “bract ” or “ bracteole ” of Griffith. I have not considered it advisable to retain the name of spathellule for the involuerophore, because this does not correspond morphologically to what has been termed the spathellule in the male spikelets and because it appears to me to be an organ of axial rather than of appendicular structure. The name spathellule implies a morphological agreement with the spathel to which it should bear the same relation- ship as the spathel does to the spathe or as a secondary spathe does to a primary one. As a matter of fact, however, the “involucrophore” appears to represent the shortened or contracted axis of a small spikelet, provided with its spathel and bearing | besides its appendicular organ which is the involucre. This structure is very evident in those species that have the spikelets inserted at the base of their own spathe, as In €. Burckianus, C. Zollingerii, C. Grifithianus, etc., when the involucrophore both emerges from the base of its own spathel and is more or less pedicellate ( PLATE II, figs. 6-9a). | ! The precise place of insertion of the involucrophore is not. always at the base of its spathel, but is much more frequently just at its mouth or a little above or below the mouth. When the involuerophore is inserted outside its own spathel, it generally seems attached to the axis of the spikelet; but in fact it adheres laterally ^to the base of the spathel above that in the axil of which it ought theoretically to have * 24 INTRODUCTORY ESSAY. originated (Prate II, fig. la) The essential part of the involucrophore is a small calyx, cup, cuplet or disk which corresponds to a spathel and is most usually sessile; in several species, however, it is more or less narrowed to the base, as in C. Grifithtanus and C. Zollingerü (Puare II, figs. 6-8a) or is even distinctly pedicellate, The latter condition is very evident in C. symphysipus, C. heteracanthus and allied species, in C. evilis and allied species, and in C. unifarius var. Pentong (PrarE II, fig. la) The pedicel of the involucrophore in the species mentioned clearly demonstrate the axial origin of the involucrophore. The involuerophore is never absent from the female spikelets and presents important diagnostic characters in its shape, and in its mode of insertion, whether sessile- or pedicelléd, included in or exserted from its own spathel, free from or partially adnate to the spathel above its own. As the involucrophore represents a contracted branchlet, or the axis of a rudimentary spikelet, emerging from the axilof its spathel, it reproduces the peculiarities inherent in all the other divisions of the spadix and just as at the insertion of the spikelets there is a specialized swelling or callus, which I have supposed to be nectariferous, we find this same callus with its transverse cleft or rima repeated in the axil of the involucrophore (Pare II, fig. 1i). Within the involucrophore and usually moulded on this is the proper involucre of the flower which corresponds exactly with the involucre of the male spikelets. Had we made use of the term spathellule for the second appendicular organ of the male spikelets, we ought to have used this term also for the involucre of the female spikelets and not for the involucrophore, a usage that must have led to confusion. ° The involucre in female spikelets is usually concave and cupular’so as to admit of the reception of the base of its flower; its margin is truncate and usually entire, but on the outer side it is more or less distinctly marked by two small teeth, between which the margin is more or less deeply lunately excised. Sometimes the involuere is almost explanate and discoid or even appears as if made up of two bracts, which in a few instances are almost separate, the apices of these bracts corresponding with the teeth of the involucre when it is cupular, On the involucre of the female flower externally, on a peculiarly shaped surface corresponding in position to the lunate excavation of the margin, is invariably inserted a neuter flower (Prats I, figs, 8, 10, 11e). | The small usually sharply defined surface on which the neuter flower is inserted I have termed the “areola” of the neuter flower. This areola is most usually lunately shaped, somewhat depressed or developed more in breadth than in height, with the horns corresponding to the two marginal teeth, Sometimes, however,. it is more developed vertically than horizontally and assumes a more or less ovate or lanceolate shape as in (€. Griffithianus (PLATE II, fig. 8c),.or is more or less concave or subinvolucriform, as in C. deerratus. The areola is not, however, always sharply defined; in some cases it is depressed or linear, or its place is simply marked by a small callosity (Prate II, fig. le) or by a punctiform scar, Sometimes. too its place is taken by a very short pedicel which supports the neuter flower, as in C. adspersus and C. Henryanus; this pedicel evidently represents there a second joint of the small and contracted spikelet from which the fertile and the abortive flowers spring. In C. Cumingianus and €, nitidus the neuter flower has at its ^ THE FLOWERS. 29 base very small bractecles of its own which may be taken as representing the rudiments of a second involucre., The involucre of the female flower, as has been already pointed out, is evidently fornted by the coalescence of two appendicular organs or bracteoles, Proof of this origin of the involucre is derived from the disposition and direction of its nerves, which converge to two opposite points on the margin and precisely to the two small horns or teeth of the areola mentioned above. Moreover, the nerve, which ought to correspond to the mesial costa of the bract, is often prominent and forms a keel so that the floral involucre besides being bidentate is often even 2-keeled on the side next to the axis. In some few instances two fertile flowers originate from each spathel, as in C. fertilis, C. didymocarpus, C. pachystachys, C. siamensis and occasionally also in €. deerratus, In C. fertilis at each spathel a single involucrophore bears two cupular involueres of equal size and similar shape (PraArE 230, fig. 5), and each involucre bears its own sterile flower; the same condition occurs occasionally in C, didymocarpus (Prats II, fig. 3). In the other species mentioned the second fertile flower arises from what ought to have been the areola of the neuter flower and in this case the areola itself is larger, deeper and subcupzlar. In the female spikelets each female flower is accompanied by a sterile one; when, therefore, as in ©, tenuis this is highly developed, the spikelets, immediately prior to the opening of the flowers, display two distinct series of fertile and two of sterile flowers, or four series in all. (C. siamensis, which has two fertile flowers to each spathel, has therefore four series of fertile flowers, and as each pair of fertile flowers is accompanied by one rather distinct neuter flower, the total number of series of flowers is six. Finally in O. fertilis, where each of the two flowers belonging to each spathel is accompanied by a sterile one the total number of series of flowers is eight. | XV.—The Flowers, i I have already explained how in the male spikelets the flowers arise from or near the azil of every spathel and how they are arranged right and left in two longitudinal series, these being in one plane or being more or less assurgent. I have also pointed out that in the male spikelets the flowers are solitary on each spikelet, and that in only a few cases ( C. viminalis, C. pseudo-tenuis ) in place of a single flower there is a glomerule of flowers representing a contracted secondary spikelet, | The bifarious arrangement of the flowers on one plane is the most usual, and in this case the spikelets are broad, flat and straight or slightly curved; in several species however the spikelets assume a scorpioid tendency, because the two series of flowers are assurgent and point upwards and are secundly arranged as in the Scorpioid cyme of a Heliotropium; this arrangement obtains in species of the groups of O. exilis, of C, heteracanthus and of OC. Huegelianus. In the female spikelets also a solitary female flower usually arises from each spathel, but this is accompanied by a neuter flower. If this be of large size and long persistent, the spikelets appear to have four series of flowers, as in C. tenuis, 0. javensis, etc.; but when the neuter flowers are very small, and after the neuter flowers fall, the female spikelets also appear to have biseriate flowers, For the Ann. Roy, Bor. Garp. CarcurrA Vor. Xi. 26 INTRODUCTORY ESSAY. exceptions to this rule see the preceding section under the heading “Female Spikelets. ” Male Flowers.—The flowers of both sexes in Calamus are small and unobtrusive. The male flowers are most frequently larger and more elongated than the female ones, but even the largest, as in C. erectus, C. longisetus, C. arborescens, | C. Flagellum, €. Thwaitesii, etc., rarely exceed 1 cm. in length; they are green or yellowish and are always of a coriaceous or pergamentaceous texture. The calyx of the male flower is shortly tubular, cupular or urceolate, and is always more or less deeply 3-toothed or 3-lobed (PrarE I, fig. 4). The corolla is always considerably longer than the calyx, and is divided almost to the base into 3 narrow segments (PrarE I, figs, 4 and 5). The stamens are 6 in number and most frequently are uniseriate, and have subulate filaments with a more or less distinctly inflexed apex. In C. pachystemonus and allied species the stamens are distinetly 2-seriate, tbree of them being, moreover, shorter than the other three, while their filaments are not inflexed at the tip, so that the anthers in bud as well as during flowering are erect and basifixed. When the filaments are inflexed the anthers are versatile and deeply divided at the base (Puare I, fig. 17, and Prarx II, fig. 5). In all the male flowers that I have examined I have always found a rudimentary ovary composed of 3 small elongated bodies or rudimentary carpels (Prate I, fig. 0», and Prare II, fig. 5s). I have never met with the faintest sign of a nectary in the interior of a male flower. Female Flowers.—'lhe female flowers are usually shorter and stouter than the male ones and are ovate or conic in shape; their calyx is cupular—urceolate or subcampanulate and always more or less deeply 3-toothed or 3-lobed ( Prare II, figs. 6-11). The corolla is always more deeply partite than the calyx, but its divisions very seldom exceed the teeth of the calyx; in a very few cases, as for instance in €, javensis, the corolla of the female flower is, conspicuously longer, and in C. adspersus is slightly shorter than the calyx. The stamens of the female flower are six in number and are always sterile; the bases of their filaments are slightly connate to form a cup or urceole which envelopes the ovary and is crowned by 6 more or less elongated teeth; these teeth bear sagittate basifixed but abortive anthers (PrarE II, fig. 11). . The ovary is globular, ovate-conic or even turbinate, and does not differ essentially from that of the other Lepidocaryew, of which the chief characteristic is a “lorica” formed of imbricated scales (Prare II, fig. IL). The ovary is: trilocular, with the 3 cells separated by very thin dissepimeuts; each cell has a solitary ovule which is anatropous, basilar, and inserted on the inner angle of the cell. - Most usually only one of the ovules grows to maturity, and of the others only the remains may be traced in fertilised ovaries. Perhaps the dissepiments of the cells are frequently incomplete from the first formation of the ovary. and the three ovules stand erect in the centre ofthe ovary ; this at all events is what hasseemed to me to be the case in dry specimens of C. Zollingerii, | i The style is usually very short and comparatively stout; the stigmas are 3 im number and are usually rather stout, elongate-trigonous, acuminate or subulate. y THE FLOWERS. oT lamellose inside, spreading and recurved when the flower opens (Prare II, figs. 6, 7, 8, 10 and 11), and are usually persistent even when the fruit is mature, i Neuter Plowers.—ln the female spadix the fertile flower is accompanied by & neuter one (Pr:rE I, fig. 107, and PrarE II, figs. 6, 7f) which is frequently very small and very soon deciduous, but not rarely is similar to a perfect male flower, though thinner and with attophied anthers and abortive ovary ; this is formed—as in the male flowers—of three very small bodies, the representatives of three carpels (Puate II, fig. 12”). The neuter flower is always inserted on the outer side of the involucre in the centre of the speclal area or niche which I have termed the areola of the neuter flower. The neuter flower is almost always sessile, but I have found that it is provided with a distinct stalk in OC. Henruanus and C. adspersus, In some species, as C. saliifolius and C. Rotang, the neuter flowers do not differ in appearance from the fertile male flowers, but I do not know if the pollen of their anthers be perfect and capable of fertilizing. In C. tenuis, C. Guruba and again in C, Delessertianus and C. Ridleyanus ihe neuter flowers are also well developed, and are only slightly smaller than the female ones; as they are only deciduous when the female flowers are on the point of expanding, the female spikelets immediately prior to the opening of the flowers have these arranged in four very distinct and almost similar series. From the manner in which the neuter flower is inserted outside the floral involucre and on account of the small axial part with which in a few cases it is furnished, we can readily recognise that, along with the corresponding female flower, it forms part of a very small secondary contracted spikelet. The neuter flower may therefore be considered to be a male flower which is rendered functionless by a retrograde process and becomes depauperated owing to the greater development of the fertile one or perhaps it may, from another point of view, be considered one that has never attained the full structure fitting it for repro- ductive functions. It happens sometimes in nature that certain structures, which have made ha appearance at an early date (in the evolutionary sense) under the stimulus of a definite need of the organism, and have been capable of fulfilling a definite function, later on, under altered conditions of existence, have become useless and have consequently been modified and reduced by atrophy, but still prsiest, possibly because the disuse began to be experienced when the malleability of the organism was already greatly diminished. All this is in accordance with my theory of variability restrained by the force of heredity in the plasmatical era.* These neuter flowers therefore appear to me to be a striking proof of the presence in an organism of useless structures that do not now exercise any function, or that perhaps never even in bygone ‘times exercised any function. The neuter flowers of Calami usually never open; they have stamens and an ovary that, according to the species, are mure or less atrophied, and that are deciduous before the fall of the female flowers. There seem however to be a few exceptions to this rule, as for example in C. Grifithianns (PrarE IL, fig. w whero | ibe the ealyx and corolla apparently expand, | : pacio anui lay cack Bei Nis Pene d Bemesdoteit, coe + eel sad dus. dios, Dor, Gand. Casoria Vi, X1: ^ —— 077 28 INTRODUCTORY ESSAY. XVI.—The Fruiting Perianth. In all the species of Calamus the perianth formed by the calyx and corolla is persistent till the fruit reaches maturity, and usually it is unaltered or at most slightly hardened, The fruiting perianth affords a good diagnostic character, as in some species the calyx and corolla are split to the base and their segments remain spread under the fruit where they form a kind of six-rayed star. In many other species the tube of the calyx is more or less cylindric, hardens a little after the fertilisation of the ovary, becomes callous and even sometimes slightly swollen at the base and as a whole gives rise te a kind of stalk or pedicel to the fruit. Whether the fruiting perianth be expianate or pedicelliform, we may always, owing to its different parts remaining unaltered, recognise from it the structure and the size of the female flowers. When the fruiting perianth is pedicelliform the fruit is usually provided at its base with a small caudicle penetrating into the cavity of the perianth. XVII—The Fruit. The fruit of Calamus does not essentially differ from that of the other Lepidocaryeae It has by Martius been termed loricate or mail-clad, on account of its pericarp which, owing to its being composed of numerous, regularly arranged, thin, resistent, retrorse scales, forms a kind of plated armour for the seed. This Structure is unique in the vegetable kingdom. The fruit of Calamus is more or less globular, ovoid, ellipsoid, oblong, conical- ovoid or even slightly turbinate; it is always comparatively small, and varies from 4-5 mm. in length in ©. microsphaerion, C. siphonospathus, C. microcarpus, etC., to 4 em in C, erectus. Among species that possess a large fruit may be men- tioned — C. Flagellum, ©. khasianus, C. nambariensis, C. Manan, C. simplez, C. ornatus, etc. The fruit of Calamus remains the same in the dry as in the fresh condition, but on account of the scaly nature of the pericarp it varies somewhat in size according to its degree of maturity, inasmuch in the fresh fruit the seed is often enveloped in a fleshy integument, which when the seed is mature exerts a degree of pressure against the pericarp that causes an expansion of this, which is comparable to the expansion of the skin of a snake when its body is distended with food. And, in reality, the hard scales of the pericarp in a Calamus fruit are joined together by an elastic and extensible tissue exactly as the scales of the skin of a snake are, The scales of the fruit of. Calamus afford good diagnostic characters by their shape, their colour, the peculiarities of their surface and margin, their number and their disposition, Martius has discussed the phyllotaxis: of the scales of the fruit of vertical rows, termed * orthostichies,” according to , which th | : ey are apparent! arranged round the fruit. The smallest number of these vertical rows observed THE SEED. 29 me is 12 in 0. Kunzeanus, O. nematospadix, C. digitatus, and C. microcarpus, while the largest is 24-27 in C., castaneus. The number of the series or rows is usually a multiple of 3, but in some fruits one of the series may at times be — partially missing, so that the rule does not always hold good. The number of orthostichies is as a rule rather constant, and the variation that occurs in each species is confined to narrow limits, The greatest difference in the number of orthostichies observed by me has been in C. fasciculatus, in some varieties of which I have counted 20, in others 14. As has been already said, the number varies from 24 to 27 in C. casíaneus. | The morphological nature of the fruit-scales of the Calam, as of other Lepidocaryew, is not thoroughly known. A. Braun has considered these scales to be leafy structures; they appear to me however to be rather hypertrophic or hyperplastic products or outgrowths of the epidermal tissue, analogous to the spinules, bristles, hairs and such like growths so frequent on the surface of the leaf-sheaths, the spathes and even the leaves of every Calamus. Martius considers (Hist. Nat, Palm. vol. III, p. exlix) their very regular and surprising phyllotaxis to be against this hypothesis. According to my view the scales of Lepidocaryee correspond to the spicule which grow on the fruits of some Cocoineae, as on those of some species of Astrocaryum and Bactris. In connection with this opinion it must be called to mind that the scaly coating or spinosity of the fruit in Palms always occurs in plants which are abundantly furnished with spines in other parts of the plant, as if its epidermal tissue were endowed with the faculty, of producing spinous hyperplasia on the homologues of the leaves, as the three carpels composing the ovary of a Calamus morphologically are. My view then is that the scales of Jepidocaryee are no more than the homologues of the spinules, hairs or bristles that are to be observed on the nerves of the leaves in almost every species of Calamus. The hyperplastic epidermal origin of the scales of Lepidocaryee is almost evident in the fruit of Myrialepis Scortechinii, where the scales are excessively minute and numerous and are reversed in the fruit, but are falcate, ascending, and inserted normally on the surface of the young ovary, as is seen when a longitu- dinial section of the ovary is made. XVIII. The Seed, Under the scaly peiranth there is usually a solitary seed; this, in the fresh state, is enveloped by its proper integument which is sometimes thin and dry, but is not infrequently, considerably developed, often fleshy, mucilaginous or acid, and in this case is not unpleasant to the taste; or is even loaded with tannic substances and astringent. In the dry fruit the integument is usually thin and adherent to the seed and more or less crustaceous, and often brittle when it originally was fleshy. The integument of C. aquatilis is of a very special structure because, besides the fleshy tissue of which it is composed and which after maturity is in time absorbed, it. contains numerous persistent short fibres attached normally to the testa of the seed and entirely clothing its surface with a velvety-pubescent covering, & $ 30 INTRODUCTORY ESSAY. In. a few species, like C. /rispermus and C, manillensis there are three fairly equally developed seeds in each fruit; C. Burckanus also has as a rule three seeds, but occasionally only one developes fully and only a conspicuous rudiment of each of the two others remains. When there are three seeds they are convex externally and have two flat facets which meet at a central obtuse angle. In C. Huegelianug and C., digitatus there are occasionally two seeds and then they are plano-convex. Excluding these few exceptions, the fruit of Calamus as a rule contains only one seed, and the remains of the other two are absorbed; this seed is commonly globular or ovoid or even slightly flattened and sublenticular ; exceptionally it is very irregularly shaped and angular as in C.. paspalanthus, C. gonospermus and C, ornatue. The seed of Calamus is always- erect in the fruit, and even when it is flattened or lenticular its longer axis is the vertical one. In most seeds of Calami it is possible to distinguish a dorsal and a ventral side, the dorsal being usually convex, while the ventral has generally in the centre a circular or elliptic’ depression (Prate 230, figs. 16, 18, and 19) termed the “chalazal fovea,” which in the seeds of several species is represented by a longitudinal furrow or even by a small inconspicuous rib. The exact situation of the chalazal fovea, though sometimes only faintly indicated, is always distinguishable evon when the seed is globular. The surface of the seed is rarely, as in ©. paspalanthus, quite smooth ; iaost frequently itis marked by small impressions, pits or alveoli, and furrows with corre- sponding ridges, wrinkles, small tubercles, and similar irregularities; these ridges and furrows often radiate from the chalazal fovea. à The proper integument of the seed, as has already been noted, contains at times tannie substances, and this is usually the case when the surface of the seed is irregular and especially when it is pitted and the integument penetrates more or less into the substance of the albumen. When the intrusions of the integument are superficial, the albumen cannot be considered ruminated, but in not a few cases the depressions on the surface of the seed are very deep, and sometimes these are developed into true narrow channels so that the albumen is rendered typically ruminate. This is a condition that obtains in the seeds of species of the allied genus Daemonorops ; in this latter genus, however, a reminated seed is the rule, whereas in Calamus it is the exception; in both cases the channels are filled with a brown astringent tannic substance. In the integument of the seed of C. Flagellum, when dry, I have observed numerous oblong or fusiform small bodies of a garnet-red colour visible even to the naked eye, but of course much more obvious with the aid of a lens; these bodies seem to be mucilaginous masses infiltrated with tannic acid, which fll corre. sponding lysigenetic cavities of the integument. The rumination of the seeds of C. cilaris, C. ezilis and allied species is of a very special nature. In these the seed is deeply and boldly plicate and. has a cerebroid appearance ; the integument is very thin, penetrates into the folds, many of which radiate from the chalazal fovea, and is formed of a few layers of parenchymatous cells filled - with very small green corpuscles partly soluble in water, to which they impart their colour and a very EXTRA-FLORAL NECTARIES. 9l bitter taste. Dr. "Veturia Bartelletti, who has made a special stady of this sub- stance,* states that it is probably a derivate of tannic acid. In the genus Calamus the rumination of the seed is not of great taxonomic value, because it happens that of two species which, from many points of view, may appear very nearly allied, one may have homogeneous, the other ruminated albumen. It is not therefore possible to make use of the character of homogeneity or rumi- nation of the albumen as a means of dividing Calamus into two principal sections. Leaving out of account its rumination, the albumen of the seed of Calamus is always horny or bony.. In the descriptions of the albumen it is termed homogene- ous not only when it is really so, but often also when the intrusions of the integument are very superficial. The rumination of the seed is very apparent in C. erectus, C. Flagellum (and its allied species) C. Huegelianus, C. Gamblei, C. gracilis C. melanacantkus, C, Diepenhorstii, C. macrosphaerion, etc. The embryo of the seed of Calamus is most commonly situated at or near its base (Puare 230, fig. 13), but it is lateral and opposite to the chalazal fovea iu C. exilis, C. ciliaris and other species of the group. It is also lateral in C. gracilis and C. melanacanthus which have, besides, a ruminated albumen, while in C, Kunzea- nus it is lateral and the albumen is homogeneous (PrarE 230, fige. 18, 19). XIX.—Extra-floral nectaries. So far as I know nectarifluous surfaces of any kind have seldom} been observed in the flowers, and never hitherto in any of the other organs of Palms. It appears to me, however, that certain special swellings or callosities, which frequently exist in certain definite positions, in not a few species of Calamus, should possibly be considered extra-floral nectaries. I have already alluded more than once, in passing, to the existence of these supposed nectaries, which are to be met with: Ist, at the insertion of the leaflets on the rachis in their upper axil; and, in the hollow formed by the folding of the leaflets at their base, just at their insertion in the lower surface; 3rd, at the insertion of the spadices and flagella; 4th, in the axil of every branching of the spadix and its subdivisions, such as the spikelets and the involucrophora. I have never had any opportunity of studying these nectariform surfaces, as they may be termed, in a fresh condition; in dry specimens it is difficult to investigate their true nature. These surfaces always have the appearance of small swellings of a lighter colour than the adjacent tissue and are formed by two lips, more or less tumescent, separated by a cleft or rima, which may be supposed to be, in some cases, nectarifluous. Most Calami are provided with these: structures, which are particularly conspicuous in the axils of the leaflets of €. perakensis and C, ramosissimus, and are extremely devoloped at the junction of the partial inflorescences with the axis of the. female spadix and. in the axil at the insertion of the spikelets in C. paspalanihus. As good. examples of the supposed nectariform structures situated in the axils of tho involucrophora I may cite C. Gamblei, C. unifarius var. Pentong, €. adspersus, etc. and. * Bull. Soe. Bot. Ital. 1904, p. 309. | a t Vide Delpino im Nuovo Giorn: Bot. It. II (1870), p. 61.- Gk dicenda 32 INTRODUCTORY ESSAY. for those found in the fold of the blade of the leaflets at their base in the lower surface may be quoted O, aruensis and C. Hoilrungit. | Usually we observe axillary nectariform callosities on the spadices when the partial inflorescences are situated outside the mouth of their respective spathes; and most commonly, where the) larger branches of the spadix are provided with an axillary callus, these appear also at the successive divisions and even at the insertions of the spikelets and of the involucrophora. The -essential requisite for a nectarial structure is the secretion of sweet fluids; therefore in Calamus, the transverse rima of the callus, if this be really a nectary, ought to be the opening for the exudation of such fluids; but in the herbarium Specimens examined by me, I have never been able to discover the slightest sign of their presence, nor have I seen any indication of their having been. resorted. to by insects of any kind, in search of nectar. It is only from observations on living plants that wel may acquire any definite knowledge of the nature of the nectariform surfaces of Calamus. X X,-- Calami under Cultivation, Although the species of Calamus, the cultivation of which bas been attempted in the hot-houses of extra-tropical countries are pretty numerous, the number of those which have become permanently established is small, owing to its being very difficult to provide them with conditions of existence like those enjoyed in their native countries, The Calami in our hot-houses therefore give but a faint and poor idea of the elegance of their foliage as it appears at the summit of a long slender and climbing stem. Young plants of Calami are, however, considerably appreciated by. horticulturists on account of their highly ornamental, bright green, graceful pinnate leaves, so that they are frequently offered for sale in commercial catalogues of living plants, But the names by which cultivated Calami are known to horticulturists are, with hardly an exception, incorrect; and since, on the other hand, horticulturists are in the habit of putting on the market small seedlings or very young non-characterized plants, the foliage of which usually exhibits much uniformity in the various species and always differs considerably from that of the adult plant, it becomes very difficult for a botanist to reduce the species rashly proposed by horticulturists as new to their’ true position in scientific nomenclature. | : I have been able, from specimens of leaves of cultivated plants preserved in the Herbaria at Kew and Berlin, to establish the fact that many of the supposed Calami are species of Daemonorops, for it appears that some species of this genus are more | easily cultivated than most species of Calamus, but I have very seldom been able from specimens of this kind to determine the actual species to which they belong, In Very many cases the exact naming of these horticultural Calami has been rendered kae e owing to the fact that their appearance in European hot-houses has n qute ephemeral, and that many with newl sed names have disappeared for ever from cultivation, : S £t : In the French edition of Nicholson and Motetts Dictionnaire d’ Horticulture ZT cultivated species of Calamus are enumerated and many of them bear the names of USES AND NATIVE NAMES. 33. very well-known species, but, as it appears to me, in almost every instance misapplied In the * Report on the Progress and Condition of the Royal Gardens at Kew” for the year 1882, 37 species of Calamus are recorded as under cultivation at Kew. So far as I know the only species that have succeeded in producing flowers in Europe are C. ciliaris (Gard. Chron., Feb. 6, 1897, p. 86, fig. 23) and C. javensis of which last I have seen a specimen from a plant that flowered at Kew under the name of C. írinervis. At Kew also Daemonorops Jenkinsianus has reached a fair flowering state. | The plants of Ca/ami never have deep roots, and they acquire a luxuriant habit in the superficial layer of humus of the tropical forests, when this overlies a silieious sub-soil, because Calami, like almost all other kinds of Palms, avoid a calcareous sub-soil. : In cultivation Calami thrive best in a compost of equal parts of sandy loam and vegetable soil formed by decomposed leaves, They require a warm moist atmosphere and copious watering. I have however to observe that Calami grow in very different situations, from marshy plains at the level of the sea up to an elevation of 2,000 metres in the mountains, so that in the cultivation of Calami, as in that of any other plant, it is necessary to know beforehand the natural conditions of their existence and to modify their. cultural conditions accordingly. If this be borne in mind, it may be found that probably not a few of the mountain species of Calamus, as for instance those of the Himalaya and Assam, will thrive better in the temperate than in a warm hot-house, while others should receive the treatment of aquatic plants in warm water, So far as my knowledge goes, no attempt has been made to cultivate any of the economic sorts of Calamus in their native countries or in countries with an analogous climate. Only C. khasianus appears to receive a primitive cultural attention in Assam for the sake of its fruits, which are eaten as a substitute for those of the Areca Palm, XXI— Uses of,* Trade in, and Native Names of Calamus. The long and slender stems or canes of Calami when divested of their sheaths ‘are usually known by the Malay name of ‘“ Rotang,” commonly altered into “Rattan” or ‘Ratan,’ and are put to various uses according to their size, length, flexibility, elasticity and toughness. | The most slender canes are those produced by C. javensis and its varieties, and by allied species, They are employed entire for binding purposes, and in making chairs, blinds, mats, wicker- or basket-work, fishing implements, etc. The largest and more resistent canes, such as those of OC. rudentum, C. ornatus * As regards the Economie value and domestic employment of Calamus, besides the great works of Blumo (Rumphia) and Rumph (Herbarium. Amboinense), the following may be consulted; with, however, the caution that the names assigned by many authors to the economic species are not to be relied on:— J. Forbes Royle: The Fibrous Plants of India, p. 92. Brandis: Forest Flora of North-West and Central India, p. 558. Gamble; A Mannal of Indian Timbers, New edit. (1902), p. 734. .Hasskarl: Aantekeningen oer het nut, door de bewoners van Java, etc, p, 61, species of Java, De Mercado: Libro de Medicinas, in the Flora de Filipinas, Gran Edicion, vol. iv, p. 80, especially with reference to the Arn. Ror. Bor, Garp. Carcurra Vor. XI. 324 . INTRODUCTORY ESSAY. C. palustris, C. albus, etc., nre made use of entire as cables by the natives of íhe Malay Archipelago, Cochin China and neighbouring countries; large cables are also often made of many small canes twisted together; these, which are often ot extraordinary strength and durability, are much employed for native craft such as Prahus and Junks. C. palustris is much used entire, in Burma, for tying timber in rafts and making the cables which stretch across the river at the Salween rope station (Gamble). To the same use are put the entire canes of C. ornatus, which are employed in Java for moving ferry boats. Usually, however, for many purposes the rattan canes are split throughout their length into 2-4 or more strips from which the inner soft brittle and spongy portion is removed by means of a knife or same other instrument, so as to leave the ‘external portion, which is hard, tough, flexible, elastic and has its outer surface very clean and smooth as if it had been varnished, The process of cleaning and redncing the canes of Calami into fine strips is termed by the Malays * raut” and from this word is derived “ rautang,” whence * Rotang," that is to say the object which it is possible to reduce to strips by the * raut.” The strips vary in width according to the use to which they are to be put; those for delicate work, as for the network of furniture, for small bags, hats, eto, are from 1-3 mm, in breadth; those employed as lashings in natiye house- building or in fastening the removable head of the Malay axe to its handle are from 5-6 mm, wide, The natives often dye these strips of Rotang red when they desire to give a ‘more elegant finish, by means of variegated patterns, to their work,* The canes of some Rotangs are used entire and cut into pieces of appropriate length to make handles, rods, walking canes and the like. For this purpose the erect and slender stems of the non-scandent species, such as C. bacularis, are best suited; but the basal portion of certain of the scandeut ones is sometimes employed, According to Gamble, C. inermis furnishes in the Sikkim Himalaya the best alpen- stocks. The well-known “ Malacca canes” or “ Pinang Lawyers” which are largely exported to Europe are cut from the stems of €. Scipionum. Their chief merit, apart from their toughness, their elasticity, and the beauty of their surface, lies in the extraordinary length of their internodes, a single one of these being sufficient to make a very elegant walking cane, : : Rattans are a natural product of the primeval forest, Those that reach Europe are largely exported from Singapore, whence come all those that have been -gathered in the forests of Sumatra, Borneo, Celebes, the Malay Peninsula, ete. . . The process of collecting and preparing Rattans is very simple. When a Rotang plant has been found in the forest, the stem is cut near the ground and is detached from the trees, from which it is suspended by its hooked thongs, by taking a strong hold of its base and thus pulling down the entire plant with its leaves, In order to clean the plant of its leaves and especially of the spinous sheaths that clothe *The process of dyeing rattan’ canes red in the Philippines has been described by Father Llanos in the “Flora de Filipinas," Gran Edicion, vol, iv, p. 49. USES AND. NATIVE NAMES. 95: the stem the top of the plant is cut off and then, handling it from the upper end, the stem is forcibly drawn in the opposite direction between two pieces of wood. In this way the stem is easily stripped of its spiny coverings and is then cut into lengths of about 5 metres each. These pieces are bent into two equal parts and fastened into bundles; in (this state the canes are brought to the market. The Hotang which are thus prepured, and are most valued by traders, are not thicker than a man’s little finger and have a fine polished straw-yellow glassy surface. | Some of the finest and most slender canes are preserved entire and are put on the. market rolled up like coils of thick iron wire. I have no reliable data as to the quantity of Rattan canes imported to Europe from Eastern ports. It is not even known what quantity is produced in each country. It is however certain that Borneo must supply a very large proportion of those that are put on the market, as we find from the Report of the Trade of Sarawak that from that town alone 27,784 piculs of Rattan were exported during the year 1899, and 31,200 during 1900, while some years before the supply had been still greater. With the exception of the Malacca cane it is not exactly known from what species the ratíaus of commerce are obtained. The indigenous names which I have always recorded when they were known to me may assist to a] certain extent in the identification of the economic and commercial species, but it is a very well-known fact that native names are seldom to be relied on in a rigorously scientific sense. Nor must we believe that all the species of Calamus produce a serviceable cane; in not a few of them the stem, though very long and flexible, is soft and brittle and therefore useless, Among Indian species to which this remark applies are (€. leptos- padis and C. Flagellum. On the other hand, the canes of C. tenuis, C. Rotang, C. fascicularis and ©. palustris, and in lower India also those of C. pseudotenuis, are of very good quality and much appreciated, In the Nicobar Islands there is a great demand on the part of traders for the rattan of C. nicobaricus; C. longisetus is much used by the Andamanese. In Java the most useful canes are those of C. javensis, €. viminalis, C. melunoloma, C. heterotdeus and C. Reinwardtii, especially the two last. which, however, are less esteemed than those imported from Sumatra and Borneo. In the Philippines the rattan of C. mollis seems to be one of the most commonly used, In the Malayan Peninsula there are certainly many species of Calamus that produce valuable canes; but, though the species of that region are comparatively well known scientifically, we are almost entirely ignorant to which of them the commercial canes that are brought by the natives fom the forests and sold to traders correspond. A Calamus which is termed ¿there **Retang Segu,” and which 1 have identified with Blume's €. caesius, corresponds to the ^ Rotang Segah” of Borneo, also termed in Sarawak “Rotang Buluch," which is C. optimus Becc., a very near ally of C. caesius, - In Sarawak this is undoubtedly the most esteemed Rotang among all the species known to the Malays; being one of those that are easily reduced to long strips it is much used for finer work on account of its very neat straw-yellow polished Axw, Roy. Bor. Garp. Carcurra Vor. XI. 36 INTRODUCTORY ESSAY. surface. Another species also much valued in Sarawak is the ‘ Rotang baroh, ” C. ertoacanthus Becc. Among the best Rotangs of Sarawak I have to mention the “ Rotang Perdas,” €. javensis and C. zonatus; the “ Rotang Sakkat,” C. muricatus ; the *'Rotang Berman," €. /fabellatus; the *Rotang Jangut” named also ''Rotang batu” or *Totang Kawat,” C. filiformis, all of which produce a very long and resistent rattan. The last named is the smallest of all. The Malay name Rotang is of rather wide significance and includes almost all the species of climbing Lepidocaryeae, especially those of the genera Daemonorops and Korthalsia, but the rattans which these last named Palms produce are not very slender and never have a vitreous polished surface, On this account they are not sought after by traders in spite of their being frequently very resistent and therefore largely used by the natives for many of the purposes which the canes of the true Calami subserve. The natives of Malesia occasionally take advantage of the silicious coating of the stems of Calami for striking fire, and, moreover, Mr. E, H. Man tells us that the Andamanese contrive. to make knives of pieces of the stem of C. palustris cut in such away as to present a sharp edge. The ruminated seed of certain Calami, being astringent, is sometimes eaten as a substitute for that of Areca. For this purpose the seeds of C. erectus and C, Flagellum are often used by the poorer people of India. The fruits of C. khasianus seem also to be employed in Assam in the same way, and according to a note by Mr.C. B. Clarke attached to a specimen in my herbarium, this species appears to be cultivated at Naughedam in Assam, The pulp which envelopes the fruit of some of the larger species, such as C. ornatus and C. Manan, is edible, being white, acidulous and refreshing. According to E. H. Man tne Andamaneese eat the cooked seeds of C. longisetus. The long large leaflets of the fronds of C, andamanicus are employed in the Andamans for thatching. | A few, probably hypothetical, medicinal properties have been attributed to some species of Calamus, Rheede says that the fruit of C. Rheedei dried and powdered, heals ulcerated knees, and Rumph- tells us that the ashes of the stem of C. viminalis mixed with Arak, are a remedy against the bite of venomous animals. The J avanese, according to Blume, attribute to the roots of C. ornatus, bruised in water, the property of alleviating the pains of labour, while Amboinese ladies, according to Rumph, use the water which flows abundantly from the stems of certain species when they are cut across, as a cosmetic to prevent the fall and increase the growth of their hair. | : Blanco asserts that the fruit of C. mollis is considered in the Philippines to be poisonous, but l think this can hardly be possible. If true, this would be the only instance of a Palm containing a poisonous substance. | E | | ig SPECIES OF THE PHILIPPINE ISLANDS, 37 The native generic names for Calamus are numerous. Besides the universally known name of Rotang, we have the following :— , Uri, in Ternate; Boucan, in Makassar; Ua, in Banda and Amboina; Bejuco (a Spanish name), in the Philippines; Penjalin, Hoéh and Hooek (Huyk), in Java; Khóe and Khiéea, in China; Máy, in Cochin China; Bét, in Hindustani; Vetra and Vetus, in Sanskrit. As a philological curiosity I may suggest that apparently the names of many climbing plants, and especially of such as are employed for the purposes of tying and binding, like the Latin * Vitis,” the Italian '*Vetrice " (Ligustrum), '**Vimine" and * Vinco” and the Latin “Juncus,” the Anglo-Saxon *'Welig," the English * Willow," etc. may have a common origin with the Sanskrit name for Calamus. XXII.—Note on the Species of Calamus of the Philippine Islands. Father Blanco in his “Flora de Filipinas” published in the year 1837 has given a very incomplete description of four. species of Calamus, growing in these islands and named ©.. mollis, C. gracilis, C. maximus and C, usitatus. Besides these, three other Calami are mentioned, namely C. Curag, the Calamus of the island of Negros, and another with the native name of “ Limoran.” As there are no authentic specimens left of any of the plants of Blanco, we must be content, in trying to identify the species above noted, to work with only Blanco’s descriptions. C. mollis is apparently a common species in the Philippines for it has been met with again and again by all modern botanists and we may consider its identifica- tion as assured. C. gracilis had. its name changed to C. Blancot by Kunth owing to the name gracilis having already been applied by Roxburgh to an Indian species. There is, it is true, no sufficient evidence that C. Bluncoi of Kunth, to which have been referred the specimens of a Calamus distributed by Cuming imdát No. 1225, exactly corre- sponds with Blanco's plant; in tue absence, however, of any type specimens we ‘can agree to accept them as the same. €, maximus I have found to be a local variety of C. ornatus Blume, and C, usiía- ius has been reduced by me to Daemonorops Gaudichaudit Mart. C. Curag Miq., which is stated by Blanco to grow on Mount Angat, and to be a non-seandent plant, is not recognizable from the brief statement in which it was published by its author. The Calamus of the Island of Negros is said to have black stems used in making walking canes and in the Novissima Appendix to the ‘Flora de Filipinas’ p. 276, it has been reduced to C. Scipionum var. maculatus ; but the only thing, I believe, that is certain about it, is that it is a quite different species from 0, Scipionum. 38 INTRODUCTORY ESSAY. The Calamus with the native name of “Limoran” is supposed to be a Zalacca (vide Nov. Append. p. 274 under the heading of Zalacca edulis), De Mercado (Fir Filip. gran edicion, vol. iv, p. 50) considers it to be Daemonorops Calapparius, but on what ground it is difficult to say; I have found however the Tagala name of ‘*Limoran” applied by recent collectors to €. crnatus var. philippinensis. In the “Novissima Appendix” at pp. 275 and 356 there are 14 species of Calamus and 14 of Daemonorops recorded but happily no new names are proposed for them and all are reduced to already known species. Their identification however is based in most cases on the inspection of a portion of their dried canes, a most extraordinary proceeding, for every botanist who has made a study of Calamus knows how difficult it is to name the speties even when in possession of good specimens, Owing therefore to the impossibility of recognising the species mentioned in the ‘‘ Novissima Appendix” I have avoided quoting them, There are now 17 species of Calamus (and 3 of Daémonorops) more or less: completely known from the Philippines; all of them are endemic in these islands except C. ornüius, a Palm which grows also in Java, Sumatra, the Malay Peninsula and Singapore but which is represented in the Philippines by a special form. It is fairly certain that many more species of Calamus still remain to be discovered in the numerous islands which compose the Philippine archipelago, because there’ as elsewhere the species seem to he much localized. Three other species, apparently different from those already known but represented by specimens too incothplete to admit of description, have been collected by Dr. Warburg in the Island of Mindanao. e -There is, however, now a hope that in a comparatively short time we may have an almost complete knowledge of the Philippine Palms, through the methodical botanical survey of the entire Philippine Flora, already begun by the American Goverhtient under the saperintendence of Mr. Elmer D. Merril], | p "Many of the species of Philippine Calami produce valuable canes, but as yet we do not know the purposes for which each is most suited or its corresponding scientific name, That the species of Calamus in the Philippines are probably very numerous is indicated by the fact that some of the species discovered by early collectors, as for example O. Cumingianus, O. manilensis, and ©. discolor have not been found again by modern explorers. XXIII.—Identification of Loureiro's Cochin-China Calami. 13 . Loüreito in his Flora Coehinehinensis, of which the first edition made its. appearance in the year 1790, gives rather detailed descriptions of six species of Calamus, One of these, C. Seipionum, is not a native of Cochin-China; this species I have most certaiuly recognised in numerous complete flowering and fruiting specimens. ~ from the Malayan Peninsula and neighbouring countries. : The five Cochiü-Chinése species bear the following names := 0, petraeus, 0. rudentum €, deris; C. amarus and O. dioiews, No authentic specimens of any of these species He known to exist in hefbaria, and their identification is therefore always open to doubt beciuse Loureito’s descriptions, though wonderfully good for their time, are far from sufficient to ensure accuracy of naming in the case of a species of Calamus. 4 ROXBURGH'S SPECIES. 39 I believe that I have recognized C, dioicus and also C. rudenlum in some specimens collected in Cochin-China by Mr. L, Pierre. I cannot say the same of C. petraeus. lf the description of this species be correct, this cannot be a Calamus, asthe spadix is said to be terminal and we ought to suspect it to be a Plectocoma or a Korthalsia rather than a Calamus ; the citation, however, of Rumph's Palmijuncus Calapperius (Herb. Amboin., v, t. 41), adduced by Loureiro, is against this opinion because Rumph’s plate represents an easily recognizable Daemonorops, C. petraeus therefore still remains an enigma. Another enigma is C, verus. By its short spadix and oblong spathe, and also by the plate cited from Rumph (Palmijuncus verus Herb. Amboin, v. t. 94) we might believe that we had to deal with a Daemonorops of the section Piptospatha ; and by its leaflets, which are described as ovate-lanceolate, it would approach D. didymophyllus. Probably then C. verus is a Daemonorops not found again by modern botanists. C, amarus appears to me to correspond to C. tenuis Roxb. (vide observations under that species); but as there is not complete evidence of this identification, and as there never can be absolute proof owing to the non-existence of authentic : specimens, I haye not proposed the adoption of the name C, amarus instead ‘of C. tenuis, although the former be more ancient than the latter. XXIV.—Note on Roxburgh’s Species of Calamus. In his Flora Indica (ii, p. 778, et seg.) Dr. William Roxburgh gives descriptions of the following species of Calamus to which I have added when necessary the corresponding correct name :— Calamus Zalacca Willd. =Zalacea sp. 5, humilis Roxb. Perhaps the young M of €, latifolius. T erectus Roxb, © Draco Wild. =Daemonorops, " latifolius Roxb. » Trudentus Willd. (not C. rudentum Lour.) =C, albus is 5j extensus Roxb. =Daemonorops Jenkinsianus Gr. ? 3 quinquenervius Roxb. =C. palustris amplissimus Beec, T Rotang Willd. » faseiulatus Roxb. =C., viminalis fasciculatus Becc. » tenuius (tenuis) Roxb. , pencillatus (sic) Roxb. =C. javensis Bl.? » gracilis Roxb. „ monoicus Roxb. =C. Kotang. Of the species proposed as new by Roxburgh, the following species :—-C. erectys .Q,. latifolius, C. gracilis and C. tenuis, haye therefore been definitely anette 4 FT additions to science. Owing to authentic specimens not being now available, the following neenajn more or less doubtful:--C. extensus, C, quinguenervius and C. pencillatus ; reference may be made to what has been said regarding each of these under the heading — of doubtful species, 40 INTRODUCTORY ESSAY. XXV.—-Note on Rumph’s species of Calamus and Daemonorops, Subjoined is a conspectus of the species of Calamus and Daemonorops figured in the fifth volume of Rumph’s Herbarium Amboinense. * Prate bl. Palmijuncus Calapparius = Daemonorops Calapparius Bl. " niger = Daemonorops niger Bl. " 53. 5 albus = Calamus albus Pers. 54-1. » verus = A young plant of a species of the group of Calamus palustris or perhaps of C. pisicarpus ? 54-2. ” verus angustifolius = Calamus Rumphii Bl. (Daemonorops Rumphit Mari.) 55-1. j verus angustifolius = Calamus pisicarpus Bl. w 55-2A.B. ,, viminalis = Calamus viminalis Willd. 56. » equestris = Calamus equestris Willd. ; 57-1. » equestris = Calamus Cawa Bl. T 58-1. » Draco = Daemonorops Draco Mart. 58-2. Rotang Assam = Calamus acidulus Bece. The Rotang Assam his been referred by Blume to C. barbatus, but to me it appears to be quite distinct nct only from this species, but from any other hitherto known. The species distinguished by an asterisk (*) have been definitely recognised. The others represent, I believe, very well-marked species which will be recognised at some future time because, considering the period at which they were made, Rumph’s figures are very good and the descriptions, if properly understood, are quite reliable. I have therefore no doubt that these species will be found again in the Moluccas when these islands are better explored. This belief cannot be certainly entertained regarding the species which are incidentally mentioned or imperfectly described by Rumph. These are as follows:— Calamus Buroensis Mart.; C. viminalis var. repens Bl.; C. graminosus Bl.; C. maritimus Bl. For information regarding these, reference should be made to the detailed account of them given in the section that deals with doubtful species. XXVI.—Note on the date of publication of the species of Calamus and Daemonorops named by Martius, Griffith and Blume. The fact that Martius, Griffith and Blume were at work contemporaneously on the genera Calamus and Daemonorops has led to great confusion in the synonymy of these genera. The great work of Martius, ‘‘ Historia Naturalis Palmarum," and the “ Rumphia ” of Blume were published in parts at long intervals, and I have not been able in all cases to ascertain the precise date of publication of each part. Moreover, some of the pages of the third volume of the “Historia Naturalis Palmarum," those num- bered 179 fo 230, were originally printed on paper of a different colour from the rest, These pages were, affer an interval of many years, reprinted and the reprint substituted for the original. But the reprint of these pages was in reality a new edition— very different from the original, for the author took advantage of their publication to entirely change the text and to make numerous additions. In the first edition of these pages (179—230) the generic characters of Calamus are given at page 203, and in the pages that follow six species of this genus, of t With reference to this subject, the note by Martius himself in the Errata at the end of vol, iii may be consulted, SPECIES OF MARTIUS, GRIFFITH AND BLUME. 41 which in all 46 species are recorded, are fully described. The genus Daemonorops is entirely omitted in the pages above mentioned, and there is no citation in them of the second volume of Blume’s ** Rumphia," which bears on the title-page the date 1836. It may be observed with regard to this volume of Blume’s work that it contains the figures of several species of Daemonorops the descriptions of which appeared many years later in the third volume of the same work. From these data we may conclude that the first edition of pages 179-230 of the “Historia Naturalis Palmarum ” was published before 1836, the more so because in these pages there is no reference to the first edition of Blanco’s ‘“ Flora de Filipinas,” which bears the date 1837. There is little or no doubt that the portion of the second volume of ‘‘Rum- phia," containing plates 171—173, where the species of Daemonorops, the descriptions of which first appeared in the third volume, were figured, was not available to the publie before the year 1843, although the title-page of the second volume is dated 1836. For this reason I believe that the first edition of pages 179-230 of the * Historia Naturalis Palmarum " should be cited before the plates 71-137, as this first edition in all probability made its appearance in 1836 and the plates in 1843, In the second edition of pages 179-230 there are descriptions of 13 species of Daemonorops and of only six species of Calamus. As regards these pages, we might suppose that they were not published later than 1846, because there is no feferente in them to Griffith’s paper on Indian Palms which appeared in 1845 in the * Calcutta Journal of Natural History.” It seems, however, that the true date of these pages is the year 1849, because there is a notice in Hooker's “ Journal of Botany ,” 1, page 221 (1849), in which mention is made of the publication of one of the later parts of Martius’ work, containing also 49 pages, belonging to part 7, which were presented to subscribers in substitution of others, the paper of which had changed colour. That the enumeration of the Calami in the third volume of the ** Historia Naturalis Palmarum" was completed in the year 1849 was explained by the illustrious author himself in a note appended to page 828 of that volume. It is also stated there that though the Introduction in Volume iii bears the date 1847, the volume was not available to the publie before 1849. This circumstance may possibly account for the fact that in the second edition of pages 179-230 of the “ Historia Naturalis Palmarum” there is no reference to the text of the third volume of ‘ Rumpbhia. " The dates of publication of the species of Calamus and Daemonorops, which are mentioned or described in the works of Martius, Blume and Griffith, so far as I can make out, are as follows :— 1842 ? Martius; Hist. Nat. Palm, iii, fírst edition of pp. 179-230, 1843 Blume; Rumphia ii, pp. 93-173 and pl. 71-187. 1845 Griffith ; Palm. Brit. Ind. in Caleutta Journal of Natural History, v. 1845 ? Blume; Rumphia, iii, pl. 134-154, without descriptions. 1849 ? Martius : Hist. Nat. Palm. iii, 2nd edition of the pp. 179-230. : m 1849 Martius ; The enumeration of Calamus at the end of vol. iii of the Hist. Nat. Palmarum. 1849 Blume; Rumphia, iii, with the descriptions of Calamus, which bears on the title-page the date 1837. 1850 Griffith ; The Palms of British East India : op. post. : Ann. Roy, Bor. Garp. OarcurrA Vor. XI. UN E 42 INTRODUCTORY ESSAY. XXVIT.—On the classification of the species of Calamus. I must admit that I consider the systematic arrangement of . Calamus proposed by me to be far from satisfactory; the species are, however, aggregated in groups whose affinities are in most cases real and not artificial. The order in which the groups succeed each other is not wniseriate; that is to say, it must not be supposed that a group is directly derived from the one which precedes it. The whole series of species of Calamus in my arrangement lays no claim to rep- resent a true phylogenetic conspectus; indeed, I believe such a classification to be in reality as impossible for the genus Calamus as it is for almost any other group of organisms, I have no belief in the formation of the species of which a genus is composed by a gradual and successive modification from a single primitive archetype. I believe therefore that it would be impossible to compile a complete phylogenetic system of existing organisms even if we could actually examine all the transitory forms of each genus which may have existed, but have now disappeared. My hypothesis is that in the remote epoch which I have called “ Plasmatical,”? reproduction may have been possible even between two organisms of very diverse nature.* I therefore assume that a few primitive dissimilar types may have been eapable of producing fertile offspring, participating in the characters of their parents, and I do not think the manifestation of this phenomenon impossible at that remote epoch, if it be assumed that the power of heredity, or of that force which causes the transmission to their offspring of the universal qualities acquired by their parents, was constantly more active the further back we go in the history of the development of the organic world; while the “ plasmatical force," or that which has given to organisms the faculty of assuming new forms or colours more fitted to changed conditions of existence, very powerful in remote geological times, has been constantly less active the more we advance towards the present era, In other words, I suppose that in the remote past fertile hybridism may have occurred between different species of the same genus, between representatives of different genera and, in the very earliest times, even between organisms of a nature so different that now we should eonsider that they belonged to different families. According to this theory the connecting links between two species would not have been the result of innumerable forms gradually appearing by continuous slight variations, but would be due to the sudden intervention of hybridism. In my systematic arrangement of the species of Calamus l have tried to aggregate the species in groups in accordance with their natural affinities, but I have not been always able to effect my purpose, partly because in some species the affinities are not very clear, partly because our knowledge of some species is defective. If we attempt to establish the chief divisions of the genus upon the more con- spicuous biological characters presented by its constituent species, as for example on whether the leaves are furnished with, or are destitute of, a cirrus; whether the leaf- sheaths are flagelliferous or not; whether the seed has a ruminate or an equable albumen: it very often is found that we are separating widely certain species that in other respects are manifestly closely allied. For instance C. erectus is evidently related * Beccari: Nelle Foreste di Borneo, p. 300, DIAGNOSTIC CHARACTERS. 43 to C. Flagellum, though it has an erect non-seandent stem with non-flagelliferous leaf- sheaths, while C. Flagellum is a lofty climber, provided with very long clawed flagella, Even in the formation of the groups I have found it almost impossible to assign to each division constant and precise characters as it almost always happens that some of the species exhibit aberrant peculiarities. With reference to the specific value of the new forms proposed by me I may observe that I have followe a middle course, neither differentiating the species excessively nor amalgamating them too boldly, I admit that all of the species I have proposed as new are not of equal value; this, however, is an aim that is impossible of attainment, so many and varied are the natural gradations between specific entities. For example C. khasianus, C. nambariensis and CU. inermis are three very closely related species which some botanists might perhaps consider to be different forms of one. They certainly exhibit far fewer differences among themselves than occur for instance between C. erectus and C, Flagellum. This unequal degree of specific differentiation exists, however, in every generic group of the organic world, and is one of the capital difficulties encountered by the systematic naturalist in dealing with the fauna and flora of every country. Among the diagnostic characters for the groups it sometimes happens that characters are mentioned of which it is impossible to verify the presence in individual species owing to the incompleteness of the available material; this deficiency, however, is almost always made up for by the correlation of characters, whereby we may, from the parts that are actually before us, deduce the nature of those that are missing. For instance, if we have a spadix which ends in a long robust clawed flagellum, we know for eertain that the leaf rachis is not prolonged into a similarly clawed cirrus; if the leaf sheaths be flagelliferous, we know the leaves are not cirriferous and, on the contrary, if the leaves have a distinct clawed prolongation at their apex, the leaf sheaths are not flagelliferous and the spadices in all probability are panicled, comparatively short and broad, and. devoid of a long terminal prolongation. It has not seemed advisable to add to the conspectus of the species an- artificial key which would not only have been very difficult .to compile but very difficult to employ, owing to the. imperfect knowledge that we possess of many of the species and on account of the universal incompleteness of the material usually collected or present in Herbaria. Moreover, the species of Calamus being usually very localized, or found within very limited geographical areas, and the number of species of each region being comparatively limited, a study of the geographical conspectus, with the assistance of the chief subdivisions and their diagnoses, and, above all, of the plates, should I think render the identification of a Calamus a matter devoid of serious difficulty. XXVIII.—Diagnostic characters of the General Calamus and Daemonorops, There is no precise and easily grasped diagnostic character which enables us to distinguish at once a Calamus from a Daemonorops. Nevertheless these two genera are so completely distinct by such an assemblage of. characters that one is never uncértain as to which of the two a particular Palm belongs. i Ann. Roy. Bor. Garp. CarcurrA Vou. XI. 44 INTRODUCTORY ESSAY, The leading difference between a Calamus and a Daemonorops is’ of a biological character and resides essentially in the function subserved by their spadices or spathes; these organs in fact are so modified as either to be of some assistance to the plant in climbing, or, when this is not the case, the spadices appear to be direct morphological derivates, and the function of assisting the plant to climb is then subserved by the leaf cirri. In Daemonorops the spadices never serve as climbing organs and their spathes, from the very first, are utilized to enclose and protect the flowers, being besides always short in comparison with those of a Calamus. In Calamus the spathes, even in spadices that are not cirriferous at their summits and that are comparatively short, are elongate and persistent, at least in their basal part which is always tubular; they are also, like the spiral part of the spadix, more or less armed with hooked prickles—the kind of spines that aid the plant to climb. If, as sometimes happens, the spathes of Calamus are not tubular but are open, flat, and laminar, the axis of the spadix is nevertheless elongate and more or less armed with claws, at least at its apex. In Daemonorops all the spathes, with the exception in some cases of the basilar one, are deciduous, usually broad and open thoughout their entire ventral aspect and, more especially, are never armed with hooked prickles; they have therefore no organs at all to assist the plant in climbing; moreover, the axis of the spadix, which is usually short, never bears claws and never shows any tendency to become flagelliform. | Among the species of Calamus I know only O. hypoleucus, a very anomalous Palm, which hes short spadices and broad cymbiform unarmed spathes, a good deal resembling those of a Daemonorops. C. Lobbianus, O. conirostris and C. brachystachys have also spathes somewhat resembling those of a Daemonorops, but in these species, which also form a group standing alone in the genus, the structure of the spikelets and flowers is that of true Calami. In fact, the spikelets of Calamus, especially the female ones, are somewhat different from those of Daemonorops; in these the spathels are less developed and generally are reduced to a mere scale or to a short membranous ring, and the involucrophore is elongate and stout so that usually the fruit of a Daemonorops appears distincly stalked. Moreover, female flowers of Calamus are distinguishable at a glance from those of Daemonorops : in the former genus they have the calyx distinctly 3-toothed or 3-lobed and the corolla hardly longer than the calyx: in the latter, as a general rule, the calyx is almost truncate at the mouth and the corolla is conspicuously longer than the calyx. The fruit in the two genera-is very similar, but a fruit with a seed in which the albumen is homogeneous can never be that of Daemonorops, though there are not a few Calami with a ruminated seed such as, without a single exception, we find in Daemonorops. | No Daemonorops ever bears leaf sheath flagella, nor are its leaves paripinnate towards the apex, which is always cirriferous. | Therefore a climbing Palm which has flageliferous leaf sheaths, flagelliferons . spadices, leave paripinnate to the apex, homogeneous seeds and tubular spathes never THE NAME OF THE GENUS. 45 can be a Zaemonorops, With this ensemble of characters we can always succeed in distinguishing a Calamus from a Daemonorops and in most cases we can do this even if the specimens are in a sterile condition, at any rate if they belong to an adult plant. XXIX.—Calamus or Palmijuncus ? Dr. Otto Kuntze in his “ Revisio Generum Plantarum,” p. 173, believes that the name Calamus Linn. must give way to that of Palmijuncus Rumph; accordingly all the species of Calamus known to the illustrious author are registered in pages 731-734 of the above quoted work under the generic name of Palmijuneus. As the adoption of the latter name involves a vexed question of nomenclature regarding which I am unwilling to express an opinion, I have left matters as they have long stood. I have only to remark that the substitution ofthe name of Palmijuncus for the name Calamus does not help to simplify the already very intricate synonymy under Calamus. I have therefore continued to use the generic name. Calamus, thinking that it will be a quite easy and at the same time a not unpleasant task for some one who may take delight in adding his own name to new species, to change into Palmijunci all those placed by me under the genus Calamus. With regard to this subject, however, I have to point out that the genus Pal- mijuncus, as understood by Dr. O. Kuntze, includes both Calamus and Daemonorops, two genera which I consider it convenient to keep separate. As Dr. O, Kuntze appears to have taken as the type of the genus Palmijuncus the first species published by Rumph in the “Herbarium” under that genus, and as the species in question is Palmi- juncus Calapparius, which is a typical Daemonorops, the generic name Palmijuneus, if resuscitated at all, ought to correspond to the name Daemonorops, and not to the name Calamus. | XXX.—Geographical Distribution. More than 200 species of true Calamus are at present known; these without a single exception are natives of the Old World, Their chief home is in the primeval hot and humid forests of tropical and subtropical Asia, and of the Asiatic Archipelagos ; & few only are African and Australian. The regions of the world where Calami are most abundant are:—the Malayan Peninsula, with 31 species; Borneo, with 30 species; Burma, with 24 species; Siam, Qochin-China and Lower China, with 18 species; New Guinea, with 18 species; the Philippines, with 17 species; Java, with 14 species; Southern India, with 12 species; Ceylon, with 11 species; Sumatra, with 10 species. After these, arranged in accordance with the number of species that they possess, come Tropical Africa, Celebes, the Moluccas, the Sikkim Himalaya, the Assam Hills, Eastern Bengal, Australia. | The five or six known Australian species are chiefly confined to the coast of Queensland and only one, C. Muellerii, extends further south, to the northern part of the ccast of New South Wales, where it has been found on the Clarence River at about Lat. 29° 30’ S. This is in fact the most southern representative of the genus and is the only one that grows outside the tropics in the Southern Hemisphere. In the Northern Hemisphere no species of Calamus extends north of Lat. 30° N. - C. tenuis, which is one of the most western of the Asiatic species is also the one that - 46 INTRODUCTOBY ESSAY. extends furthest north, its extreme limit being Kumaon in the Western Himalaya. In Eastern Asia the most northern point is attained by C. formosanus, which has been collected at Kalung in North Formosa, in Lat. 29° 30' N., but it is possible that some other species of Calamus may occur further to the north and east in the Japanese Archipelago of Liu-kiu, for I have. seen, in the Leiden Herbarium, some fruits of a. Daemonorops labelled “ Japonia, van Siebold.” We may therefore expect that one or more species of Calamus also occur there. An unnamed Calamus, apparently allied to C. Moseleyanus, has been discovered by H. N. Moseley in the Admiralty Islands ; this and C. vitiensis, which has been found in the Island of Taviuni, Long. 180° E., in the Fiji group, are the only species of Calamus- as yet known to occur to the east of New Guinea. C. vitiensis is undoubtedly the most eastern species of the genus and is the only one that occurs in the remote Pacific Islands, We know of eight species from Tropical Africa. Their area of distribution ranges from the mouths of the rivers Senegal and Gambia to the White Nile where a species, €. Schweinfurthii, has been discovered by Dr. Schweinfurth near tie equator in the Niam-Niam and Monbuttu country. The other known African species have their home on the coasts of the Gulf of Guinea north of the Equator. So far no Calami have been recorded from the very large and botanically unexplored region which lies between the mouth of the Niger and the home of ©. Schwetnfurthii, It is almost impossible to believe that there are no representatives of the genus in this wide tract, “extending as it does over 29 degrees of longitude. The African Calami do not differ strikingly from some of the Asiatic species of the flagelliferous and non-cirriferous groups, and form a distinct group of closely related species, very difficult to distinguish from each other. Summing up our knowledge of the general geographical distribution of the species. of Calamus we may say: that this genus occupies the very large tropical and sub-tropical area in the Old World which ranges in latitude from 30°N. to 30°S. and in longitude. from 17° W. to 180° E., or a good deal more than half the circumference of the globe. Generally speaking, nearly all the species of Calamus have a very limited geo- graphical distribution, Each botanical region and sub-region has numerous species. peculiar to itself and there are extremely few species that occupy a very wide area, The few that are at all widespread, such as ©. viminalis and ©. palustris, affect the frequently flooded forests of low-lying lands near the sea. C. palustris, with its numerous varieties, extends from the mouths of the Ganges to Cochin-China, and appears to be the species from which have originated others endemic in the southern islands of the Philippines and in the Moluccas. C. ornatus is also represented by distinct varieties in the maritime regions of Java, Sumatra, the Malay Peninsula, Borneo and the Philippines; this species is perhaps indebted for its wide. area to the dispersal of its fleshy fruits by birds. Of the inland species C. tenuis is the one most frequently met with; it extends right across Northern India from Kumaon eastward to Burma, and occurs also in Cochin-China. I have already remarked that each region possesses species peculiar to itself, but the main head-quarters of endemic Calami are evidently the Malayan Peninsula, GEOGRAPHICAL DISTRIBUTION. V AT Borneo, Littoral Burma, the Philippines and New Guinea; among these centres perhaps Borneo and the Malayan Peninsula hold the leading places. Java possesses various species in common with Sumatra, but some uncertainty exists as regards the exact habitat of several of the Javan and Sumatran species, owing to the fact that some confusion and admixture appear to have taken place among the specimens ‘preserved in Dutch herbaria, and gathered by early collectors in these countries. I imagine besides that many Calami still remain to be collected in Sumatra, especially ‘in the basins of the large rivers that reach its east coast. As an instance of the great localisation of Calami I may mention Borneo, where, out of the 30 species known to oecur, only five grow also in the Malayan Peninsula, in spite of the similarity of the two floras. Nor are these five absolutely identical ; they are represented by geographical varieties. Ail the species known to occur in Celebes, the Moluccas, and New Guinea are also endemic within their respective areas, As has already been said, the number of species of Calamus at present known to exist, exceeds 200. Possibly, however, this number only represents about two-thirds -of the species that actually exist, a very large tract of forest-land in the Malayan Peninsula, New Guinea, Borneo, Sumatra, the Philippines, Burma, Siam, Cochin-China, Equatorial Africa, North Australia, etc., being still botanically unexplored. The Rotangs or Palm Lianes, including in this category, besides Calamus, the other scandent Lepidocaryee, such as Daemonorops, Korthalsia, Plectocomia, &e., form one of the striking features of the virgin tropical forests of the Old World. In America climbing palms, though represented by several species of Desmoncus, are far less abundant than in Asia and Indonesia. The Rotangs are never gregarious, but always grow isolated in the forest, and none of the species are ever so abundant as to give a special character to the forest vegetation. In certain localities, however, especially in deep valleys where the soil is rich in humus at the foot of the mountains, several species of ‘Calamus may often be found growing in company within a very limited area, and J recollect having collected about 20 species of Palm Lianes ( Calamus and Daemonorops) in narrow valley at the base of Mount Mattang in Borneo. | The graceful fronds of Calami may often be seen arching downward from trees nd rocks along the banks of rivers, but generally Palms of this kind prefer the deep shade of the primitive forest of the plains and of the slopes of the mountains, some times reaching their tops when these do not exceed 1,500-1,800 metres in elevation, Rotangs are never to be niet with in the secondary forest, or in that which grows up after the primary forest has been destroyed. C. salicifolius and €. tonkinensis, two Cochin-Chinese species, are perhaps an exception to this rule, for they apparently grow in open situations and consequently have less of a forestal character than other Calami; they are not scandent but bushy, and their leaves are of firm texture and have a glaucous appearance. A few non-scandent species of Calamus form part of the undergrowth of the great forests, along with several other Palms. Among these may be enumerated €. castaneus, €. Grifühianus, C. bacularis, C. Lobbianus, C. iy C, eins C. ramosissimas and probably a few others. | sal a 48 . "INTRODUCTORY ESSAY. In Java the species that reach a considerable altitude on the mountains are C. javensis, C. heteroideus and C. adspersus; the last mentioned has been gathered by Zollinger on Mount Semiru at an elevation of about 1800 metres. In Sumatra I have obtained C. opacus at 1700 metres on Mount Singalang. In Borneo and in the Philippines a few species of Calamus have been found growing up to an elevation of 1900. metres, and in New Guinea C. Cuthbertsonii occurs at an elevation of 2500 metres. E The ‘species which exhibits what is perhaps the greatest altitudinal range is C. javensis, for it extends from the level of the sea to the tops of mountains of moderate height in Java, Borneo and the Malayan Peninsula, where however it assumes peculiar and very slender forms. I have not found Calami so frequent in New Guinea as in Borneo: still, the total number of species known from that island is considerable. It seems that a few representatives of this genus occur more or less scattered over the whole Papuan region, and as the species there appear strictly localized in small areas, it is probable that, when the flora of the island is thoroughly explored, the number present in New Guinea will be found not to be smaller than the number present in Borneo. In India several species of Calamus contribute to the luxuriant vegetation of the . forest-clad sub-Himalayan slopes of Nepal, Sikkim and Bhutan, ©. Flagellum has been collected at 1370 metres and C. acanthospathus at 1800 metres in Sikkim; this is the highest altitude that I find recorded as attained by a Calamus in the Himalaya. The same altitude of 1800 metres is reached by €. Huegelianus and 1900 metres is reached by C. Brandis in the Nilgiri mountains in Southern India. In the most completely explored forests of the districts of Perak and Malacca in the Malayan Peninsula, several species of Calamus, such as C. Diepenhorstii, €. javensis O, bubuensis, C. exilis, C. luridus, etc., have been found growing on Gunong Tamban Batak, Gunong Bubu, Mount Ophir, on the Taiping hills, etc., at elevations of from 1,000 to 1,900 metres. As regards the geographical distribution of the Calami, taking into consideration only those species that occur within the boundaries of the British Indian Empire, I have to observe that the Calami found on the southern slopes of the Himalaya following the Burmese “littoral” from Chittagong to Pegu and Tenasserim, pass int the Malayan Peninsula, which is the chief head-quarters of endemic species. In the Indian Peninsula the head-quarters of the genus lie in its most southern extremity. The Indian species so far known number 83 or about 40 per cent. of all known species; of these 71 are endemie and 12 extend’ to other countries. As already stated, the Indian region that is the richest in Calami is the Malayan Peninsula, which possesses 31 species, or 40 per cent. of those belonging to the Indian flora and 16-5 per cent. of the known species of the genus. Of the 31 only one species, C. palustris, has been found north of the Isthmus of Krau, and 23 are endemic. ° $ In Northern India the two principal head-quarters of Calami are Sikkim ana Assam. The following are the species grownig in Sikkim :—C. erectus, C, schizospathus C. Flagellum, C. ieptospadiz, C. acanthospathus, C. latifolius, €. latifolius niarmoraius, C. inermis; the following are those of Assam :—C. erectus, €. leptospadiz, C. Ardini: C. Kingianus, C. viminalis, C. tennis, C. Garuba, C. gracilis, C. khasianus, C. nambariensis, 5 GEOGRAPHICAL DISTRIBUTION. 49 . Burma is rich in Calami, especially in its littoral regions. Tenasserim and the Mergui Archipelago, though botanically very little known, have supplied very peculiar forms, many of which belong to the group with long open flat spathes, such as C. platyspathus, C. nitidus, ete. Especially characteristic of the littoral regions of Burma are (OC. hypoleucus, C. leucotes and C. myrianthus which have their leaflets coated on their lower surface with a more or less white pulverulent or chalky indumentum. Of the 9 species growing in the Andaman and Nicobar Islands, C. longisetus is also found in Burma, C. viminalis and C. palustris are widespread species, and C. unifarius Peníon; isa curious geographical form of a Javanese species. The others are endemie, but of these C. pseudo-rivalis, C. Helferianus and C. andamanicus appear somewhat alied to certain Ceylon species. i One of the principal conditions of existence for Calami being an abundant humi- dity of atmosphere, they are absent from the dry and hot regions of the Indus and Upper Gangetic Plains, and from the Central Provinces. Southern India, including Sir J. D. Hookers provinces of Malabar and the Deccan, has 12 species: of these C. Rotang is found also in Ceylon unmodified three others also occur in Ceylon, but are represented on the continent by varieties of, geographical species—C. Thwaites var. canaranus, C. pseudo-tenuis, C. Metzianus—( = 0, rivalis ? ); the remainder are endemic. - Ceylon has 1l species, of whicb 7 are endemic and 4 are represented by varieties or reappear unchanged on the Continent. The South Indian and Ceylon Calami taken as à whole are nowhere to be found north of Lat. 17? N. Of the Ceylon Calam, C. delicatulus appears allied to C. nicobaricus and C. Helferianus, C. rivalis to C. pseudo-rivalis, and C. ovoideus to C. andamanicus. The Indian non-endemic species of Calamus, not including the Malayan Peninsu- lar, are as follows:— C. viminalis, C. tenuis, C. palustris, C. unifarius. If the Malayan Peninsular species be added to the four just enumerated, we have the following non-endemic:—C. javensis, C. paspalanthus, €, Diepenhorsti C. ornatus, C. Scipionum, C. viridispinus, C. caesius, C. Ozleyanus. Of these non-endemic species of the Malayan Peninsula, we find in Java C. viminalis, C. javensis, C. ornatus ; in Sumatra C. Diepentorstii, C. ornatus, C. Scipionum, C. viridispinus ; finally in Borneo C. javensis, C. puspalanthus, C. Diepenhorstii, C. ornatus, C. Sceipionum, C. caesius. The Indian Calami that have the widest geographical distribution are C. tenuis, C. viminalis, C. palustris. i The following is a summary of the geographical distribution of the species of Calamus in India, according to the botanical provinces proposed y Sir J. D. Hooker in his “Sketch of the Flora of British India”:— Province. Specie, I. Eastern Himalaya es eis eos II. Western Himalaya e vii is 1 III. Indus Plain bes vid is à IV. (a) Upper Gangetic Plain - E as e (b) Bengal and Sundrabans T. n sig 5 V, VI. Malabar and Deccan i cis bi 12 VII. Ceylon s $i. ee ies 11 VIII. Burma A ee oes on 22 1X. Malay Peninsula - 222 ET 31 Ann. Ror, Bor. Garp. Carcurta Vor. XI. hA. FS 50 INTRODUCTORY ESSAY. Geographical distribution of the WESTERN HIMALAYA. EASTERN HIMALAYA, LOWER BENGAL. BURMA. MALAYAN 1 North Burma (Assam, e.g.). Central Burma. South Burma (Tenasserim, e.g.) Continent, Penang. . | C. erectus, v. schizos- pathus. Flagellum, leptospadix, acanthospathus, latifolius, — m mar- inermis. C. viminalis, » tenuis. » Guruba, » ustris v. gn Men » latifolius, C, erectus. » Flagellum. » leptospadix, » "floribundus." v. depauperatus, » Kingianus, » viminalis. » tenuis. , acanthospath- us. Guruba, » gracilis. » nambariensis, ^» ” » erectus v, bir- manicus, Dane. » arborescens, longisetus, viminalis, tenuis. Guruba, hypoleucus, leucotes, palustris, Doriaei. polydes mus, ” ” viminalis, concinnus, Helferian- us,* Feanus. Guruba, nitidus. platyspath- us, C. ” kid castaneus, Griffithianus, radulosus, rugosus, javensis (peninsu- laris), filipendulus, luridus, perakensis, ramosissimus. paspalanthus peninsularis, Diepenhorstii. v. exilis, C. javensis v. ” pinangi- anus, viminalis v, pinan- gianus, Diepenhor- stil. Martianus, P WARE Pr, ME T GEOGRAPHICAL DISTRIBUTION. 5l Indian species of Calamus. PENINSULA. | SOUTHERN INDIA, : ANDAMANS. NICOBARS, CEYLON., Singapore. Malabar Coast, Nilgiri. | Coromandet Coast; — | C. luridus. C. longisetus. C. dilaceratus, C. Thwaitesii. C. Thwaitesii v. | C. pseudo-tenuis, C. viminalis — v, canaranus, bengalensis, » Diepenhorstii v.|,, viminalis v. anda-| ,, pseudo-rivalis. Pachystemonus, i , , " * » Delessertianus,* i singaporensis, manicus. » Metzianus. » pseudo-tenuis, ^ » nicobaricus. | ,, digitatus, Brandisii » Ridleyanus. » Helferianus.* » PSeudo-tenuis. » Hookerianus, » andamanicus, radiatus, H 2 " ” " uegelianus. » € v. horrid- | „ andamanicus. +» travancoricus, : » Botang. : » palustris, » rivalis. Gamblei. Wem, | + Bhosle Ju ” aa s» pseudo-tenuis. " » Ppallidulus, » delicatulus. » Oxleyanus. » Rotang, » Lobbianus, » Zeylanicus. » Ovoideus. » Polystachys.* r] n i | 1 j Ayn. Roy. Bor. Garp, Oatcurra Vor. XL TRODUCTORY ESSAY, Prospectus of the species of Calamus in each of the Java. SUMATRA. BORNEO. CELEBES. MoLUccas, NEW GUINEA, C, Burckianus, C. schistoacanthus, C. ruvidus. C. symphysipus. C. Cawa. C. Papuanus. » javensis, » Opacus, » muricatas, » kandariensis, » equestris. » interruptus. ei 4 v. docilis, * » v. exilis, » Diepenhorstii. » zOnatus, » plicatus, » Rumphii. ic dix + Zebrinus, minalis. E » ciliaris. » flabellatus, » Minahassae. > albus. k », serrulatus, » Reinwardtii. ia : 3 P rhomboideus V. | » javensis v. t » ingerii. » pisicarpus. T " ^ uberrimus. : stichus. wem * qs » barbatus, >» heteroidens, » v. sublevis, » — V.&cieularis, | ,, macrosphaerion. . 5. + » v. depau- | ,, spectabilis v, su. : " - » vestitus, peratus, matranus. » filiformis, » pachystachys. SES ” » v.psllens| ,, ornatus v, su-| ,, corrugatus. T matranus, » didymocarpus, — , macroc . » horrens. » gonospermus, : d » Scipionum, » acidus, "e 3 gogolensis. -» ciliaris, TE j "m schistoacanthus,* ” » Viridispinus v, su- Sp. 9. s matranus, » heteracanthus. » rhomboideus, » nematospadix, » Manan. ide >, Cuthbertsonii, ” spectabilis, z ” . Sp. 10, » Warburgii, +» adspersus, » Myriacanthos, » ornatus javanicus, » Dygmaeus, » aruensis, » asperrimus. » bacularis, » Mollrungii, » unifarius, » paspalanthus, z — s melanoloma, » marginatus, ET Sp. 14, » hispidulus. » Macgregorii. \ » Pilosellus, P aa » sarawakensis, Sp. 18. . *& » Blumei. : , Ornatus v. mitis. » Scipionum, » mucronatus, » érioacanthus, + Optimus. ,, Caesius, ! » mattanensis, » subinermis, » brachystachys, » ferrugineus, Bp. 30. - mi [ 3€ Ww "1 2 D. m. tr. 2. p | r3 very EUN? GEOGRAPHICAL DISTRIBUTION. 53 principal partial extra Indian Floras. SIAM N. PHILIPPINES. , aprox amus £xp MINOR SUNDA ISLANDS. AUSTRALIA AFRICA. C. — siiis C. dongnaiensis, C. scabridulus, C. Muellerii, C. deerratus LI] . . . . » Meyenianus, r — » Scipionum. » Caryotoides. » Barterii, » Blancoi, etim: » Manan. , australis, » Heudelotii, 2. vininali di » Oumingianus, j pote 3 w Oxia. » Moti. » falabensis, » Ornatus v. philippinensis. ,, siamensis. Sp. 4. » radicalis, » Leprieurii, » Merrillii. » tenuis, i Bp. 5. » Perrotte » Moseleyanus, » Godefroyii. mpra » 8Spinitolius. » Faberii. » Schweinfurthii, », trispermus, » tonkinensis, ehe » manillnsis. |. , » salicifolius, microsphaerion. ” » - V. leiophyllus. ,9 . ditiis » tetradactylus, E . » Vidalianus, í— ÓÁ ». . Siphonospathus. p coma » » Ve farinosus, 1 alustris v, cochin- p " " v nbn. 1°? chinenala, | » » v. oligolepis v ollgo oris » platyacanthus, ) LJ a minor, » " v. polylepis. | ,, Henryanus, s» microcarpus, s thysanolepis, » dimorphacanthus, » Kunzeanus, » discolor, Sp. 18, \ ‘ Sp. 17, - s — M Eum E PB ——— M—— a ——— —— ae — naa " 54 INTRODUCTORY ESSAY. General distribution of the species of Calamus. E d£ HF HEMHEHE «11 c FEE g bee slat RR Ti i E HHFHIBFHE "13 v. £ £ E AaKSENE- HE Tis THEIHHHBMEHFEHHBHHHHBHHEEHHE HEIE HEFIEIEHEIHHBHEHEHEHEFIHEEE HHBBEEBEBHEHBEHHBBSHBBHBEPHHEBBSHEHEFHBE CALAMUS, » — v.Schizospathus (Gr.) Beco. vee | coe a FF Tin. | ote | cee | ose | ove | one soe | oeo f tee | oon F oo, | sen | ee | son | one | cee | cee | coe | ase » wWenejus Je a - wDepereult^teleleleleleleLlelwietwlmlelisltwle]e]e foe 2 | Flagellum Grif. es pee m | em *i*jle-l-l[-l-[-[-9l1-]-l]l9Lb-1llel-1]e9]-b-9-pb-bp-le€ Fi v. karinensis Becc. eis b Pepe IMFPGlwepepertnlelwlwuwdl€«€lelulselwelwlw E ee T a 3 | arborescens Grif. we x. de Sw eae mp1 eck) dl ed ne ee ee en ee eee oe ee a 4 | dongnaiensis Pierre pom per -— e| vee J coe | coe | ose | cos | coe | ceo | TF le€ljv€-€1lv-1lelv€VkeObe.lselelel«ebeljeLle] p |longieus Gef. 0 M cen fone fee | coe | Ee | sce | see | ore Love | 008 | cee a | eo | o> | oc E | soo | ovo | eve | coe | oe | ove | eos e pW wees Bases om es om eon Ea ata Psy | ae | woe Pe Ete Pies Pa ome Foe Tn Bk Tocca | wwe [om | ocr | oe | a ose » v.canaranus Beco, edad font he Cd on ee ee eee oe eee ee, od eee Oe Ge fee ee ee ee L^ 7 [radentam Lower, ce o m — wef cee Face | one | cee | coe | cee | ove | Mt age | woe Bonn Bias, prove | oe To, LoD sae | con foe | ove f oso ove | com estospadiz Grif... a ae nee caw fe vee vee | HL Lee | one Pace | 000 | ane | cee | woe Pane | coe 1 oe | cee | sce | coe | coe | coe | coe] coe | ove | coe 9 | dilaceratus Beco, Cs a Ged Kd we ee dd ee one bo ey ae re oe Gere pe Leeds Le 10 |osstaneus Grif, se see soe ene [RR f ooe db Rb ooo Rl eR f ooe F noa | ore | coe Fase | ome bh soei | nse L cca | cee | ove | soe | coe | coe | cee | oe n |Grifthianus Mart, se se e mm f ooe F uor f ooo | oeo | | RIS | see | ot Lee | nce Poe Lom | sce Lace | ave [ove | ove | ove | eee | one | one 13 | Burckianus Beco, ee coe aan | oe | one | ove | aoe | oom | one | on | vee | one | oo | ose | see | FE om | coe NEN NS D m 13 |deerratus Mann. & Wendi. se sos son | see | soo | vee | soo | vee | ooo | ooe f eR F ana f ene | coe lul-l-l.. Po gra pal o ra aT E a 16 |Berterli Boots 0 m [m Pee IRR RS IR a | re | a a 1 oe a | oo aTa | see | ove | oer | one | 15 | Hondelotii Beco, ws sp soe ose RR f ooe | oso | see | soe RIS DR | ane RIRs | ove | vee | om | oe | falabensis Becc, ... m m € €IpweperepepwerepererlrwerebfrtertetuaeltlilwbseluluriwloeOre- 17 |LepeuriBes. o see € € san | one J nee | one | cae | cae |e | ase | see | coe | 00 | ose | cee | cee bb eee'f a coe | veo | ean | ove | oe | oe | F 18 |Perrottetii Broo. e se o m sa | vee | wee | oe f sen | coe | ve | one | oe | ove | eo | ce | cae bese bo Wace | cee | ce | ove | ove | one | oo [oe | 19 |akimensio Booo. qe me a e mj on | vee | vse DR | vee | ee | vce | vee IRI ee | cee fase | coe Toot | see 1 ccs | eee | cee lle w | + 90 | Gohweinfurthil Bogos se see. [m | nse | coe | seg | cee | eee | veo | vo | aps | om | cee | geg Powe | oes fn, Shia aha bowl lw © M |pechysemonns The, s ai x f Hf | cee | ise | one | ee | one | cee | ose [eee DR lem | cca Pcs | one | cco | sce | cco | 000 | eee | one MB |digitatus Bec. x wee | Hf wee | one | ase | one | ase | vee | ore | cae | oot | ove | ave | RI cot | cee | cco | vee | exe ae sn d un | om Dj radiates The, p — 0 m9 |I gj Fen | cet Fite | ene | ef we | oe Pn dae Doe Ee Fs Ps ce be be | we fe | % | ruvidus Beco. (09 0 c3 |j ee ae i OP et de be i bes bm . 9$ |sesbridulus Beco, as see x sae vee | noe | oe | owe | ase | cae | con | coe | coe | ee | ore | coe Lace | oe | | ce pe eR es eee ed ee 26 | muricatus Beco. e e se con te ben Can ad ea aa a Die a a a a E N | €- |smstus Bo. — we — 0 M aja f i a a aa Ea aaa face | wee | ooo | oom | oe | we £5 [mishbe DA se ms, es | hh ee e a aa oa a Pie aa De Te bi oe Ba Pe C ae Ba a ee ae 29 | rugosus Bece, meo « .-—- ow e (ond! cond feast oo) Goods (ttl all Gnd) (oc) eed ne ee ee ee ae 30 | Mabellatus Becc, ... m we; oe ee Le Dee aan er Pee ee ee te Pd ee ke a a a a E a si javensis Bl. (insularis) ove - eos an an Whaat ends Condi Gell east Goad) Soe atl Maeda ae ee ee ee subv, exilis oe ove feed L908 | cee Ess Teen | woe Eon Poe de PO es Ba a webiste Peale " v. peninsularis Beee. "- wee | coe oe f eoe F ooo | ceo | P F see | cee f wee f oe | cee F uoo | soe Ecce | cee woe | e* | woe | cee | tee | cee | vee | one suby. purpurascens Bece. a anit Samet Mandl Koad Snot Mall Noell Good! edi est an de ee Pe ee ie piosugianus Bese... | cos | oe: | ome | sce | sce | H | coe | coe | ce | cco | vee Piece | ome | cee | a ote | sen | ase Tine Peee | oe te | ob » Vatotrastichus BE oe ose | one | ee | ore | ee | ono | ooo | oao f vee | one | oe | I ae | ove | ove | ove on fas | fe eo ae M , V. tenuisimus Beco.: e wJpv[wIrelwiwert9i-rebje[j€-1v€1-5l€9fel«talelatlstlstela e "Wl . e san] ws J tee | tee | nse | sce | ase | f cee Fe | oe | © Pere fe | cae | ue Fn foe | ce ne f see Fees | ta} oe » V polypiyihe Dia e tl ee | ee ce ed She im ol hed fil dl ool a ee ee ee en ee ee ie 4...2. Y. intermedius Bepe.. me oat B al a d Gael Ke dl ar ee sl en [eel one | oss | cos bs | lala (species VOR e - ll ed a bee | te ba a bed ee Oe be woe | oos | wee | coo | one f oos | ooe f oe | gee ps filiformis Bece. .. "m e — c€-- —— c£] [9 eae | wee eR ER f wee | RB | eI FimlimiwiwPablwmiwlaltaiate oa $$ |corrugatus Bece, se xe o» o wee | cae | n de wee f sos | soo f sos eb€r*iel€.1lelwlelwelelelele GEOGRAPHICAL DISTRIBUTION “worssy [eordoaz "spuvis] HIA wee j "t ^" Mos "e penmy "- see P, Bo mI ‘spuse nay on LLLI E ‘vamp MON IST on i sump MON uvüiop) d | oe - v =f ee fom "wouinp) AON Yopncy Ld + + "suono "e oe Jos [ov Do. LE *$9qe[oo LIII = |] v po [o - >a "spuvisy wpung 100r LI LLLI Ct Gendt Mot oe oe od oe eo | LI "Av f LLI Losd oon E *wjwung ooo ] s» ow | re "- LS f -— * d j ee *oou.togp “ *- | oe = | ww "-— dou domm onm L2) EI p on dm = f =» 'səujddjyd P: oo | rd coed cond © bene Tom | vee] mse | cot’ woe | oe | Ot Deed peed G | cee eon.) e | "* sæ j -" ad ooo | wee | fee "wusujüaq uw£w[v]c af S1BQO0IN pug SUBLIBpPUY * nae on du |] ** | ,u | * | e | "vppu] YOK “~ - elt Hoon) edt et Baal Bo) et ot oe ee ATTE ‘urpu qqnog “mop sag, LIII Lond Roxb.) Beco. LU Lud wo f so e. > bud f wee | vee w» | + | a. f v ] o ] v 4 "o. ] o£. d - ^- -] s | | — LOB EISE bad -] coe | o] t -— oo ian beet Son FOO. Een Fie A cen B peo tts A tend ee | @ Lowe | ae | ed cee itt Toot) Tiel *»^ E ] »» 1 o D o ] v. nee -——À + - - E LLLI ritil T depauperatus Bece, pallens (BL) Bece, Hilb4 -f = ~~ 45 |siamensis Bece. 55 | nematospadix Bece, 5323869t9 € 56 INTRODUCTORY | ESSAY. Andamans and Nicobars Malayan Peninsula, South India, North India, Burma, Ceylon, Siam. Minor Sunda Islands, German New Guinea, British New Guinea. | 70 71 "i a d a a SSSBese3R2e28 892955898 &5ur800522uS]i1 ae 8 BREESss & | myriacanthus Becc. pygmaeus Becc. ... ese "T | barbatus IMP. æ - o£ vestitus Becc. Wl ww ralumensis Ward. Qi ie macrochlamys Bece, ... =o gogolensis Bece. see . .. — tenuis Roxb. ade Wd. a horrens Bl. Med am. ae Godefroyi Bece, s. ee ee Rotan D uou ae Walkerii Hance .. . Faberii Bece. ve WE ae tonkinensis Becc., se — ... .« Delessertianus Bece, — .. — .« Brandisii Bece. |... — se salicifolius Becc, .. — .. v. leiophyllus Bece. » tetradactylus Hance — .. s acanthospathus Grif. .. we Feanus Bece. da cl o WE baeularis Bece. se — ae perakensis Becc, se ss ove ramosissimus Griff. aw ae psspalanthus Bece. seats eae ” Guruba Ham. Wie DM, ME niddusMar. .. 4. platyspathus Mart, st 0 myrianthus Bece. n.e se © ons hypoleucus (Kurz) Bece. a leucotes Becc. iu, a Fn travancoricus Bedd. — .. «u Rheedei Griff. se se su Huegelianus Mart. H ae Gamblei Beco. xe . wue ^" gracilis Rozb. oon .... ... melanacanthus Mart, .. .. Diepenhorstii Miq. LLLI oe CALAMUS—continued, marginatus Mart. — 4. s whem n 4 aiwa o a (ow hispidulus Beco. se — .« — pilosellus Becc, .. .. ase | sarawakensis Becc. eso CHA. rhomboideus Bl .. E eee | eee | ee oes oe m sss | 000 | ose . PI soe LI LII LJ LLLI LI LIII 7 LIII ee oe .. bà eh Se fh Ee ee eee oon s.. LL Ll oe #14 e P b.. . L vet * see on «ew doe | e^ LI m I m oo LII K s ve s.. | oo LLII " * oe LLLI eee ... * * ede] T Lid oo T T A S E. * e bua Pos T O a Back aa IET De m Ceo ceed E ee a Ss og oe WI wr tj abe Pa FO n] 9 s e Edu. ed ee cre uclcBlIa Iv pap o. met tats ` aub. Gg, E S T ig ni T A S ERE OTEN GEAN Do PET AT EG he OG a i Pu ics [ae be be bw Pvc, Pies Pon lw bulle oo oo - ... oot | v e Tus uc Doi un Tos woal af api p . Li E [zi 5 S : ic 2 e ` a S m i R= À E © eiu & E ; 4123 S|d|s8 o - .|18 H zZ Sielalica ajo] z|8 ES sls a g = I z $ 3 8 S - S K am SISiS/8/]ais C5] 2 "lal, s - > E 3 z 5 5 A à sl&lsigi!s E] & = O|R a | o|[s8|an 4|4|R oe eee | oor + eee eee LLLI D a. mI b.. oo oo LII m . + m mm eee LLI LIII + eee .* e "mE oon + oor eee eee oo oon oe + LIII eee one = m ene vee | oe * eon | oe ve oo LI - m D eee Hja oe + ^ aoe oe oo n + EI LII vee oes a- a bed T LII LI eee oo phere) T" T - sos oo - LI eee .. LLLI .. ^. vee oon oor LLLI LLLI TII eee oon oo one oo oon oo oon b. .. oe oe + .. oe .. b. vee ... T LI m eee | oe see eee DI ... oo + oe EI toe tee tes E oo tee aes oe P m toe e* om s.. oe oe oe LLLI ml El oon oor sop ee oe b. .. . n m eee m ebedebtTleebpeebe ]ce]|seebee oes ooo | eee aad LLLI ses eee eee LLLI LIII v. oe e LIII LLLI cow oor - LLLI ses woe on sse oe see ote oe + m s.. P bur J aoe wee wee m toe LII Ll — ... eee oon LIII LII LII oor oe oo re LL eee eee LLLI oo coe LLLI Lg .. ae oes oe von = ane | ose eee oes .. P mI LI m LLLI * be: LLLI LLLI ... isha oon LLLI ie ... "m EJ m ac Ll P - - kad .. oo oo E oe .. one t s.. ^" sas eee eee mI ae .. s SA e.» mI vee ee E oo oe m sss | eee one E sa om s. PM m s.. vo oo] ee .. wes | sns | oe s. \ ese .. oo eee + | eee see - Ll on m I eee s.. | o m oe oo La s ome see $e. eee oes E -- ml m oe *- e oo Lr . "T oe m p oe e] o I ae eee | oss p m .. eas prey EI see eee s oo LIII oo vee LII s. .. LLLI oon LE ee "- - one sco | t oo oo ... | 9 EI LII m oo ELI oe oo LI oe | s (onm + oe E one DE . LI LI .. m - one eee EI mn mI oo vee oe ses | ase LI .. oon te D EI oon oo oo + oe oon LLL ... LLLI oe LL so. oo LLLI oe ooo oo aoe LLLI oe oe + LLLI oo LLLI LII LIII LLLI or m cee s.. sos oon ... ... .. oo b.. LIII s.. bs. ... Lind Lll mI DII - isse | coe | e 1 ese. | ose | soe | coe | o] «| es | e | eve | eve | e | eso LLLI oo eee + LLLI PN e.. oe oon LLLI eee oo eee eee or eee oo on - on Ll ... oe ... LLLI ... ss s.. ooo | e one waa | see | 4, ooe T Lr eee | en wee | tee sas | see sea | see | ooo mI s | e] m ^ ose | e ses | eso | vee sss | s s.es | eos | cee | t GEOGRAPHICAL DISTRIBUTION. South India. North India, Burma, Andamans and Nicobars. | Malayan Peninsula, Siam, Cochin China. China. Philippines. med w e tomentosus Becc. ... Blumei Bece. spectabilis Bl. Martianus Bece. insignis Grif. ornatus Bi, (javanicus) ... CALAMUS— continued. Ve oon LI v. sumatranus Bece., » Bousigonii Pierre... — .. | heteracanthus Zipp. |... .. symphysipus Mart, Gig ie Cumingianus Bece. VAT vitiensis Warb. ie .. kandariensis Becc. R adspersus Bi, eis w wie plosns Bh, ii a 4A as Minahassae Warb, pba tau RWW Sea oW om equestris Willd. sw .. su Cathbertsonii Bece. dee. e |spathulatus Bece. ..^ .. — ee v. robustus Bece. ... » . V. horridus Bece. ies » Vv.sumatranus Becc. ... » vmi x s» . V.philippinensis Becc. | Seipionum Lowr. s» .. .. densiflorus Becc. s .. .. Ridleyanus Bece, ne .. .. zeylanicus Bece. ” polystachys Becc. OR n andamanicus Kurz eee aes Zolingerii Becc. ... .. w Merrillii Bec. — uL aquatilis Ridley... un ous Warburgii K.Schum, ... .. Moseleyanus Bece. WE i uds formosanus Bece. se — .. .. Rumphii Bl. M i A palustris Gn. at iU v. amplissimus Bece. ... v. malaccensis Bece, ... intermedius Bece. LIII kortha!siaefolius Bece. ... Minor Sunda Islands. Dutch New Guinea. German New Guinea. British New Guinea. = mM H kz] a £z mn 8 [] o at Q e un - z — Lo-4 - = ; AE E: wee m . ð LI .. . eee Em see ids eee see A oe oo oo yes 1.115 oo see soe toe | tee E m^ ^ ase ips oe s.. one m aoe .. T" sss | t ... ed .. nm oo aoe D LE "m ELE vee wee In DM + m see oes LII] * wee m E oe aoe m ves ] bey oe aoe eee s.. oo s. ilo oo s. m r ore oo eee oes és. s. | tee m T I m one wee see vo | FL see e | eee | woe | see on ^ + mn oer, wee see LIII LJ oon oe eee + -* et aoe one m sev vee | ee aoe one oe wee — E eee vee ee vw LLLI wer oe LLLI LL one dd LIII soe en, aoe eee "m ... P vee m woe LIII LIII "Tw ee oer vee oon LIII LLLI P LIII ves aoe ore oe | Australia, | ‘Tropical Africa. 58 INTRODUCTORY ESSAY. axillaris Bece. — ... |negleetus Becc, ... bubuensis Bece, viridispinus Bece. mucronatus Becc, asperrimus Bl. ... erioacanthus Becc. optimus Bece. le caesius Bl, oe simplex Bece, "ue Doriaei Bece. ——... polydesmus Bece, khasianus Bece. nambariensis. ecc. inermis T, And. Manan Miq. E giganteus Bece, | platyacanthus Warb. albus Pers. 2 | pallidulus eec, | macrosphaerion Bece, mattanensis Becc. | microsphaerion Bece. ramulosus Bece. unifarius Z, Wendl. subinermis H. Wendl. pisicarpus Bl. s aruensis Bece. ..... Hollrungii Bece. Vidalianus Bece, pachystachys Ward. | didymocarpus Warb. melanoloma Mart. siphonospathus Mart. microcarpus Bece. | dimorphacanthus Bece. conirostris Becc. Lobbianus Bece. brachystachys Bece, Henryanus Bece. | thysanolepis Hance | ferrugineus Bece. CatamMis—continued, “we Oxleyanus Teysm. 4. Binn. soe oe r.. A v. Pentong Becc. ... m v. sumatranus Bece. Š v. sublevis Bece, Ra v. farinosus Becc. 3 v. oligolepis major Bece, Ll Br v. polylepis JBece. |... 5» v. oligolepis minor Bece. ... br g > 4 LJ ela E ;|[ $48 2/8 E $|8|8 z| E BEI L- a : si S/O /16 g 3: 13 2|'8 s Ole r T S | & « o S a ms D E m a} EE: m | py a E E » Eig iis Sja =] = L o 5 3 5 Š = Z g s BI "SEE Pa g E: = -| 8 areca tes als$lz 3 s 8|2|s|R|5 "ZESHIEHHEHPHHBHEBHEIEEILPHEIE mip Du ee > T 3 "--—94- pe Selig @ | cc HHIEHIHEFKFFHELFHFHHEHEBEEBHEBEEHDBEHBHEHBHE ove eee os LIII LLLI oe + LIII eee LLLI oon LIII eee LII LLLI eee oF in one LII eee oe LIIJ aoe eee LI LLLI eee oe + oe LIII LIII oon LIII .- LIII oor .. LIII eee | LIIS LIII oer LLLI LIII ... ose ... .. ooo see . T LI oon eee roe oe .. .. LLLI .. aoe ore .. ... ... eee LI LLLI epu UTI eee een pE te a . m E . e | vee | coe | woe | see ot "| f sss | soo | vee | oos n LLLI eee oe albumen. Hasirat.—The plant originally described by Roxburgh as C, erectus was a native of Sylhet, and it seems very common on the not very distant Khasia Hills (Griffith, Hooker f. & Thomson in Herb. Kew, and G. Mann in Herb. Becc.); from theso hills I have also received good specimens collected by C. B. Clarke at Lakkat. (80 m.), at Monsto (760 m.) at Mahadeo (900 m.); in Upper Assam (0. collinus Griff. near Koreahparah, one of the Bhutan Duars; in Manipur at Kassome at an elevation of 900 m. (Watt Nos. 5122, 5135 in Herb. Kew). Kurz writes (Journ. As. Soc. Beng. xlii. pt. 2, 1874, p. 210) that this species is found in Pegu, but as this author has confused C. erectus with C. longiseius, this locality very probably belongs to the last mentioned species. A Calamus (in Herb. Kew) gathered by Hooker f. & Thomson at Seetakoond in Chittagong, near the seashore, does not seem to me to differ from the most typical specimens of C. erectus from the hills. Roxburgh says that in Sylhet “sun-gutta” is the vernacular name of this plant, and that the poorer natives use the seed as a substitute for that of Areca, j Ann. Roy. Bor. Garp. Catcurra Vor. XI. 124 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA [C. erectus. Oxservations.—The characteristics assigned by Roxburgh to his C. erectus are not many, but they are sufficient for a sure identification. I have based my description of this species mainly on the specimens collected by Mr. C. B. Clarke and by Mr. G. Mann in the Khasia Hills. Kurz (Journ. As. Soc. l. c.) refers C. longisetus Griff. to C. erectus, from which, however, it is perfectly distinct (see observations under (C. /ongisetus) The same author refers also OC, macrocarpus Griff. to C. erectus, and in this he is certainly correct. Griffith had based the description of C. macrocarpus on a fruiting specimen gathered by Major Jenkins near Koreahparah, one of the Duars of Bhutan. The figure of C. macrocarpus in Griffith’s large work (fig. I, [plate 186 A) evidently represents the summit of a spadix clawed in the attenuated portions of its main axis, but this may be also very frequently observed in the most typical specimens of C. erectus. The very peculiar lacerate marcescent spathes of C. erectus are fairly represented in Griffith’s quoted figure of C. macrocarpus. The fruits of this species in fig. I of the said plate are 4 cm. in length by 25-27 mm. in diam. and look as if they were perfectly ripe. In this stage the fruits of many species of Calamus increase considerably in volume by the pressure of the fleshy and watery tissues of the interior which stretch the scaly coating. Some of the fruits of C. erectus in the specimens gathered by Hooker f. & Thomson at Seetakoond are hardly smaller than those of C. macrocarpus as figured by Griffith, while they exactly agree in size and shape with those of C. macrocarpus as figured by Martius (l. c.) The fig. II of plate 186 A of Griffith’s work represents another fruit of C. macrocarpus which in no way differs from those of the most typical specimens of @. erectus. From all these considerations I cannot consider C. macrocarpus even as a variety of C. erectus. Griffith founded his C. collinus on a fruiting spadix of C. erectus and some portions of a leat of a Zalaeca, very probably of Z. secunda, as I have been able to ascertain from QGriffth's authentic specimen in the Herbarium at Kew, where the portion of leaf preserved seems to be the very one represented in plate 186 of Griffith’s work. Griffith had acknowledged the affinities between C, collinus and C, schizospathus {and hence with C. erectus), and to this he alludes when at the foot of the description of C. collinus he says: ‘‘ This species appears to be closely allied to the succeeding,” that is to say, to C. schizospathus, not to C. maerocarpus, as the latter has been added by the publisher of Griffith’s posthumous work, the name ‘macrocarpus’ not being mentioned in the original paper on Palms published in the Caleutta Journal; moreover, Griffith not having seen the leaves of his C. macrocorptis could not have added to the description of this species *' the terminal part of the leaf of which (viz. C. schizcspathus) differs however from this species (viz, C, collinus), etc." The fruit of the authentic specimen of C. collinus is of the same shape as tbat of the true C. erecíus, but a little smaller, being 3 cm. by 18 mm. The seed of this fruit is not quite ripe and is only a little smaller than, but is otherwise perfectly like that of C. erectus. The spadix looks more compact and shorter than is usual in the specimens of ©. erectus I have examined, and is not aculeate on its axial portions and the partial inflorescences are formed by single spikelets ; but probably what was considered by Griffith to be 'an entire spadix is only a branch or partial inflorescence; besides, the spadix in the figure of Q. D. erectus. | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 125 collinus mentioned is terminated by a short appendix clothed with diminutive spathes; bu: ths termination of the spadix of C. erectus seems very variable. Amongst the numerous and complete specimens of C. ercctus which have been tent to me by Mr. G. Mann from the Khasia Hills, there is an entire male spadix without any flagelliform appendix at its summit aud with flowers more slender than usual; on the other hand another spadix is terminated by a flagelliform appendix 35 cm. long. Some of the female spadices have many partial inflorescen- ces, of which the largest, the lowest, bears on each side 8-10 spikelets; other spadices have 4-5 partial inflorescences only, and of these the lowest are composed of 3-4 spikelets, and the uppermost of one. From the foregoing considerations, I fee] inclined to consider C. collinus also as not even a variety of C. erectus. A large specimen of C, erectus received from Dr. Treub and taken from a plant cultivated at Buitenzorg has a leaf 4 m. in length, including the petiole, which is 80 em. long, three-fourths-terete, 2 cm. in diam., narrowly channelled above, armed, in the lower portion of the rachis, with nearly complete, rather remote (5-10 cm. apart) horizontal or slightly oblique pectinate whorls of flat pale spines, 2-3 cm, long and confluent at their bases. The largest leaflets are 75 cm. long. The spadix is nearly 3 m. long inclusive of a flattened peduncular portion 1 m. in length, and a terminal rudimentary slightly aculeate flagellum 50 cm. long. A very remarkable character in €. erectus is furnished by the two large and prickly auricles formed by the division of the ocrea at the mouth of the leaf-sheaths; but as that organ is deciduous, they are wanting in old leaves. PrarE 1.—Calamus erectus Rozd. The figure on the right side above is taken from a specimen collected by G. Mann in the Khasia Hills, and represents the upper portion of a leaf-sheath with the peculiar auricles of the ocrea and the basal portion of a mele spadix; from the same place and collector are the two fruits and the seed near the left corner. The figure in the middle represents a portion of a male spadix from Lakkat (C. B. Clarke). The figure on the left side is the lower portion of a leaf from the Khasia Hills (G. Mann) The figure on the left upper corner is a spikelet with ripe fruits and seeds frem Monsto (C. B. Clarke). The fruits in the middle are also from the Khasia Hills (C. B, Clarke). The fruits in the lower right-hand corner accompanied by an entire seed, one longitudinally cut through the embryo and snother in transverse section, are from Monsto (C. B. Clarke). CALAMUS ERECTUS Roxb, var. scuizospaTHus Bece. €. schizospathus Griff. in Cale. Journ. v, 32 and Palms Brit. Ind. 41, pl. clxxxvii; Mart. Hist. Nat. Palm. iiij 332; Walp. Ann. iii, 482 and v, 829; T. Anders. in Journ. Linn. Soc. xi (1869) 71; Gamble Man. Ind. Timb. 423; Becc. in Rec. Bot. Surv. Ind. ii, 197 C. erectus Bece. (partly) in Hook. f. Fl. Brit. Ind. vi, 438. Descrirtion.— Leaflets more or less distinctly :j—5-costulate; the secondary nerves sometimes more conspicuous than in the type, and one of these on each side of the mid-costa furnished like it with a few bristles on the lower surface and sometimes also on the upper one mainly towards the summit. Male spadiz, spathes and spike- lets as in typical O. erectus, but the spathels more distinctly striately veined, 126 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA, (C. erectus. the flowers more immersed in the involucres, the calyx half-projecting from the involueres and strongly veined, the corolla twice as long as the calyx. Fruit as in the type. Hasitat.—India: Sikkim in Upper Camon and Lower Singbik and Rhu (Hooker f. & Thomson in Herb. Kew); at Currssiong (Kurseong) 1,400 m. elev. (T. Anderson in Herb. Boiss.); Sivoke on the Teesta, 1,000 m. elev. (Brandis in Herb. Bece.); hills near Sivoke, Gamble (l. c.) writes that this has a stem about 5 cem. in diam. with hard wood and closely packed fibrovsseular bundles, very close, as usual, towards the edge; the canes, however, are useless (Gamble Mss.) It is called “Rong” and ‘ Reem” by the Lepchas, but it is known also in Sikkim by the name of *Phekri Bet," but probably these names are also applied to the typical form. Oxservations.—I had considered C. schizospathus quite the sanie as €. erectus (Becc: in Hook. f. Fl. Brit. Ind. l. e.) following Hooker and Thomson who on the labels of the distributed specimens of the Herb. Ind. Orient., had already united the two species. Nevertheless T. Anderson (Journ. Linn, Soc, xi, p. 8) considers C. schizospathus ` as a species distinct from C. erectus, and writes as follows: “@. erectus of Roxburgh from Silhet and perhaps from Chittagong is a nearly allied species; but its ripe fruits are nearly half as large again as those of C. schizospathus,” which is found on the steep northern slopes of the valleys of the rivers Teesta and Rungeet, where micaceous slate abounds. A good specimen of C. erectus schizospathus from Sikkim sent to me in September 1902 by Lieut.-Col, Prain has ripe fruits 32 mm. long and 20 mm. iu diam., and consequently not smaller than those of the type with which they perfectly agree also in shape and all other characteristics, in the leaflets of this specimen I have not been able to find any difference from those of the typical C. erectus from Khasia; the sheathed stem 7 cm. in diam. and the naked canes 4'5 cm. In the male specimens. from Sikkim I had at my disposal, the spikelets are larger than in the specimens coming from Khasia and Sylhet, and have more broadly infundibuliform and more distinctly striate spathels, the flowers are more deeply enclosed in the involucres more ovoid or less elongate, and with the corolla relatively shorter and the ane more distinctly striate. I consider C. erectus schizospathus to be only a local form of C. erectus, peculiar to the hot and damp sub-Himalayan region in Sikkim and Darjeeling, while the type abounds more to the east, mainly in Assam, Sylhet, Khasia, Manipur, ete. Prate 2.—Calamus erectus Roxb, var. schizospathus Bece. Portions of a leaf from an adult plant, and partial inflorescence of the lower part of a male spadix, from. a specimen collected by Dr. Brandis at Sivoke (1000 m.) on the Teesta. CALAMUS ERECTUS Roxb. var. srrmanicus Bece. in Rec. Bot. Surv. Ind. li,. 197. / Descrivtiuy.—Female spadix more slender than in the type and produced into a. rather long (75 em.) flagelliform aculeate appendix; also the fruit smaller (20-30 mm. by 14-16 mm.). Hanrzar.— Burma: on the Karen mountains at an elevation of 1000-1200 m., collected by Sig, L. Fea in Dec. 1887. : €. Flagellum.) BECCARI MONOGRAPH OF THE GENUS CALAMUS. 127 Pirate 3.—Calamus erectus Rozb. var, birmanicus Bece, Portion of a leaf and upper part of a spadix in fruit. From Sig. Fea’s specimen. 2. OCaríAMUS FLAGELLUM Griff. in Mart. Hist. Nat. Palm. iii, 233, pl. 176, f. ix; Griff. Palms Brit. Ind. p. 48; Walp. Ann. iii 484 and v, 830; T. And, in Journ, Linn. Soc. xi (1869), 8; Gamble Man. Ind. Timb. 493; Hook. f. Fl. Brit. Ind. vi, 439; Becc.. in Rec. Bot. Surv. Ind. ii, 197. C. Jenkinsianus Griff. Palms Brit, Ind. 40, pl. clxxxvi A. fig. iii (not p. 89). C. polygamus Koxb. Fl. Ind. ii, 780? Description,—Scandent, robust and large. Sheathed stem 4-5 cm. in diam.; naked canes 2°5-3 cm. in diam. Leaf-sheaths gibbous above, flagelliferous, densely armed with scattered or sometimes confluent, spreading or somewhat deflexed, laminar, extremely acuminate spines, which are usually 3-4 cm. long (those near the mouth even 6-7 cm.) and intermingled with innumerable others of all sizes also scattered. Ocrea membranous, dry, exsuccous, exterding at the sides of the petiole into two small unarmed rounded auricles ultimately marcescent and deciduous, Leaf-sheath flagella very long (sometimes 7 m.) closely armed with half-three-fourths whorls of dark-tipped claws. Leaves very large not cirriferous, petiole very stout, 2-3 cm. thick, 30-45 cm. long, broadly channelled above, rounded beneath, where irregularly armed, mainly at the sides and more sparingly along the middle, with variable straight spines; rachis in its first portion flattish or slightly concave above with the side angles acute and spinulous and with broad side-faces where are inserted the leaflets; upwards acutely angular and with two side faces above and armed beneath up to the summit with a central series of solitary claws; leaflets numerous, equidistant or very nearly so (4-7 em. apart), alternate or sub-opposite, rather firm, green, almost shining on both surfaces, slightly paler beneath, broadly ensiform, alternate and deeply plicate at the base, gradually acuminate at the apex; the mid-costa stout, furnished on both surfaces, but mainly above, with few subspiny bristles; secondary nerves sometimes rather strong, naked on both faces; transverse veinlets approximate and distinct; margins ciliate with short spiny bristles which are rather remote lower down and approximate at the summit; the largest leaflets 60-70 em. long and 35-4 cm. broad; the upper ones shorter; the two of the terminal pair the smallest and confluent at the base. Male spadiz excessively long 4-5 m, and more), flagelliform, simply decompound or slightly and partially supradecompound, prolonged at the apex into a long, strongly clawed flagellum and armed on the back on the very long unsheathed portions between two partial inflorescences with half-whorls of very stout claws; primary spathes tubular, very closely sheathing, very long, coriaceous, longitudinally split, lacerated and fibrous at the summit; the lowest somewhat compressed and acutely two-edged, more or less armed with claws on the back and with straight auricles at the edges; upper primary spathes cylindraceous, strongly clawed on the back; partial inflorescences very few, very remote (even 1 m. apart), nodding, with 3-4 spikelets on each side; secondary spathes unarmed, tubular, narrowly infundibuliform, obliquely truncate at the mouth and extended at one side iuto a triangular, speedily withered and lacerated tip; spikelets 10-25 em. long, flexuose, slightly compressed, bearing 18-30 distichous and rather remote flowers on each side; spathels fugaciously scaly furfuraceous, broadly 128 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. re. Flagellum asymmetrically infundibuliform with truncate and discoloured subscarious margin, more or less produced at one side into a split and lacerated tip; involucre cupular, truncate, shorter than its own spathel. and laterally adnate at the base of the one bove. Male flowers elongate, acute, 8-11 mm. long, 3 mm. broad, outwardly curved, half-projecting from the spathel; calyx divided down almost to the middle into 3 lobes; corolla not quite twice as long as the calyx, almost entirely divided into $ lanceolate, acute segments; stamens with subulate filaments not inflected at the apex and united at the base with the corolla; anthers large, broadly linear, attached to the middle; rudiment of the pistil formed of 3 small acicular bodies. Female spadix very nearly the same as the male; partial inflorescences very few (5-6 at most); spikelets very remote, elongate (20-35 cm. long), subcylindrical, vermicular, curved downwards, inserted near the mouth of their respective spathes; spathels infundibuliform, rather elongate with the margin truncate and withered, produced on one side into a triangular and lacerated point; involucrophorum unilaterally cupular, truncate, nearly exserted from its own spathel and attached at the base of the one above; involucre cupular, truncate, almost entirely immersed in the involucrophorum; areola of the neuter flower distinctly lunate. Female flowers conical-ovoid acute, relatively large (about 7 mm. long); calyx ovate, very shortly 3-dentate; petals lanceolate, acute, a little longer than the calyx; stamens with triangular, subulate filaments highly connate at the base; anthers sagittate, blunt; ovary ovate, crowned by the trigonous stigmas. Neuter flowers narrower than the female ones, but nearly as long, soon deciduous. Fruiting perianth explanate (not pedicelliform); its calyx completely split; its corolla mareescent. Fruit large, when perfectly ripe about 3 cm. long by 20-22 mm. in width, broadly ovoid, rounded at the base, abruptly and shortly beaked ; scales very large, in 12 longitudinal series, channelled along the middle, of a dirty straw colour with a dark intramarginal line, lighter finely erose margins and obtuse apex. Seed regularly ovoid, rounded at the base, slightly narrowing towards the obtuse apex, circular in section, 20-22 mm. long and 13-15 mm. in diam., covered with a dry (at first fleshy) adherent coat, and minutely pitted on the surface beneath the coat; chalazal fovea very superficial and indistinct; albumen deeply ruminate; embryo 3-4 mm. long, basilar. HaBrrar —N. E. India. The type specimens, which I have seen iu the Herbarium at Kew, were gathered by Major Jenkins at Golaghat on the River Dhunseeri in Assam. It grows also on the Khasia Hills at Churra (Hooker f. & Thomson in Herb. Kew), in Cachar in the Daarbund pass and at Shapore (Keenan in Herb. Kew’; in Eastern Bengal on the Soorma Hooker f. & Thomson in Herb. Kew); in Sikkim, where according to Anderson it is by far the commonest Calamus, from the level of the plains as high as Kurseong (1370 m.); amongst the Podocarpus (Hooker f. & Thomson in Herb. Kew); Dulka-Jahr (Gambie). The canes, though very long (this Palm reaching to the top of the highest trees) are soft and useless (Anderson). It receives the vernaculur names according to Gamble) of ‘Rabi Bet” by the Nepalese, of “Reem” by the Lepchas, and of ‘‘Nagagola Bet" in Assam. The fruit is edibie (Hooker f.). OssrnvaATIONs,— Griffith has given a long and accurate description of the male plant of this species of which I have examined the original specimens in the Herbarium at Kew. Q; Flagellum] BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 129 The fruit represented in fig. III of plate CLXXXVI A in the work of Griffith with the name of C. Jenkinsianus is that of C. Flagellum, as may be proved by the fig. IX, plate 176, of Martius Hist. Nat. Palm. where the same fruit bears its true name. C. Flagellum is very closely allied to C. erectus by the structure of the flowers and seeds, but it is easily distinguished by the scandent stem, the flagelliferous leaf-sheaths and very elongate flagelliferous spadices (which are strongly armed in the unsheathed axial portions with half-whorls of stout claws) and the partial very remote inflorescences. . Furthermore, the leaf-rachis is not armed on the back with straight long spines, but is clawed throughout to the extreme apex, The seed is | very similar to that of C. erecius in size, shape and rumination. ©. Flagellum may be considered to be C. erectus transformed into a climbing plant. A very large specimen from a plant cultivated in the Botanic Garden at Buitenzorg has the sheathed stem 5 cm. in diam.; the petioles 2 cm. thick; the leaf-sheaths, petiole, rachis and spadix covered with a grey fuscescent fugacious scurf. An entire leaf is 2 m. long, the largest leaflets measure 60 cm. by 4 em, The male spadix is simply decompound or in large specimens furnished with some additional spikelets in the lower partial inflorescences. One spadix I measured was 35 m. in length and terminated by a very formidably armed flagellum which was as long; partial inflorescences very few (only 2-3), 40-70 cm. apart with few (3-6) spikelets on each side and ending in a Sspikelet longer than the side ones, which are 10-20 cm. long. Of the female plant I have received from Lieut.-Col, Prain an entire upper portion collected in Sikkim and bearing very unripe fruits, which are 3 cm. long, narrowing into a conical beak with small connivent stigmas, PLATE 4.—Calamus Flagellum Grif. Portion of the upper part of a leaf seen from the lower surface, and terminal part of a male spadix from the already- mentioned plant cultivated at Buitenzorg. | Prate 5.—Calamus Flagellum Grif. Basal portion of a leaf from the cultivated male plant mentioned above; basal portion of a female spadix in flower with an entire partial inflorescence from Assam (Herb. de Cand.); spikelet with immature fruit from Sikkim (Herb. Becc.). CALAMUS FLAGELLUM Griff. var. KARINENsIS Becc. Drscription.—Large and scandent. Sheathed stem 5 cm. in diam. Leaf-sheaths flagelliferous, slightly puckered above, densely armed with dark, very acuminate, laminar spines which are light at the base, darkening towards the apex, 4-6 cm. long, usually deflexed, solitary or confluent and forming interrupted series, and intermingled with innumerable criniform prickles disposed in crowded transversely irregular series, Ocrea in full-grown leaves inconspicuous, Leaves large, not cirriferous (only one seen and perhaps not of the upper part of an adult plant), about 2:5 m. long; petiole very stout, 2-3 cm, thick, rather short (20-25 cm. long), broadly chanelled above, rounded beneath, where, as in the first portion of the rachis, it is regularly armed at the sides and along the middle with straight laminar spines; in the middle portion the rachis is hattish or slightly concave above, with the side-angles acute and spinulose and with broad side-faces, where are inserted the leaflets; upwards towards the apex the Ann. Roy. Bor. Garp. CarncurrA Vor. XI. 130 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [0. Flagellum rachis: is trigonous with two side-faces above and armed beneath with solitary or ternate, . black, deflexed, usually straight or slightly hooked spines; leaflets numerous, equidistant, rather firm, green, almost shining on both surfaces, slightly paler beneath, broadly ensiform, alternate and deeply plicate at the base, gradually acuminate towards the apex; their midribs stout, furnished on both surfaces, but mainly above, with few subspiny black bristles; secondary nerves slender, naked on both surfaces; transverse veinlets approximate and distinct; margins ciliate with relatively strong black spinules which are more crowded towards the apex; the largest leaflets, those a little above the base, about 70 cm. long and 3-4 cm. in width, the upper shorter, the two of the terminal pair the smallest. Male spadix not seen. Female spadiz: primary and secondary spathes not seen; in one specimen a partial inflorescence is about 60 cm. long with 5-6 spikelets on each side and terminates in a slender tail-like unarmed appendix which is about 10 cm. long; secondary spathes tubular, slightly enlarged above, closely sheathing, coriaceous, 3-4 cm. long, unarmed, obliquely truncate at the mouth where it is produced at one side into a triangular point and at length decays; spikelets inserted at or a little below the mouths of their respective spathes, vermicular, subcylindrical, flexuose 10-15 cm. long; spathels infundibuliform, smooth, truncate, entire or ultimately somewhat decayed at the margin, slightly produced at one side into a short point; iuvolucrophorum unilaterally cupular, truncate, almost projecting from its own spathel and attached at the base of the one above; involucre cupular, truncate, almost entirely immersed in the involucrophorum ; areola of the neuter flower distinctly lunate. Fruit large, when per- fectly ripe 30 mm. long, 22-24 mm. broad, broadly ovoid, equally rounded at both ends, very suddenly apiculate at the apex with the vestiges of the very small and connivent stigmas; scales very large, very gibbous and deeply channelled along the middle, of a dirty yellowish or straw colour with narrow scarious and finely erose margin and obtuse apex. Seed regularly ovoid, rounded at both ends, 2 cm. long, 15 mm. in diam., circular in transverse section, enveloped in a dry (at first fleshy) adherent coat and when freed from this minutely pitted on the surface; albumen rather deeply ruminate ; embryo perfectly basilar or sometimes slightly to one side. Hasitat.—Collected by Sig. Leonardo Fea in January 1888 in the Valley of Yado in Burma at an elevation of 1200-1400 m. . Osservations.—This variety differs from the type in the leaf-sheaths covered with - distinctly seriate spines; in the inflorescences with more numerous and shorter spikelets . in the spathes and spathels less lacerated at the margin and in the fruit scales not bordered with a dark line and more deeply furrowed along the middle. The leaf- rachis armed on the back not with a series of solitary claws, but with deflexed, black, flat, often binate or ternate spines, is a peculiarity probably depending on the leaf not being collected from the upper part of ‘the adult plant. If this were a constant character this variety ought to be raised to specific rank, : PrarE 6.—Calamus Fagellum Griff, var, karinensis Bece, Portion of a sheathed stem; portion of a leaf (under surface) near the apex; partial inflorescence with mature fruit; seed, entire, longitudinally eut through the embryo and iu transverse section. From Sig. Fea’s specimen in Herb. Beco, © © — C. arborescens ) BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 131 3. CALAMUS ARBORESCENS Griff. in Cale. Journ. v, 33, and Palms Brit. Ind. 42, t. clxxxvii A. B. C.; Mart. Hist. Nat, Palm. iii, 332; Walp. Ann, iii, 483, and v, 829; Miq. Fl. Ind. Bat, iii, 113; Teysm. Cat. Hort. Bot. Bog. 71; Kurz in Journ. As. Soc. Beng. xliii, pt. 2, 208, t. xxii, and For. Fl. Brit. Burma ii, 516, and Rep. Veg. Pegu (1875), 90; Gamble, Man. Ind. Timb. 423; Hook. f. Fl. Brit, Ind. vi, : - 489, Becc. in Rec. Bot. Surv. of Ind. ii, 198. | C. hostilis Hort. Cale. DrscRIPTION.— Not scandent, czspitese. Stem erect, 4-6 m. high, 6:5 cm. in diam. (Griffith); with annular green internodes about 18 cm. long (Griffith). Leaf-sheaths short, not tubular, open on the ventral side, not flagelliferous. Leaves very large (as long as 5 m.) not cirriferous, forming a crown at the apex of the trunk; petiole very long ('6-1'5 m.) and very stout, deeply channelled above, rounded beneath and powerfully armed with very large, thinly laminar, elastic, black, polished and very acuminate spines, of which the largest are 7-8 cm. long and 6-8 mm. in width at the undilated base, intermingled with smaller ones, which are disposed in oblique and often nearly complete series ; rachis covered when young with whitish scurf, then glabrous, armed beneath with laminar, confluent, long straight, deflexed spines, which are in every respect similar to those of the petiole and are verticillate and pectinate in the lower portion, becoming upwards ternate or binate and at last solitary towards the apex but always straight and very long; above, the rachis is trigonous with two-side faces above from the middle upwards and smooth; leaflets numerous, equidistant, alternate or subopposite, rather distant, very nar- rowly lanceolate or broadly ensiform, very long (up to 1 m.) and 3-5°5 em. in width, narrowed and deeply plicate at the base, gradually acuminate into an acute apex which is bristly-spinulous at the sides, conspicuously two-coloured, green and shining on the upper surface, and white and dull beneath, with the mid-rib stout and bristly mainly towards the apex on both faces; lateral nerves rather slender, one on each side of the mid-rib sometimes sparingly spinulous beneath; transverse veinlets sharp and crowded above, indistinct beneath; margins very bristly and pungent. Male spadiz very long (1'5 m., Griffith), pendulous, not flagelliferous at its apex, supradecompound, with many approximate partial inflorescences; primary spathes not very long (20-30 em.) tubular, rather closely sheathing, thinly coriaceous, lacsrated and fibrous in their upper part; armed with very slender, black» needle-like, scattered or subseriate spines; the lowest spathe somewhat compressed, the upper cylindraceous, slightly enlarged above; partial inflorescences very long (60-90 cm.), the lower decompound, the upper simple, with rather numerous spikelets on each side; secondary spathes tubular at the base with a clavate or subinflated split and lacerated limb, which is often blackened, withered and usually armed with scattered, spreading, subulate, black spines; spikelets not pedicellate, inserted near the mouth of their respective spathes, large, 10-15 cm. long, flattened, with 20-30 perfectly bifarious and regularly pectinate flowers; spathels very closely imbricated, short, broadly infundibuliform, horizontally truncate and produced at one side into a very short triangular point, which is usually split; involucre hidden in its own spathe and inserted at the base of the one above, dimidiately cupular or like a swallow’s nest, obliquely Any. Roy. Bor. Garp. CarcurrA Vor. Xl. - 132 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [ €. arborescens truncate, flat, two-keeled, emarginate and bidentate on the side next to the axis. Male flowers inserted at an angle of 45°, elongate, large, 10 mm. long and 2:5-3 mm. thick, often slightly outwardly curved; calyx ovate, divided down to about the middle into 3 semi-ovate acute lobes, not striate; corolla two and a half to three times as long as the calyx, divided into 3 broadly linear, or narrowly lanceolate acute segments, entire where it is enclosed in the calyx; stamens united to the corolla in its undivided basal portion; filaments linear, subulate, inflected at the apex in the bud ; anthers broadly linear attached by the middle, versatile, their cells parallel, shortly discrete at the base; rudimentary pistil long, consisting of three angular, elongated, acute bodies united to the middle. Female spadix with partial inflorescences 40 cm. long (at least the one seen, which perhaps is not entire), with 5 spikelets on each side; secondary spathes as in the male spadix, unarmed, tubular, enlarged and somewhat inflated above, withered and lacerated upwards and transversely zoned ; spikelets flexuose, spreading or recurved, 15-18. cm. long with 18-20 distichous flowers on each side; spathels infundibuliform, truncate, produced on one side into a short ultimately decayed point; involucrophorum obliquely cupular, nearly entirely ‘included in its own spathel at the base of the one above, flat, 2-keeled and 2.toothed on the side next to the axis; involucre hardly longer than the involucro- phorum, cupular, truncate, entire or superficially 3-toothed; areola of the neuter flower depressed-lunate, sharply defined. Female flowers about 7 mm. long, Fruiting perianth explanate, the calyx split into 3 broadly ovate acute parts, not or hardly callous at the base; the corolla with the segments narrower but a little longer than the lobes of the calyx; the stamens with filaments united as usual by their bases, triangular in the free portion and a little shorter than the calyx. Fruit 20-22 mm. long, 14-15 mm. broad, obovoid oblong, suddenly and stoutly beaked; scales in 12 longitudinal series, rather broader than long, deeply channelled along the middle, dirty yellowish or reddish-brown, with a very narrow, darker intramarginal line and finely ciliately fringed margins, especially near the rather obtuse tip. Seed, when freed from the integument, 12 mm. long, 6 mm. thick, with a very uneven, almost facetted surface on the raphal side and with a deep central chalazal fovea; albumen bony, equable; embryo exactly basilar. Hasirat.—Burma: in marshy places in Pegu. At Bassein, Myaungmya Division at Kyetsha, Walkema Subdivision and at Rangoon (J. H. Burkill), Kurz writes that it is “frequent in marshy beds of choungs, in the moister and evergreen tropical forests of Pegu, on the sandstone,” and that it is called ** Thanoung” by the Burmans. Gamble (J. c.) gives the Burmese names of “ Danoung" and ‘‘Kyenbankyen” and Burkill those of ‘‘ Damon” and “ Danoung Thain.” OssERvATIONS.— Griffith who had described this species diffusedly from male plants cultivated in the Botanic Garden at Calcutta, says that it is “a very elegant palm, in . Some cases stoloniferous, forming at the base, apparently from offsets, very thick leafy tufts, from which arise elegant stems fifteen or twenty feet high, two and a half inches in diameter” and that it is ‘a very handsome and well-marked species distinguished by its erect stems, dark brown almost black spines and the leaves which are white underneath,” and I may add by the want of claws or short hooked Spines on the “N dongnaiensis | | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 133 spadix and on the leaf-rachis, where only black straight flat or needle-like Spines are to be found. Very few Calami have the leaflets so decidely white underneath as this. A very large and complete specimen from a plant cultivated at Buitenzorg and sent to me by Dr. Treub has a leaf 5 m. long, including the petiole, which is L5 m. in length and is as thick as a man’s wrist; the largest leaflets are more - than 1 m. long; the male spadix measures 5 m. and terminates in a rather long tail-like, aculeate (not clawed) appendix. The partial inflorescences are very large and the lower ones decompound. PLATE 7.—Calamus arborescens Grif. Portion of a leaf and apex of a male partial inflorescence. From the above mentioned plant cultivated at Buitenzorg. 4, CALAMUS DONGNAIENSIS Pierre Mss. ex Becc. in Rec. Bot. Surv. Ind. li, 198, Description.—Tufted and nearly stemless, 1-3 m. high (Pierre) Leaves very large, not cirriferous; petiole (the one seen, probably from a radical leaf) nearly terete or very obscurely trigonous, sparsely armed with straight, rather short and slightly deflexed spines, which are seated on a broad base; rachis (of the upper portion of the leaf) trigonous, bifaced above, flattish beneath, where marked with deep impressions left by the pressure of the spines during prefoliation, and where it is rather densely armed with long (4-5 em.), , flat, elastic, black-tipped spines, which are paler and yellowish at the base; leaflets inequidistant, ensiform, very gradually acuminate, green and sub- shining on both surfaces with their mid-rib furnished above near the apex with few distant spinules and beneath, especially from the middle upwards, with some stiff spadi- ceous bristles (10-15 mm, long); secondary nerves slender, only one on each side of the mid-rib furnished with a few spinules on the upper surface and occasionally also on the lower one; transverse veinlets very many, rather distinct and interrupted ; the largest leaflets seen are 50-60 cm. long, 2:5-3 em. broad; the upper shorter; the two of the terminal pair the smallest and united at the base. Male spadiz simply decompound, very long, the one seen 2:5 m. long, thinly covered with a very easily removable dark scurf and prolonged into a long, caudate, sterile, sheathed not clawed appendix, but armed with needle-like, black, straight and slender spines or totally unarmed ; the peduncular portion of the spadix is rather long, flattened and unarmed throughout ; lowest primary spathe elongate, flattened and two-edged, unarmed; the upper spathes tubular, cylindraceous at the base, enlarged above into a somewhat inflated, lanceolate, much lacerated or longitudinally split and withered limb, which is of a greenish-straw colour and is covered with, a thin, deciduous, brown scurf and is more or less distinctly marked with few transverse zones or slightly raised ridges; the attenuated and lower portion of the spathes is armed with straight, flat, subulate, unequal spines, of which some are 2 cm. long and arise solitary or fascicled from a pale tubercle; partial inflorescences 5-6, the lowest, the largest, about 40 cm. long, with few (4) remote spikelets on each side; secondary spathes elongate-infundi- . buliform with an inflated sub-auricled limb, which is speedily withered and lacerated, zoned as in the primary ones; spikelets not pedicellate, inserted near the mouth of their respective spathes, large, 10-15 cm. long, with up to 20 flowers on each side, flexuose, flattened, scaly-furfuraceous; spathels short, broadly and asymmetri- cally infundibuliform, truncate, entire, not ciliate at the margin, extended on one 134 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (ÇC. /ongisetus side into a very short point; involucre half-exserted from its own spathel and laterally attached to the base of the one above, two-keeled and deeply emarginate on the side next to the axis, otherwise entire and obliquely truncate. Male flowers perfectly bifarious, very regularly alternate and 5-6 mm, apart, obtusely trigonous, 8-10 mm. long, 3 mm. thick, occasionally slightly curved, narrowly oblong, gradually attenuated from the middle upwards into an almost acute point; calyx striately nervose, subcampanulate, divided down about to the middle into 3 large triangular almost acute lobes; corolla twice as long as the calyx or nearly so, divided down nearly to the base into 3 elliptic-lanceolate segments; stamens with filaments complunate in their lower part, subulate upward, not inflected at the apex; anthers narrowly sub- sagittate, with cells deeply discrete at the base; rudimentary ovary columnar, a little shorter than the filaments, divided into three subulate bodies. Other parts unknown. Hasrrat.—Discovered by M. L, Pierre in March 1877 in the northern part of the River Dongnai at Chiao-xhan in Lower Cochin-China (Herb. Pierre No. 4829). The fruit is eaten by the natives (Pierre). OnsERvATIONS.—I have seen only a male spadix and some portions of leaves of this very fine species which is closely related to C. arborescens Griff., from which it differs in the leaflets being of the same green eolour on both surfaces and not white under- neath, and in the limb of the secondary spathes being greenish and when withered and decayed not nearly black. The spikelets and the flowers are very much the same as those of C. erectus, but from this C. dongnaiensis differs in its inflated spathes, C. longisetus is also another allied species, but this is climbing and is powerfully clawed in the attenuated axial portions of the spadix, while C. dongnaiensis bears only straight spines throughout, Prate 8.—Calamus dongnaiensis Pierre. Apex of a leaf and leaflets of an intermediate portion of leaf; basal portion of a male spadix with an entire partial inflorescence and the terminal portion of the same spadix. From the authentic specimen of Pierre in Herb. Becc. 5. Carawus LoNGISETUS Griff. in Cale. Journ. Nat. Hist. v, 36, and Palms Brit. Ind. 44, t. clxxxix. A.B.; Mart. Hist. Nat. Palm. iii, 333; Walp. Ann. iii, 483, and v, 830; Miq. Fl. Ind. Bat. ii, 114; Hook. f. F] Brit. Ind. vi, 440; Becc. in Rec. Bot. Surv. Ind. ii, 199. - C. tigrinus Kurz in Journ. As. Soc. Bengal, xliii, pt. 2 (1874), 211, t. xxv and xxvi and For. Fl. Brit. Burma ii, 519. Kc à DescripTion.—Large and scandent; unsheathed stem 9:5—3 em. in diam. Leaf-sheaths (probably flagelliferous) fearfully armed with whorls and half-whorls of broad, flat, sharp glossy, fuseous or black spines (about 3 cm. long), intermingled with shorter d thinner ones (Kurz). Leaves very large (3-4 m., Griffith) not cirriferous; petiole long 2-2°5 cm. thick, deeply channelled above, round and armed beneath with ihe similar to those of the sheaths and of various sizes (some of them 4 em. long) seat tered or variously aggregated in more or less complete comb-like whorls; rachis of the C. longisetus | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 135 middle and anterior portion subtrigonous, bifaced and not very acute above, rounded and armed along the middle below with solitary, strong, short, black, deflexed spines; leaflets inequidistant or approximate in not distant groups of 2-3 in the lower portion of the rachis, subequidistant towards the summit, green and shining on both surfaces, faintly paler beneath, ensiform, subulately acuminate; mid-costa acute and remotely spinulous towards the apex above, furnished below with some very long (sometimes even 3 cm.) blackish bristles; secondary nerves not strong but distinct, of these one on each side of the mid-costa occasionally bristly above, all naked beneath; margins ciliate, mainly near the apex, with short, black, somewhat approximate, spiny bristles; transverse veinlets fine, very distinct and much interrupted; the largest leaflets 60-65. cm. long, 3-4 em. broad, the two of the apical pair shorter than the others and con- fluent at the base. Male spadix very long, flagelliferous, simply decompound, with few, elongate, (40-60 cm. long), partial inflorescences ending in a short, flattened, unarmed, caudiform appendix and bearing 5-6 distichous, remote, erect spikelets on each side; the elongated part of the axis between two inflorescences and the apical, very long and stout flagellum are armed with rather approximate half-whorls of black-tipped claws with swollen and confluent bases; primary spathes elongate-tubular, rather loosely sheathing, with lacerated decayed limb, armed with short strong claws, often confluent, mainly near the base; secondary spathes unarmed, shortly tubular at the base, and with a somewhat inflated lanceolate limb, at first truncate acuminate at one side, but later decayed and lacerated in the upper portion, the basal’ still living portion being sharply defined from the dead one by a distinct dark transverse line; spikelets inserted inside their own spathe but not pedicellate, very large, flattened, 10-12 cm. long, and 2 cm. broad, when covered with fully developed flowers, which number 20-22 on each side and are very regularly distichously and closely set; spathels fugaciously fur- furaceous, closely packed, short, asymmetrically and broadly infundibulitorm, truncate, apiculate at one side ard split under the flower; involucre nearly entirely enclosed in its own spathel, and laterally attached to the base of the one above, dimidiately cupular or like a swallow's nest, obliquely truncate and entire on the front side, deeply emarginate and two-toothed posticously next to the axis. Male flowers amongst the largest in the genus, 8-11 mm. long, 3-4 mm. thick, oblong, obtusely trigonous, some- times slightly curved, somewhat attenuated at the apex; calyx thinly pergamentaceous, finely striately nerved, with 3 short, broad, triangular, very acute or apiculate lobes; corolla more than twice or nearly three times as long as the calyx, shortly tubular at the base, with oblong or elliptic segments; filaments of the stamens stout, not much shorter than the segments, inflected at the apex, united at the base with the undivided part of the corolla; anthers large, broadly linear, acute, the cells shortly discrete and obtuse at the base, rudimentary ovary small, formed by three very minute subulate bodies. Female spadiz simply decompound; primary and secondary spathes as in the male spadix; spikelets flexuose, 10-16 cm. long, ‘thicker and larger than the male ones; spathels ultimately lacerated and decayed at the apex; involuerophorum unilat- erally infundibuliform, not exceeding its own spathes and attached to the base of the one above, deeply emarginate, two-toothed and acutely two-keeled posticously next to the axis; involucre cupular, nearly entirely exserted from the involucrophorum, truncate, emarginate and toothed on the side of the neuter flower, of which the areola is rather deep, lunate, but often somewhat vertieally evolute, and sharply defined, Female flowers large, about 1 em. long. Fruiting perianth explanate under the 136 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. longisetus fruit and not pedicelliform, its calyx 3-toothed; its corolla with the petals lanceolate, as long as the calyx; its stamens with filaments broadly triangular, suddenly subulate and not very highly connate at the base. Fruit elliptic-ovate, when fully developed 30-33 mm. long and 15-20 mm, in diam., tapering at both ends, but more towards the apex, which is regularly conical and acuminate; scales in 12 longitudinal series, slightly convex, not channelled along the middle, longer than broad, yellowish in the unexposed part, marked across the centre with a conspicuous, lunate, blackish, shining band (giving the fruit the appearance of being mottled like a tiger-skin), and prolonged into a triangular, pale, scarious apex, which is conspicuously fringed at the margin. Seed oblong, 18 mm. long, 11 mm. broad and a little less in thickness, flat at the base, longitudinally 5-7.costate and superficially channelled on the back; smooth on the raphal side, where marked with a narrow circular chalazal fovea penetrating nearly to the centre; albumen equable; embryo in the middle of the base. Hasrrar,—Pegu (Grifith), Rangoon (McCicliand in Herb. Kew, Burkill); Hill jungle at Port Mouat ravine, Kada Kachan, and Ali Masjid in South Andaman (Sir G. King’s collector in Herb. Cale.) Vern. name “Leme” (Burkill). Kurz writes that this is common in the evergreen tropieal forests from the eastern slopes of the Pegu Yomah and Martaban down to Tenasserim and the Andamans, and that it’ is called “Lémé” in Burma and “ Umdah” by the Andamanese. From the Audamans I have received some very good specimens through Mr. E. H. Man, who says that the natives eat the fruit cooked, and employ the leaflets for coverings, and that they give it the name of “ Am,” The samples of the canes, which also seem to be employed by the Andamanese, - have a polished yellow surface and are from 2:5-3 em. in diam. with joints 25-30 cm. long. One sterile specimen from the N icobars, which I have seen in the Caleutta Herbarium probably belongs to this Species. OBSERVATIONS.— Griffith, who has drawn up the description of this Calamus from a male plant native of the forests of Pegu and introduced into the Botanic Garden at Calcutta, says that it is a “ tufted stemless species, with the habit of young Specimens of C. arborescens,” but he adds that it has “the flagellus humifuse or trailing over the neighbouring shrubs and armed with the usual prickles.’ From this it might be argued that this palm begins to flower when still in a young and stemless state and becomes a climber as it grows older. Kurz has described his C. tigrinus ( which without hesitation I consider the same as C. longisetus) on fruiting spadices only, and says that it is a large scandent rattan. It is a very distinct species by its peculiarly spotted fruits, very large male spikelets and flowers very much like those of OC. erectus. The male spadix simply decompound is not of very common occurrence amongst the Calami. PrarE 9.—Calamus longisetus Grif. Male partial inflorescence from a specimen in the Cale. Herb. collected at Port Mouat ravine in the South Andamans; partial inflorescence with young fruit and lower portion of a leaf from the Andamanese specimens forwarded to me by Mr. Man; the portion of a spikelet with ripe fruit, €. Thwaitesii | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 137 the entire seed laterally seen and in vertical section, from Kurz’s authentic specimen of his C. tigrinus. 6, Cavamus Tuwarrestr Bece. in Hook. f. Fl. Brit. Ind, vi, 441, and in Rec. Bot. Surv. Ind. ii, 199. C. longisetus Thw. Enum, Plant, Zeyl, 330 (not of Griffith). Description.—Large and probably high scandent. Leaf-sheaths not seen, very probably sometimes flagelliferous, Leaves large, not cirriferous; petiole armed with straight, black, laminar, seriate spines which are very much the same as in C, longisetus ; rachis in its first portion channelled above and with spinulous side-angles, from the middle upwards trigonous, bifaced above, flat beneath, where armed throughout ‘with solitary rather distant claws, at least in leaves of the upper part of the plant; leaflets large, irregularly fascicled in groups of 2-3 on each side, the groups nearly opposite, with rather long vacant spaces between each pair of groups, upwards, mainly near the summit, more regularly set and nearly equidistant, rather firmly papyraceous, longitudinally plicate, shining above, green on both surfaces, very narrowly lanceolate or ensiform, gradually narrowed to the base, the margins ciliate throughout, almost from the base, with approximate, short, black, spreading, subspiny bristles; the mid-rib acute and prominent above, weaker below, furnished on both surfaces, but chiefly on the lower one, with black, short, subspiny bristles ; secondary nerves (coinciding with the plice) slender, 1-2 on each side of the midrib, naked on both. surfaces; transverse veinlets minute and crowded; the largest leaflets seen by me ( probably the mesial) 60-70 cm. long and 4-4°5 em. broad, the upper gradually shorter and narrower, the two of the apical pair the smallest and confluent at the base. Male and female spadices very similar and simply decompound, elongate, flagelliform, with the partial inflorescences very remote ( 60-80 cm. apart ) and the axial portion interposed amongst them flat or rather concave on the inner side and convex on the back, where it is strongly armed with rather approximate half-whorls of stout black-tipped claws which gradually become smaller and more scattered on the spathes; primary spathes very long, tubular, closely sheathing, thinly coriaceous, armed with short-hooked black-tipped prickles, withered and lacerated near the mouth; partial inflorescences up to 60 cm. long, bearing 4-6 (and perhaps sometimes more) spikelets on each side and terminating in a rather long and slender appendix; secondary spathes tubular-infundibuliform, about 4 em, long, somewhat enlarged above, obliquely truncate at the mouth and acuminate at one side where ultimately decayed and lacerated, usually sprinkled upwards with very short tubercular spines, but occasionally smooth; male spikelets inserted inside near the mouth of their respective spathes at an angle of about 45°, large, flattened, 10-15 em. long, with 20-30 distichous flowers on each side; spathels broadly and asymmetrically infundibuliform, truncate, split under the flower and not prolonged into a point; involucre almost entirely included in its own spathel and attached at the base of the one above, cupular, obliquely truncate, two-keeled and lunately emarginate on the side next to the axis. Male flowers large (not seen fully developed), ovate when in bud, with the calyx three-toothed, very finely striately veined as are the spathels, Female spikelets 10-30 cm. long; spathels infundibuliform, truncate, prolonged at one side into a triangular, ultimately marcescent point ; Axx. Roy. Bor. Garp. Carcurra Vor. XI. 138 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. Thwaitesii involucrophorum short, cupular, truncate, exserted from its own spathel and attached at the base of the one above, acutely two-keeled on the side next to the axis; involucre regularly cupular, rather deep, scarcely longer than the involucrophorum, truncate, entire or faintly emarginate on the side of the neuter flower, of which the areola is distinctly lunate, rather concave and sharply bordered. Female flowers ovate and 7-8 mm, long; the calyx with 3 ovate acute teeth; the corolla divided almost to the base into three lanceolate acute segments slightly longer than the calyx; the stamens with filaments united by their bases and forming a [short ring, triangular and subulate in the free portion. Fruiting jperianth not pedicelliform; its calyx somewhat hardened and callous at the base, but almost entirely split into three pieces, Fruit ellipsoid-obovoid, subturbinate and slightly tapering towards the base, suddenly contracted at the apex into a stout and rather long conical beak, 22-25 mm, long, 12-15 mm. in diameter; scales in 12 longitudinal series, the largest 7 mm. long and not quite so much in breadth, convex, broadly channelled along the middle, yellowish-brown or yellowish-reddish with a dark marginal line and very finely erosely ciliated, not fringed, margins, Seed (when freed from the dry, certainly once fleshy, coat) ovoid, very slightly compressed, 15 mm. long, 10 mm. wide, 7 mm. thick, with obtuse apex and flat base, uneven or indistinetly pitted on the back convex on the raphal side and with a very deep narrow cireular chalazal fovea above the centre; albumen equable ; embryo perfectly basal. Hasrrar.— Ceylon: in the Hantani district of the Central Provinces at an eleva. tion of 2,000 to 3,000 feet (Thwaites C. P. No, 287 3). OssERvATIONS.— This is quite distinct from C. longisetus by the different shape of the fruit scales. It seems more nearly related to ©. arborescens than to C. longisetus but this is an erect species with leaflets white beneath, and with the axial portions of the spadix not clawed. From C., erectus and allied species it is separated by its seed with equable albumen. The spadix of the Ceylon plant (No, 2873 in the St. Petersburg Herbarium) . is accompanied with a portion of a leaf which seems a radical one (or that of a young plant) with the upper part of the petiole; this is terete and is armed, as in the first portion of the rachis beneath, with three-fourths whorls of numerous slender long straight spines; the petiole and rachis are covered with removable grey scurf. Prate 10.—Calamus Thwaitesii Bece. Lower portion of a male partial inflor- escence, with flowers not fully developed and lower portion of a (radical ?) leaf, from C. P. No. 2873 in St. Petersburg Herbarium. Pirate 11.—Calamus Thwaitesii Bece, A partial inflorescence with not quite ripe fruit and a portion of a leaf from near its base (seen from the upper surface), from C, P. No. 2873 in Herb. de Cand. a spikelet with full-grown fruit ; detached fruits; one seed entire and another in longitudinal section from C. P, No. 2873 in St. Petersburg Herbarium. | CaLaMUs T'uwarresu Becc. var. CANARANUS Becc, Description.—Leaves as in type, but slightly paler beneath than above; in the small portion seen by me the leaf-rachis is armed beneath with small Straight ternate C. rudentum | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 139 deflexed black spines. Male spadiz elongate flagelliform; its lower partial inflor- escences elongate (in one specimen 70 cm, long with 11-12 spikelets on each side) and terminating in a small, unarmed, inconspicuous, tail-like appendix; secondary spathes tubular, narrow and closely sheathing at the base, somewhat enlarged above where usually split on the ventral side, and embracing with their expanded, exsuccous, triangular apices the base of their respective spikelets ; spikelets large, the lower ones 10-12 cm, long, 1 cm. broad, with very numerous and very approximate flatly bifarious flowers (as many as 40 on each side); upper primary inflorescences with fewer and sometimes much shorter spikelets, and these often with a scorpioid tendency ; spathels very shortly and broadly asymmetrically infundibuliform. Male flowers (full. grown) narrowly oblong, 8-10 mm. long, 3 mm. broad, obsoletely trigonous, conically narrowed to the summit and sometimes slightly curved; the calyx tubular-campanulate shortly and broadly 3-toothed; the corolla (during anthesis) almost three times as long as the calyx. Female spadiz as in the Ceylon plant; spathes tubular-infundibuliform, truncate and entire at the mouth, very slightly prolonged at one side into a broadly triangular point. Fruit more or less obovoid, turbinate, very suddenly beaked, 20-25 mm. long (including the beak and the perianth), 15 mm. thick. Seed suborbicular, 19 mm. long, 11 mm. broad, somewhat flattened or lenticular, 6 mm. thick with a rather acute margin all round. Hapirat.—Southern India; Canara district, at Marmagoa, near seashore, (Talbot No. 28544); Ainshi Ghat, (Talbot No. 2855), and Keldra, (Talbot No. 2856, in Herb, Kew). OxsERVATIONS.—The male spadix of this variety at first sight looks very different from that of the Ceylon plant, on account of its spikelets which are shorter and at the same time with more numerous and more approximate flowers. The fruit how- ever is externally exactly similar to that of the Ceylon plant, as already stated by Sir J. Hooker (Fl. Brit. Ind. vi, 441), but the seed is more flattened, almost lenticular and with a rather sharp edge all round. The portions of leaves which accompany Talbot’s specimens of the spadix have grouped leaflets and the rachis armed beneath not with claws but with small, slender, ternate, deflexed, straight, black spines, probably because the said portions belong to radical leaves or to those of the lower part of the plant. The specimen of a male spadix with curved male flowers, 4 inch long, that Sir J. Hooker mentions under C. pseudo-tenuis (l. C., 445), and that I have seen, belongs to the var. canaranus of C. Thwaitesii. PLATE 12.—Calamus Thwaitesii Becc. var. canaranus Bece. Partial male inflor- escence and portion of a leaf (under surface) from Marmagoa (Talbot No. 2854 in Herb. Kew); spikelet with mature fruit and seed in lonigtudinal section through the embryo, from Ainshi Ghat, N. Kanara, (Talbot No. 2855 in Herb. Kew). 7. CALAMUS RUDENTUM Lour. Fl. Cochinchin. Ist edit. i, 209, and 2nd edit., 260 (excl. Rumph. syn.); Willd. Spec. ii, 203; Lam. Encycl vii, 304; Rees Cycl, No. 2; Roem, & Schult. Syst. vii, 1327 ; Martius Hist. Nat. Palm, iii, 1st edit, 211 (excl. all syn.) and 340; Walp. Ann. ii, 49, and v, 831; Miq. Fl. Ind. Bat. iii, 139; H. Wendl. in Kerch. Palm. 237 Becc. in Rec. Bot. Surv. Ind. ii, 199. Ayn. Roy. Bor. Garp. Carcurra Vor. XI, 140 ! ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (C. rudentum Description.—Scandent. Stem very long and robust. Leaf-sheaíhs not seen, but very probably flagelliferous. Leaves very large, non-cirriferous; petiole . . . . .; rachis trigonous in the mesial and upper portion, with acute and smooth angles and two flat side-faces above, flat and armed beneath with solitary, rather strong, straight, slightly defiexed spines, which rest on a swollen base; leaflets numerous, equidistant, alternate, very long, linear-ensiform, the largest 65 cm. in length by 15-18 mm. in breadth, stifily papyraceous, plicate along the secondary nerves, shining and green on both surfaces, narrowed to the base, gradually acuminate towards the apex which is acute in the lower leaflets and obtuse acutely keeled above and bristly penicillate at the apex in the upper ones; the mid-rib acute and spinulous towards the apex. above, and from the middle upwards furnished below with some very long (1-2 cm.) spadiceous bristles; secondary nerves few, slender, naked on both surfaces ; margins acute, bearing short adpressed spinules 10-12 mm. apart, closer, longer and. more spreading upwards; transverse veinlets very crowded, rather sharp and frequently anastomosing; the two apical leaflets smaller than the others and shortly connate at the base. Male spadiz ultradecompound, flagelliform, excessive- ly long; the one measured, though without its basal portion, 3°5 m. in length and further lengthened out into an apical flagellum measuring. another metre; the vacant spaces between two inflorescences very long, and armed (as is the flagel- lum) with half-whorls of stout black-tipped claws resting on broad and confluent bases; partial inflorescences rather many (6 in the spadix mentioned above), the uppermost rudimentary, the lowest, the largest, 60 cm. long, twice branched near the base and with single spikelets in the upper portion; primary spathes . Cylindrical, very long tubular, closely sheathing, often ultimately split longitudi- nally, strongly clawed especially in their lower portion, terminating in a lanceolate sparsely aculeate limb which is not lacerated, but often at length decayed and partially deciduous at the summit; secondary spathes finely furfuraceous, unarmed or sparingly spiny-tuberculose, 15-20 mm, long, tubular, slightly enlarged above, truncate, entire and not ciliate at the mouth where lengthened on one side into a short triangular acute point; male spikelets reddish-brown when dry, attached imme- diately above the mouth of their own spathe, spreading or recurved, flattened, 7-11 em. long with 420-30 bifarious contiguous flowers on each side, inserted at an angle of about 459; spsthels finely furfuraceous, short, broadly and asymmetrically infun- dibuliform, truneate and entire at ihe mouth, shortly apiculate at one side; involucre slightly longer than its own spathel and attached to the base of the one above, like a swallow’s nest or dimidiatel y cupular, obliquely truncate, emarginate, two-keeled and two-toothed on the side next to the axis, Male [flowers 6 mm. long, elongate- ovoid, obtusely trigonous, somewhat attenuated at the apex; calyx campanulate, striate- ly-nerved, with three broad apiculate lobes; corolla twice as long at the calyx, divided down to the base into three lanceolate rather obtuse segments; stamens with filament nearly free at the. base, subulate and slightly inflected at the apex; anthers linear with their cells discrete from above the middle; rudimentary ovary very: small, formed by three narrow scales not quite 1 mm, long. Female spadiz simply decompound, with very elongate partial inflorescences, of which the largest, the lowest, is 1 m. in length, with 10-12 spikelets on each side; the upper 40-50 cm. long, with fewer spikelets; secondary spathes a little longer than in the male spadix, 2:5-3-5 cm. loug, C. rudentum | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 141 often split upwards; spikelets not pedicellate, inserted at the mouth of their own spathe, arched downwards, ihe largest, the lowest, 20-25 cm. long with 20-25 flowers on each side, those of the upper and more slender branches not more than 7-8 em. in length with proportionally fewer flowers; spathels asymmetrically infundibuliform, truncate, entire, apiculate at one side; involucrophorum half-projecting from its own spathel and obliquely attached at the base of the one above, dimidiately cupular, two-keeled, emarginate and two-toothed on the side next to the axis; involucre not or slightly exceeding the involucrophorum, cupular, rather shallow, entire, superficially emarginate on the side of the neuter flower of which the areola is depressedly lunate. Female flowers inserted at an angle of 45°, ovate, 6 mm. long; the calyx divided more than halfway down into three, ovate, rather thick, somewhat obtuse lobes; corolla divided almost to the base into three, ovate-lanceolate, acute, striate segments a little longer than the calyx; stamens forming by the united bases of their filaments a cup which. reaches to the middle of the corolla and is crowned by six, triangular, elongate and subulate teeth; fertilized ovaries subovoid, suddenly narrowed into a conic beak; scales in 12 series, shining, not channelled along the middle, straw-coloured, with a darker intramarginal line and ciliolate fim- briate margin, especially near the tip. Fruit not seen ripe; in the very young fruit the perianth is already split and not pedicelliform. Hasitat.—Cochin-China (Loureiro). Rediscovered by Pierre at Dian-lau-me in the province of Bien-hoa in Lower Cochin-China (Herb, Pierre No, 4846); also near Tan Kiyen on the River Dong-nai (No. 4845) and near Mount Pong-lu in the province of Binh-Thuan. According to Pierre the Annamite and Moi name for this Calamus is “ Ke- wang.” No. 4845 bears that of “ May-chia-wang,” which has some resemblance with that of “ May saong”? assigned by Loureiro. The Rotang produced by this Calamus is of a good quality and much used by the natives for cables. Loureiro writes also of the uses of this species: Pro- navium rudentibus; ad magna onera trahenda, ad elephanthos indomitos coercendos et allizandos." OssERVATIONS.— This species appears to be closely allied to C. dongtsetus, from which however it is certainly distinct by the much smaller male flowers, thinner female spikelets and much narrower leaflets. It appears to me that we can recognize in the specimens of Pierre, Nos. 4845, 4846, the C. rudentum of Loureiro, not so much from the characters given by this author, which are quite insufficient for the recognition of any species of Calamus, as from the general notes. Loureiro assigns to C, rudeníum the extraordinary length of 500 feet, but this is certainly an exaggeration. This species seems rather variable, and amongst the specimens of Pierre those bearing the No. 4845 differ somewhat from the others, and have the leaflets more crowded towards the apex, shorter, broader (34-50 by 2:5 cm.) and less acuminate than in No. 4846; the rachis bears the impressions of the spines, which at least in the apical biton are ternate, the mesial 3 cm. long, the longest, Straight, flat, narrow, yellowish with black tiv. Some spikelets, mainly those near the apex, 149 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. leptospadix in a female spadix, bear male flowers only and are apparently fertile, but are more slender than those of the male spadix. The Calamus named by 'Thwaites C. rudentum (Enum, Pl. Zeyl. 330) I have considered to be a new species (C. zeylanicus Becc.) C. rudentum of Roxb. (Fl. Ind. iii, 76) I have reduced to @. albus Pers. as to the Ceylon plant. C. rudentum, of Mart, (Hist. Nat. Palm, iii, 340) is C. rivalis Thw. PLATE 13.—Calamus rudentum Zour. The terminal part of a male spadix; portion of a female partial inflorescence with ovaries in course of development; summit of a leaf seen from the lower surface ; another portion of a leaf seen from above; very young fruits from Pierre's specimens in Herb. Bece, 8. CALAMUS LEPTOSPADIX Griff. in Cale. Journ. Nat. Hist. v, 49, and Palms, Brit. Ind. 60, t. exciy A. B. C.; Mart. Hist. Nat. Palm. iii, 339, t. 175, f. ii and t. Z. xviii, f. xiii; Walp. Ann, iii, 485, and v, 830; T. And. in Journ. Linn. Soc. xi, 8; Gamble Man. Ind. Tim, 423; Hook. f. Fl. Brit. Ind. vi, 441; Beco: in Rec. Bot, Surv, Ind., ii, 199. DerscRIPTION.—Scandent, long and slender. . Sheathed stem 12-15 or at most 20 mm. in diam.; naked canes 8-10 mm. thick, with internodes 18-30 cm. long, slightly thickened at the joints, smooth. and rather polished on the surface when freed from the scurf with which they are covered when with the sheaths on, .Leaf-sheaíhs flagelliferous, ferrugineously scurfy, puckered above, armed with very variable, very short or 15-25 mm. long, straight, flat, narrow, subulate, horizontal or slightly deflexed, solitary or aggregate and even subwhorled spines. Ocrea very distinct, sub- coriaceous, persistent, about 15 mm, in length, prickly-hispid especially on the margin, .Leaf-sheath flagella slender, filiform, very long, flattened and smooth in the lower portion, in the upper part armed with weak scattered claws. Leaves rather delicate, from 60 cm. to 1 m. in length, more or less covered when young with a cottony-ferrugineous deciduous scurf on the petiole and mainly on the rachis; petiole short (in full-grown plants), rounded beneath where armed with few deflexed spines, channelled above near the base, then flat, armed on the margins with some rare straight spines, otherwise nearly smooth; in young plants the petiole is longer, spar- ingly armed with straight, horizontal, (2-3 em. long ) spines; rachis with a very acute and raised angle and two flat side-faces above, rounded below in its lower portion, flattish upwards and armed throughout along the middle with long, straight deflexed, always solitary spines which gradually pass into claws towards the apex; this however, occasionally smooth ; leaflets numerous, rather closely set and very regularly equidistant, alternate or subopposite, broadly linear or linear-ensiform, slightly attenuated at the base, gradually acuminate into a bristly subulate apex, almost the same colour on both surfaces, distinctly 3-costate above; the mid-costa stronger than the side ones, all three equally furnished with closely set bristles; margins densely ciliate with m adpressed hairs, whieh are closer and longer towards the apex; the largest leaflets those not very far from the base and in the largest specimens 20-30 cm. long by 10-14 mm. ; the upper gradually shorter and very often opposite, less acuminate at the apex, but more bristly there than the others; the terminal pair free from the base and very small Male spadiz simply decompound, very long, excessively slender and C. leptospadix | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 143 prolonged into a filiform clawed flagellum; partial inflorescences not many, very distant, very strict, slender, 20-40 cm. long, bearing on each side 10-20 adpressed spikelets; primary spathes narrow, tubular, closely sheathing, the lowest flattened, acutely two-keeled, scarcely armed on the back with some short spines; the secondary less flattened, very narrow and also very long, the upper cylindraceous, 30-40 cm, long, narrowed towards the base, where flat and smooth on the inner side, rather strongly clawed and externally produced at the apex into a lanceolate acute limb keeled on the back; secondary spathes tubular at the base and enlarged above into an ovate-lanceolate, long-acuminate, auriculiform, smooth limb which is keeled on the back and embraces the base of the spikes and terminates in a small brushed deciduous appendix 2-3 mm. long; spikelets curved or distinctly scorpioid, especially when young, arising distichously from the interior of their respective spathes, short (1-2 cm. long), the lowest the largest, with 6-12 bifarious, closely set and nearly horizontal flowers on each side; spathels very crowded, membranous, very broad, spoon-shaped, amplectent, acute, often penicillate-furfuraceous at the apex; involucre cupular, rather deep, obliquely truncate, flat and sharply two-keeled on the side next to the axis. Male flowers (full-grown buds) oblong, rather obtuse, 4-4'5 mm, long, about 1°5 mm. broad; calyx tubular-campanulate, striate, divided down about to the middle into 3 broad, finely striately veined and rather obtuse or apiculate lobes; corolla twice as long as the calyx at most, divided down to two-thirds into ' three oblong acute segments; filaments of the stamens subulate with inflected apex and united by their base to the tubular part of the corolla; anthers versatile, linear, sagittate, apiculate; rudimentary ovary formed by three subulate bodies, reaching to the base of the anthers. Female spadiz very similar to the male, the one measured 4 m, in length (including the slender filiform flagellum 60 cm. long ) and with five partial inflorescences; spikelets arising erect from the spathes and slightly outwardly curved, 15-20 mm. long, with the flowers when young distinctly arranged in four series, of which two are of fertile female and two of sterile or neuter flowers, these last scarce- ly smaller than the others, each series composed of 5-8 flowers in the largest, the lowest, spikelets; spathels as in the male spadix, but deeper; involucrophorum anti- cously split and posticously discrete as if formed by two small bracts connate by their bases; involucre rather deep, obliquely truncate and laterally evolute on the side of the neuter flower, of which the areola is broadly ovate or nearly circular with sharply ` defined borders. Female flowers ovoid, about 3-3°5 mm. long, with the corolla a little longer than the calyx. Fruiting perianth not pedicelliform, split and spreading under the fruit. Fruit globose or globose-ovoid, about 10 mm. in diam, mammil- late-apiculate; scales in 18 series, not channelled along the middle, yellowish with a con- spicuous reddish-brown marginal line, broader towards the point, of which the extreme tip is pale, scarious, obtuse and not fimbriate; margins unequally erose. Seed irregularly globular, 8 mm. long and 6:5 mm. thick, shining, yellowish-brown when freed from the scanty coat with wbich it is enveloped, convex and coarsely alveolate on the back, marked on the ventral face with some superficial furrows radiating from the chalazal fovea, which is central, circular, rather deep and continued down to the base in a channel ending at the embryo; albumen equable, Hasrar.—N. E. India: Khasia Hills in Assam (Grifith, Hooker f. & Thomson). G. Mann from Assam sent me some complete specimens gathered at about 1000 m 144 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (Ç. dilaceratus - above sea-level. In Manipur and. on the Naga Hills (G. Watt No. 7459). Sikkim (Hooker f. & Thomson; Treutler in Herb. Kew and Herb. St. Petersb.; Gambie) Buxa Reserve in the West Duars (Gambie). T. Anderson (l. c.) says that ‘‘this species occurs in most of the deep valleys of the outer ranges of the Sikkim Himalaya, and extends into the interior along the course of the larger rivers and their tributaries did . . The canes are soft and useless. The Lepcha name of this Calamus is ‘Lat. In Nepal according to Gamble it is called ‘Dangri Bet’ and at Kurseong ‘Rani’ or ‘Rabi Bet’ (Burkill). OnskRVATIONS.—AÀ very well-marked delicate species, distinguished by its very long, simply decompound, extremely narrow spadices very much the same in both sexes, and with long, strict, partial inflorescences bearing numerous short spikelets which are decidedly scorpioid in the male spadix. Anderson (l. c.) says.of this Calamus that ‘sit is somewhat gregarious in its habit, The long slender stems, when lying on the ground, send out short leafy shoots from their joints and form a thicket of prickly leaves. The flowering extremities of these prostrate stems ascend the trees by the assistance of the strongly barbed straight tendrils springing from the sheaths of the leaves.” PrarE 14,—Calamus leptospadix Grif. Intermediate portion of a leaf (seen from above) and partial inflorescence (both on the right-hand side), from a specimen collected by Sir D. Brandis on the Teesta (Herb. Bece.); apex of a leaf (on the left side) seen from the lower surface; leaf-sheath and a nearly entire spadix from the Khasia Hills (Herb. Beec.); apex of a leaf-sheath with large ocrea and base of the petiole from a young plant (on the right hand lower corner). Fruits and seeds (from the Calcutta Botanic Garden). i 9. CALAMUS DILACERATUS Becc. in Rec. Bot, Surv. Ind. ii, 198. DzscRrPTION.— T'ufted, probably not scandent, Leaves: the only one seen supposed to belong to the spadix hereafter described, and probably a radical one, is 3 m. long and has the apical portion in a decayed condition, but apparently not cirriferqus. Its sheath is not completely tubular, but opened on the ventral side and is densely armed throughout with very thin, black, criniform, friable, straight spicules of unequal length (the largest 4-5 em. long), confluent at their bases and arranged in rather erowded oblique series; the petiole is very long (about 1 m.), sablersts faintly channelled on the upper surface of the lower portion, sparingly tuberdled: spioulous; the rachis is rounded below and sprinkled there with some small subspinous tubercles, with an acute angle and two flat side-faces above; leaflets very many, equidistant, rather crowded (about 2 cm. apart), linear-ensiform, very acuminate, thinly papyraceous, the largest, the mesial, 35-10 cm. long by 1°5 cm. green, subconcolorous on both surfaces, with three rather acute cost above, of which the central is bristly-spinulous near the apex and the lateral furnished with few short black sete; beneath, all nerves are very faint, the central rather densely, and one on each side of this sparingly setiferous. Male spadiz . .. Female $padiz decompound, rather large, panicled, not flagelliferous, with many partial inflorescences; primary spathes not very long, tubular-infundibuliform, dry, somewhat D. castaneus | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 145 enlarged in the upper half. or third, where in the fruiting stage discoloured, decayed and lacerated, armed with small, straight, somewhat deflexed spicules which rest on a bulbiform base; partial inflorescences erect-spreading, 25-35 cm. long, bearing distichously on each side 8-10 spikelets which are 4-7 cm. long and furnished with a slender flattened pedicellar portion (1-1-5 em. long) attached to the bottom of their respective spathe; secondary spathes thinly coria- ceous or submembranous, tubular-infundibuliform, 1°5-2 cm. long, more or less split down to the base, naked or sparingly spiculose, extended at one side into a rather long point; spathels shortly infundibuliform, almost auriculiform, acute at one side, ultimately more or less split; involucrophorum inserted at the bottom of its res- pective spathels, not laterally attached at the base of the one above, spathaceous-auri- culiform, acute, dorsally two-keeled ; involucre also spathaceous-auriculiform, acute at one side; areola of the neuter flower elongate, lanceolate, sharply bordered. Female flowers about 6 mm. long. Fruiting perianth not pedicelliform; its calyx and corolla almost similar, divided down to the base into three lanceolate finely striate lobes of equal length and breadth; staminodes persistent, reaching to about the middle of the corolla with the filaments acuminate from a very broad base and furnished with a small sagittate anther. Fruit small, broadly ovate, 12 mm. long by 9 mm., very suddenly contracted into a short but thick beak or mucro; scales small, in 21 longitudinal series, pale-yellowish, shining, channelled along the middle, with brown and finely fimbriate tip. Seed subglobose, slightly compressed, flattish or slightly depressed on the raphal side and convex on the back, equally rounded at both ends, with equal and not pitted surface, 8 mm. long by 6-65 mm. thick; albumen homogeneous; embryo basilar, All parts of thé spadix, the secondary spathes, spathels and other appendices and parts of the flowers in the fruiting stage acquire a chestnut-brown colour when dry, and show traces of a tobacco-coloured scurf and are more or less split or lacerated. HasrrAT.— The Nicobar Islands, where it was found by Mr. E. H. Man, who sent me the specimens described above in August 1888. OssERvATIONS,— The leaf I have described was detached from the specimen of the spadix, without any reference to this; but the spines with which the petiole is armed are very much like those that cover the spathes and show almost certainly that this leaf belongs to that spadix, and besides that this Calamus is erect or bushy and not scandent. Pirate 15,—Calamus dilaceratus Becc. The upper part of a spadix with almost ripe fruit, from Mr. Man's specimens in Herb. Becc. 10, CALAMUS CASTANEUS Griff. in Cale. Journ. Nat. Hist. v, 28, var. B, and Palms Brit. India, 37, t. clxxxv. C. (and A. B.?); Mart, Hist. Nat. Palm. ii, 331, t. Z xviii, f. xxii and Z xxi, f. ii and Z xxii, f. xv; Walp. Ann. iii, 482, and v, 829; Mig: Fl. Ind. Bat. iwi, 112; H. Wendl. in Kerch. Palm, 235 (excl. syn.); Hook. f, Fl. Brit. Ind. vi, 440; Becc. in Rec. Bot. Surv. Ind, ii, 198, Description.—Erect, 1-15 m, high. Sheathed stem 7-10 cm. in diam. Leaf- sheaths not flagelliferous, truncate at the mouth, not ocreate or ligulate (at least Ann. Roy. Bor, Garp, CarcurrA Vor. XI. 146 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. castaneus in full-grown leaves) very densely armed with flat, very acuminate, often sinuous, dark-grey spines which are as much as 6-7 cm. long and are intermingled with smaller ones of the same shape. eaves not cirriferous, large, 1°5-2 m. long; petiole very stout and long (60-70 em.), 2 em, thick, faintly channelled at the base and flat upwards above, where more or less armed with straight erect spines, rounded underneath where armed along the middle with straight horizontal spines and on the margins with the same kind of straight, long and closely set spines pointing in different directions and intermixed with much shorter ones; rachis in the intermediate and upper portion trigonous, acutely bifaced and smooth above, roundish near the base below, where armed with distant solitary, short, horizontal or slightly deflexed spines, flat and unarmed near the apex; leaflets green on both surfaces, slightly paler beneath, numerous, equidistant, alternate or subopposite (at least in the portions seen), narrowly lanceolate or broadly ensiform, somewhat narrowed at the base, gradually acuminate to a spinulous apex; mid-costa acute above, bearing on both surfaces, mainly towards the apex, few short bristly spinules; secondary nerves 2-3 on each side of the mid-eosta, rather distinct and naked on both surfaces, margins finely and acutely spinulous, transverse veinlets indistinct; the largest leaflets 60 cm. long and 4-5:5 cm. broad; the terminal pair smaller than the others. Male spadiz . . . . . Female spadiz not seen entire; partial inflorescences about 40 cm. long, bearing 7-8 spikelets on each side; secondary spathes thinly coriaceous, elongate-infundibuliform, loosely sheathing, unarmed, prolonged into a lanceolate-auriculiform limb, ultimately withered and decayed at the apex, but not fibrous; spikelets large and thick, 12-17 cm. long, spreading, furnished with a peduncular part 1-2 cm. long, by which they are attached inside and at the bottom of their own spathe; spathels covered with a very thin adherent brown indumentum, broadly and symmetrically infundibuliform, and having a rather loose entire or later irregularly split limb, which is prolonged at one side into a triangular ultimately decayed point; involucrophorum spathaceous, irregularly split, attenuate at the base, attached inside and at the bottom of its own spathel, flattish and two-keeled on the side next to the axis; involucre not or slightly exceeding the invojucrophorum, subcupular, more or less irregularly split ; areola of the neuter flower vertically ovate and acute, Female flowers 5 mm. long. Fruiting perianth not pedicelliform; its calyx not thickened or callous at the base, entirely split into three lanceolate acuminate lobes; the segments of the corolla scarcely longer than the calyx; filaments with stamens highly connate at the base and in the free portion elongately triangular, Fruit ellipsoid, ovate or obovate, rounded at the base and suddenly narrowed at the top into a rather long beak, 13 mm. broad and 22-24 mm. long (including the beak), of uniform chestnut-brown or chocolate colour; scales small, in 24-27 longitudinal series, about 2 mm. broad. and scarcely less in length, narrowly channelled along the middle, dull and under a strong lens finely scabridulous, with slightly paler, finely erose margins and obtuse tip. Seed not seen perfectly mature. The plant acquires in drying a reddish-brown colour. HasrraT.— The “Malayan Peninsula; in* thick jungles about Malacca (Grifith, Maingay No. 1533 in H. K.) and in the district.of Perak on the hills of Larut between 150-200 metres above the level of the sea (Herb, Cale. No. 5880). is €. Grifithianus] BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 147 Oxservations.—My description of this species is founded on the fruiting specimen No. 5880 of the Calcutta Herbarium, not taking into account all the others, not being sure that some mixture had not occurred among the different parts of this species with those of C. Grifithianus. The fruit of C. castaneus figured by Griffith, which I have seen on a portion of the authentie specimen preserved at Kew, is more globose than that I have described, but the shape of the fruit in this species, as in C. Grifilhianus, seems very variable in the different stages of its development. Griffith has described and figured the male spadix of C. castaneus ; but as this author had not distinguished this species from the closely allied C. Grifithianus, it is not exactly known to which of the two species such description and figures belong; but probably the male spadix differs very little in the two species. The leaflets of the said specimen No. 5880 belong to a full-grown leaf and do not show any trace of the peculiar thin purpurascent coating on the lower surface, which may be seen in the young leaves of C. Griffithianus. The leaflets of OC. castaneus are broader than those of C. Grifithianus, bristly only on the mid-costa, and probably equidistant ; but the fragments of leaves seen are too short to be sure of this characteristic, Sir George King's collector in the label to the specimens of C. Grifithianus and C. castaneus had also pointed out the main differences between the two species. Indeed C. Grifithianus has a stem creeping at first and then ascendent and has terete petioles; - on the contrary C. castaneus has a short erect stem and petioles channelled at the base and flat upwards above and rounded beneath. Pirate 16,—Calamus castaneus Grif. Upper portion of a leaf-sheath with the base of the petiole; portion of a leaf from about its intermediate part; portion of the female spadix with unripe fruit: all these figures from a specimen bearing the No. 5880 in Herb. Calc. The portions of male spadix (on the right hand side) and of the female one (on the left), the ripe fruit, the seed entire and the one in section in the upper part of the plate are from Scortechini’s specimens which I doubtfully refer to C. castuneus, these parts not being accompanied by the leaves and it being very difficult to distinguish C. castaneus from (C. Griffithianus by the reproductive organs only. 1l. Carawus GmrrrrrHIANUS Mart. Hist. Nat. Palm. iii, t, 332, Zxviii f. xxvii, and Zxxi, f. xiv; Walp. Ann., iii, 482 and v, 829; Miq. Fl Ind, Bat. iii, 112; Hook. f, Fl. Brit. Ind, vi, 440; oie in Rec. Bot, Surv. Ind. ii, 198. C. castaneus (partly) Griff. in Cale, Journ. Nat. Hist. iv, 29, and Palms Brit. Ind. 38; Ridley in Trans. Linn. Soc. Bot., iii, pt. ix, 392, DrscRIPTION.— Stem creeping, rooting at the joints and then ascending erect, about 3 cm. in diam. Leajf-sheaths not flagelliferous, armed with straight, long, unequal, laminar, subulate spines, which are scattered or more or less aggregate and remotely obliquely seriate. Leaves large, 2-3 m. long, not cirriferous; petiole from 60 cm. to 1 m. in length, subterete (at least in its upper portion), 12-13 mm. in diam., armed with laminar, elastic, horizontal spines of various sizes, 2-5 cm. long and sometimes considerably longer, very irregularly scattered or, as on the leaf-sheaths, aggregate and more or less seriate; rachis from the middle upwards trigonous, bifaced above and Ann. Roy. Bor. Garp. CarcurrA Von XL 148 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. — [Q. Griffithianus Often but not always spinulous on the upper angle, more or less armed below, chiefly in its lower portion, with solitary aggregate or pectinate straight spines; leaflets arranged in alternate groups of many, being in each group equidistant and not clustered, elongate-ensiform and glossy, 45-65 cm. long and 15-35 mm. broad, green above, paler underneath where—at least when young—covered with a very thin mealy-violaceous coating; mid-costa prominent above, bristly towards the apex on both surfaces; secondary nerves rather distinct, one on each side of the mid-costa naked above and conspicuously covered beneath with rather long bristles; transverse veinlets fine, much interrupted; margins appressedly and finely spinulous. Male spadiz 0°8-1m. long, ultradecompound, not flagelliferous, with many partial inflorescences or primary branches, variously spreading or nodding, inserted inside their own spathe and pedicellate; each branch also pedicellate and inserted inside its own spathe, divided again into 6-10 secondary branchlets or compound spikes which are 8-15 cm. long and bear distichously 2-5 recurved or scorpioid spikelets and decrease in length from the base of the branchlet towards its apex; the apical spikelet, which is the continuation of the axis is much longer and larger than the side ones and has larger spathels and flowers; primary spathes relatively short, tubular at the base, enlarged and somewhat inflated upwards, with an auriculiform often longitudinally split limb, half-decayed upwards where usually sharply defined from the still living basal portion by a transverse slightly prominent line; secondary spathes (spathes of the compound spikes) short, broadly and asymmetrically infundibuliform, truncate, with the poirt somewhat decayed and split; side spikelets 2-6 cm. long, curved or scorpioid, flattened, about 1 cm, broad (including the flowers), their spathels very closely packed, shallow, nearly boat-shaped, prolonged on one side under the flower into a broadly triangular acute point; involucre cupular, half- immersed in its own spathel and attached to its base, aeutely two-keeled, deeply emarginate and two-toothed on the side next to the axis, anticously obliquely truncate or more often split and apparently formed by two connate bracts, Male flowers slender, 5-55 mm. long; calyx tubular, slightly angular from pressure, with three rather acute finely striate lobes, ultimately cleft down to the base; corolla a little longer than the calyx, deeply divided into three linear-lance- olate segments; stamens with filaments united to the base of the corolla, rather thick, subulate with inflexed apex; anthers sagittate, acute; rudimentary ovary composed of three elongate subulate bodies. Female spadiz short, erect, 40-50 cm. long, panicled, not flagelliferous or appendiculate, with a flattened peduncular part, 20-25 cm. long, closely sheathed with short obliquely truncate spathes, unarmed or nearly so; partial inflorescences few (2-4), each bearing subdistichously or somewhat irregularly 5-6 spreading flexuose spikelets, 8-12 cm. long, inserted inside their own spathe and pedicellate; spathels broadly and asymmetrically infundibuliform or obconic, occasionally more or less split, prolonged on one side under the flowers into a triangular ultimate- ly decayed point; involucrophorum inserted inside its own spathel and shortly pedicellate, spathaceous, unilaterally auriculiform, anticously split and opened, acutely keeled on the side next the axis; involucre included in the inyolucrophorum, subcupuliform, truncate, ultimately radiately split, the areola of the neuter flower vertically lanceolate-ovate, acute and sharply bordered. Female flowers elongate conical, finely striate; their calyx divided almost to the base into three ovate- lanceolate acute lobes; corolla very similar to the calyx and almost as long but C. Griffithianus] BECCARI, MONOGRAPH OF THE GENUS CALAMUS. 149 with narrower segments; stamens as long as the corolla, with filaments highly connate and forming a long campanulate urceolum which reaches and even surpasses the middle of the corolla and is crowned with 6 triangular teeth; anthers broadly sagittate ; stigmas thick, lamellose, recurved. Neuter flowers finely striately veined externally, very similar to and about as long as the male ones (6 mm. long) but a good deal thinner; the calyx at first tubular and shortly 3-toothed, then cleft almost to the base into 3 lanceolate parts; the corolla slightly longer than the calyx and divided down almost to the base into 3 linear-lanceolate acute segments ; stamens with the filaments subulate, united by their bases and shortly inflexed at the apex, almost as long as the petals; anthers linear, abortive, versatile ; rudimentary ovary narrowly elongate, 3-dentate, slightly shorter than the stamens, Fruiting pertanth not pedicelliform. Fruit ovate-oblong or subglobose-obovate, very suddenly and conspicuously beaked or rostrate, 20-22 mm. long and 14-16 mm. in diam. ; scales chestnut-brown, in 18-24 longitudinal series, narrowly channelled along the middle, dull and under a strong lens finely scabridulous with slightly paler finely erose margins and short obtuse point. Seed, when freed from the dry (once fleshy) coat, lenticular, 14-15 mm. long, 12-13 mm. wide, 7 mm. thick, somewhat convex, sinuately rugose and superficially pitted on the .back, with a small, round, shallow, chalazal fovea in the centre ef the flattish raphal face; albumen equable ; embryo basilar.—All parts of the male and female spadix, including the flowers and fruits, of a uniform brown chestnut colour when dry. Hasitat.-—-The Malay Peninsula near Malacca (Grifith; Ridley No. 842); near Perak (Scortechini); at Larut in open jungle on hilly rocky localities between 60 and 250 metres above the sea (King’s collector No. 3040 Herb. Hort. Cale.) ; in Selangor (Ridley Nos. 3417 and 3478). OssERVATIONS,—I have derived my description from Scortechini’s incomplete specimens and from the No, 3040 of the Calcutta Herbarium. In some of Scortechini’s specimens the spines on the petiole (probably of a leaf from a young plant) are excessively long (7-15 cm.) and the rachis in the upper angle is not spinulous. O. Grifithianus differs from O. castaneus in its smaller dimensions, ascendent stem, less densely and seriately armed leaf-sheaths, but chiefly in the petiole which is nearly cylindric upwards, and in the leaflets with 3 nerves, which are bristly underneath, while in C. castaneus the mid-costa only bears bristles there and the petiole is plano-convex in transverse section. The leaflets in C. Grifithianus are also distinctly disposed in large groups and probably in C. ' castaneus are equidistant, but of this last species I had not the opportunity of seeing an entire leaf. The number of longitudinal series of the scales of the fruit seems a very variable character in this species as is also the more or less roundish shape of the fruit. Some specimens of a Calamus from the western side of the Malayan Peninsula sent to me by Mr. H. N. Ridley undoubtedly belong to C. Grifithianus; conse- quently I suppose that the C. castaneus from Tahan woods quoted by that author in his memoir on the “Flora of the Eastern Coast of the Malay Peninsula” 150 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. Burckianus (Trans. Linn. Soc. London, Botany iii, pt. 9, 392) may possibly be reduced to €. Grifithianus. C. Grifithianus seems a much commoner species than C, castaneus. The natural affinities of C. castaneus and C, Griffthianus are probably with C. ovoideus and C. andamanicus, which, however, have cirriferous leaves. PLATE 17.—Calamus Griffithianus Mart. Basal portion of a leaf with the upper part of the petiole; three leaflets from near the apex showing the upper surface; an entire fruiting spadix; flowers, fruits and seed (one in longitudinal section) from the spadix quoted; all figures from the No. 3040 in the Calcutta Herbarium, 12. Carawus Burcxianus Becc. in Rec. Bot. Surv. Ind. ii, 198. | Description.—Probably bushy (not scandent). Stem and /eaf-sheaths not seen. Leaves (including a portion of the unarmed terete-subcompressed petiole measuring 25 cm.) a metre in length; rachis glabrous, trigonous, convex below where smooth (or sometimes armed with small rudimentary claws?), the margins and the superior angle acute; leaflets thinly papyraceous, numerous, alternate, regularly equidistant (10-12 mm, apart), linear-ensiform, somewhat attenuate towards the base, subulately acuminate at the apex, green on both surfaces when dry, the lower surface slightly paler; upper surface distinctly 3-costate, the mid-costa acute, prominent and bearing on its apical half some dark bristles, the side costs bristly along their whole length; lower surface with nerves indistinct and only the mid-costa sparingly bristly, or sometimes smooth or one nerve on each side bristly; the margins with a few distant adpressed spinules, more numerous towards the apex, the lower somewhat thickened by an intramarginal nerve; transverse veinlets rather distant and indistinct, sinuous and much interrupted; the largest leaflets, the lowest, 28-30 em. ong and 12-13 mm. broad, the others gradually decreasing in size upwards, the two apical (the smallest) 12 cm. by 6-7 mm,, slightly connate at the base. Male spadiz . . . + + Female spadiz not seen entire; the few partial inflorescences seen 50 cm. long, ending in a caudate, rigid, unarmed appendix, a few cm. long, bearing on each side few (4-7), alternate, distichous, erecto-patent spikelets ; primary spathes not seen; secondary spathes tubular-infundibuliform, unarmed, closely sheathing, 3-4 cm. long, the mouth obliquely truncate and naked, but at one side prolonged into a triangular, acute, erect point ; spikelets attached to the bottom of their own spathe by a slender 2-3 em. long pedicellar part, vermicular, rather thick and rigid, flexuose, 12-15 cm. long, with 10-15 distichous, rather distant (1 cm. apart) flowers on each side; spathels shortly tubular-infundibuliform, usually split in the fruiting stage, the mouth horizontally truncate and naked, apiculate on one side; involucrophorum distinctly pedicellate, arising from the bottom of its own spathel, obliquely truncate, sharply two-keeled next to the axis, where not adnate ; involucre cupular, slightly longer than the involucrophorum, truncate and obscurel 3-toothed ; areola of neuter flower lunate Fruiting poriondh expl i à . ! planate (not pedicelli- form); calyx not callous at the base, split almost to the base into 3 ovate, acute segments ; corolla with three segments almost as long as those of the calyx, but narrower and more acute. Fruit, when not quite ripe, globose and 9-10 mm. in diam., suddenly apiculate, its scales in 24 longitudinal series, brownish, rather dull, slightly channelled along the middle, small (1-5 mm. broad), their apices obtuse ; C. deerratus | BECCARI, MONOGRAPH OF THE GENUS CALAMUS. | 151 their margins pale, very finely and minutely erosely ciliate. Seeds (immature) usually 3, with smooth surface, convex back and two flat ventral facets, occasionally one or two of the seeds tending to abort. Hasrrat.—Java (Kurz in Herb. Munich and Zeysmmann in Herb, St. Petersb.)— Javanese name “Hooy buluk-buk”—vide Blume, Rumphia iu, 30. OpsERVATIONS.—I have seen of this only one leaf, apparently a radical one and deprived of its sheath, and some portion of a spadix bearing not quite ripe fruits ; these specimens were sent to the St. Petersburg Herbarium by Teysmann. Another more incomplete specimen, perfectly like the preceding, is preserved in the Herbarium at Munich, sent there by Kurz. On these specimens alone is based the description above, but I consider as conspecific with the specimens mentioned another which I have received from the Leyden Herbarium, In this the spikelets are 8-12 cm. long, the fruit is perfectly globular or a trifle longer than broad (10-12 mm. in diam. ), rounded at both ends, but surmounted by a very short beak; the scales are in 23 rows and have a very narrow intramarginal line. In all the ripe fruits which I have examined I have found only a single seed fully developed; of the other two seeds only inconspicuous traces were found. The seed has an even, dark, opaque, not alveolated nor pitted surface, is orbicular, somewhat depressed, 9-9:5 mm. in diameter, somewhat concave on the raphal side and with a shallow chalazal fovea; the albumen is equable and the embryo is situated near the base of its ventral and convex face. A leaf in the Leyden Herbarium, labelled ** Java: Hooy belock-buk, Hasskarl,” apparently belongs to the fruiting spadix just described, but is armed on the back along the middle of its rachis with a few small short solitary and remote claws. PLATE 18.— Calamus Burckianus Bece. Partial inflorescence with unripe fruits and apex of a leaf seen from the upper surface with a small portion at its base seen from the lower; these parts are from Teysmann’s specimen in the St. Petersburg Herbarium. ‘The spikelet with mature fruits, the seeds and the portion of leaf on the left-hand side are from the specimens in the Leyden Herbarium described in the observations, 13, CALAMUS DEERRATUS Mann & Wendl. in Trans. Linn. Soc. xxiv, 429, t. xl, F; H. Wendl in Kerch. Les Palm. 236; Drude in Bot. Jahrb. xxi. iii; Beco. in Rec. Bot. Surv. Ind. ii, 199; Wright in Fl Trop. Afri, viii, 109 (partly). | DescairTion.—Scandent, rather slender, or of moderate size, 5-10 metres long. Sheathed stem 18-25 mm. in diam, .Leaf-sheaths rather thickly coriaceous, more or less partially fugaciously scaly-furfuraceous, gibbous above, rather densely armed with brown, rigid, flat, very thin, lanceolate, subulate spines which arise from a broad but not tumescent base and are often divided or laciniate or with the margins deeply cleft, spreading or slightly deflexed, solitary or aggregated in small series ; the spines are more abundant, longer (as much as 2 cm. long) and more distinetly seriate on the ventral side of the upper portion of the sheaths, where from a horizontal position they gradually become erect near the base of the petiole. Ocrea 152 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. {[(. deerratus very conspicuous, produced laterally at the base of the petiole into two papyraceous or dry membranous, ultimately lacerated auricles, which are 4-5 cm. long and entirely covered with some spines similar to those of the sheaths, mixed with others which are more slender or bristle-like. Leaves rather large, the one seen entire 1 metre in length, not cirriferous; petiole very short (5-6 cm.), rather robust, flattish above and armed at the margins with some straight, needle-like, ascending spines, 1-2 cm. long, rounded beneath where more or less armed, at east along the middle, with some straight spines passing into claws; rachis flat above in the first portion and channelled laterally where the leaflets are attached, acutely bifaced above and trigonous in cross-section upwards, rounded near the base and flat upwards beneath, where armed throughout along the middle and occasionally at the sides with dark-tipped solitary claws; leaflets rather numerous, subequidistant 2-3 cm. apart (aggregated in young plants, according to Mann and Wendland), rigidulous, papyraceous, linear-lanceolate, or lanceolate-ensiform, attenuated towards the base, where deeply plicate, gradually acuminate into a subulate and caudate spinulous tip, glabrous and subshining above, rather distinctly paler beneath, where dotted or more or less sprinkled with brown scales, and under the lens finely striate ; mid-costa acute and prominent above, accompanied on each side by 2-3 secondary nerves, of which one is a little stronger than the others, but not so much so as to render the surface distinctly 3-costate; all the nerves in the upper surface smooth, or sometimes the mid-costa spinulous (as exceptionally and very sparingly are the side-nerves); on the under surface the mid-costa and 2-3 slender nerves on each side of it occasionally are furnished with few, small, short, spreading, spinulous bristles, which rest on a sub-bulbous base; oftener, however, the bristles are closer and stronger along 3 of the 5 nerves, but chiefly on the mid-costa, while on the side ones they are small, very scarce or wanting; margins slightly thickened by a weak secondary nerve and rather densely aculeolate-serrate; trans- verse veinlets slender, very interrupted; the largest leaflets, those a little above the base, 35-38 cm. long and 15-3 cm, broad; the upper rather abruptly shorter, but not narrower, and with a less acuminate tip, which is indented or notched on the lower margin, but in a lesser degree than in the basal ones; the two of the apical pair 15-16 cm. long and almost entirely free at the. base. Male and Female spadices simply decompound, elongate, flagelliform, 70-80 cm. long with very few partial inflorescences (1-3) and besides lengthened out into a clawed flagellum of equal iength; primary spathes very narrow, tubular, elongate, closely sheathing; the lowest usually split longitudinally (as the upper ones), somewhat flattened, with not very acute and smooth or slightly spinous edges; the upper cylindraceous, unarmed or nearly so, obliquely trurcate at the mouth and produced at one side into a short triangular point; male partial inflorescences 15-20 cm. long, attached inside their respective spathes with a rather elongate peduncular portion and furnished with 7-9 distichous approximate spikelets on each side; secondary spathes tubular-infundibuliform, unarmed, obliquely truncate and ciliate at the mouth, produced at one side into a short point; spikelets inserted just below the mouth of their own spathe and not callous at their axilla, 4-5 cm. long, bearing 8-12 distichous flowers on each side, the upper spikelets shorter and with fewer flowers; spathels very crowded, brown, strongly striately-veined, obliquely and very broadly infundibulifbéns, extended at one side into an acute — point ; C. deerratus | BECCARI, MONOGRAPH OF THE GENUS CALAMUS. 158 involucre subdimidiately cupular, very obliquely cut off posticously, lunately emarginate, and acutely 2-keeled, attenuated at the base and attached to the bottom of its own spathel. Male flowers 5 mm. long, ovoid-oblong; calyx ovoid, submembranaceous, strongly striately veined, rather deeply divided into 3 acute lobes; corolla one-half longer than the calyx, divided down almost to the base into 3 oblong, acute, striate segments ; stamens all of the same length, the filaments subulate, inflexed at the apex and shortly united at their base; anthers lanceolate, acute, with deeply separated cells; rudimentary ovary rather conspicuous, formed by 3 subulate bodies which are unit- ed by their bases and are nearly as long as the filaments. Female spadiz very similar to the male; spikelets with spathels a little larger than in the male spadix; involucrophorum obliquely cupular, truncate, posticously 2-keeled, inserted at the bottom of its own spathel and entirely included in this; involucre irregularly cupular with the margin often split or lobate; areola of the neuter flower large and deep, sometimes subcupuliform, occasionally with a fully developed flower and therefore with two nearly equally developed flowers in one spathe. Female flowers ovoid, about the same size as the males; calyx ovate, thinly coriaceous, striately veined, acutely 3-toothed at first, ultimately split down to the base; -corolla a little longer than the calyx, divided almost to the base into 3 ovate-lanceolate acute segments; filaments of the stamens united at the base into a not very high ring and in the free part elongately triangular with sterile sagittate anthers; ovary oblong, tapering towards the base, crowned by 3 thick, trigonous, acute stigmas which are strongly lamellose inside, Fruiting perianth split and explanate under the fruit. Fruit ovoid, 15-17 mm. long, 10 mm. in diam., rounded at the base, conically narrowing at the apex, crowned by the bases of the stigmas; scales in 21 series, rather shining, somewhat convex, very faintly channelled along the middle, yellowish-brown, with a broad, brown-chestnut, intramarginal line; margins and tip broadly scarious, beautifully and finely fimbriate. Seed oblong, slightly com- pressed, rounded at the base, apiculate at the apex, 1 cm. long, 7 mm. thick, rugose or wrinkled on the back, with a shallow, elongate, chalazal fovea in the centre of the raphal face, from which irradiate a few superficial ridges; albumen equable; embryo in the centre of the base. Hasitat,—West Tropical Africa: on the rivers Bagroo and Cameroons (G. Mann Nos. 891 and 2147 in Herb, Kew). OBSERVATIONS.—From the accurate study of this species I am able to state that no remarkable character separates the African from the Asiatic Calami; C. deerratus is indeed strikingly related to some of the Asiatic species of the fifth group. The involucre of the male flowers and the involucrophorum of the female Ones are attached to the bottom of their own spathel by means of a very small basilar point, and are consequently almost stalked and not laterally adnate to the base of the spathel' above its own, but are completely free from it. This, however, is not a character peculiar to the African Calami, but it is one which they have in common with some Asiatic species (0. Zollingerii, castaneus, Grifithionus, ete.); in these, however, when the involucre of the flowers have such a structure, the spikelets are stalked by a peduncular portion arising from the bottom of their respective spathes, In C. deerratus, on the contrary, the spikelets, though issuing Ans. Roy. Bor. Garp. CarcurrA Vor. XI. 154 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. Barterii from inside, are attached very near the mouth of their own spathe. No leaf.sheath flagella were present in the specimens I have seen, but very probably the plant is furnished with them in its upper part when not bearing spadices, In the specimens from the Cameroon River (Mann No. 2147) all the nerves ni the upper surface of the leaves are smooth; in those from the Bagroo River (No. 891) the mid-costa is spinulous, and exceptionally a few spinules are to be Seen on the secondary nerves. A male spadix from Cameroon is 1:35 m. long, and bears only one partial inflorescence issuing from the lowest spathe, the remaining portion forming the flagellum. Another male spadix from Bagroo has three inflorescences. Rather frequently two flowers of equal size, and probably both fertile, come out from one spathel ; when this is the case, each flower is sometimes furnished with its own involucre, but more frequently there is a normal cupular involucre accompanied by a smaller one at its side. Putare 19.—Calamus deerratus Mann & Wendi, Leaf-sheath with the base of a leaf and a male spadix from a young plant; spadix with mature fruits; apex of a leaf-sheath from a full-grown plant (in the upper right-hand —— All ' the figures from Mann’s specimens in the Herbarium at Kew. 14, CaraMus Barrer Becc. in Herb. Kew.; Drude in Engl. Bot. Jahrb. xxi 134 (partly); Becc. in Rec. Bot. Surv. Ind. ii, 199; Wright in Fl. Trop. Afr. viii, 109 (partly). DescripTion.—Scandent, slender. Sheathed stem as thick as a man’s finger, 10-15 metres long (Barter). Leafsheaths flagellierous, thinly coriaceous, scaly furfuraceous (fugaciously ?) longitudinally striated, unarmed, very slightly gibbous above. Ocrea conspicuous, liguliform, 12-15 mm. long, shortly bilobed, split on the outer side, where, as at the apex and in a lesser degree at the margins, bristly- spinulous. Leaf-sheath flagella slender, filiform, terete, about 60 em. long, armed with small solitary or sub.aggregate claws. eaves not cirriferous, 45-50 cm. long ; petiole 6-8 cm. in length, very obsoletely trigonous, channelled above, armed at the margins with some patent, straight, relatively robust spines of variable length, and beneath along the middle with solitary small claws, which appear at distant intervals throughout the entire length of the rachis; rachis partially furfuraceous, slender, trigonous, bifaced with acute and smooth angle above; leaflets few, 9-10 on each side, grouped in rather distant opposite fascicles of two to four on each side, very spreading and sometimes nearly horizontal; sometimes single leaflets have the corresponding one on the opposite side or remain solitary ; four slightly shorter than the others are grouped at the apex, and the two of the terminal pair are completely free at the base; they are all linear-lanceolate, narrowed at the base and very gradually acuminate into a long, filiform, bristly-ciliate tip, dull-green (when dry) above, slightly paler beneath, where somewhat rusty-furfuraceous near their insertion, but otherwise glabrous, not or very sparingly sprinkled with brown sealy dots, thinly papyraceous and rather flaccid, subherbaceous; the mid-costa in the- upper surface not very strong, acute and smooth; the side-nerves slender and also smooth; on the under surface the mid-costa mug one nerve on each side of it C. Heudelotii | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 155 furnished with some very smail spinules; transverse veinlets rather remote and much interrupted; margins very closely spinulous with a slender nerve running along them ; the largest leaflets, those near the base, 15-18 cm. long, by 10-13 mm. in breadth. Other parts unknown. HanrrAT.— West tropical Africa at Onitscha on the River Niger (Barter No. 110 in Herb, Kew). OssERVATIONS.— Of this species I have seen the upper portion of a sterile plant with the sheathed stem 6 mm. in diam. Amongst the African Calami, this seems well defined by the well-marked clustered arrangement of the leaflets. Judging from the general structure of the leaf.sheath flagella, which are morphologically sterile spadices, we may suppose that the spadices in C. Barter ought to be very like those of C. deerratus, but this is a much larger plant with strongly armed sheaths and numerous subequidistant leaflets. I have considered as a new species (vide C. falabensis) the specimens described by Drude (l. c.) as the male plant of 0. Barterii. The canes are much employed in the lower part of the River Niger for tying. The fruit is said to be small and dark brown when ripe (Barter). Pirate 20.—Calamus Barterii Bece, The entire Barter’s specimen No. 110 in the Herbarium at Kew. 15, Catamus Heuperorm Becc. in Herb. Kew.; Drude in Engl. Bot. Jahrb. xxi, (1896), pp. 112 and 134; Becc. in Rec. Bot. Surv. Ind, ii, 199 Wright in Fl. Trop. Afr. viii, 109. : DrscnrPTION.—.Slender, not very high, scandent (2-3 m., Heudelot). Sheathed stem about 1 cm. in diam, .Leaf-sheaths flagelliferous, partially and fugaciously furfuraceous greenish even when dry, longitudinally striate, slightly gibbous above, armed with scattered, solitary, very small, horizontal, semiconical spines. Leaf-sheath flagella filiform, slender, about 1 m. long, their lowest spathe flattened, scantily aculeate on the sharp edges; the succeeding spathes cylindraceous, more or less clawed; the apical portion nearly terete, armed with 2-3-nate claws. Ocrea conspicuous, externally produced into a ligule which is 2 cm. long, obliquely cut like the mouth of a beaked flute, entire, with smooth margins and sirgularly ornamented externally with closely seriated laminar, lacerated or comb-like, 4-5 mm. long spines. Leaves (not seen entire) about 60-70 ecm, long, not cirriferous; petiole short (7-8 cm. long) flattish on its upper face, acute at the sides, where armed with some straight horizontal spines which become hooked upwards (as in the first portion of the rachis), rounded on the lower face near the base, and armed there along the middle with a few strong, rather long (10-12 mm.), solitary, straight, somewhat deflexed, black-tipped spines, which rest on a large swollen base and gradually diéfthso: in length and are transformed into claws along the rachis; the rachis is flat and deeply channelled laterally (where are inserted the leaflets) in its first portion, and upwards is bifaced with an acute naked angle above; leaflets not very numerous, perhaps 18-20. on each side, very patent and sometimes horizontal in fully developed leaves, more or less irregularly grouped in fascicles Ann, Roy. Bor. Garp. Catcurra Vor. XI. 156 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. — [Q, Heudelotii of 2-3, occasionally 4, on each side, the fascicles of one side opposite or alternate with those of the other side and with short or long vacant spaces among them; in some the rather long portions are sometimes nearly equidistant ; they are all linear, lanceolate or lanceolate-ensiform, narrowed at the base and subulately acuminate into a hairy ciliolate tip, papyraceous, rather rigid, subshining, of about the same colour on both surfaces, but a little paler beneath, where they are glabrous, not scaly or dotted and with all the nerves faint and naked or with a few very small spinules along the mid-costa; the upper surface indistinctly 9-eostulate, or 1-costate with 1-2 slender nerves on each side of the mid-costa, which is rather acute and spinulous above; the side-nerves naked or sometimes very sparingly spinulous ; the transverse veinlets much interrupted and rather distant ; the margins slightly thickened by a fine nerve and finely spinulous-serrate ; the largest leaflets, those a little above the base, 25-28 cm. long by 12-16 mm. in breadth; the two extreme ones much smaller and quite free at the base. Male spadiz not seen. Female spadiz (not seen entire)—its axial unsheathed part very slender— strongly armed with very sharp, solitary or aggregate black-tipped claws which rest on a swollen light base; primary spathes tubular, elongate, closely sheathing, the lowest flattened aculeolate at the sides ; partial inflorescences short (15-17 cm. long), rising erect and then arched downwards, not callous at the axilla, with 5-7 spikelets on each side, secondary spathes elongate-infundibu- liform, smooth, obliquely truncate at the mouth and produced on one side into a triangular, acute, patent point; spikelets inserted just at the mouth of their own spathe, slightly callous at their axilla, arched and strongly recurved, short, sub- cylindraceous, 2-4 cm. long, bearing 3-7 flowers on each side; spathels infundibuli- form, narrowed at the base, smooth, truncate at the mouth; involucrophorum cupular, very obiiquely truncate, narrowing to the base and attached at the bottom of its own spathe where it is almost entirely enclosed, very acutely 2-keeled on the side next to the axis; involucre obliquely cupular, entire, rather deep; areola of the neuter flower concave, ovate, sharply defined. Fruiting perianth not pedicelliform, but explanate under the fruit. Female flowers about 4 mm, long; the calyx striately veined, 3-toothed, ultimately entirely split; the corolla one-third longer than the calyx, divided down into three narrowly ovate, nearly obtuse segments, smooth outside ; the stamens with filaments connate by their bases, elongate-triangular in the free part, anthers linear. Fruit ovoid, roundish at the base and gradually narrowing upwards into a conical beak, about 15 mm. long, 9 mm. thick; scales in 15-16 series, channelled along the middle, shiny, yellowish-brown with a darker rusty- reddish intramarginal liue, rather acute tip and erosely-toothed scarious margins. | Seed elongately-ovoid, about 9 mm, long, with an almost smooth surface and a not very deep chalazal fovea in the centre of the raphal face; albumen equable; embryo basal. : . Hasrrat.—Senegambia: abundant on the Islands Cayaye and Souloubolon of the River Gambia (Heudelot No. 372 in Herb. Webb, Kew and Delessert); River Gambia (Ingram in Herb. Kew.). O y A ce rdi g to a note (the copy of Heudelot’s original one ?) accompanying 2 specimen in Webb’s Herbarium at Florence, this is a plant not surpassing 2-3 m, in height. This specimen consists of the upper part of a leaf C. falabensis) BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 157 and of a small portion of a female spadix with only one partial inflorescence charged with nearly mature fruits. Ingram’s specimen is sterile and consists of the upper leafy portion of a stem. à The description of the leaf-sheath and of the ocrea is from a sterie specimen in the Herbarium at Kew, collected like those of Heudelot on the banks of the River Gambia. Heudelot's original note, with the date of 1836 annexed to his No. 372 in Delessert’s Herbarium, says that it is “a palm 2-3 m. high with reclining stems which are 3-4 cm. in diam. and furnished as well as the leaves with hooked spines. It was in flower and had fruit in March, Found on the Islands Cayaye and Souloubolon, where ‘it grows in such abundance'on the banks of the river (Gambia) as to render it very difficult to penetrate into the interior of those islands," : Heudelot's specimens are all very fragmentary, consisting of portions of the leaves and detached partial inflorescences with almost ripe fruit. — Prare 21.—Calamus Heudelotii Becc. The upper portion of a young plant (on the left side) from Ingram’s specimen in Herb. Kew; apex of a leaf and partial inflorescences with mature fruit from Heudelot’s No. 372 in Herb. Webb at Florence. 16. CALAMUS FALABENSIS Becc. sp. n. Description.—Scandent and apparently slender. Leaf-sheaths . . . » . Leaves not cirriferous; petiole. . . .; rachis in the upper portion flattish beneath, where armed with remote, small, black-tipped claws; leaflets inequidistant, inserted at a rather acute angle, some of them no more than 2 cm. apart with vacant spaces 4-7 cm. long, papyraceous, rather rigid, dull-green, slightly paler beneath, very narrowly lanceolate, somewhat narrowed to the base and from the lower third part upwards gradually acuminate into a very fine point, bristly spinulous at the sides, unicostate or sometimes sub-tricostulate ; the mid—costa rather slender, furnished with a few bristly brown rather distinct spinules on both surfaces; the secondary nerves are two on each side of the mid-costa, with an additional one on each margin—all are naked above and one on each side of the mid-costa is spinulous-bristly beneath; transverse veinlets fine, not very crowded and much interrupted; the largest leaflets (the lowest of the small portion of the leaf seen by me) 20 cm. long and 18-20 mm. in breadth; the upper rather suddenly shorter, the two terminal very narrow, free. at the base. Male spadiz elongate-flagelliform (not seen entire); the basilar axial portions between ‘wo partial inflorescences narrow and armed on the back with more or less aggregate claws; upper primary spathes very narrow, very long, tubular- cylindraceous, very closely sheathing, almost polished, but striate laterally, obliquely truncate at the mouth and extended at one side at the apex into a tri point; partial inflorescencés inserted above the mouth of their own spathe, not callous at their axilla, arising at first erect next to the axis, then nodding, in one specimen 35 cm. long with 12-13 spikelets on each side and ending in a cylindra- ceous tail-like appendix—about 4 cm, long, sheathed with unarmed spathes; secondary spathes tubular-infundibuliform, narrowed to the base, unarmed, finely striately veined, entire and obliquely truncate at the mouth, prolonged at one side into a rather elongate triangular point; spikelets attached near, but inside the mouth of their \ 158 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [6. Leprieurif own spathe, not callous at the axilla, somewhat arched and flexuose, spreading or recurved, flattish, 8-10 cm. long (the uppermost slightly shorter than the lower ones) with 19-20 flowers on each side; spathels infundibuliform, narrowed a good deal at the base, rather approximate and with the base of the one partly included in the one below, not very distinctly striately veined, entire at the mouth and extended at one side into a very acute and patent point; involucre cupular, rather deep, almost totally concealed in its own spathel, very obliquely truncate externally, very acutely two-keeled and almost two-winged, bi-dentate and deeply emarginate on the side next to the axis. Male flowers perfectly bifarious, inserted at a rather acute angle and half concealed by their respective spathels, elongate, very slightly falcate, very acute; the calyx tubular-cylindraceous, slightly striately veined, with three short acute triangular teeth; the corolla one-third longer than the calyx with acute segments, polished externally.—Other parts unknown. Hasirat.—Sierra Leone: in swampy places near small streams on the laterite plateau of Falaba (Scott Elliot No. 4460 in Herb. Berol). OsnsERVATIONSs.— Of this species I have seen only a specimen of one partial inflores- cence and the upper portion (35 cm. in length) of a leaf. This same specimen was regarded by Prof. O. Drude (l. c.) as belonging to €. Baríerii, from which however it differs in the different arrangement of the leaflets, which in C. Barterii are, as in the Asiatic C. gracilis, distinctly clustered into a few patent spreading groups, and besides are thin and almost herbaceous in texture. In (€. /falabensis the leaflets are simply inequidistant, rigid and firm in texture, and inserted at a rather acute angle. C. Heudelotii seems to me a nearer ally to C. falabensis than €. Barterii, but of C. Heudelotii the female plant only is known, while we have only the male of falabensis, and consequently it is difficult to make an exact comparison of the two; but in €. Heudelotit the leaflets have the nerves naked beneath and only the mid- costa is spinulous, while in C. falabensis 3 nerves are spinulous beneath. It differs from C. Leprieurit, of which the male spadix much resembles that of @. falabensis, in the faleate male flowers and in the more rigid and more inequidistant leaflets. Puare 22.—Calamus falabensis Bece. Partial inflorescence and apex of the leaf of the type-specimen in the Herbarium at Berlin, 17, Catamus Leprizurit Becc. in Rec. Bot. Surv. Ind. ii, 200. Description.— Very probably scandent and not of large size. Stem . . . Leaf-sheaths . . . . . Leaves short, 35-40 cm. long, not cirriferous ; patiolé® 7-10 em. long, rather thick, subcylindric and longitudinally wrinkled (when dry), more or less covered with greyish, removable, furfuraceous scurf, armed beneath and at the sides with some strong, straight, horizontal, dark-tipped spines; rachis more or less furfuraceous as is the petiole, rigid, and relatively thick, acutely bifaced above, roundish beneath, where armed one. the middle with solitary, straight, horizontal or slightly deflexed spines, which change into small claws towards the apex; leaflets rather many and crowded, inserted at an angle of about 45°, more or less inequidistant or interruptedly equidistant, being sometimes, mainly near the apex, divided by short vaeant spaces into 2-3 groups, where each leaflet is regularly about 15 mm. apart; furthermore the leaflets. are papyraceous, narrowly lanceolate, somewhat attenuate to C. Leprieurii ) BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 159 the base, gradually subulately acuminate at the apex, opaque, slightly paler beneath than above, their mid-costa sparingly spinulous on both surfaces or smooth beneath ; the secondary nerves slender and naked; the transverse veinlets sharp, much interrupted; the margins rather closely spinulous, not or very slightly thickened by a marginal nerve; the largest 15-17 em. long, 15 mm. broad, the upper ones shorter ; the two of the terminal pair quite free at the base. Male spadiz flagelliform, very elongate, simply decompound, the one seen (wanting the base) | m. long and in addition produced into a rather long finely clawed flagellum; primary spathes very narrow-tubular, cylindraceous or somewhat flattened, strictly sheathing, chartaceous, sparsely aculeolate, obliquely truncate at the mouth and produced at one side into a triangular acute point ; partial inflorescences terminating in a caudate sheathed unarmed appendix, in the spadix seen by me 5 in number, of which the largest, the lowest, is 30 cm. long with 7-8 spikelets on each side, the others shorter and with fewer Spikelets; secondary spathes elongate-infundibuliform, unarmed, finely striately veined, obliquely truncate and naked at the mouth and produced at one side into a broad triangular acute point; spikelets attached a little below the mouth of their own spathey narrower at the base than above (with the spathels there very closely imbricate, flowerless and smaller than higher up), flattened, arched, spreading, the largest 6 em. long with about 12 flowers on each side, which are erecto-patent and half immersed in their own spathels; spathels very asymmetrically ^ infundibuliform, subspathaceous, entire, truncate and naked at the mouth; finely striately veined, pro- duced at one side into a rather acute tip; involucre dimidiately cupular or shaped like a swallow's nest, anticously truncate, posticously deeply lunately emarginate, acutely two-keeled and bi-dentate next to the axis. Male flowers narrowly ovate, obtuse, slightly curved or asymmetric; the calyx striately-veined, with 3 broad lobes; the corolla one-third longer than the calyx—its segments polished externally, apiculate or almost obtuse. Female spadiz elongate with rather many remote (about 40 cm. apart) partial inflorescences; primary spathes as in the male spadix, the lowest about 25 em. long, split longitudinally on the ventral side and acutely bicarinate on the back, the carin; armed with small remote ascendent spines; axial portions of the spadix (between two partial inflorescences) concave on the inuer side at their base and convex and sparingly clawed dorsally; partial inflorescences very elongate, 30-50 cm. long, terminated by a short, filiform, unarmed, tail-like appendix, the largest with 12 spikelets on each side; secondary spathes elongate-infundibuliform, unarmed and almost polished, truncate at the mouth and extended on one side into a broadly triangular and rather acuminate point; spikelets inserted a little inside the mouth of their respective spathes, rather thick, arched and recurved, the largest, the lowest, 10-20 cm. long with 10-18 flowers on each side, the upper ones gradually decreasing in length and number of flowers, the uppermost half the length of the lower ones; spathels infundibuliform, unarmed, indistinctly striately veined, truncate and entire at the mouth and prolonged at one side into a short triangular ultimately withered spreading point; involucrophorum inserted at the bottom of its own spathel, subspathaceous or obliquely cupular, acutely bicarinate, bi-dentate and emarginate on the side next to the axis, almost entirely enclosed in its own spathel; involucre unilaterally cupular, rather deep, my with the areola of the sterile flower ovate, sharply defined, Female flowers about 5 mm, long, ovate. Fruiting pertanth not 160 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. Perrottetii pedicelliform, split down to the base into 6 equal lanceolate strongly striated lobes (3 of the calyx and 3 of the corolla). Fruit (immature) ovate, conically narrowed at the apex; scales in 18 series, yellowish near the base and with a broad intramarginal sharply defined band and a scarious brown fimbriate-denticulate margin, Seed . . » . . —The spikelets and flowers acquire in drying a chocolate-brown colour. Hasar, —West Tropical Africa: Senegambia, Leprieur. OssrRvATIONS.—Of this species, which has remained more than 70 years unnamed in Herbaria and has not been found again by modern travellers, I have seen in the Paris Herbarium a specimen of a mule spadix, accompanied by two entire leaves, labelled: *Senegambia—M. Leprieur, 1830—Herbier d'Adrien de Jussieu, donnó au Muséum par ses enfants en 1857," and another specimen, apparently of the same gathering and with immature fruit, in Delessert’s Herbarium at Geneva. A third specimen, consisting of two partial female inflorescences and one leaf, is preserved in the Leyden Herbarium and was also gathered by Leprieur in Senegal. The leaflets of the female specimen in the Leyden Herbarium have a small and short but relatively strong spinule at their base on the upper side next to the rachis. This spinule is scarcely visible in the two leaves of the Paris specimens, which leaves are respectively 35 and 40 em, long and are very similar to that of the above-mentioned Leyden specimen, but in one the leaflets are distinetly grouped, while in the other they are almost equidistant and with short vacant spaces only near the apex of the leaf, The mid-costa is sparingly spinulous above and quite smooth. beneath in both, PLATE 23.—Calamus Leprieuri Bece. An entire leaf (undersurface); male spadix apparently almost entire. From Leprieurs specimen in the Paris Herbarium, PLATE 24.—Calamus Leprieurii Bece. Leaf and portion of a female spadix with very young fruit, From Leprieur’s specimen in the Leyden Herbarium. 18. Carawus Perrotretit Becc, in Rec. Bot, Surv. Ind. ii, 200. Description.—Slender, scandent. Leaf-sheaths armed with scattered laminar rather small, 8-10 mm. long, black-tipped spines. Ocrea rather elongate, furfuraceous horizontally truncate, very sparingly spinulous at the base and not on the etes side. Leaves (in one specimen) 60 cm. long, not cirriferous; petiole rather long (18 em.) rather thick, subcylindric and wrinkled longitudinally when dry; armed irregularly all round with some straight, horizontal, rather strong, 1 cm, long, dark- tipped spines; rachis more or less fugaciously furfuraceous as is the petiole, rigid and relatively thick, acutely bifaced and smooth above, roundish beneath, where armed along the middle with solitary, straight, horizontal or slightly deflexed spines, which change into small claws towards the apex; leaflets rather many and crowded inserted at an angle of about 45°, interruptedly equidistant, viz. divided te groups by short vacant spaces, but equidistant in each group, papyraceous, linear- lanceolate, somewhat attenuate towards the base, gradually subulately acuminate io the apex, opaque, slightly paler beneath than above, their mid-costa acute and sparingly spinulous above or smooth on both surfaces; the secondary nerves slender and naked; the transverse veinlets sharp, much interrupted; margins rather closely spinulous; the largest leaflets, those a little above the base, 18-20 cm. long, 12-14 C. Perrottetii] BEOCARI MONOGRAPH OF THE GENUS CALAMUS. 16t mm. broad, the uppermost shorter, the two of the terminal pair free at the base, Male spadiz ... . . . Female spadie apparently as in C. Leprieurii ; primary spathes . . . . .; partial inflorescences elongate, with many distichous spikelets on each side and terminating in a short, sheathed, unarmed, tail-like appendix ; secondary spathes elongate-infundibuliform, unarmed, finely striately veined, entire and obliquely truncate at the mouth and extended at one side into a broadly triangular, acute point; spikelets inserted just at the mouth of their respective spathes with a rather distinct axillary callus, rather thick, arched and recurved, the largest (the lowest in a portion of an inflorescence) 7 em, long, with 10 pairs of female flowers on each side (as it seems that each spathel subtends two equally well-formed female flowers) ;. the uppermost spikelets half the length of the lower ones; spathels approximate, broadly infundibuliform with a very narrow base, striately veined, extended at one side into a broadly triangular, acute or acuminate, erect, amplectent point; involucrophorum almost entirely immersed in its own spathel and attached at the base of the one above, sub- spathaceous, enveloping the neuter flower, acutely bi-carinate, bi-dentate and deeply emarginate on the side next to the axis; involucre deep, cupular, unilaterally evolute, sub-auriculiform, immersed in the involucrophorum ; areola of the neuter flower very conspicuous, ovate, concave, with raised and often winged borders. Female flowers ovate, 5 mm. long; the calyx striately veined, cleft into 3 concave, ovate, acute parts ; the segments of the corolla concave, acute, ovate-lanceolate, opaque, striately veined externally, slightly longer than the calyx. Neuter flower apparently not: differing from the female one, which is in the usual position. Hasrrar.—Senegal: at the mouth of the River Casamance. OssERvaTIONs.—In the year 1902 I had given the name of C. Perrottetii to a Calamus preserved in the Herbarium Delessert at Geneva and collected by Perrottet (No. 761) iu the damp forests of the west coast of tropical Africa on: the 10th of April 1829, near the village Sedhiou on the river Casamance in Senegal. Later I have received another specimen of this same species from Dr. Schweinfurth and collected by Leprieur in 1826, also on the river Casamance near the village of Montsor at Cape Rosso. This last specimen consisted of the apex of a stem with a portion of the leaf-sheath and an entire leaf, and of the apex of a partial inflorescence 20 cm. in length with 6 spikelets on each side, This is the specimen I have described. C. Perrottetii is extremely like C. Leprieurii, but its spikelets have a peculiar facies on account of the large, broadly infundibuliform, spathaceous spathels which embrace the flowers; the leaflets have not at their base the spinules so often seen in C. Leprieuríói and are more elongate than in this last. In C. Lepricurii the companion neuter flower at each spathel is always sterile, whilst in C, Perrottetii the two flowers during anthesis seem perfectly alike, but I have seen no spikelets after fertilisation. Nevertheless €. Perrottetii must be considered as a rather doubtful species, and must be compared again with C. Leprieurüi when more complete materials have come to hand, Prate 25.—Calamus Perrottetii Bece. The entire Perrottet’s type-specimen in Herb. Schweinfurth. Ann. Roy. Bor. Garp. Catcurra Vor. XI. 162 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (C. Schweinfurthii 19, CALAMUS AKIMENSIS Becc.: sp. n. DEscRIPTigy.—Apparently scandent. Siem . ` . . . Leafsheaths . . . .. Leaves . . . . . Female spadiz: partial inflorescences rather large, in one specimen 50 em. in length, elongate-pyramidal in outline, with 15 gradually shortening spikelets 'on each side; secondary spathes infundibuliform, usually split longitudinally in their} upper part and prolonged at one side into a rather elongate-triangular acu- minate point; spikelets thick, vermicular, inserted near the mouth, but inside their own spathe ; the lower ones, the largest, 15 cm. long with about 25 flowers on each side, slightly sinuous; the upper ones gradually shorter and with fewer flowers, strongly arched; those near the apex 7-8 cm. long with 7-8 flowers only on each side; spathels finely striately veined, very broadly and obliquely infundibuliform, extended at one side into a broad triangular point; involucrophorum cupular, bi-dentate and acutely two-keeled on the side next to the axis, inserted at the bottom of its owr spathel and entirely included in this; involucre entire, subauriculiform or obliquely cupular, viz. more elongate on the side of the neuter flower of which the areola is very distinct, vertically elougate and with a very acute margin. Female flowers about’ 5 mm. long. Fruiting perianth split and explanate under the fruit; the segments of the corolla lanceolate, acuminate, about as long as the lobes of the calyx and slightly narrower than these. Fruit conically ovoid from a round base or gradually tapering towards the apex into a conic and rather thick beak, about 2 cm, long and 1 cm. broad; scales in 15 series, shining, broadly and not deeply channelled along the middle, light-brown with a rusty-red irregularly fringed margin, and an acute point. Seed narrowly oblong, round at the base, somewhat apiculate at the apex, li mm. long, 6 mm. thick, coarsely, irregularly and superficially grooved on the surface, its chalazal fovea elongate on the centre of the raphal side; albumen equable; embryo basal. l | HanrraT.—Discovered in December 1899 by W. H, Johnson at Kibbi in the Akim district of the Gold Coast (Herb. Kew). OssERvATIONS.— Of this species nothing is known beyond the partial inflorescence with mature fruit described above. Closely related to C. deerratus, but distinct by its larger partial inflorescences with numerous spikelers, which are also larger with broader or more spathaceous spathels. The fruit is longer or more grad ually narrowed into a conic beak and with the scales in 15 series (in €. deerratus they are in 21) with the margins coarsely and irregularly (not very finely) fringed. By its subspatha- ceous spathels it resembles also C. Perrottetii a good deal, ia Prate 25A.— Calamus akimensis Becc. The entire type-specimen in Herb. Kew. 20. CALAMUS SCHWEINFURTHII Becc. in Ree. Bot. Surv. Ind. ii. 200. T C. secundiflorus (not of Beauv.) Schweinf. Beitr, Fl. Aethiopiens, 291; Drude in Englers Bot. Jahrb. xxi. 131 (1896). Description.—Scandent, slender. Sheathed stem 10-15 mm. in diam. Leaf-sheaths elongate, cylindrical , armed with small, scattered, solitary, deflexed, flat, laminar subulate, blackish, shining spines, which rest on a small, tubereuliform, light base ud are about 1 cm. in length. crea liguliform, 2-3 cm. long, prolonged externally and C. Schweinfurthii] BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 163 obliquely cut like the mouth of a beaked flute, membranous, dry, bristly spinulous along the middle of the ventral face. Leaves not cirriferous, those of the upper part of the adult plant 1°2 m. in length including the petiole; this 25-30 em. long and 6-7 mm. broad, finely longitudinally striate, flat and smooth above, the margins. acute and feebly armed near their base with slender, straight, needle-like, black-tipped spines and upwards with a few small claws, rounded beneath ; rachis remotely clawed beneath along the middle as is the petiole—the claws extending, solitary and rather small, to its apex; smooth, acute and bifaced above; leaflets rather numerous, about 20 on each side, inequidistant, usually 15-20 cm. apart, but sometimes interrupted by 2-3 longer vacant spaces, linear-ensiform, somewhat narrowing to the base, gradually attenuate towards the apex into a subulately acuminate spinulous-ciliolate point, thinly chartaceous, almost shining, about the same colour on both surfaces, but slightly paler beneath; the mid-costa acute, sparingly’ bristly-spinulous near its apex above, Jess prominent but more spinulous beneath; the side nerves all slender, one on each side of the mid-costa furnished on the upper surface with a few, rigid, dark, bristly spinules; on the under surface 2 and sometimes + secondary nerves more or less. sprinkled (as is the mid-costa) with spinules shorter than those of the upper surface ; transverse veinlets very distinct, much interrupted; margins minutely and closely spinulous ; the bristles as well as the spinules on the nerves and on the margins ` have all a dark-brown point and a light bulbous base; the largest leaflets, those near the base, 30-38 cm. long, 15-20 mm. broad; the uppermost shorter, less acuminate, the two of the terminal pair quite free at the base; the leaves of young shoots in not fully-grown plants are shorter, have the sheaths covered with a thin rusty-furfuraceous indumentum, the liguliform ocrea shorter and smooth, the petiole even 50 cm. long, subterete and armed with longer straight spines, the leaflets shotrer and relatively broader, subequidistant and less spinulous on the nerves. Male . padit- . . . Female spadix (not seen entire); partial inflorescences with many approximate distichous spikelets; the largest I have seen 35 cm. long with 13 spikelets. on each side, terminating in a short, rather thick, unarmed, sheathed, tail-like appendix; other inflorescences, which probably are from the upper part of the spadix, are much shorter and with fewer spikelets; secondary spathes rather short, infundibuliform, unarmed, thinly coriaceous, polished, truncate at the mouth, usually longitudinally split, prolonged at one side into a broadly triangular, acute point ; spikelets thick, rigid, strongly arched and deflexed, attached inside the mouth of their respective spathes; the lower ones the largest, about 7 cm. long, with 14-15 flowers ‘on each side, the uppermost slightly shorter; spathels approximate, broadly infundibuliform, thinly coriaceous, not distinctly veined and almost polished, horizontally truncate and entire at the mouth, shortly extended at one side into a triangular point; involucrophorum almost entirely immersed in its own spathel and attached to the base of the one above, cupular, bi-dentate and acutely two-keeled on the side next to the axis; involucre irregularly cupular, unilaterally evolute, sub-auriculiform, polished and smooth internally, rather thick and subcoriaceous in texture; areola of the neuter flower large, ovate concave, with sharp raised borders. Female flowers about 5 mm. long, the corolla very slightly longer than the calyx. Fruiting perianih not pedicelliform, split into 6 almost equal, ovate-lanceolate, acute parts. Fruit ovoid, rounded at the base, conically narrowed at the apex or slightly. contracted into a Ann, Ror. Bor. Garp. CarcurrA Vor. XI. 164 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (C. Schweinfurthii broad conic rather obtuse point, 16-18 mm, long, 11-12. mm. broad; scales in 15-18 series, shining, deeply channelled’ along the middle, yellowish-brown with a darker, narrow, rusty-reddish, intramarginal line, the tip rather acute and, like the margins, erosely toothed. Seed ovoid or oblong, 10-12 mm. long, rounded to both ends or slightly apiculate, convex and obsoletely furrowed longitudinally on the back, slightly flattened, with an elliptic, not very deep central chalazal fovea on the raphal side; albumen bony, equable; embryo basal. Hasrrat,—Central Africa: discovered by Dr. Schweinfurth in the Niam-Niam country at Mansilli (No. 2860, 7th February 1870) and at Nabambisso, 6th May 1870 (No. 3703); at Lado on the White Nile, collected in fruit by Dr. Emin Pasha, 1883 (Herb. Schweinfurth); Ussororo, collected by Dr. Stuhlmann (Emin Pasha. Exp., 1891, No. 2531 in Herb. Schweinfurth), OssrRVATIONS.—I have been able to write an almost complete description of this, Dr. Schweinfurth having most liberally placed in my hands all his specimens of Calamoid palms from Central Africa, which apparently belong to only one species (that described above) and to an Eremospatha, apparently new but allied to E, Hookerit (E. Schweinfurthii Becc.; 3 Schweinfurth No. 3675). Schweinfurth's — specimens of C. Schweinfurthi are of a few entire leaves detached from full-grown plants and of the terminal portion of some young leafy shoots ; while those of Dr, Emin Pasha consist only of some partial inflorescences with ripe fruit without leaves. There is not therefore any absolute evidence that the leaves described by me and the fruit belong to the same species, still I have little or no doubt about it, as- both fruit and leaves are very similar to the corresponding parts of the other true African Calami and especially to OC. deerratus and C. Heudelotii. , €. Schweinfurthii differs from C, deerratus in its leaves having a very long petiole and in its fruit having larger not fimbriate scales; from C. Heudeltii in its larger and thicker spikelets with broadly infundibuliform spathels and in the different orna- mentation of the ocrea. The fruit and the seed in the three mentioned Species are very similar; certainly they are very nearly allied species. I first assigned the name of C. Schweinfurihi in 1892 to a specimen (Schweinf. No. 2860) which I had seen in the Herbarium at Kew. Now Prof. O. Drude in a paper on the Palms of Tropical Africa in Engler's “ Botanische Jahrbücher xxi, 1896," mentions my C. Schweinfurthii and insists on referring it to Ancistrophyllum secundiflorum, basing this opinion on the supposition that all non-cirriferous leaves in Calamus must belong to young plants, and adding that these leaves ought not to be collected and preserved in good collections ; but it is quite certain that entire sections of Calamus never have cirriferous leaves, and to this class belong all the African Calami known to me. Moreover, the leaflets of Ancistrophyllum are slightly sigmoid, while in @, Schweinfurthii, as in all true Calami (African or Asiatic), the leaflets are straight, The diagnostic characters of O. Schweinfurthii are the elongate leaves with the petiole very long, flat above and round beneath; the numerous inequidistant, not distinctly fascicled, narrowly ensiform leaflets: the ocrea prolonged externally and spinulous on the ventral face; the female spikelets thick, with broad infundibuliform C. pachystemonus] ^ BECCARL MONOGRAPH OF THE GENUS CALAMUS. 165 spathels; the fruit scales broad, in 15-18 longitudinal series, with narrow erosely toothed margin. | Pirate 26.—Calamus Schweinfurthii Bece. Basal portion of leaf from a young plant (Sehweinfurth No. 3703) on the right hand side; basal portion and apex of a leaf from an adult plant (Schweinfurth No. 2860); partial inflorescence ‘with mature fruit from a specimen collected by Emin Pasha ( Herb. Schweinf.); seed from the dorsal side; seed longitudinally cut through the embryo. 21. CALAMUS PACHYSTEMONUS Thw. Enum. Pl, Zeyl. Addenda, p. 431; Hook. f. Fl. Brit. Ind. vi, 442; Becc. in Rec. Bot. Surv. Ind. ii, 201. €. gracilis (not of Roxb.), Thw. l.c., 330. DzscRiPTION,—-Seandent, very slender. Sheathed stem about 7 mm. in diam, Leaf-sheaths slightly gibbous above, sparingly armed with straight, solitary, scattered, unequal, very short and conical spines, which are occasionally 5-8 mm. long, horizontal and subulate. Ocrea in full-grown leaves very short, horizontally truncate, unarmed, Leaves short, 30-40 cm. long, not cirriferous, pauci-jugate; petiole very short, 3-4 cm. long, sub-terete, obsoletely channelled above, more or less armeg beneath and at the sides with scattered unequal, mostly conical, straight, very short spines; rachis slightly furfuraceous, acute and bifaced above, more or less convex below where aculeate throughout up to the apex mainly along the middle; the lower aculei usually straight, the upper ones shorter and hooked or transformed into claws: Leaflets very few, 2-3 ‘on each side, . with. a terminal pair, the side-leaflets irregularly set, 3-7 cm. apart, chartaceous, rigidulous, concavo-convex, almost shining above, slightly paler beneath, lanceolate, ovate-lanceolate or lanceolate-elliptic or oblanceolate or sub-obovate, 12-18 cm. long and 8 cm. broad (the lowest the smallest), tapering towards and acute at the base, rather suddenly narrowed upwards into an acuminate and bristly tip, and furnished with 3-5 fine and acute costæ which run from the base up to the apex with minor nerves between them, all naked above ; the mid-costa stronger than the side ones, usually furnished with very few, short, erect spinules beneath, where the other nerves are smooth; margins acute, smooth (not ciliate or spinulous), the lower one usually bordered on the upper surface with a rather broad, brown, polished band which is occasionally accompanied by a few others running along the main nerves; the two terminal leaflets larger and broader and more suddenly acuminate than the side ones, connate up to about the middle. Male spadiz as in C. digitatus, with very few partial inflorescences (only two in the spadices seen) and prolonged into a slender aculeolate flagellum; primary spathes very narrow and long, cylindrical, strictly sheathing as in C. digitatus; partial inflorescences 10-14 cm. long, narrow, dense, of equal breadth at the base and at the apex, ascendent at first then arched and nodding, inserted inside their own spathe; secondary spathes tabular- infundibuliform, finely striate, produced at the apex at one side into a deflexed point; spikelets short, 1-1-5 cm. long, strongly arched downwards or subscorpioid, inserted at the mouth of their own spathe with a distinct axillary callus and bearing 5-15 very closely packed flowers on each side; spathels scale-like, concave, strongly veined, acute; involucre cupular, shorter than the spathels, obscurely and broadly 3-toothed, strongly veined. Male flowers slender, cylindrical, 5 mm. long and 2 mm. thick, slightly curved, obtuse and somewhat callous at the top; calyx campanulate, strongly striately 166 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. digitatus veined, with: 3- short. broad acute lobes; corolla twice as long as the calyx, divided down to the lower third of its length into 3 linear, callous, apiculate segments, which are strongly longitudinally striato outside; stamens in two series of unequal length, their filaments united as high as the middle of the corolla, in the free portion thick near the base, subulate and not inflected at the apex; anthers sagittate-lanceolate, acute, the connective and the filament black when dry; rudimentary ovary very small, formed by three small approximate clavate bodies which are shorter than the filaments. Female spadiz and fruit unknown. Hasrrat.—Ceylon: in the neighbourhood of Galle, Thwaites. I have seen a good specimen of this species in the Herbarium at Paris, gathered at Caitura by Leschenault in July 1820. OBSERVATIONS.—The specimens of the ‘Ceylon Plants,” which exactly agree with the description of C. pachystemonus of Thwaites, bear the No, 2334 and were prepared from plants originally found in the jungle near Galle and introduced into the Botanical Garden at Peradeniya; but with the same No. 2334 specimens of C, digitatus have been also distributed. And indeed C. pachystemonus is closely related to @. digitatus with which it has been amalgamated by Thwaites, but from which it is easily distin- guishable by the pinnate leaves, as described above. The male flowers are also larger n €. pachystemonus than in €. digitatus, and are callous at the top, with the filaments ‘of the stamens thicker, shorter and more agglutinate. C. pachystemonus, C. digitatus and C. radiatus form a small group, peculiar to Ceylon , distinguished chiefly by the elongate curved flowers with biseriate stamens and filaments. not inflected at the apex and with erect anthers when in the bud. Leschenault’s specimen has the leaf with two leaflets on each side of the rachis “besides the terminal pair; the leaflets have not the small spinules on the mid-costa on the lower surface, as in Thwaites’s specimens. Prate 27.—Calamus pachystemonus Thw. An entire leaf seen from the lower surface; the base of another seen from the upper one and an entire male il from Thwaites’s specimen in Herb. Kew. 99. CaraMUs DiIGITATUS Becc. in Hook. f. Fl. Brit. Ind. vi, 442; Becc. in Rec. Bot. Surv. Ind. ii, 201. €. pachystemonus (partly) Thw. Enum. Pl. Zeyl. 431. Description.—High scandent, very slender, Sheathed stem 5-8 mm. in diam. Leaf- sheaths. very slightly gibbous above, more or less armed with subulate, elongate or short, slender, horizontal spines or sometimes almost unarmed or only transversely rugose, Ocrea smooth or spinulous, at first liguliform, 5-7 mm. long, very soon withered and deciduous, therefore apparently short and very obliquely cut off. Leaves very short with very few sub-digitate or indistinctly sub-pinnate leaflets; petiole subterete channelled above, 7-8 cm. long, sometimes almost unarmed, usually furnished near the base laterally with some straight, rather strong, ascendent spines and armed below with a few distant irregular claws of which a few sometimes also appear on the very short rachis; leaflets mostly only two, sometimes 3-4, but whatever be their number the two of the terminal pair more or less confluent at the base C. digitatus] BECCARL. MONOGRAPH OF THE GENUS CALAMUS. 167 and the side ones, when these are present, quite free and very approximate to the terminal pair, or exceptionally 15-20 mm, apart, and . therefore sub-pinnately et; furthermore the leaflets are oblong-spathulate, or oblanceolate, slightly narrowed at. the base, enlarged upwards, where somewhat convex above and very suddenly: contracted into a short bristly-brushed tip, 20-30 cm. long and 3-65 cm. broad (the lower ones usually slightly narrower than the upper ones), chartaceous, rigidulous, shining and acutely 3- or exceptionally in the terminal leaflets 5-costate above (the mid-costa the strongest), with intermediate, often rather prominent, secondary coste and other minor nerves; all nerves naked on both surfaces; margins acute, smooth ; the lower margin hated with a polished band as in C. pachystemonus ; transverse veinlets slender, sharp and crowded, Male and female spadices very much the same, very slender, 1-1°5 m. long, flagelliform, terminating in a very slender, filiform, aculeolate appendix with a callous swelling at their insertion and a transverse rima in their upper axilla, simply decompound, with 2-6 partial inflorescences; primary spathes tubular, very closely sheathing, very narrow, the lowermost flattened, spinulous near the base, the upper ones cylindrical, aculeolate, truncate at the mouth, acute or acuminate at one side and often split at the apex; partial inflorescences straight, elongate, 8-15 cm. long, narrow, dense, bearing distichously. 10-20 short approximate spikelets; secondary spathes tubular infundibuliform, glabrous, striately veined, longitudinally truncate and not ciliate at the mouth, apiculate at one side; spikelets inserted just at the wouth of their own spathe, arched or subscorpioid and strongly deflexed, with a distinct callus and a transverse rima in their upper axilla. Male spikelets 8-10 mm. long, with 5-8 (seldom more) flowers on each side; spathels very crowded, concave, scale-like, broadly-ovate, acute; involucre dimidiately cupular, obliquely truncate, flat and two- keeled on the side next to the axis. Male flowers very closely packed, slender, — cylindrical, curved or subfaleate, rather obtuse, 4 mm. long, 1 mm. thick; calyx campanulate, strongly striately veined with 3 short, broad, acute lobes; corolla two and a half or three times as long as the calyx, divided down almost to the base into 3 linear, acute, striate segments; stamens. 6, arranged in two series, 3 longer than the others, their filaments thickened at the base, subulate and not inflected at the apex; anthers lanceolate-sagittate, dorsally attached (erect and not versatile during the anthesis ?) ; their connective perfectly black when dry: rudimentary ovary formed by 3 small clavate bodies which are shorter than the filaments. Female spikelets larger than the male ones, the largest 2 cm. long, with 8-10 very approximate flowers on each side; spathels very short and broad, subspathaceous, strongly striately veined, acute at one side; involucrophorum obiiquely infundibuliform, truncate, almost completely sunk in its own spathel and attached to the base ot the one above; involucre eupular, rather deep, entire, obliquely truncate; areola of the neuter flower very «large, broadly ovate, acute, deep, sharply defined by a raised border, Female flowers ovoid, acute, about 3 mm. long; ealyx divided into 3 ovate, acute, striately-veined lobes; corolla about one-third longer than the calyx, its segments striate, lanceolate, acute; stemens with filaments united at the base and dentiform in the free part. Neuter flowers scarcely smaller than the fertile ones. Fruiting perianth explanate under the fruit, not pedicelliform. Fruit globular, 9-10 mm. in diam., very shortly mucronate ; scales in 12 series, distinctly longer than broad, faintly channelled along the middle, N 168 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (C. radiatus yellowish, neatly bordered with a narrow reddish-brown band which is a little larger towards the shortly prolonged, obtuse, dentieulate tip; their margins erosely denticulate. Seed subglobose, 6:5 mm. long, with a slightly depressed chalazal fovea on the raphal side, otherwise with even surface; albumen equable; embryo basal.— One fruit had two seeds, which were flat on the ventral face and convex on the back. Hasrrat.—Ceylon, in the southern part of the Island. Distributed by Thwaites with the same number as C. pachystemonus (C. P. No. 2334). It was first discovered by Major-General Walker, according to a specimen in the Kew Herbarium. Another specimen collected by Gardner is in Webb's Herbarium at Florence.—Singalese name ‘ Kookool-wel.' | OnssERvaTIONS.— Very closely related to C. pachystemonus (see observations on} this species) and C. radiatus. From the last it differs in the fewer, broader and many- nerved leaflets, and in the fruit with scales in 12 instead of 15 series. Puare 28.—Calamus digitatus Bece. Fruiting specimen from St. Petersburg Herbarium. PLATE 29.—Calamus digitatus Becc. Female specimen in flower (on the right- hand side) from a specimen in De Candolle’s Herbarium ; male specimen in flower on the left side from Webb’s Herbarium. 99. CaLamus RADIATUS. Thw. Enum. Pl. Zeyl. Addenda, 431; Hook. f. Fl. Brit. Ind. vi, 442; Becc. in Rec. Bot. Surv. Ind. ii, 20, Descrrption.—High scandent, very slender, Sheathed stem 5-7 mm, in diam. Leaf.sheaths sometimes flagelliferous, not distinctly gibbous above, more or less densely covered with straight, elongate or short, subulate, slender, horizontal spines, which are solitary . or confluent by their broad bases and subseriate, longer, more numerous and pointing upwards near the mouth, Ocrea short and obliquely truncate in full-grown leaves, Leaves not pinnate, but with 5-8 digitate or radiate leaflets grouped at the apex of the petiole; petiole 5-7 em. long, subterete, narrowly channelled above, sparsely spinulous throughout and sparingly clawed on the back, leaflets rigidulous, chartaceous, about 25 cm. long, and 15-22 mm. broad, the two of the centre united at the base, all about of the same length, very broadly linear (the outer usually narrower than the central ones), shortly attenuate at the base, very suddenly contracted. at the apex into an acuminate and very sparingly bristly-spinulous tip, shorter in the centre leaflets, shining above, paler beneath, with the mid-costa slender but very acute and with 3-4 slender secondary nerves on each side of it; transverse yeinlets slender, sharp, rather crowded, much interrupted; margins smooth; the mid-costa and nerves smooth on both surfaces. Male spada . . . . . Male flowers (as described by Thwaites) cylindraceous-falcate; calyx twice as long as broad, with 3 short, acute, triangular lobes; corolla about three times longer than the calyx, divided down almost to the base into 3 linear, acute segments; stamens with filaments thickened in the lower half. Female spadiz simply decompound, inserted near the mouth of the leaf-sheath with a distinct basal £. ruvidus } BECCARI. MONOGRAPH OF THE GENUS CALAMUS, 169 callus and a transverse axillary rima, flagelliform, very slender, with very. few partial inflorescences (2-6, Thwaites), and prolonged into a filiform minutely and densely clawed flagellum; primary spathes tubular, very narrow and long, very closely sheathing, obliquely truncate at the mouth, rather densely armed with scattered small claws; the lowest flattened, the upper ones cylindraceous; partial inflorescences small, delicate, 5-8 cm, long, inserted at or above the mouth of their own spathe with 2-4 spikelets on each side; secondary spathes very narrowly tubular, slightly enlarged above, glabrous, longitudinally striately veined, truncate and apiculate at one side at the mouth; spikelets inserted just at the mouth of their own spathe with a callus and transverse rima -at their upper axilla, very short (1-2°5 em.) distichous, strongly deflexed, the largest—the lowest—with 6-7 flowers on each side; the upper somewhat smaller and with fewer flowers; spathel obliquely infundibuliform, much narrowed at the base, apiculate at one side at the mouth where some strong nerves converge; involucrophorum shortly infundibuliform, obliquely truncate, attached at the base of the spathel above its own; involucre deeply cupular or sub-infundibuliform, entire, obliquely truncate at the mouth, strongly striately veined, callous at the base; areola of the neuter flower broadly ovate or nearly round, with a very sharp border. Female flowers ovoid- acute, about 3 mm. long; calyx acutely trilobate; corolla divided into 3 lan- ceolate, acute segments, one-third longer than the calyx; calyx and corolla strongly striately veined; stamens with the filaments united by their bases, elongately triangular in the free portion. Fruiting perianth explanate (not pedicelliform). Fruit globular, 10-11 mm. in diam., supported by the somewhat pedicelliform involucre and tipped by a distinct mucro; scales in 15 series, distinctly broader than long, faintly channelled along the middle, yellowish, sharply bordered with a narrow reddish-brown band which is a little broader towards the shortly ‘ prolonged, obtuse, denticulate tip; margins finely erosely dentieulate, Seed irregularly globose, about 8 mm. long, with almost even surface; chalazal fovea small, slightly depressed on the raphal side; albumen equable; embryo basal, Hasirat.—Ceylon : in the southern districts of the Island (Walker in Herb. Kew; Thwaites C. P. No. 3805).—Singalese name ‘ Kookool-Wel.’ OBSERYATIONS.—À very elegant species remarkable among all Asiatic Lepidocaryee by its digitate leaves, resembling those of the American Lepidocaryum, It is also easily distinguished from the two related species, C. pachystemonus and C. digitatus, by the unicostate and peculiarly arranged leaflets, Puate 30.—Calamus radiatus hw. Portion of a plant with a female spadix in flower, from Thwaites’ No. 3805 in De Candolle’s Herbarium. 24, CaLaMUs ruvipus Becc. in Rec. Bot. Surv. Ind. ii, 202. Descriprion.—Scandent. Leaf-sheaths . . » + + Leaves not cirriferous; petiole . » + » +, rachis acutely bifaced and smooth above, armed beneath throughout with rather strong, solitary, scattered claws; leaflets very few, remote, alternate, oblanceolate, rather concave beneath, somewhat suddenly acuminate, long-attenuate at the base, chartaceous, rigid, about the same colour perfectly glebrous and Any. Roy. Bor. Garp. CarcurTA Vor. XI. 170 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (CC. ruvidus. shining on both surfaces, 28-30 cm. long and 3:5 cm. broad, with 5- costa, Which are acute above but prominent also on the lower surface and naked on both; transverse veinlets very fine, approximate and continuous; margins smooth; the two terminal leaflets a little shorter but broader than the side ones and united in the lower third of their length, suddenly contracted at the apex into a bristly-penicillate tip. Male spadix > « . . Female spadix not flagelliform, paniculate, with not many, approximate, partial inflorescences and terminating in a long spikelet, which bears at its base a finely clawed fili- form appendix 7 cm. in length (a rudimentary flagellum); primary spathes not very elongate (5-10 cm.), closely sheathing in their lower portion, somewhat enlarged and loose above, finely striated longitudinally, thinly coriaceous, glabrous, greenish-brown, armed with small scattered claws in their lower portion, decayed and falling to pieces (not fibrous) in their upper part, the dead part sharply defined from the living; partial inflorescences approximate, all (except the uppermost which is smaller) of about the same size, 15-18 cm. long, inserted inside their own spathe, at first ascendent, then arched, bearing 8-10 spikelets on each side and terminating in a spikelet longer than the side ones; secondary spathes strongly and densely scabrid-papillose, tubular-cylindraceous and strictly sheathing in their basal portion, suddenly enlarged near the mouth and extended at one side into a. rather long, triangular, subulate, decayed point; spikelets vermicular, inserted just at the mouth of their own spathe, slightly callous at their upper axilla, horizontal and somewhat arched, the largest, the lowest, 6-7. cm. long, with 15-20 very approximate. flowers on each side; the upper a little shorter; spathels very densely scabrid-pupillose, with a very short tubular basal part and suddenly expanded into a concave subcymbiform limb, which is prolonged at one side into a triangular, acute, spreading or deflexed tip; involucrophorum shallowly cupular, attached almost outside its own spathel at the base of the one above; involucre more or less regularly — cupular, often asymmetrically evolute, strongly striately veined; areola of the neuter flower very large, flattish, almost circular, very sharply bordered, sometimes only slightly smaller than the involucre. Fruiting perianth explanate; the calyx divided into 3 broad indistinctly veined lobes; the segments of the corolla lanceolate, one-half or one-third longer than the calyx, smooth outside. Fruit (unripe) very small (7 mm. in diam.), spherical, very shortly beaked; scales in 16 series, yellowish-brown, convex, very faintly channelled along the middle, with scarious finely erose margins and tip, where sometimes they are marked with an indistinct intramarginal line. Hasrrat.—Borneo; Sarawak, (Lobb in Herb. Kew). OssERvATIONS.—I have seen of this only one specimen (preserved at Kew) consisting of the upper part of a leaf and the apex of an immature fruiting spadix. This portion of spadix (probably the greater part of it) is 40 em. long and bears 4 partia inflorescences. C. ruvidus is a near ally of C. scabridulus and C. radulosus by its very scabrid secondary spathes, spathels and involucres; it differs however from both in the leaves having very few, 5-costulate, somewhat concave leaflets, which are oblanceolate or broadest above the middle, without bristles or spines, and in the rather compact female spadix with few short and approximate partial inflorescences. The characters assigned by me to the leaf-sheaths in the diagnosis of C, ruvidus in LC. scabridulus | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 171 the Records” l. c. belong to a male specimen preserved at Kew, which I doubtfully referred to that species, but which I now think it safer not. to take into account. This last specimen was gathered by Motley at Banjarmassing in Borneo. | PLATE 31.—Calamus ruvidus Bece. The complete type-specimen preser ved at Kew. ' 25. CALAMUS scaBripuLus Becc. in Rec. Bot. Surv. Ind. ii, 208. Descriprion.—Scandent, slender or of moderate size. Leaf-sheaths . M Be Leaves not cirriferous; petiole . . . . o ; rachis of the upper part of ‘the leaf armed beneath with rather stout solitary black-tipped claws; leaflets not very nnmerous, subequidistant, rather remote, 4-7 cm. apart, flat, papyraceous, rigidulous, narrowly lanceolate or .ensiform, almost equally narrowed to both ends, very gradually acuminate into & subulate apex, 40-45 cm. long and 2-2°5 cm, in ‘breadth, the upper ones somewhat shorter, the two of the terminal pair shortly connate at the base, 23-25 cm. long, but a little broader than the others; all almost concolorous, shining and very finely longitudinally striate under the lens on both surfaces, furnished above with 3 rather acute bristly-spinulous coste and with in addition another more slender naked nerve near the margin; on the under surface the 3 coste less prominent than above and sparingly spinulous near the apex; transverse veinlets very distinct, sinuous and interrupted; margins indistinctly, remotely and appressedly spinulous. Male spadiz . . . . . ” Female spadiz (not seen entire) with very slender and long (50 em.) partial inflorescences, these terminating in a short (3-4 cm.) filiform, very acute, scabridul- ous appendix and bearing about 10 spikelets on each side; secondary spathes very narrowly tubular and very closely sheathing, somewhat clavate in their upper part, flattish inside at the base, 1°5-2°5 em. long, scabrid, finely aculeolate upwards on the back, entire, obliquely truncate and ciliated at the mouth, where produced. at ‘one side into a lanceolate and acuminate point; spikelets vermicular, slender, slightly arched, attached just above the mouth of their own spathe and deflexed by the pressure of a very distinct axillary callus; the lowest spikelets, the largest, 6 cm. in length, with above 20 distichous flowers on each side, the upper ones gradually smaller, those near the apex 2°5 em. long with proportionately fewer flowers; spathels very broadly infundibuliform or concave and sub-bracteiform, pro- longed at one side into an acute tip, densely scabrid-papillose and strongly striately veined; involucrophorum unilaterally subcupular, almost exserted from its own spathel and attached at the base of the one above ; involucre asymmetrically cupular, obscurely lobulate; areola of the neuter flower relatively large, broadly ovate, sharply bordered with a discoid subcircular callus in its centre. Female flowers small, 2:5 mm. long (perhaps when not fully developed), conic-ovoid, acute; the calyx with a callose base, strongly striately veined, broadly 3-toothed; corolla twice as long ss the calyx. Fruit not seen. Hasrrar.—Billiton Island in the Java Sea (Riedel 1876, in Herb. Becc.)—Malay name ‘Rotang mengkekeran’ (Keker = a file). : OzservaTions.—This species is very nearly allied to 0. ruvidus and especially to C. radulosus, and the secondary spathes and the involucres are very ‘scabrid as in ANN. Roy. Bor. Garp. Catoutra Vor. XI. 172 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [Q. muricatus those two. From C. ruvidus it: dffiers in the more numerous, flat and not more or less concave ensiform leaflets, which are bristly spinulous on 3 coste, and in the spadix with elongate partial inflorescences which bear many remote spikelets. From C. radulosus it differs in the leaflets which slightly decrease in length in the upper part of the leaf and have the two ultimate leaflets larger than the lower ones, and in the much more slender partial inflorescences and spikelets. Pirate 32.—Calamus scabridulus Becc, The terminal portion of a leaf ( upper surface); the portion of the same following ( under surface); two partial inflor- escences with female flowers.—From Riedel’s specimen in Herb. Bece. 26. CALAMUS MURICATUS Bece., Nelle Foreste di Borneo 609, and in Rec. Bot. Surv. Ind, ii, 203. : DescripTion.—Slender, scandent. Stem with the sheaths 13-16 mm, thick, Lea/- sheaths flagelliferous, slightly gibbous above, obliquely truncate at the mouth, very conspicuously armed with straight, horizontal, subwhorled spines which are 4-5 mm. long and rest on approximate, prominent, annular ridges, these being alternate with other smaller, sinuous, interrupted ridges or wrinkles, which are armed with very small spines or are simply scabrid on their crest; the base of the petiole or the apex of the sheath bears many such wrinkles, Ocrea very short, inconspicuous. Leaf-sheath flagella filiform, very slender, armed with very fine, small, ternate or half- whorled claws. Leaves not cirriferous ; petiole rather long (40 cm.), sub-biconvex in section in its upper part, slightly flattish above and convex beneath near the base, sparingly aculeolate on the back and at the margins upwards; rachis flat and above bifaced and sniooth, glabrous, armed with small, solitary, scattered claws beneath; leaflets not very many, 14-15 on each side, thinly papyraceous ‘ very inequidistant, scattered, not distinctly grouped, linear, very elongate, the largest, those a little above the base, 35 cm. long, 13-14 mm. broad, gradually narrowed to the base, and gradually acuminate at the apex into a subulate bristly tip, about the same colour on both surfaces, sub-3-costulate, or with the mid- costa rather acute and one distinct acute nerve on each side of it, all the 3 sparsely bristly-spinulous above, beneath, the mid-costa only bristly ; margins very adpressedly and inconspicuously spinulous ; transverse veinlets very minute, much interrupted ; the two terminal leaflets smaller than the others and free at the base.—Other parts unknown. Hasrrar.—Borneo; on Mt, Mattang, near Kuching in Sarawak (Beccari P. B, No. 1928.)—Malay name ‘Rotang sakkat.’ OsservaTions.—Allied to C. zonatus, but distinct in its larger size, in the more powerfully armed leaf-sheaths, in the longer petioles and larger leaflets which are setose on 3 nerves above. The diagnostic characters of ©, murieatus are the armature of the leaf-sheaths formed by whorled short spines which rest on annuler raised ridges alternating with interrupted wrinkles; the leaves with a long petiole and the not numerous, inequidistant, very narrow leaflets, which are bristly on 3 nerves above, Puare 33,—Calamus muricatus Becc. The entire (sterile) type-specimen in Herb. Beccari. : e. zonatus | BECCARL MONOGRAPH OF THE GENUS CALAMUS. 173 97. CALAMUS ZoNATUS Becc., Nelle Foreste di Borneo 609, and in Rec. Bot. Surv. Ind. ii, 203. i DescripTion.—Very slender, scandent. Sheathed stem T-8 mm. in diam. Leaf- sheaths flagelliferous, gibbous above, striate longitudinally, conspicuously ornamented with approximate, sinuous, unequal, sometimes interrupted, annular ribs or prominent wrinkles which are scabrid on their crests or sometimes furnished with pungent warts or rudimentary spinules, ZLeaf-sheath flagella very long, slender and very finely clawed. Leaves delicate, not cirriferous, 50-70 cm. in length; petiole short, 2-5 cm. long, flattish or slightly channelled above, rounded beneath where rather strongly aculeate along the middle and at the sides; rachis glabrous, acutely bifaced above, finely irregularly clawed throughout beneath; leaflets not many, 10-12 on each side, inequidistant, scattered, not distinctly grouped or fascicled, elongate-linear, the largest (those a little above the base) 20-28 cm. long, 10-12 mm. broad, all almost equally attenuate at both ends, acute at the base, gradually acuminate at the apex into a subulate bristly tip, thinly papyraceous, rather rigid, of the same colour and sub-shining on both surfaces, distinctly 3-costulate, the mid-costa sparingly bristly above; all nerves naked beneath; margins smooth, only the apex ciliate; tranverse veinlets fine, much interrupted; the two terminal leaflets smaller than the others, more or less shortly connate at the base. Male spadiz partially ultra-decompound, elongate, filiform, flagelliform, with few remote, very delicate partial inflorescences ; primary spathes very narrow and long, very closely sheathing, the lowest sligh’ly flattened, obliquely truncate and entire at the mouth, almost unarmed; the upper ones cylindraceous, extremely narrow, aculeolate mainly in their upper part; un- sheathed axial portions of the spadix between two partial inflorescences very slender, clawed on the outer side; partial inflorescences 20-40 em, long, inserted above the mouth of their own spathe, divaricate, with a conspicuous axillary callus; the lowest inflorescedces (the largest) decompound, with 1-3 branches on each side near the base and 5-6 simple spikelets (also on each side) in the upper part; secondary spathes very narrow, very closely sheathing, slightly clavate, scabridulous, obliquely truncate at the mouth, acute at one side, the lowest 4-5 cm. long, the upper ones gradually shorter, the uppermost 15-20 mm, long; spikelets 3-6 cm, long, very slender and delicate, filiform, straight, horizontal, attached above the mouth of their own spathe with a distinct axillary callus and bearing 18-20 horizontally inserted flowers on each side; spathels tubular-cylindraceous at the base, suddenly expanded into a short, infundibuliform, truncate, scabrid limb; involucre completely exserted from its own spathel and laterally attached to the base of the one above, subdiscoid or shortly cupular, with a short, obsoletely bidentate, strongly striately veined Timb,—Other parts unknown. HanrraT.— Borneo ; Mt, Mattang, near Kuching in Sarawak (Beccari P. B. No. 1921.)— specimen gathered by Lobb, probably also in Sarawak (Herb, Calc.), bears a male spadix, but all the flowers have fallen. This species produces a very slender Rotang of great toughness, used by the Malays of Sarawak for fastening the iron blade of their axes to the handle, whence its name of ‘ Rotang perdas’ (perdas=the handle of the Malaysn axe). 174 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA, [C. radulosus OssERvaTIONS,—My specimen No. 1921 is sterile. The Calcutta specimen ‘does not essentially differ from it; only the annular ribs or ridges of the leaf-sheaths are a little more prominent and bear here and there some rudimentary spines in the shape of conical pungent warts; furthermore the leaflets are quite smooth in the upper surface, and no hairs or spinules are visible on the mid-costa. It is very closely allied to €. muricatus, but is more slender and delicate in every part and has the leaf-sheaths spineless or almost so. Prats 34.—Calamus zonatus Becc. The Calcutta specimen with a male spadix devoid of flowers. : 28. QCarnAMUS RADULOSUS Becc., in Hook. f. Fl Brit. Ind. vi, 443, and in Rec. Bot. Surv. Ind, ii, 203. i Descrrprion.—High scandent, of moderate size. Sheathed stem 2:5-3 cm. in diam. Leaf-sheaths obliquely truncate and naked at the mouth, very densely armed with very short and broad (4-6 mm. long), sub-triangular, scattered spines, which are obliquely inserted and ascendent (never deflexed), conspicuously swollen above and flat beneath. Leaf-sheath flagella as long as the leaves. Leaves rather large, 1:2-1°5 m. long; petiole obsolete; rachis bifaced above, armed below with black-tipped always solitary claws, which are numerous and pluriseriate lower down and on a single line in the upper portion; leaflets numerous, equidistant, rather closely set (2-3 cm. apart), papyraceous, green, glabrous, rather shining and almost of the same colour on both surfaces, under the lens finely longitudinally striately veined on the lower one, ensiform, finely acuminate at the apex, gradually narrowing at the base, furnished with 3 acute but not very strong coste and with another faint secondary nerve near the margin, the mid-costa alone sometimes bristly, the 3 coste rather closely bristly beneath towards the apex, but fainter than above; transverse veinlets very sharp, much interrupted; margins apparently naked, but under the lens very adpressedly, minutely and remotely spinulous; the largest leaflets, the mesial 45 cm. long and 25 mm. broad at most, the lowermost a good deal narrower uS the mesial, the upper gradually decreasing in length and breadth; the two of the terminal pair the smallest, 13-15 cm. long and 8-12 mm. broad, shortly connate at the base. Male spadiz ultradecompound. Female spadix very long (3:5-45 m., Scortechint) and robust, with many remote partial inflorescences, sub- flagelliform at the apex and terminating in a slender, tail-like, unsheathed Moers about 20 cm. long; primary spathes very elongate, closely sheathing thinly coriaceous; the basal one about 50 cm. long, somewhat flattened, a 15 mm broad, acutely two-edged, obliquely truncate at the mouth, armed, mainly on "hs outer side, with small scattered hooked spines; the upper spathes cylindraceous, often split longitudinally in the upper part, where usually decayed and faling to T but not fibrous, aculeate throughout, but not seabrid; unsheathed axial portions of the spadix, between two partial inflorescences, elongate, flat on the inner and convex on the outer side where armed with small scattered claws; partial inflorescences very long, the lowest as much as 1 m., the uppermost 30 cm, in length t distichously many (10-15 ən each side) spikelets; secondary syátlien « diis infundibuliform, somewhat enlarged above, where later usually split longitudinall on the inner side, obliquely truncate, searious and lacerated at the moni Q. rugosus) BECCARI. MONOGRAPH OF THE GENUS UDALAMUS. 175 at one side, strongly scabrid throughout and occasionally more or less aculeolate; spikelets vermicular, inserted just above the mouth of their own spathe, hod and. deflexed, the largest, the lowest of each inflorescence, 8-9 cm. long with 20-23 distichous flowers on each side, the upper shorter and with fewer flowers; spathels short, bracteiform, concave or broadly boat-shaped, strongly striately veined, apiculate at one side, very scabrid-papillose; involuerophorum shallow, subcupular, almost exserted from its own spathel and attached to the base of the one above; involucre more or less asymetrically cupular, scarious at the margin; areola of the neuter flower large, ovate, with acute scarious borders. Female flowers closely packed, small (about 8 mm. long). Fruiting perianth explanate (not ^ pedicelliform); the calyx deeply striately veined, broadly 3-lobate; the corolla with lanceolate segments, smooth outside, one-third longer than the calyx. Fruit (when still very young) very small, pisiform, globose, rather long-beaked ; scales not channelled along the middle, yellowish-brown with a reddish line across the base of the searious lacerated tip. |. Hanrrar.—Malayan Peninsula; Perak (Scortechini No, 468" in Herb. Becc.); Goping (King’s Collector No. 8171 in Herb. Calc.)—Malay name ‘Rotang Kikier.' OBsERVATIONS—AÀ. very distinct species, remarkable for the armature of the sheaths and the scabridity of the secondary spathes, spathels and involucres, sharing, however, this last peculiarity with C. ruvidus and C. scabridulus. It differs from both in the leaves with the leaflets gradually decreasing in length from the middle towards the apex, the two apical leaflets being the shortest and the narrowest. It differs besides from C. ruvidus in its long spadices with very long partial inflorescences. The female spadix of C. radulosus seems very much the same as that of C. scabridulus, but in this the partial inflorescences and the spikelets are much more slender. Amongst Scortechini’s specimens of C. radulosus there is & male spadix with very young flowers. It does not differ from the female one, and bears a partial inflorescence (the lowest) 1'2 m. in length, with many very long com- pound spikes on each side, which again bear distichously) many simple spikelets; the apex of the inflorescence bears only simple spikelets; the secondary spathes, the spathels and the involucres are scabrid as in the female spadix. The fruit has been described from No. 8171 of the Caleutta Herbarium. | Prare 35.—Calamus radulosus Bece. Part of the sheathed stem with the base of two leaves and the basal portion of a spadix; an entire partial female inflorescence; the apex of a leaf seen from the lower surface; two detached leaflets with a portion of the rachis, seen from the under surface and taken from near the base of the leaf.—From Scortechini’s specimen No. 468" in Herb. Beccari. 99. Carawus RuGosus Becc. in Hook, f. Fl. Brit. Ind. vi, 443, and in Rec. Bot. Surv. of Ind. vii, 209. Descriprion.—Very slender, scandent. Sheathed stem 8-10 mm, in diam. Leaf sheaths flageiliferous, gibbous above, obliquely truncate at the mouth, armed with confluent, sub-whorled, triangular, short (4-5 mm. long at most), laminar, sub-horizontal (not deflexed) spines, and further ornamented with many small more or less interrupted annular ridges or wrinkles, which are fringed on the crest with very small confluent spinules. Ocrea inconspicuous. Leaves about 70 em. long; petiole 176 ds ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [Q.flabellatus rather long} (14-17 cm. in length), slightly channelled above near the base, flat upwards, its margins very acute and armed—as is the lower rounded surface—with scattered, short, straight aculei; rachis acutely bifaced and smooth above, regularly and closely armed beneath with solitary, rather small claws; leaflets numerous, equidistant, rather closely set, linear-lanceolate, gradually finely acuminate into a subulate bristly tip, somewhat attenuate at the base, where suddenly plicate at their insertion on the rachis, thinly papyraceous, about the same colour on both surfaces, shining above, distinctly and rather densely bristly above on the acute mid- costa and on one rather slender nerve on each side of it; beneath only the mid-costa is bristly; margins distinctiy and appressedly spinulous; the largest leaflets, those a little above the base, 15-17 em. long and 8-10 mm. broad; the two ter- minal slightly smaller than the others and quite free at the base. Male spadiz à . . Female spadiz very slender, filiform, about as long as the leaves; with very few partial inflorescences; primary spathes very elongate and very narrow, very closely sheathing, the lowest somewhat flattened and acutely two-edged, the others cylindraceous, finely and densely aculeate, truncate and entire at the mouth; partial inflorescences spreading, very small and short (5-6 cm, long), inserted above the mouth of their own spathe, and with very few (usually 2) spikelets on each side distinctly callous at their upper axilla; secondary spathes scabrid, tubular, slightly enlarged above or very narrowly infundibuliform, closely sheathing, ob- liquely truncate at the mouth, where acute at one side; spikelets very short (15-20 mm. long) and relatively thick, horizontal, inserted at the mouth of their own spathe with a distinct axillary callus, and bearing 4-5 rather remote flowers on each side; spathels cylindraceous at the base, scabrid, striately veined and ' slightly infundibuliform in their upper half, horizontally truncate and entire at the mouth; involucrophorum sub-cupular, almost hollowed laterally into the base of the spathel above its own; involucre asymmetrically sub-cupular, striately veined limb; areola of the neuter flower rather large, parts unknown, with an irregular concave.— Other Hasrrar,— The Malayan Peninsula. Discovered in the district Scortechini (Herb. Beccari); and in the same district found a Thaiping Hills (Ridley No. 11314). of Perak by gain by Ridley on the OBSERVATIONS.— This is allied to C. zonatus, spinous leaf-sheaths and the numerous equidistant leaflets, which are besides 1-costate and not 3-costate. The female spadix I have seen was an adult one, but without. flowers; it was 70 cm. long and bore 3 partial inflorescences (all about the same dimensions), each with 4 spikelets in all. but is quite distinct because of its ` Prate. 36.—Calamus rugosus Bece. Two portions of with an entire leaf, and a female spadix without fruit.—Fro in Herb. Beccari, the sheathed stem, each m Scortechini's specimens 30. CALAMUS FLABELLATUS Bece, Malesia iii, 62, and in Ree. Bot. Surv, Ind. ii, 201. | Description.—Scandent, very, long and slender, Sheathed stem 5-6 mm, in diam, Leaf-sheaths flagelliferous, not or very indistinctly gibbous above, very obliquely Q. javensis] BECCARI, MONOGRAPH OF THE GENUS CALAMUS, i 177 truncate at the mouth, distinctly striate longitudinally, quite unarmed. Ocrea very short, almost obsolete. Zeaf-sheath flagella filiform, very slender, finely aculeolate callous at the base, Leaves simple, not pinnate, 45 cm. long, 10 cm. brosi, elongate-flabelliform, papyraceous, green, of the same colour on both surfaces prada ally attenuate and acute at the base, furcate or divided in the upper third bust Gt their length into two broadly lanceolate, acuminate, acutely 6-7-costate lobes; the doei very acute, naked on both surfaces all of the same strength and all reaching the apex, and lower down meeting at different heights and at a very acute angle along the mid-costa of the blade; beneath, the mid-costa (or rachis) rather robust and sparingly clawed; margins acute, not ciliated or spinulous; transverse veinlets rather approximate and sharp; petiole 3-4 cm. long, slender, sub-trigonous, striate, unarmed.—Other ports unknown. Hasirat.—Borneo; on Mount Mattang, near Kuching in Sarawak (Beccari P. B. No. 1911.)— The Rotang is of very good quality and is called in Sarawak Rotang Berman. | OssERvATIONS,— The only specimen I have preserved of this-very peculiar Calamus is sterile, but it belongs to a nearly full-grown plant. Other species have undivided or furcate radical leaves, but this is the only one I know with furcate or flabellate leaves on the upper part of the stem. Its affinities are apparently with the species of the group of C. javensis, PLATE 37.—Calamus flabellatus Becc. An intermediate portion of the leafy stem from P. B. No. 1911. 31. Carawus savensis Bl. Rumphia ii, 137 D. and iii, 62 (var. a and B firmus); Mart. Hist. Nat. Palm. ii, 337; Walp. Ann. iii, 487, and v, 831; Miq. Fl. Ind. Bat. ii, 125, in Pl. Jungh. 159, in Journ. de Bot. Neerl. i, 22, in Prodr. Fl Sum. 256 and in De Palmis, 27; Teysm. Cat. Hort. Bog. 75; H. Wendl. in Kerch. Les Palmiers, 236; Becc. in Rec. Bot. Surv. Ind. ii, 201. C. equestris Bl. (not of Willd.) in Roem. et Schult. Syst. Veg. vii, 2, 1330 (the Javan plant only); Mart. Hist, Nat, Palm. iii, t, 113 and t. 198 and 203 (lst edit.) and 207 (2nd edit.) partly; Kunth Enum. Plant. iii, 204 (partly) C. equestris? — Zolling. Syst. Verzeich. 79 and Exsie. No. 1597? (non vidi) and No. 3696, Description.—Scandent, slender. Sheathed stem 5-8 mm. in diam. Leaf-sheaths distinctly striate longitudinally, green, glabrous, sparingly armed with straight 1-5 mm, long, solitary, horizontal or slightly deflexed spines. Leaf-sheath flagella with the lowest spathe usually smooth. ^ Ocrea 5-12 mm. long, ciliate, maked or fugaciously furfuraceous. Leaves short, 30-50 cm. long; potiole very short or almost obsolete ; leaflets very few, 3-6 on each side, more or less inequidistant, usually opposite or subopposite, oblong-lanceolate or lanceolate-elliptic, cuspidate, rather variable in size and in relative breadth, with 3 often spinulous coste; the two leaflets near the base smaller than the others and spreading, the two of the terminal pair two-thirds connate. Male and female spadiz as in var. peninsularis. Fruit 15-16 mm, long (including the beak), 9-10 mm. in diam., ellipsoid-ovoid, with - Axx, Rox. Bor, Garp, Carcurra Vor. XI. 178 ANNALS OF THE ROYAL BOTANIC GARDEN, .CALCUTTA. (G. jauensis roinded base and vertex topped by a cylindrical beak, this 1:5-2 mm. long; scales in 20-21 series. Seed about 1 em. long and 7 mm. thick. HaBITAT.—Javá and Sumatra. It seems common on the mountains between “950-1500 m. in Western Java, where it receives the name of ‘Huy (Hooe) Omash ' or ‘Ommas,’ (Blume); at Tapos in the forests of Mt. Patuha (Junghuhn), where it is known by the name of 'Rotang Tjatjing or Chaching’; in Bantam at Pasir Orai (Forbes No. 258 in Calc. Herb.). In Sumatra (Forbes No. 2507 in Calc. Herb.).—It produces one of the more slender and more esteemed Rotangs, much employed for tying, basket-making, matting, etc. OxsseRvations.—This is a very variable species, perhaps the most polymorphic of the entire genus, and with a relatively wide geographical distribution, growing in the Malayan Peninsula and Borneo as well as Java and Sumatra, The type must be considered the Javan plant described and figured by Blume; the variety & of this ‘author seems to me one of the usual forms of the type. A form with narrower leaflets than usual is that figured in Plate 113 of Martius’ Hist. Nat. Palmarum; and indeed C. javensis even in Java varies very much in the size and relative breadth of the leaflets. I have not thought it necessary to give a detailed descriptive of the type of C. javensis as I have more fully described the var» peninsularis of which I had more complete specimens at my disposal and with which it agrees in most of its characters. : CALAMUS JAVENSIS subvar. EXILIS Bece. DESCRIPTION.—Sheathed stem very slender, 4-5 mm, in diam, `Petiòle ‘elongate (17 cm.). . Leaflets very narrow (1'5-2 cm. broad). Hasrrat,—Java, Reinwardt in Martius’ ‘Herbarium at Munich. OBSERVATIONS.—I do not know if this is a constant or a transitory form. The only specimen I have seen probably belongs to a rather young plant growing on high mopntains. This variety may. be considered. as the Javan representative of var. fenuissimus of the form peninsularis, CALAMUS JAVENSIS var. PENINsULARIS Becc. in Hook. f. Fl. Brit. Ind. vi, 442, and in Rec. Bot. Surv. Ind. ii, 201. C. penicillatus Roxb. Fil. Ind. iii, 781. DescripTion.—Scandent, very slender, 3-10 m. high. Sheathed stem 5-10 mm. in diam. Leaf-sheaths sometimes flagelliferous, slightly gibbous above, always more or less distinctly striate longitudinally, armed with solitary scattered rather numerous or scanty spines, which are short or 7-8 mm. long, usually straight with a broad base, flat beneath, horizontal or sometimes with a tendency to become hooked, often scurfy at the margins. . Ocrea membranous, 10-15 mm. long, unarmed, and with long fibrous cilia at the margin when young, brittle and falling to pieces later, Leaf-sheath flagella inserted near the mouth of the sheath in opposition to the petiole, callous at the base, filiform, rather long, with the lowest spathe slightly flattened and usually smooth, but sometimes sparsely aculeate on the back. . ‘hort, 30-50 em. long, not cirriferous ; petiole very ‘short or nearly wanting; rachis " d C. javénsis] - BECCARI. MONOGRAPH OF THE GENUS CALAMUS: 179 obsoletely bifaced above, rather densely armed: beneath along the middle and at the sides with rather slender claws, which are usually scattered or 2-3-nate (mainly upwards) and sometimes near the apex half-whorled.. Leaflets very few, 3-6 on each side, inequidistant, often opposite or subopposite, but never grouped on one side, of very variable shape but always relatively large in proportion to their length, lanceolate, oblong-lanceolate, and more frequently elliptic or ovate-elliptic, ‘the largest (the intermediate) 15-18 cm.: long, and 3-6 cm. broad, rigidulous, . papyraceous, glabrous, green or occasionally vinous-purpurascent, faintly paler beneath, narrowed te the base, where acute, rather suddenly. acuminate into an acute tip (this bristly- penicillate when young), furnished, at almost equal distances, with 3 primary costz; these all reaching the apex and of about the same strength, acute and raised above, less. prominent beneath, smooth (not spinulous or bristly) on both surfaces, with one slender but distinct secondary nerve interposed between each of them and the margins; transverse veinlets distinct, very crowded and continuous; margins acute, smooth throughout except’ at the extreme apex where ciliate; the lower margin often bordered on the upper surface with a narrow polished band; the two leaflets of the terminal pair somewhat longer, broader and more enlarged towards the apex than the side ones, connate in the lower two-thirds of their length; the basal pair, and sometimes the next, smaller than the upper ones and inserted very near the mouth. of the sheath, distinctly callous and as if they were articulated at the. base, strongly deflexed, concave and often the two connivent and almost embracing the stem. Male spadix simply decompound or partially ultradeeompound, inserted with a conspicuous basal callus near the mouth of the sheath opposite the leaf, or nearly so, very variable in length (from 60 cm. to 98 m.) flagelliform, . very delicate and slender, with few (3-4) or in very robust specimens, even 10-12 partial inflorescences and not terminated by a flagellum, but by a very slender filiform aculeolate appendix, which is a few em. long, shorter and more slender than the nearest inflorescence; primary spathes very narrow and long and very strictly sheathing; the lowest slightly flattened, more or less acute at the side, almost unarmed, terminating in a very narrow acuminate limb keeled ^on the back; the upper cylindrical, more or less aculeolate throughout, very long-attenuate at the base, where the axis is reduced to a slender thread and is flat on one side and armed on the convex back with scattered or confluent, but not regularly half-whorled claws; partial inflorescences divaricate, horizontal or. deflected by a very conspicuous axillary callus; they vary in length from 10 to 50 em. and have the axis slender, straight or slightly sinuous, bear 2-7 spikelets on each side and. terminate in a spikele& larger than the side ones; secondary spathes very narrowly tubular-infundibuliform, striately veined, smooth or aculeolate, obliquely truncate at the mouth, where usually but not always ciliolate,; more or less prolonged at one side into an acute point; spikelets usually straight or slightly flexuous, inserted horizontally a few mm. above the mouth of their own spathe, slightly deflected by a distinct axillary callus, delicate, flattened, 3-5 cm. long or at most in very robust plants 8-10 cm. and with 10-20 mp to 40-50 flowers. on each side; spathels rather crowded, shortly and very broadly infundibuliform, rather strongly striately veined, produced at one side into ‘a short acute point; involucre cupular, truncate, acutely bidentate and two-keeled .on Ann. Ror. Bor. Garp. CarcurrA Vor. XI. 180 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [Q,. javensis the side next to the axis, attached to the base of the spathel above its own. Male flowers inserted at an angle of 45°, rather variable, perfectly bifarious, the one close to the next or more or less distant, 3—4:5 mm. long, narrow, cylindrical or obsoletely trigonous, rather acute; calyx urceolate-campanulate, more or less distinctly striately veined, divided down to a little above the middle into three broad acute lobes; corolla about twice as long as the calyx or even longer, divided down almost to the base into 3 oblong acute segments; stamens very shortly united at the base, the filaments slender, filiform, subulate, much longer than their anthers, inflected at the apex; anthers versatile, narrowly sagittate, acute, their cells deeply discrete at the base; rudimentary ovary conspicuous, reaching about to the middle of the corolla, formed by a short pedicel (the ovary) and 3 linear connivent bodies (the stigmas), Female spadix . very variable in size, filiform, very elongate and with very few partial inflorescences, otherwise very much the same as the male one; primary spathes as in the male spadix; partial inflorescences 17-20 cm. long in large specimens, or 5-6 cm, only in small ones, terminating in a small and short more or less developed slender appendix, horizontal or deflexed and ,with a conspicuous axillary callus, their axis more or less zig-zag sinuous with 2-4 remote distichous spikelets on each side; secondary spathes very narrowly tubular, slightly enlarged above, very closely sheathing, smooth or aculeolate, obliquely truncate and ciliate at the mouth, prolonged at one side into an acuminate point; spikelets 2-5 cm. long, inserted at the mouth of their own spathe by means of a large axillary callus, horizontal or deflexed, rigid, with zig-zag sinuous axes; the largest with 10-13, the smaller with 5-6 flowers on each side; spathels strongly striately veined, very broadly sub-infundi- buliform when the flowers are rather remote, bracteiform and boat-shaped when approximate, always acute at one side; involucrophorum short, shallow, sub-cupular or almost explanate, attached laterally to the base of the spathel above its own; in- voluere shallow, irregularly cupular, strongly striately veined as is the involucrophorum and scarcely longer than it; areola of the neuter flower large, roundish, often deeply emarginate above, with acute and sharply defined margins. Female flowers almost horizontal, perfectly bifarious, not very remotely set, alternate, 3 mm. long, flat at the base, very slightly conical; calyx superficially striately veined, divided down almost to the middle into 3 very broad acute lobes; corolla twice as long as the calyx or a little less, divided down almost to the base into 3 lanceolate acute segments, polished outside; stamens almost equalling the corolla, with filaments united in the lower portion into a rather long tube, free, triangular and subulate upwards; anthers flattened, sagittate. Ovary columnar with the stigmas triangular, large and showing among the segments of the corolla. Neuter flowers large, thinner but scarcely shorter than the fertile ones, deciduous when these have been fertilized. Fruiting perianth sub-pedicelliform, somewhat hardened and callous at the base. Fruit globose or a little longer than broad, about 8 mm. in diam., topped by a cylindric 29:5 mm. long beak; scales in 18 series, light coloured, yellowish or greenish, flattish or slightly depressed in the centre, channelled along the middle, almost obtuse, with a pale erosely toothed margin and a faint intramarginal line, which is more distinct across the rather obtuse point. Seed sub-dimidiately globose, flattish and with a circular and rather deep chalazal fovea on the raphal side, roundish on the back, where the surface is obsoletely C. javensis] .BECCARL MONOGRAPH OF THE GENUS CALAMUS. 181 © facetted, the facets plane or slightly concave; albumen equable; embryo almost basal or slightly sheathed towards the dorsal side.—The different parts of the plant when young are more or less covered with a rusty scurf, more permanent on the spikelets and flowers. Hasirat,—The Malayan Peninsula, where it seems rather frequent. I have seen many specimens gathered in the district of Perak at an altitude of 100-260 m. above the level of the sea (Herb. Cale. No, 1996 ( ¢) and No. 2678 (3) and No. 7932; also Scortechini No. 236 in Herb. Beccari), OssERVATIONS.—The typical form peninsularis of OC. javensis must be considered that which, more than any other of the numerous forms of this species, resembles the Javan plant and which grows at no very great elevation above the level of the sea. Probably the stem of this species acquires a great length, but being a species much sought for by the natives for its slender and valuable Rotang, only young plants which have not. attained their full development are usually met with in the jungle. It varies in the size of the stem, in the number and shape of the leaflets, in the length of the spadices, in the number of partial inflorescences, in the size and length of spikelets, and in the first (or lowest) pair of leaflets more or less deflexed and having & tendency to embrace the stem. A large and complete spadix which I measured was 2-5 m. long including 2 m. of peduncular portion, and with only 2 inflorescences, respectively 17 and 20 cm. long and bearing 4 spikelets on each side. Another much smaller spadix had only one inflorescence 5 cm. long and altogether had 5 small spikelets 15-20 mm. long. The variety peninsularis differs from the Javan type in the leaflets which are not spinulous on the coste, but mainly in the more roundish fruit which has fewer scales arranged in 18 instead of 20-21 longitudinal series, Probably to C. javensis var. peninsularis must be reduced Roxburgh's penicillatus to which this author attributes the leaves with “thirty-four” pairs of leaflets; but the old botanists had not the habit of giving the exact number of the organs when these were very numerous, and very probably ‘“‘thirty-four” is a misprint for ‘‘three or four,” and if this be the case the description of C. penicillatus agrees pretty well with that of C. javensis.* | CALAMUS JAVENSIS var. PENINSULARIS, subvar, PURPURASCENS Bece. in Hook. f. Fl. Brit. Ind. vi, 443. Descriprion.—Much resembling in general habit and size the type peninsularis, and the var. fetrastichus, Leaflets broader than usual, the two basal strongly deflexed, embracing the stem and harbouring ants; sheaths moderately spiny. All parts, but specially the leaflets, conspicuously purpurascent. Hasrrat.—The Malayan Peninsula: Pulo Penang (Curtis); Perak (No. 7932 in Herb. Calcutta.) * Colonel Prain states that the manuscript of Roxburgh’s diagnosis shows 3-4, not 34 leaflets. [ Editor. ] 18% ANNALS OF. THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. javensis. CALAMUS JAVENSIS var. PENINSULARIS, subvar, PINANGIANUS Becc. in Hook. f. Fl. Brit. Ind. vi, 443, and in Rec. Bot. Surv. Ind, ii, 201. ^ DrscnrPrTi0N,— Sheathed stem 6 mm. in diam. Leaf-sheaths mottled with dark green and light patches and covered with numerous slightly deflexed spines of the usual form but with finer tip. Leaf-sheath flagella with the lowest spathe aculeolate on the back. Leaves with the petiole very short almost obsolete; rachis more strongly aculeate than in the type; leaflets usually more or less evidently. clustered, rather narrow, lanceolate. acuminate, 15-18 em. long, 2°5-3 cm. broad, not spinulous on the carine; the two of the lowest pair inserted very near the mouth of the sheath, spreading (not deflexed). Male spadiz very long, the one seen with 3 partial inflorescences, of which the lowest, the largest, has 3 spikelets on each side and one at the apex, which is longer than the side ones; the other inflor- escence has 6 spikelets in all, and the uppermost only two. Male flowers 4 mm. long, cylindraceous, obtuse. Hasrrar.—Pulo Penang: collected by Mr. Curtis at an elevation of about 650 m, (Moniot’s Rd.) in Jan. 1886. : OssEkVATIONs.— lhe characteristic notes of this variety are: the mottled densely spinous and apparently not striate leaf-sheaths; the narrow subclustered leaflets, of which the two of the basal pair are not deflexed and do not embrace the stem. CALAMUS JAVENSIS var. TETRASTICHUS Bl. Rumphia iu, 62; Beco, in Rec, Bot. Surv. Ind. ii, 20) $ C. tetrastichus Bl. Rumphia iii, t. 153; Mart. Hist. Nat. Palm. iii, 337; Walp. Ann. iii, 488 and v, 831; Miq. Fl. Ind. Bat. iii, 126, and De Palmis, 27 (with the var. borneensis Miq.); H. Wendl. in Kerch. Las Palm., 238. | | Q. amplectens Becc. Malesia ii, 78 and 278, pl. lxiv, f. 4. | C. borneensis Miq. Anal, Bot. Ind, 4 and Fl. Ind. Bat. iii, 120. Descriprion.—Sheathed stem 6-7 cm. in diam. Leaf-sheaths longitudinally striate, rather densely armed with straight, flattened, subulate, horizontal or slightly defloxed spines which are swollen at the base and sealy-barbed at the margins when young, Ocrea 8-15 mm. long, more or less woolly-furfuraceous and with long cilia at the margins when young, finally brittle and falling to pieces. Leaflets few, oblanceolate, suddenly acuminate into a long tip, distinctly 3-costate (the costæ and nerves naked on both surfaces), the two of the terminal pair connate for two-thirds of their length, the two basal usually quite deflexed, concave and embracing the stem. Leaf-sheath flagella with the lowest spathe aculeolate on the back or smooth. Female spadiz with the primary spathes rather strongly aculeolate, especially at the npex; partial inflorescences with 3-4 spikelets, each 4-5 cm. long. Fruiting perianth almost entirely explanate under the fruit, its calyx subapiculate at the base. Fruit 12-13 mm. long (including the beak) ellipsoid-ovoid with rounded base and vertex, page by a cylindraceous or obscurely trigonous beak, this 1°5 mm. long and bearing the BONUS Ee eee ee Seed 1:5-8 mm. Ai with irregular rather concave facets. €. javensis] . . BECCARI, MONOGRAPH OF THE GENUS CALAMUS. 183 Hasitat.—Borneo: in, the southern parts of the Island on the River Batang Sengaleng (Mueller in Herb. Leyd.); Bandjarmassing (Moiley in Herb. Kew); in the N. W. part, in Sarawak, at Kuching (Beccari), T OnsERVATIONS.— C. éetrastichus Bl, is certainly nothing more than the Bornean form of €. javensis, to which must also be reduced C. amplectens Becc. of Sarawak. Of C. -tetrastichus I have seen in the Leyden Herbarium an authentic specimen entirely agreeing with plate 153 of Blume's Rumphia, except in the ocrea which is not so densely hispid as is represented in that plate. Motley's specimen js perfectly like Mueller's one. That the flowers in the female spadix are arranged in four series is not a peculiarity of C. javensis. In all true Calami the female flowers are accompanied by a sterile or neuter one, and at a certain period of their development in many species the disposition in four series is very evident; but as the sterile flowers are very soon deciduous, then the fertile appear biseriate only. Miquel’s €. borneensis, which afterwards by the same author was considered to be a variety of C. tetrashchus, has been founded on male specimens in no way differing from @. javensis var, tetrastichus, Blume accorded some importance to the number of primary spathes sheathing the peduncular portion of the spadix; but the number of these, as well as that of the inflorescences, has very little value as a specific character; the differences depend chiefly on the conditions of more or less exuberant vegetation of the plant. From typical ©. javensis of Java, the variety tetrastichus differs in the lowest or basal leaflets being quite deflexed, concave, completely enclosing the stem and forming an ant-harbouring receptacle; in the leaf-costze never being spinulous; and in the more armed leaf-sheaths, where the spines have also a tendency to become hooked. The fruiting perianth is wholly explanate and not subpedicelliform as in the Javan form. In the fruit I have not found any important difference between the Javan and the Bornean plant, although perhaps the fruit of var. ¢etrastichus is slightly smaller but with an equally long beak. The seed also is the same. | From the Malayan forms of C. javensis, VAR. tetrastichus differs in the fruit having a shorter beak and in the seed having more concave facets. Prate 38.—Calamus javensis var. tetrastichus Becc. Portion of the leafy stem with a flagellum (the lower figure) from a Sarawak specimen, P. B. No. 1694 in Herb. Beccari. CALAMUS JAVFNSIS var. TENUISSIMUS Becc. in Hook. f. Fl. Brit. Ind. vi, 443, and in Rec, Bot. Surv. Ind. ii, 201. E Description.—Stem excessively slender, 925-4 mm. in diam. with the sheaths on; naked canes ^2 mm. only. Leaf-8heaths armed with small scattered slightly recurved prickles. Ocrea 10-12 mm. long. Leaves about 40 cm. in length; petiole 5-8 cm. long, roundish, sparsely aculeolate underneath, narrowly channelled above; leaflets very few, only 3 on each side, almost opposite (the couples rather inequidistantly remote ), narrowly lanceolate or oblanceolate about 2 em. broad, the two terminal the longest (as much as 20‘ em. long), connate to above the middle, the two lowest smaller than the others, spreading, not deflexed 184 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. javensis and not callous at their insertion. Male spadiz l-l:3m. long, with 3-4 partial inflorescences, each of which bears 8-10 spikelets. Female spadiz exceedingly slender, filiform, the one seen about 60 cm. long with only one partial inflorescence, which bears 2-3 spikelets 15 mm. long, otherwise agreeing in the minute characters of the flowers and their involucres with those of the typical form, peduncular portion of the spadix sheathed by three spathes. Fruiting perianth almost entirely explanate under the fruit. Fruit ovoid-ellipsoid, more elongate and more conspicuously beaked than in the type, 1 cm. long without the 3 mm. long beak, and 7 mm. in diam.; scales in 15 series, almost flat or slightly con- cave, narrowly channelled along the middle, the margins pale, the tip marked by a transverse dark line, Seed more elongate than in the type, its facets slightly concave. | Hasrrar.— The Malayan Peninsula, on the summit of Gunong Tambang Batak (Scortechini No. 648 "*)—Malay name 'Rotang Pseh or Seh? OssERVATIONS.— This seems to be the alpine form of €. javensis (form peninsularis ), corresponding to the variety ezilis of the Javan form. It differs from the type in the exceedingly slender stem; in the spines of the leaf-sheaths having a tendency to change into claws; in the small number and narrowness of the - leaflets; and in the long-beaked elongate fruit. PLATE 39.—Calamus javensis var, tenuissimus Bece. Portion of the plant with a leaf and an entire male spadix, from Scortechini’s No, 236 ^**, in Herb. Beco.; another portion of the plant with a fruiting spadix from Scortechini’s No. 648 *s, in Herb. Beccari. CALAMUS JAVENSIS var. SUBLAEVIS Becc. Description.—Sheathed stem 7-3 mm. in diam. Leaf-sheaths green, almost polished, distinctly striate longitudinally, smooth or with very few short straight spines. Leaves 45-80 cm. long with a petiole sometimes very long (as much as 30 em.); leaflets inequidistant, 4-6 on each side. Ocrea elongate, smooth, finally brittle and deciduous. | HasrrAT.— Borneo; at Kutcing on Mt. Mattang (Beccari P. B. No. 1694). OssERvATIONR.— The very scantily armed or almost smooth leaf-sheaths distinguish this variety from tetrastichus. I have two leaves of it, the one with the petiole 5 cm. long, the other more than 30 cm., probably because the latter leaf belongs io a young shoot. PLarE 40.—Calamus javensis var. sublevis Bece. Portion of the stem with a leaf (on the right-hand side) from P, B. No. 1694 in Herb. Beccari. CALAMUS JAVENSIS Var. POLYPHYLLUS Bece. in Hook, f. Fl. Brit. Ind vi, 443, and in Rec. Bot. Surv, Ind. ii, 201. `’ DrscRrPrION.—S/e5 more robust than in the type. Leafsheafhs rather densely covered with straight, horizontal, 5-6 mm. long spines. Leaves with relatively numerous leaflets (one leaf 55 cm. long had 10 leaflets on each side, and the two C. javensis] BECCARI, MONOGRAPH OF THE GENUS CALAMUS. 185 of the terminal pair highly connate) almost equidistant, lanceolate, the basal pair inserted very near the mouth of the sheath, and deflexed. Male spadiz with many partial inflorescences ; spikelets more numerous and nearly twice as long as in VAR. peninsularis, very slender, sometimes branched "in the lower portion and with very many flowers. Male flowers 3 mm. long; secondary spathes now and then spinulous. Female spadiz not seen, Hasirat.—The Malayan Peninsula: on the summit of Gunong Tambang Batak in the district of Perak (Scortechini No. 651).—To this variety probably belong some incomplete specimens gathered by Sir G. King’s collector between 700 and 1,000 m. at Larut, also in the district of Perak (Herb. Calc, No. 6312). OxservVATIONs,—Distinct from the Javan plant as well from the var. peninsularis (normal form) by its more robust habit and its numerous subequidistant leaflets. The No. 6312 in the Calcutta Herbarium is more robust than Scortechini’s specimens and has male spikelets more robust but not so long; the male flowers are also 4 mm. long, and the leaflets, at least in the portions of leaves I have seen, do not look exactly equidistant, though always more numerous than in the normal Javan form; the largest leaflets are 23 cm. long and 3 cm, broad or a little more. In Scortechini’s specimens they vary from 15-17 cm, in length and are not more than 3 em. broad. PLATE 40.—Calamus javensis var. polyphyllus Bece. Portion of the stem with a leaf; a male spadix on the left-hand side of the plate, from shore cni 8 specimen No. 651° in Herb. Beccari. CALAMUS JAVENSIS var. INTERMEDIUS Becc. in Hook. f. Fl. Brit. Ind, vi, 443 and in Rec. Bot. Surv. Ind. ii, 201. Description.—Sheathed stem 5-7 mm, in diam. ZLeaf-sheaths armed with many, small, short, straight spines. eaves 40-50 cm. long with a petiole 2-5 cm. long and with 6-7 leaflets on each side, alternate or subopposite, not quite equidistant, lanceolate or oblanceolate, the mesial ones 15-17 cm. long and 2 cm. broad; the two of the terminal pair connate up to above the middle; the basal pair slightly smaller than the others, 2-5 cm. remote from the mouth of the leaf-sheath, not distinetly callous at their insertion on the rachis and not deflexed. Hasrrat.—The Malayan Peninsula, in the district of Perak (Scortechini No. 236). OBSERVATIONS.—More slender than var. polyphyllus and with fewer leaflets; the stem is slightly larger than in var. /enuissimus and the leaves have more numerous leaflets. Prate 38.—Calamus javensis var. intermedius Bece. Portion of a leafy stem (upper figure) from Scortechini’s No. 236 in Herb. Becc. CALAMUS JAVENSIS var. ACICULARIS Becc. Description.—Very slender. Sheathed stem 6 mm. in diam. Leaf-sheaths armed with scattered, short or rather long, straight, horizontal spines. Ocrea in very young shoots truncate and ciliated with long filaments at the mouth, finally deciduous. Axx. Roy. Bor. Garp. CarcurTA, Vor. XI. : . 186 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (C. filiformis Leaves short, on the whole 30-35 em, long.with very few (1-2) pairs of lanceolate or oblanceolate leaflets; the latter near the terminal pair; the lowest or basal pair transformed into two opposite, straight, flat, rigid, spreading, acicular spines, which are 3-3°5 cm. long and inserted very near the mouth of the sheath; the long vacant space of rachis between these spines and the leaflets has the appearance of a petiole and is densely furfuraceous and furnished with a very few straggling slender spines, Female spadix (small) shorter than the leaves, with very few partial inflorescences. Hasitat.—Borneo in Sarawak on Mount Mattang at about 290 m. elevation: collected by (Mr. R. H, Hullet in 1890). OssERVATIONS,—Of this very curious variety of C. javensis I have seen only one specimen which was forwarded to me by Mr. H. N. Ridley. This variety is so unlike any other that I should have been much tempted to assign to it a distinct specific name had I not found among the numerous specimens in my possession an intermediate form between this var. acicularis and the usual peninsularis. In this intermediate specimen, collected by F. Keheding in 1879 at Klang in the Malay Peninsula, the spines at the base of the petiole are shorter, but have the same morphological value and are in the same place as in VAR. acicularis, and consequently the leaves appear as if furnished with a very long furfaraceous petiole and have 2-3 pairs of leaflets approximate to the two of the terminal pair. Piare 41.—Calamus javensis var. acicularis Bece., The entire specimen described above (Herb. Beccari), 92. CALAMUS FILIFORMIS Becc. Nelle Foreste di Borneo, 607, 608, and in Rec. ^A Bot. Surv. Ind. ii, 201. DzscRrPTION.— Exceedingly slender and delicate. Sheathed stem 3-5 mm. in diam. Leaf-sheaths flagelliferous, sparingly armed with small, short, scattered, slightly hooked aculei or quite smooth. Leaves 25-49 em. long; petiole very short or almost wanting; leaflets 5-6 on each side, somewhat inequidistant, alternate or subopposite, 3-costate, very narrowly lanceolate and very gradually attenuate at the base and acuminate at the apex, 8-12 cm. long, 8-14 mm. broad, the two of the terminal pair more or less highly connate at the base, the two of the basal pair deflexed and somewhat callous at their insertions, Male spadiz . . ,. . . Female spadiz filiform, very slender, with very few (1-3) partial inflorescences which are reduced to a single very slender spikelet 5-6 cm. long with 16-17 pectinate flowers on each side ; spathels obtuse with few strong and distinct nerves; involucrophorum laterally attached at the base of the spathel above its own, strongly striately veined as. is the involucre. Female flowers small, 2:5-3 mm. long; calyx callous at the base and divided down to about the' middle into 3 broad acute lobes, indistinctly striately veined; corolla about twice as long as the calyx or somewhat less, divided almost to the base into 3 lanceolate, acute segments; filaments of the stamens connate by their bases and forming a membranous urceolum which is as long as the third part of the corolla and is crowned with 6 broad, ovate-lanceolate, subulate teeth; anthers deeply sagittate with obtuse apex. | | C. corrugatus] BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 187 Hasrrar.—Borneo; on Mount Mattang, near Kuching in Sarawak (Beccari P. B. No. 1909.)— This species is represented in the Herbarium at Kew by a female flowering specimen collected by Lobb, very probably also in Sarawak. Native names. in Sarawak ‘Rotang Jangut,’? ‘R. Battu, ‘R. Kawat? a i It produces the most slender Rotang known to me. It is very tough and of very good quality, much employed by the natives for binding purposes, basket- making, ete. OxnstRvaTIons.—This is perhaps not so much a distinct species as an aberrant form of C. javensis closely related to var. tenuissimus from which, however, it differs in the infloresences being reduced to a single spikelet and in the more numerous and narrower leaflets, of which the upper pair are usually highly connate as in the different forms of C. javensis. ! PLATE 42.—Calamus filiformis Bece. The upper part of a plant with a female spadix in flower and another intermediate portion, from P. B. No. 1909 in Herb. Beccari. 33. CALAMUS CORRUGATUS Becc. in Rec. Bot. Surv. Ind. ii, 201. Description.—Scandent; very long and slender. ^ S/waíhed stem 4-5 mm. in diam. Leaf-sheaths flagelliferous, obliquely truncate at the mouth, unarmed, but conspicuously marked by many approximate tranverse annular ridges or prominent wrinkles. — Ocrea very short, more or less hairy-furfuraceous. Leaf-sheath flagella very slender, filiform, unamed in their basal portion and finely clawed upwards. Leaves short, 20-35 cm. long, not cirriferous ; petiole very short (about 1 cm. long), hairy-furfuraceous; rachis also furfuraceous, slender,’ filiform, almost round, striate longitudinally, armed below with weak, solitary or ternate, semi-verticillate claws ; leaflets very few (5 on each side), perfectly opposite, forming remote pairs, elliptic-lanceolate, narrowed to the. base, where acute and slightly callous: at their insertion, subulately acuminate at the apex, thinly papyraceous, about the same colour on both surfaces, plicate longitudinally and apparently many-nerved, but furnished with only three acute slender costa (which are naked on both surfaces) and some slender secondary nerves; transverse veinlets sharp and rather approximate; margins acute, smooth ; the largest leaflets, the mesial, horizontal, 12-13 cm. long, 2:5 em. broad, the two of the terminal pair a little smaller than the side ones, united up to about their middle, the two near the base, the smallest, deflexed and callous at their insertion. Spadiccs not seen. HaBrrAT.—Borneo; on Mount Mattang, near Kuching in Sarawak (Beccari P. B. No. 1910.)—There is a sterile specimen of this species in the Kew Herbarium collected by Lobb, probably also in Sarawak, in 1853. -— QOgservattons.—This is a very elegant and delicate species, which produces one of the smallest Rotangs of good quality. It is easily distinguished among those of the group of C. javensis by the ridged or wrinkled, not spinous, surface of the leaf-sheaths, and by the few, perfectly opposite and horizontal leaflets, which are approximate on each side of the rachis in remote pairs. € j - Ann, Roy. Bor. Garp. Carcurra, Vor. XI. 188 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. filipendulus Piatt 43.—Calamus corrugatus Becc. The entire specimen, described above, of P. B. No. 1910 in Herb. Becc. 34. CaLAMUS papuanus Bece. Malesia, iiij 60, and in Rec. Bot. Surv. Ind. ii, 201. Calamus sp. No. 47., Becc. Malesia, i, 88. DzscgrPTION.—Scandent, slender, very long. Sheathed stem 6-8 mm. in diam. Leaf-sheaihs ^ flagelliferous, gibbous above, faintly striate longitudinally, fuga- ciously furfuraceous, sprinkled throughout with very small tuberculiform spines which point upwards, which are more numerous along the slightly raised longitudinal line that corresponds to the side where the flagella are inserted, Leaf-sheath flagella very slender, filiform, finely aculeolate throughout and distinct- ly callous at their insertion. Leaves short, about 30 cm. long, not cirriferous; petiole short (2 cm.), subtrigonous, aculeolate; rachis subtrigonous, obsoletely bifaced above, armed beneath with scattered claws; leaflets few (13 in all in the few: leaves seen), inequidistant, clustered in about 4 remote fascicles of 3-4 each, generally disposed in opposite geminate divaricate pairs, the two near the base not opposite, spreading and not deflexed, the two of the terminal pair connate up to about their middle, all of about the same size and form, the largest 10-11 em. long, 2°5 cm, broad, glabrous, papyraceous, rigidulous, of about the same colour on both surfaces, almost shining above, quite devoid of any kind of hairs, bristles or spinules, elliptic-lanceolate or ovate-elliptic or broadly oblanceolate, acute at the base, rather suddenly subulately acuminate into a naked and not bristly apex; furnished with 5 very slender costz, of which the central is slightly stronger than the side ones, all naked on both surfaces, margins acute, smooth, transverse veinlets distinct, much interrupted.— Other parts unknown. All parts acquire a brown colour in drying. Hasitat.—Dutch New Guinea; at Ramoi (Beccari P. P. No. 421) OsseRvations.—This seems related to C. javensis. The broad, grouped, opposite, divaricate leaflets quite hairless or spineless on the nerves, on the margins and at the apex, distinguish this species from any other of the group. Prate 44.—Calamus papuanus Bece, An intermediate portion of the adult plant from P. P. No. 421 in Herb. Becc. 35. CALAMUS FILIPENDULUS Becc, in Hook. f. Fl. Brit. Ind, vi, 443, and in Rec. Bot. Surv. Ind. ii, 202. Description.—Scandent; rather slender. Sheathed stem 1~1°5 cm. in diam. Leaf- sheaths mottled when young, gibbous above, armed with scattered or slightly confluent, laminar, short and rather broad brown spines, which are 5-10 mm. long or even shorter and subtuberculiform, slightly deflexed, broad and concave beneath at the Leaf-sheath flagella filiform and very slender. Leaves not cirriferous, ‘5=1 m. long, with a very variable petiolar portion (from 4 to 30 cm. in length), smooth or strongly aculeated, at the margins and on the back, flattish or superficially channelled above; rachis more or less armed, mainly along the middle, with scattered claws; leaflets very few (5-8 in all), large and broad, in- equidistant (the tw^ of the terminal pair not differing from the others, but C. filipendulus | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 189 confluent by their bases), glabrous, rather shining and of about the same colour on both surfaces, oblong or elongate-oblong, 25-40 cm. long and 5-10 cm, broad, gradually narrowing to the base, rather suddenly narrowed above into a short apex, furnished with 5-9 coste which are devoid of bristles or spinules on both surfaces; margins naked and remotely spinulous near the apex; tranverse vein- lets crowded, more distinct in the lower surface. Male and female -spadices very similar, filiform, simply decompound, very slender, 1-l:8 m. long, differing from the leaf-sheath flagella only in the few remote partial inflorescences they bear; primary spathes very narrow and long and very closely sheathing, truncate at the mouth; the lowermost compressed, acutely two-edged, smooth or aculeolate on the cuter surface; the upper cylindrical, finely clawed, longitudinally striated, not scabrid; partial inflorescences very narrow, divaricate, inserted far above the mouth of their own spathes, callous at their upper axilla, with 8-12 very short spikelets on each side ; secondary spathes very scabrid or densely covered with very short tubercled spicules, tubular-cylindraceous, about 5 mm. long, truncate at the mouth and produced at one side into a triangular point which is deflexed under the spikelet. Male spikelets very short,. 3-15 mm, long, horizontal, arched downwards, with 3-10 very approximate flowers on each side; spathels bracteiform, concave, very broad, strongly striately veined as is the involucre, which is shortly cupular with irregular margin. Male flowers small oblong, obtuse, 2:5 mm. long; the calyx striately veined, broadly 3-toothed; the corolla also striate but polished, twice as long as the calyx; stamens with filiform filaments which are inflected at the apex; anthers linear; rudimentary ovary slender, columnar, terminated by 3 small recurved stigmas. Female partial inflorescences as the male ones, but somewhat more robust, with the lowest 3-4 spathels empty or without the usual spikelet ; spikelets very short, 7-8 mm. long, inserted at the mouth of their own spathe; callous in their upper axilla, arched and deflexed, ; with very few approximate flowers; spathels bracteiform as in the male spikelets, with very few strong nerves converging to the apex; involucrophorum and involucre shallow, sub-cupular, strongly striately veined and with irregular margin; areola of the neuter flower rather large, ovate and almost two-winged at the sides. Female flowers larger than the males, 3°5 mm. long, conical-ovoid, acute; the calyx sub-urceolate, strongly and deeply striately veined and with three broad acute lobes; the corolla longer by one-third than the calyx, with lanceolate, acute segments. Fruiting perianth not pedicelliform. Fruit small, about 12 mm, long, broadly conically ovate, or from a broad base gradually narrowing into an acute and slender beak crowned by the recurved stigmas; scales shining, not or indis- tinctly channelled along the middle, yellowish, with a narrow often indistinct intramarginal dark line and a brown scarious and almost fringed tip. Seed irregularly globular, facetted on the back, with concave facets and a shallow broad chalazal fovea; albumen equable; embryo basal. Hasirat.—The Malayan Peninsula, in the district of Perak (Scortechini No. 2312” in Herb. Bece.; King’s Collector Nos, 5659, 5773, 8019 in Herb, Calcutta). OssERvaTIONS.— This is quite distinct amongst the species of the group by its leaves with large, long, broad and many-costate leaflets; by the scabrid secondary 190 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (Ç. gonospermus spathes; by the very short few-flowered spikelets and by the conie or obturbi- nate fruit. , | Prate 45.— Calamus filipendulus Bece. Leaf-sheath with the lower portion of & leaf and a male spadix; terminal portion of a leaf, portion of a female spadix with unripe fruits, from Scortechini’s specimens No. 232^ in Herb. Beccari; two full grown fruits from the Calcutta Herbarium. 86 CaraMUs GoNOSPERMUS Bece. in Rec. Bot. Surv. Ind. ii, 202, DzscRiPTION.— Seandent. Sheathed stem about 15 mm. in diam. Leaf-sheaths flagelliferous, armed with straight, horizontal or slightly deflexed spines, which arise from a very broad base (concave beneath) and are 5-15 mm. long. Leaf-sheath flagella very | slender, finely armed with half-whorls of very sharp black-tipped claws. Leaves about 50 cm. long; petiole flattish above, rounded beneath, 17 cm. long, rather powerfully armed at the margins and on the back with unequal claws intermingled with very small ascendent or horizontal spines; rachis acutely bifaced above and furnished beneath with many irregularly set rather stout claws; leaflets very few, 6 in all, the two terminal connate about two-thirds up, two others very approximate to these and the two lowermost opposite and remote from the upper pairs; they are all about of the same size and shape (20-22 em, in length, by 45-56 cm. . in width), oblong or spathulate-oblong, some- what concaye or spoon-shaped, gradually narrowed to the base, suddenly contracted from near the apex into a caudate, linear, bristly-spinuloas tip (2 cm, long); green and of about the same colour on both surfaces, glabrous, rigidulous, ‘chartaceous, with 5 acute coste which are quite naked on both surfaces; margins smooth, except at the sides of the tip, the lower one bordered on the upper surface with a polished band; transverse veinlets numerous and rather distinct. Male spadiz . . . s ., Female spadiz short, comparatively robust, with very few short partial inflorescences, each with few, very short subscorpioid spikelets; primary spathes tubular, closely sheathing, thinly coriaceous, aculeolate, truncate and entire at the mouth; the lowest elongate, -flattish on the inner side, dorsally convex, where irregularly armed with small, straight, unequal spines; secondary spathes tubular, cylindrical, closely sheathing, truncate at the mouth, aculeolate; spikelets (or abbreviated partial inflores- ences ?) 3-5 cm. long, very dense, with 3-5 glomerules (abbreviate spikelets ?) of flowers on each side at each spathel, with the flowers pointing upwards er with a secund arrangement; spathels shortly tubular, subtrigonous, truncate at the mouth; involucrophorum and involucre almost explanate with an irregularly lobate limb. Fruiting perianih explanate, with the calyx divided into 3 broad acute lobes, and the corolla with the segments much narrower than these but as long; the calyx and corolla, as well as all the imvolucres, bard in texture, deeply and very sharply striately veined, of a rusty colour and with a broad shining, quite black, scarious margin. Fruit rather large, 20-23 mm. long, 15-17 mm. in’ diam., very closely packed, globose, ventricose, somewhat tapering towards the base, where obsoletely angular by mutual pressure, with a conical and acute top; scales in 16 series, broad, shining, very. adpressedly imbricate, convex and not channelled along the middle, brown-yellowish near, the base, broadly bordered with dark chestnut-brown, very obtuse C. floribundus] BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 191 ‘or rounded at the apex and with an erose denticulate margin. Seed very irregularly globular and acutely angular, 10-12 mm. in diam.; chalazal fovea indistinct; albumen equable; embryo basilar, Hasirat.—Borneo; near Kutcingat Siul, in Sarawak (Beccari P. B. No, 23). OpsERVATIONS.—Of this very distinct species I have seen only one very incomplete specimen consisting of one leaf and a portion of a spadix as to which I am uncertain whether if it is simply decompound with very small partial inflorescences bearing very abbreviate spikelets, or if it is supradecompound with spikelets 3-4 cm. long and bearing few glomerulate flowers at each spathel. The main characters of this species are the leaflets with few, broad, 5-costulate leatlets; the very abbreviated spikelets with very closely packed secund flowers; the rather large fruit with a conical point and not furrowed scales, and the angular seed. PLATE 46,—Calamus gonospermus Bece. ‘The entire specimen in Herb. Beccari, 37. CALAMUS FLORIBUNDUS Griff. in Macl Cale. Journ. v, 56 and Palms Brit. India, 66, pl. cxcvii; Mart. Hist. Nat. Palm.. iii, 337; Walp; Ann. iii, 487 and v. 831; Hook. f. Fl. Brit. Ind, vi, p. 444. Becc. in Rec. Bot. Surv. Ind. ii, 204. C. mishmeensis Gri. in Macl. Cale. Journ. v, 55 and Palms Brit. India, 65; Mart. Hist, Nat. Palm. iii, 337; Walp. Ann. v, 831. C. multiflorus Mart. in Wallich’s list No. 8613 (se Mart. lc, p. 387, No. 506). Description.—Gregarious, trailing at first, then not very high scandent, 3-6 m. long (€. B. Clarke). Sheathed stem 2-25 cm. in diam. or exceptionally smaller; naked canes 7-15 mm. in diam. with a polished surface. Leaf-sheaths sometimes flagelliferous, more or less gibbous above, very densely covered with spines of two kinds; some of them large, 2-3 cm. long, narrow, flat, subulate, horizontal or deflexed, scattered. and solitary or more or less confluent and subseriate; others (by far the more numerous) much smaller and often reduced to sub-spiny bristles with a sub-bulbous base; all dark. brown, at least af, their apex, and with a light base. Ocrea with a short ovate liguliform limb, 10-15 mm. long and densely covered with brown rigid bristles. Leaf-sheath flagella elongate, irregulerly armed with very unequal and sometimes long-tipped claws. Leaves comparatively short? | usually 6-1 m. in length, not cirriferous; petiole robust, channelled above in its first portion, flat upwards, rounded below, very irregularly armed at the margins with a few stout, straight, 1-4 cm. long, rigid, horizontal or deflexed spines, which are swollen or sub-bulbous at the base; in addition the petiole is rather densely covered throughout on both faces with small, short, straight spines, which are subulate from a broad conical base and sometimes reduced to small spiny tubercles; of this last kind of spine some appear also on the rachis, chiefiy at the sides; rachis spinulous on the upper aspect (or sometimes smooth ?) where it is acutely bifaced in the upper, and obtusely in the first portion, remotely armed beneath (where flattish near the apex and rounded in the remainder) with a few stout and long spines which, like those of the petiole, are sub-bulbous at the base but suddenly deflexed and intermingled with scattered solitam small claws; leaflets few, 192 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [Q. floribundus very inequidistant, 6-10 on each side, or fewer in small specimens, and rather distant, but more or less distinctly approximate into 3 remote groups of 2-3 on each side; the leaflets of the terminal group (3-7) approximate and almost digitate, the two of the terminal pair highly connate by their bases; the largest, the mesial, as much as 50 em. long and 35 em. broad, narrowly lanceolate, acuminate at the apex, narrowed down from the middle to an acute base; the uppermost shorter, but not narrower and with a bristly-penicillate, but less acuminate apex; all are sub-shining above, slightly paler beneath, usually 3-costulate, but sometimes with an additional costa near each margin; all coste (3 or 5) bristly spinulous above; beneath the mid-eosta constantly and the side ones occasionally and sparingly bristly; margins rather closely ciliate, mainly near the apex, with spreading subspiny bristles; some of the leaflets, especially the uppermost, sometimes furnished on the upper surface on the mid-costa, near the base, with a few small spinules. Male spadiz ultradecompound, flagelliform, 1:5-2:5 m. long, ending in a slender flagellum which is irregularly armed with weak claws but intermingled with straight deflexed or hooked spines; partial inflorescences not very numerous, rather remote, pyramidally paniculate, rather dense, 20-40 cm. long, and composed of 8-16 distichously alternate and gradually decreasing branches (or compound spikes) which are inserted just above the mouth of their respective spathes and are distinctly callous in their upper axilla; lowest primary spathe tubular, closely sheathing, acutely two-edged, armed with siraight or variable spines or almost unarmed, more or less split longitudinally in the upper part and terminated by a lanceolate limb; upper primary spathes subcylindrical, closely sheathing at the base, slightly enlarged in the upper part, where often split longitudinally, acute or acuminate at the apex, more or less armed throughout, but chiefly at the base, with unequal scattered or aggregate claws of various sizes, which are often intermingled with small straight or tuberculiform spines; secondary spathes (spathes of the partial inflorescences) unarmed, tubular-infundibuliform, more or less furfuraceous and sprinkled with light or brown scales, truncate and entire at the mouth, where densely ciliate at the margin and prolonged at one side into an acute or subulate, ciliate or penicillate point; tertiary spathes similar to the secondary ones but smaller, somewhat angular, pubescent when young, narrowly tubular at the base and suddenly broadened into an acuminate ciliate limb; primary or compound spikes spreading and arched, the largest, the lowest, 15-25 cm. long, and with 10-12 spikelets on each side; these spikelets horizontally inserted at the mouth of their own spathe end gradually decreasing in length and number of flowers from the base upwards, the lowest, the largest, 2-3 em, long, with 12-15 approximate flowers on each side, the uppermost very short and with very few flowers; spathels approximate, membranous, bracteiform, very broad, concave, pushed downwards by their respective flowers, prolonged into an acuminate fip, ciliate at the margins, finely striately veined; involucre. laterally attached to the axis of the spikelet, subcupular, very obviously formed by two concave, ovate, acute, finely striutely veined bracts which are united by their bases, Male flowers 3'5 mm. long, ovate, acute; the calyx’ rounded and almost smooth at the base, divided down beyond the middle into 3 ovate obsoletely striately veined acute lobes; the corolla twice as long as the calyx, its segments ovate-lanceolate, acuminate ; the stamens with subulate filaments which are inflected at the apex in the LI C. floribundus] BECCARI, MONCGRAPH OF THE GENUS CALAMUS; 198 bud; its anthers versatile, sagittate; the rudimentary ovary formed by three subulate rather elongate bodies. Female spadiz as the male but simply decompound, with few ; remote partial inflorescences which are 15-30 cm, long, and with 4-8 spikelets on each side; primary and secondary spathes as in the male spadix, scaly-furfuraceous when young, ultimately subglabrous; spikelets vermicular, slightly arched, somewhat zig-zag sinuous between the flowers, the largest, the lowest, 8-12 cm. long, with 12-16 bifarious rather remote flowers on each side; spathels furfuraceous, very broadly infundibuliform from a narrow base, truncate and entire at the mouth and prolonged at one side into a short triangular tip; involucrophorum sub-cupular with an acute tooth on each side, almost exserted from its own spathel and laterally attached at the base of the one above with a distinct supra-axillary callus; involucre eupular, usually emarginate on the side of the neuter flower, of which the areola is sublunate, rather deep and relatively large. Female flowers 4 mm, long, conic- ovate, acute; the calyx deeply 3-lobed, not or indistinctly striately voined outside; the corolla with lanceolate, acute segments as long as the lobes of the calyx; the stamens with filaments forming an urceolate tube crowned by 6 short teeth; the abortive anthers sagittate. Neuter flowers very similar to the fertile ones, but soon deciduous and thinner, with vacuous anthers and an abortive ovary formed by 3 small acute bodies. Fruiting perianih explanate (not pedicelliform). Fruit almost spheric (subobovate when immature), suddenly beaked, 9-10 mm. in diam.; scales in 15 series, obtuse, shining, superficially channelled along the middle, straw-yellow, bordered with a narrow brown-reddish line; margins finely erosely toothed. Seed suborbicular, rather convex, irregularly alveolate-suleate on the back, with a deep circular chalazal fovea on the flattish raphal side; albumen equable; embryo basal. Hasirat.—North-East India; Assam, Khasia Hills and Sylhet, (Wallich No. 8613), and (Hooker f. & Thomson in Herb. Kew). I have specimens from the Nambar Forest (G. Mann) and from the Charduar Forest (Brandis) in Assam; from the Khasia Hills (C. B. Clarke) at Sheelghat (100 m.) at Gowhatty (160 m.), and at Borlasa (1200 m.) It has been found also at the foot of the Mishmee mountains near Tapan Gam’s village in fruit in November (Grifith).—It grows. in the plains as well as on the hills, and it seems a rather common plant. Ossrrvations.—A. very variable plant in size, number of leaflets, and degree of armature of the different parts. It is very well characterised amongst the allied species by the few, grouped, relatively broad leaflets (with 3-5 costae, spinvlous above) and by the radiate arrangement of those of the terminal group. The young leaves of very robust plants are larger than those described above and may be mistaken for those of C. latifolius Roxb., but this has leaflets with smooth not spinulous nerves. Sometimes C, floribundus assumes very small dimensions, and seems almost a different species (see vAR. depauperatus). Prate 47.—Calamus floribundus Griff. Leaf-sheath with the. basal portion of a female spadix in flower and upper part of a leaf (on the right hand side) from Mann's specimens in H. Becc.; portion of a male spadix and two leaflets from Gowhatty (C. B. Clarke in H. Becc.); summit of a fruit-spadix, from Borlasa (C. B. Clarke). | Ann. Ror. Bor. Garp. Carcurra Vor. XI, 194 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. (C. interruptus CALAMUS FLORIBUNDUS var. DEPAUPERATUS Bece, Description.—Small, delicate, 1 m. high (O. B. Clarke). Sheathed stem very slender, 5-6 mm. in diam, Leaf-sheaths armed with very small, short and broad spines. Ocrea densely bristly-hispid. Leaves 35-40 em. long with 6-8 leaflets (in al) in two groups; leaflets 15-20 cm. long, 15-22 mm, broad; those of the terminal pair more or less connate at the base; petiole and rachis armed with small scattered claws. Male spadiz slender, flagelliform, almost simply decompound, with 2-3 small partial inflorescences. Hasrrat.—Gari at 400 m. in the Garo Hills in Assam, C. B. Clarke in H. Beccari. OxseRVATIONS.—This variety at first sight appears very distinct from the type, and recalls some of the forms of C. javensis. Piare 48.—Calamus floribundus var. depauperatus Becc, C. B. Clarke’s entire specimen in H. Becc. 38. CALAMUS INTERRUPTUS Becc. Malesia iii, $0 and Ree, Bot, Surv. Ind. ii, 204. Descrietion.—Scandent, of moderate ize. Steathed stem 15-20 mm. in diam. Leaf-sheaths cylindrical, almost glaucescent, fugaciously furfuraceous, slightly gibbous above, very obliquely truncate at the mouth, gradually passing into the petiole, armed with a few strong, scattered, solitary or sometimes confluent, deflexed, flattened and straight and subulate 15-20 mm. long spines, which have a broad base concave. beneath. Ocrea small, glabrous, liguliform and membranous, narrowly bordering the margins of the base of the petiole and the mouth of the sheath. Leaf-sheath flagella very long and robust, flattened and acutely two-edged in their lower portion where they are spinulose-serrulate or furnished with small aseendent prick'es at the sides, terete above and strongly armed with scattered or aggregate or hulf-whorled claws. eaves not cirriferous, 1:5 m. long; the petiole rather long (25-35 em.) very broad at the base, where deeply channelled above and with acute membranous naked margins, rounded and unarmed beneath in its first portion; higher up flat and smooth above, and armed irregularly beneath along the middle and at the sides, like the first portion of the rachis, with rather robust and scattered claws; the rachis bifaced and smooth above, and armed rather densely beneath, mainly in the upper portion, with sirong, solitary, or more or less confluent, black-tipped claws; the ieaflets not numerous, about 15 on each side, irregularly approximate into 4-5 distant fascicles of 2-3 on each side, elongate-lanceolate, gradually narrowed to an acute base, acuminate at the apex into a subulate, naked or very sparingly bristly tip, 25-35 cm. long, 2°5-3 em, broad, papyraceous, rather firm, brown when dry, shining above of the same colour on both surfaces, perfectly glabrous, their mid-costa acute ew and with 2-3 secondary nerves on each side of it, naked on both surfaces: jaan veinlets sharp, much interrupted, margins acute, smooth; the two ieadlidà of the terminal pair somewhat shorter than the others and confluent by their bases.— Other parts unknown. The leaves acquire a dark brown colour on drying. Hasirat.—N.-W. New Guinea; at Ramoi, Beccari P. P. No, 490. €. dioicus | BECCARI MONOGRAPH OF THE GENUS CALAMUS. 195 OsseRvaTIONS.—Distinguished ‘amongst the Papuan species by the cylindrical, smooth or sparingly spinous leaf-sheaths, which gradually pass into the petiole, and by the leaflets which are not numerous, narrowly lanceolate with only the mid-costa acute and 2-3 secondary nerves on each side of it quite smooth (also at the margin), and approximate into 4-5 distant groups. PLATE 49.—Calamus interruptus Becc. Portion of the sheathed stem with an entire flagellum and the base of a leaf; and intermediate portion of a leaf seen from the under surface (on the left lower corner); the apex of the same leaf, from P. P. No. 420. CALAMUS INTERRUITUS var. DOoCILIS Becc. Malesia ii, 60, C. docilis Becc. in Rec. Bot. Surv. Ind. ii, 904. Description.—It differs from the type only in the quite unarmed leaf-sheaths. Hasitat.—North-Western New Guinea; at Ramoi, with the type Beccari P. P. No. 418. Prate 50.—Calamus interruptus Var. docilis Becc. Portion of the sheathed stem with an entire flagellum; an intermediate portion of a leaf (upper surface); the apex of the same leaf, from P. P. No, 418 in Herb. Becc. 39. Catamus piorcus Lour. Fl. Cochinchin. i, 210; edit. Willd, i, 262; Lam Encycl. vi, 366; Roem. et Schult. Syst. Veg. vii, II. 1322; Mart. Hist, Nat. Palm. iii, edit. I. 213 and p. 842; Kunth Enum. Pl iii, 213; Walp. Ann. iii, 491 and v, 832; H. Wendl. in Kerch. Les Palm., 236.; Bece. in Rec. Bot. Surv. Ind. ii, 201. Description.—Very slender, scandent. Sheathed stem 4-5 mm. in diam.; naked canes 2:5 mm. in diam. Leaf-sheaths flagelliferous, gibbous above, hispid or bristly near the mouth and densely covered with very unequal, slender, straight, slightly deflexed spines which are 1-5 mm. long and rest on a broad base. Ocrea 15-20 mu, long, dry, «membranous, bristly-spinulous on the ventral side, later lacerated and deciduous. Leaf-sheaíh flagelia very slender, filiform, with a large axillary callus at the base, and with the lowest spathe flattened and spinulous. terete upwards and very finely clawed all round. Leaves short, 25-40 cm. long, not cirriferous, petiole very short or almost obsolete; rachis trigonous, clawed throughout beneath leaf-sheaths, petiole and rachis fugaciously covered with very dark scurf; leaflets very inequidistant, more or less distinctly grouped, 8-11 in all, of which 4 are very approximate at the summit with the two terminal free at the base and other 4 inserted very near the mouth of the sheath and kept. in a divergent or deflexed position by a distinct basal cailus; the intermediate ones opposite when there are 2; if 3, one straggling; all are very narrowly lanceolate, 10—20 cm. long, 11-15 mm, broad, narrowing to the base and acuminate to a bristly-penicillate apex; very finely 3- or sub-5-costulate; the coste acute and all more or less bristly-spinulous above, the mid-costa scarcely more distinct than the side ones, beneath all superficial and smooth; secondary nerves very faint; transverse veinlets much interrupted and not very crowded; margins finely and appressedly spinulous, sometimes bordered by a polished Ann. Rox. Bor. Garp, CarcurTA Vor. XI. 196 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. — [Q. schistoacanthus band. Male spadiz . . . . . Female spadiz very slender, filiform, 40 cm. long, with a large polished callus at its insertion and terminating in a filiform flagellum which is finely aculeolate all round; partial inflorescences very few, only two in one specimen and of these the lowest 12 cm. long, with 9 spikelets in all; primary spathes elongate, very narrowly tubular, closely sheathing, truncate at the mouth, fugaciously covered with blackish scales and densely armed all round with very small and very acute black-tipped claws; secondary spathes tubular-subclavate, attenuated a good deal at the base, where they are flat onthe inner side truncate at the mouth, obtusely apiculate at one side, smooth or armed with a few short claws; spikelets alternate, distichous, straight, filiform, horizontal or slightly deflexed, 2-4 cm. long, attached above the mouth of their own spathe, and with a distinct axillary callus; spathels tubular-infun- dibuliform, truncate at the mouth, indistinctly striately veined; involucrophorum shallow, sub-cupular, laterally attached to the base of the spathel above its own; involuere sub-cupular with unequal margin; areola of the neuter flower lunate with acute borders. Female fiowers small, 2°5 mm. long. Fruiting perianth distinctly pedicelli- form; the calyx glabrous, smooth, not striately veined, hardened and depressedly ventrieose at the base, divided down almost to the middle into 3 broad apiculate lobes; the corolla divided into 3 ovate, acute, smooth segments, which are longer by one-third than the calyx; stamens forming with the united bases of the filaments an urceolum which is as long as the calyx and is crowned by 6 triangular lanceolate, subulate teeth. Fruiting perianth pedicelliform. Fruit globular or a trifle longer than broad, 5 mm. long by 8-8:5 mm. in width and further topped by a cylindric beak 1°5 mm. long; scales subshining, but very finely scabridulous under a strong lens, almost flat, faintly channelled along the middle, broader than long, with an intramarginal light line and further bordered by another line of a chestnut- brown colour, their tips slightly prolonged and appressed, their margins almost entire- Seed irregularly globular, slightly compressed, 6 mm. long, coarsely alveolate; chalazal fovea roundish, shallow; albumen equable; embryo basal, Hasrrar.—Cochin-China, Loureiro, Rediscovered by Pierre in February 1879 on the Chiao-Xhoin mountains.—Native name in Moi language “Rani,” in Annamite * May Sap,” (Pierre No. 4834). O TIons.—Notwithstanding the very defective description of Calamus dioicus lefi by Loureiro, many considerations have induced me to identify with this species the specimens of Pierre described above. C. dioicus is related to C. javensis, which it resembles in the slenderness, length and toughness of the stem. It is, however, distinguished by the leaflets having 3-5 slender coste which are bristly-spinulous above and by the fruiting perianth being distinctly pedicelliform. | Prare 51.—Calamus dioicus Zour. Portion of the plant with a fruit spadix from Pierre's No. 4834. 40. CALAMUS SCHISTOACANTHUS Bl. Rumphia iii, 49; Mart. Hist, Nat. Palm. iii, 936; Walp. Ann. iii, 406 and $830; Miq. Fl. Ind. Bat, iii, 122, and in Journ. Bot. Neerl i p. 21, and De Palm, Arc. Ind. 27; Wendl, in Kerch. Les Palm, 237; Becc. in Rec. Bot. Surv. Ind, ii, 201. €. Kingianus} BECCARI. MONOGRAPH OF THE GENUS CALAMUS: 197 Description.—Slender, scandent, as thick as a man’s little finger at most. Leaf- sheaths obliquely truncate at the mouth, fugaciously tomentose, densely armed with very unequal spreading straight subulate spines, which are short or 2-3 cm. long, swollen and light-coloured above at the base, otherwise glossy, and of a leaden schistaceous colour. Leaf-sheath flagella slender, filiform, armed with small solitary or half-whorled claws. Leaves rather short (40-60 cm. long) not cirriferous; petiole short or almost obsolete, channelled above, rounded on the back, armed at the sides with some straight, 5-10 mm. long, spreading conical subulate dark- tipped spines; rachis dotted with brown deciduous scales, bifaced above, rounded beneath, where irregularly armed with small scattered solitary geminate or ternate black-tipped claws; leaflets not many, 12-13 on the whole, of which 4 somewhat more remote than the others and approximate at the apex, of these the two terminal entirely free at the base; those of the basal pair opposite; the intermediate ones irregularly and remotely alternate; they vary from 15 to 30 cm. in length and 10-12 mm. in breadth and are narrow, linear-ensiform and attenuate at the base, acuminate and bristly-penicillate at the apex, almost papyraceous, rigidulous, glabrous .and about the same colour on both surfaces, but sprinkled beneath with small scales which are visible under a lens, furnished with an acute mid-costa and 1-2 fine weaker nerves on each side of it—all naked on both surfaces; margins somewhat thickened by a slender nerve running alongside and appressedly bristly-ciliate only near the apex and smooth at the base; transverse veinlets sharp, rather remote ‘and much interrupted, Other parts unknown. Hasitat.—Sumatra, Praetorius; also in Borneo on thé River Dussoon, Korthais according to Blume. OxsERVATIONS.— Very imperfectly known. Blume says that the specimens from Sumatra agree with those from Borneo, and that only the leaves and leaflets of the first are more robust; nevertheless I entertain some doubt about the Bornee specimens, and I consider as typical those of Praetorius from Sumatra, of which I have seen one, kindly sent to me from the Leyden Herbarium, In this specimen the sheathed stem is 7-8 mm. in diameter; the leaves are almost without a petiole and bear five very narrow leaflets on each side with other four approximate at the top as described above. OC. schistoacanthus seems related to C, javensis; but its true affinities in the absence of the spadices remain uncertain. PLATE 52.—Calamus schistoacanthus B/. The entire type-specimen of Praetorius, from Sumatra, in the Leyden Herbarium, 4l. Catamus Kryeranus Becc. sp. n. Description.—Slender, probably scandent, Sheathed stem 10-12 mm. in diam. Leaf-sheaths cylindrical, strongly gibbous above, covered with a thin ashy-brown scurfy-crustaceous indumentum, very densely armed with rather short unequal slender laminar horizontal spines, which often form short interrupted very approxi- ‘mate and sometimes crested ridges. Ocrea inconspicuous, Leaves not cirriferous, 70 cm. long on the whole; petiole 16 cm. long in one leaf, slightly channelled near the ‘base and otherwise flat and naked above, sparingly armed at the sides and along 198 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. Kingianus the middle beneath with small solitary claws; rachis acutely bifaced above, rather convex beneath, where armed scantily and irregularly with scattered and solitary or 2-3:nate claws; leaflets very few, very inequidistant, approximate into a few remote fascicles of very few leaflets each; 4 leaflets are grouped at the apex, digi- tate and free at the base; they are all almost of the same size and shape, 25-26 cm. long, 3 cm. broad, lanceolate, almost equally tapering towards both ends, with an acute bristly penicillate apex, green even when dry, dull and almost of the same colour on both surfaces, papyraceous, rather rigid, distinctly and acutely 3-costate, with another rather distinct nerve near each margin and therefore sub-5-costate, the 3 larger costz on the upper surface spinulous from the middle upwards; beneath all the nerves more slender and naked; margins finely and appressedly spinulous near the base, more spreadingly spinulose towards the apex; transverse veinlets more dis- tinct on the lower surface than on the upper. Male spadiz not very large and rather rigid, simply decompound (not seen entire); lowest primary spathe tubular, elongate, closely sheathing, flattened, two-edged, armed at the sides with horizontal straight spines and, especially in its basal portion, with small often hooked prickles, obliquely truncate at the mouth, being prolonged at one side iuto a rather short obtuse point; attenuated axial portion between two partial inflorescences flat towards the base on the inner side, and convex on the back, where rather densely armed with un- equal scattered solitary claws; partial inflorescences (only on» seen and it the lowest) attached with a distinct axillary callus and transverse rima above the mouth of its spathe, 25 cm. long with 4-5 spikelets on each side and with a terminal spikelet larger than the side ones; secondary spathes tubular, slightly infundi- buliform, closely sheathing, unarmed, flattish on the inner side near the base, obliquely truncate, entire and naked at the mouth, where slightly prolonged at one side into a short point; spikelets attached outside their own spathe with a distinct callus and axillary rima, horizontal or arched downwards, slender, elon- gate; the lowest, the largest, 8-9 cm. in length with 20-22 flowers on each side, the upper ones gradually shorter; spathels asymmetrically infundibuliform, finely striately veined, truncate, entire and naked at the mouth, prolonged at one side into a triangular sometimes cleft rather acute point; involucre dimidiately cupular, attached to the base of the spathel above its own, acutely two-keeled and bi-dentate and with the margin deeply excavated on the side next to the axis. Male flowers rather remote, ovate-lanceolate, acute, about 5 mm. long; the calyx eylindraceous, obscurely striately veined, with 3 broad triangular acute lobes which form a third part of the length of the entire calyx; corolla about twice as long as the calyx with narrow acute externally polished segments.—Other parts unknown. Hasitat.—North-East India: found by Sir G. King’s collectors in Assam at an elevation of about 500 m, in February 1893 (Herb. Calc.). Ossrrvations.—Of this I have seen only one specimen of the stem with leaf and the basal portion of a male spadix; this probably was flagelliferous at its apex. It seems allied to ©. floribundus, from which it differs in the more diš- tinctly grouped leaflets and in the much longer male spikelets with remote flowers ; but its true affinities are somewhat uncertain in the absence of the female spadix l. Muelleri | BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 199 The main characters of C. Kingianus are the short leaves with a relatively long petiole; the very few lanceolate leaflets, which are distinctly fascicled, 3- or sub-5 coBtate ; the two terminal free at the base; the male spadix with partial inflor« escences terminating in a spikelet longer than the side ones which are elongate with many distant flowers. Prate 53.—Calamus Kingianus Becc. The entire type-specimen in Herb. Cale. 42. Catamus MurLLERG H. Wendl. in lett. to F. v. Muell; H. Wendl. & Drude in Linnea, xxxix. (1875), 193, pl. ii, f. 1-8; H. Wendl. in Kerch. Les Palm. 237; Benth. Fl. Austr. vii, 134; F. v. Muell. Census Austr. Pl. 119; Becc. Malesia i, 88, and in Rec. Bot. Surv. Ind, ii, 202; Bailey in Queensl, Fl. 1686. | C. australis (not of Mart.) F. v. Mueller Fragm. Phyt. Austr. v, 49 (fide Wendl. & Drude 1l. c.). : Description.—Scandent, slender, © Sheathed stem 6-8 mm. in diam, Leaf-sheaths flagelliferous, not gibbous above, rusty-furfuraceous, entirely covered with ascendent or spreading chestnut-brown (ultimately deciduous?) bristles, which are 6-7 mm. long at most, and become closer, longer and erect near the mouth of the sheath. Ocerea very short (a few mm. long) horizontally truncate, very densely bristly-spinulous like the leaf-sheaths, ^ Leaf-sheath flagella filiform, rather rigid, armed with very slender scattered claws. Leaves not cirriferous, short (about 30 cm. long); petiole almost obsolete or very short, trigonous, bristly-spinulous at the sides, bat chiefly underneath; rachis rather densely furfuraceous like the petiole, trigonous, bifaced and acute above, flattish underneath, where sparsely and irregularly armed at the sides and along the middle with slender scattered recurved spines, which change towards the summit into small claws; leaflets few (11-14 in all) very irregularly set, lanceolate or linear-lanceolate, almost equally attenuate to both ends, acute at the base, subulately acuminate at the apex, thinly papyraceous, about the same colour on both surfaces, almost shining, 3-sub-5-costulate, the mid-costa slightly stronger than the side ones, all naked on both surfaces or sometimes the mid-costa furnished near the base on the upper surface with a few (1-4) pale, weak, 5-7 mm. long spicule; trans- verse veinlets rather sharp and approximate; margins ciliated with small remote spread ing spinules; the largest leaflets, the mesial, 18-20 cm. long and 16-24 mm, broad, 8-4 of those nearer the base usually approximate and inserted very near to the sheath; the 4 uppermost also approximate and the two terminal free or more or less connate at the base. Male spadiz . . . . . Female spadiz supradecompound, elongate, flagelliform, very slender, with very few, remote partial inflorescences; primary spathes very elongate, tubular, very closely sheathing, rather densely aculeolate, bristly at the mouth; the lowermost slightly compressed, the others cylindraceous; partial inflor- escences furnished with a filiform flattened peduncular portion, the largest among them {the lowest) 10-15 cm. long with 6-7 spikelets on each side; secondary spathes very closely sheathing, tubular, slightly enlarged above, subtrigonous, smooth or aculeolate in their upper part, truncate and ciliolate at the mouth, where acute at one side; spikelets inserted above the mouth of their own spathe, with a very distinct, axillary callus; the largest (the lowest) 3-6 cm. long, with 8-12 flowers on each side; the 200 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. — (0. Muellerit appermost much shorter and with fewer flowers; spathes broadly infundibuliform, striately veined, truncate and subciliolate at the mouth, apiculate at one side; in- volucrophorum inserted at the base of the spathel above its own and with a distinct axillary callus, cupular, truncate and ultimately split; involucre cupular, often asym- metrically evolute ; areola of the neuter flower rather large, slightly concave with acute borders. Flowers horizontally set. Fruiting perianth explanate, its calyx rather coria- ceous, at first 3-toothed, later entirely split into 3 parts; corolla divided into 3 ovate- lanceolate thinly coriaceous segments one-third longer than the calyx; filaments of the stamens forming an urceolate cup which reaches to the middle of the corolla and is crowned by 6 triangular teeth. Fruit broadly obovate or subglobose, about 15 mm. long and 12 mm, broad, rounded at both ends, but topped by a short beak; scales in 18 series, dull dirty-yellowish, very faintly channelled along the middle and with a very narrow dark-brown intramarginal line margins pale, scarious ; tip short, triangular, erose-denticulate. Seed irregularly globose, 9 mm. in the largest diam. with a minutely rugulose testa and with a rather deep chalazal fovea about the centre on the raphal side; albumen horny, equable; embryo basal. HasrraT.—Australia: in Queensland, on the Brisbane River and in Moreton Bay, (Wendi, $ Drude); at Laguna Bay, where it is said to be common in certain places in the damp forest, Diels No. 8230 in Herb. Berol; Pine River, Hull & Mueller in Herb. Kew. In New South Wales on the Richmond River, Henderson in Herb, Kew; on the Clarence River, F. v. Mueller in Herb. Beccari. OnsERVATIONS,— [his seems a rather variable plant. The basal portion of a leaf from the Clarence River specimen has the petiole very short and the rachis densely furfuraceous, furnished with deflexed hooked spinules and with rigid subspiny bristles ; the basal leaflets (4 in number) are approximate, very narrow, subulately acumináfó, 11-20 em. long, 12 mm. broad, and bear on their mid-costa 1-4 needle-like, slender, straight, erect, 4-7 mm. long spines. The specimen from Pine River has a leaf 30 em. long with 11 leaflets which are devoid of spines on the mid-costa. In the specimen from Laguna Bay the leaflets are larger than in the above mentioned and only occasionally a straggling spinule may be seen on their mid-costa. The fruit I have described is from the specimen collected by Henderson on the Richmond River, Prats 54.—Calamus Muellerii Wendl. An intermediate portion of the stem with two entire leaves and the upper portion of a spadix with very young fruit; from Diels’s No. 8230 in Herb. Berol. Pirate 55.—Calamus Muellerii H. Wendi. The basal portion of a leaf with 4 leaflets, the spadix and two fruits in the lower right-hand corner, from the Clarence River specimen in Herb. Becc.; the partial inflorescence with full-grown fruit and the seeds, one entire and ha two halves of one longitudinally cut through the embryo, from the specimen from Richmond River in Herb. Kew: the entire leaf from the Pine River specimen in Herb. Kew. Canawus MUELLERIN var. MAcRosPERMUS Wendl. & Drude in — xxxix, (1875), 194, pl. II, f. 1, 9-11. . OmsERvATIONs.—1 have not seen a specimen of this variety, of which no Special locality is given by its authors, but the dimensions of the fruit assigned by them C. caryotoides] BEOCARI MONOGRAPH OF THE GENUS CALAMUS. 201 exactly agree with those I have registered above, However it must be remembered that the fruit of Calamus is somewhat variable in size according to its degree of maturity, 43. CALAMUS CARYOTOIDES All Cunn. in Mart. Hist. Nat, Palm, iii, 212 (ist edit.) and 338; Kunth. Enum. Pl. iii, 212; Walp. Ann. iii, 489 and v, 831; H. Wendl. & Drude in Linnzes, xxxix, 195; Wendl. in Kerck. Les Palm. 235; Becc. Malesia. i, 88 and ii, 77; Benth, ` Fl. Austr, vii, 135; F. v. Muell Census Austr. Pl. 119; Becc. in Ree. Bot Surv. Ind. iij 202; Bailey, Queensl Fl. 1686. Description.—Slender, scandent. Sheathed stem 5-8 mm. in diam. Leaf-sheath gibbous above, finely striate longitudinally, furfuraceous in youth, later glabrous, densely covered with rigid hairs like deciduous spiculz, which afterwards leave a sub- spiny tubercled base. Ocrea 5-6 mm. long, almost horizontally truncate, densely hispid. Leaf-sheath| flagella slender, filiform, very finely aculeolate. ^ Leaves short, 25- 40 cm, long, not cirriferous; petiole almost obsolete or very short and thick with a distinct swelling or callus at its axilla; rachis subtrigonous, bifaced above, armed irregularly, chiefly on the lower surface, with very small claws and often sprinkled with black-tipped sub-spiny tubercles ; leaflets very few (6-9 in all) very inequidistant, rigidulous papyraceous, rather shiny above, slightly paler beneath ; the two of the terminal pair are more or less united (sometimes almost to the apex) and form a broad, furcate flabellum which is cuneate at the base and has the terminal margin truncate, sinuous and premorse; side-leaflets alternate, oblong- spathulate or more usually elongate-cuneate, {gradually narrowing towards the base from near the apex, which is also irregularly truncate and premorse; the largest (the mesial) 15-18 cm. long, 3-4 cm. broad; the lowest approximate, narrower and shorter, usually divergent or even deflexed, all furnished with 5-7 slender costs diverging from the base; costz smooth on both surfaces, the central hardly stronger than the side ones, and all reaching the apex; transverse veinlets sharp, approximate and quite continuous; margins acute, more or less furnished with small and. remote spinules. Male and female spadices almost the same and simply decompound, inserted almost opposite the leaf near the mouth of the sheath with a distinct axillary callus, very slender, flagelliform, and terminating in a filiform aculeolate appendix; primary spathes very narrowly tubular, very long, and very closeiy sheathing, sparsely aculeolate, the lowest slightly flattened the others cylindrical, obliquely truncate at the mouth; axial portion between two partial inflorescences very slender and armed externally with small claws. Male spadiz 1-l:2 m. long, with 6-7 partial inflorescences, nodding, inserted inside their own spathes; the largest, the lowest, as much as 15 cm. long with 89 spikelets on each side; the others gradually smaller; the upper most reduced to a few spikelets; secondary spathes tubular-infundibuliform, closely sheathing, unarmed, obliquely truncate and with paleaceous cilia at the mouth; spathels shortly tubular at the base, suddenly enlarged into a concave limb, striately veined and prolonged at one side into a spreading acute point; involucre almost entirely exserted from its own spsthel and attached to the base of the one above, cupular, truncate, almost entire or slightly bi-dentate and bi-carinate on the side next Ann. Roy. Bor. Garp. CarcurrA Vou. XI. 202 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C. viminalis to the axis. Male flowers: narrowly oblong, 4 mm. in length; the calyx strongly striately veined, tubular, with 3 short very broad teeth; segments of the corolla shining and smooth externally, twice as long as the calyx. Female spadiz somewhat shorter than the male one, 60-80 cm, long, with 4-5 partial inflorescences which are short., and broad, the largest (the lowest) 6-7 cra. long, with 6-7 spikelets on each gide; secondary spathes as in the male spadix ; spikelets slightly arched, hori- zontal or somewhat deflexed, attached just at the mouth of their own spathe with a distinct axillary callus, 2-3 cm. long, and with 6-10 flowers on each side ; spathels shortly tubular at the base, suddenly enlarged into a spreading broad con- cave limb, aeute at one side; involuerophorum concave, with a short limb, inserted laterally almost entirely outside its own spathel, at the base of the one above; involuere cupular, entire; areola of the neuter flower lunate, relatively large and sharply bordered. Female flowers ovoid, about 3 mm. long, horizontally attached; the calyx strongly striately veined with 3 broad acute lobes; corolla with acute and polished segments one-third longer tian the calyx. Neuter flowers slightly smaller than the fertile ones. Fruiting perianth not pedicelliform. Fruit very broadly ovate or sub- spheeric, rounded at both ends, but topped by a short beak, 12-13 mm. long (including the beak), 9-10 mm. broad; scales in 18 series, dirty yellowish, not very shining, superficially channelled along the middle, with a narrow dark-brown intramarginal line and scarious erosely toothed margins and tip. Seed irregularly globose, with rather smooth surface, very convex on the back and with a slightly Gépressed chalazal fovea on the raphal side; albumen equable; embryo basal, Hasrrat.-—Australia: Queensland; Endeavour River, Allan Cunningham as from Martius; Bloomfield River, Rockingham Bay, Deliachy in Herb. Becc. from F. v. "Mueller; Russell River, W. A, Suyer in H. Becc. from F. v. Muelier; Cairns at Kamerunga, Warburg No. 19504 in Herb. Berol.;. Mt. Cook near Cooktown and Cairns, Z. Diels No. 8293, 8480 in Herb. Beccari. | OssERVATION.—This is a very characteristic species, easily distinguished from any other, so far as we know at present, by its short leaves with few irregularly truncate and præmorse leaflets which resembles those of some species of Ptychosperma. PrarE. 56.—Calamus caryotoides AU., Cunn., An intermediate portion .of the plant with a spadix bearing ripe fruit; an entire female spadix with very young fruit; the apex of another spadix with female flowers; seeds.—From Warburg’s specimens No, 19504 in Herb. Berol. | 44. Carawus viminaLis Willd. Sp. Pl. ii, (1799), 203 (not of inw. 1 Mart): Lam. Enc. Bot. vi, 306; Rees, ha s 9; ed é Schult, Syst. Veg. vii, 2*, 1328 (excl. all cit. except Herb. Kbi. pl. 55, f. 2, A. B.); Nees v. Esenbeck Plant. Of. t. A. B. (ex. Mart.) Blume Rumphia iii, 45 excl. €. viminalis 8 prostratus Bl. and excl. Q, viminalis R amplus Mart. (this is reduced by Mart. himself, 1. c. b 336, under C, buroénsis, to C. Reinwardtii); Mart. Hist. Nat. Palm. i iii 205 (1st edit.), only as to the plate in the Herb. Amb. and "| 336 under C. buroénsis; Becc. in Rec. Bot. Surv. Ind. ii, 203. C, viminalis ) BECCARI MONOGRAPH OF THE GENUS CALAMUS. 9208. C. Rotang var. E Linn., Sp. Pl, 2nd edit., p. 463 (partly). C. buroénsis Mart. Hist, Nat. Palm. iii, 336, as to the plate in Herb, Amb. not as to the plant from Buru; Walp. Ann. iii, 486, and v, 830; Mi Fl. Ind. Bat. iiij 121, and De Palmis, 27; Becc. Malesia, i, 88. i C. litoralis Bl. Rumphia, iii, 43; Mart. Hist. Nat. Palm. iii, 333; Walp. Ann. iii, 483, and v, 830; Mig. Fl. Ind. Bat. iii, 114, and De Palmis, 27; Teysm. Cat. Hort. Bot. Bogor. 75; Kurz Veg. Bangka in Natuurk. Tijds, Ned. Indie, xxvii (1864), 218. j C. gracilis Roxb.? Zolling. Syst. Verzeich., 79 and exsicc. No. 2884. Palmijuncus viminalis Rumph. Herb. Amb. v, pl. 55, f. 2 A, B. DEsCRIPTION.—Scandent, of moderate size; sheathed stem 2-3 cm. in- diam. naked canes 15-20 mm. in diam., shining, vitreous and straw-coloured on the surface. Leaf-sheaths occasionally flagelliferous, gibbous above, fugaciously mealy- furfuraceous; when young armed with straight pale flat spines, which are broad and concave beneath at the base, 1-3 cm. long, almost horizontal or slightly deflexed, scattered or not very regularly arranged into more or less approximate oblique series. Leaf-sheath flagella very long, with the lowest spathes flattened and sparingly spinous at the margins, clawed upwards. Leaves 1-1:256 m. long, not cirriferous; petiole very short, rounded beneath, flattish akove and armed at the sides with straight horizontal spines, rachis bifaced above, slightly rounded or flattish below, where armed, chiefly towards the apex, with strong straight spines, which are 1-4 cm. long, broad at the base; very acuminate, solitary, geminate or ternate, spreading, horizontal or deflexed ; leaflets very many, pointing different ways, more or less grouped in distant fascicles of 2-4 (more rarely 5-6) on each side of the rachis, the leaflets near the apex more regularly arranged and sometimes almost equidistant and on one plane; all narrowly lanceolate, attenuate at the base, gradually acuminate at tbe apex, 15-30 cm. long 1-25. em. broad, green, almost shining, about the same colour on both surfaces, with the mid-costa acute and spinulous above, this fainter and usually, but not always, naked beneath; secondary nerves weak, naked or sometimes sparingly spinulons beneath or on both surfaces; margins acute, regularly and closely ciliate-aculeolate. Male spadix opposite to the leaf, very long, decompound and ending in a long slender clawed flagellum ; partial inflorescences many, diffuse, bearing 7-8 alternate spikelets on each side; primary spathes narrow, elongate-tubular, cylindrical, coriaceous, closely sheathing, almost horizontally truncate at the mouth and very shortly apiculate at one side, armed with scattered claws; secondary spathes very narrowly tubular-infundibuli- form, 10-15 mm. long, truncate, entire, apiculate and ciliolate at the mouth, glabrous, unarmed ; spikelets very slender, filiform, 10-20 cm. long, horizontal or deflexed, inserted at the mouth of their own spathe; spathels 30-35 in every. spikelet, very small, tubulose at the base, suddenly enlarged into a broad infundibuli- form, entire, laterally acute limb; each spathel with a secondary very small subscorpioid spikelet, which is composed of very few (4-8) small bi-seriate approxi- mate flowers, rarely more; secondary spathels bracteiform, membranous, broadly ovate; involucre formed by two very small membranous, broadly ovate, acute Ann. Roy. Bor. Garp. CarcurrA Vor. XI. 204 ANNALS.OF THE ROYAL BOTANIC GARDEN, CALCUTTA. [C.. viminalis bracteole, which are united by their bases but do not form a cup. Male flowers very small (2°5 mm. long), ovate; the calyx divided almost to the base into 3 large acute lobes, of which 1-2 are keeled and more or less distinctly ciliolate on the back; the corolla twice as long as the calyx, ovate, obtusely trigonous, glabrous, divided almost to the base into 3 ovate rather acute lobes, superficially striately veined externally ; stamens with linear-subulate and—at the apex—inflected filaments ; anthers lanceolate, shortly bilobed at the base; rudimentary pistil columnar, 3-dentate, shorter than the filaments. Female spadiz about 2 m. long (including the terminal flagellum), simply decompound, with 5-6 partial inflorescences, which are shorter than those of the male spadix and bear 3-6 subdistichous spikelets on each side; primary and secondary spathes as in the ‘male spadix; spikelets more robust than the male ones, inserted just at the mouth of their respective spathe, usually 8-10, but sometimes even 20 cm. long, with numerous bifarious flowers; spathels usually furfuraceous, short, broadly infundibuliforn from a narrow base, truncate and shortly apiculate at one side at the mouth; involucrophorum very short, subdiscoid, supported and embraced by its own spathel and attached at the base of the one above; involucre orbicular, subdiscoid or shortly cupular; areola of the neuter flower depressedly lunate. Female flowers small, about 3 mm. long; the calyx divided into 3 ovate- acute, not distinctly striately veined lobes; the corolla as long as the calyx; fila- ments of the stamens forming an urceolate cup which is crowned by 6 short teeth. Fruiting perianth explanate. Fruit small pisiform, spherical or slightly depressed or sometimes subturbinate, 8-9 mm. in diam., crowned by a distinct narrow cylindric beak; scales in 15 series, dirty-yellowish, shining, channelled along the middle, not bordered by a darker intramarginal line, almost obtuse, sometimes darker at the . apex, their margins entire. Seed globose, slightly compressed, about 6 mm, broad and 4 mm. thick, opaque, convex and deeply pitted on the back, flattish on the raphal side, with a round central chalazal fovea; albumen equable; embryo subbasal. Hasrrat.—C, viminalis, with its varieties, has a rather wide geographical distrib- ution, being found in Java, India, Burma, in the Andamans and in Cochin-China, but the plant growing in Java must be considered as the type, and therefore is that which I have described above. From this island I have seen the specimens collected by Winter near Batavia (Leyden Herb.) named C. litoralis by Blume, and others from the same locality in the Delessert Herbarium ; these last probably came from Burmann’s collections. Zollinger’s No. 2884 comes from the forests of the Province of Banjuwangi, also the No. 2652 of the same collector in Herb. De Cand. belongs to this species, The native name in Java according to Zollinger is ^ Rotang Glatek.” Blume gives that of “ R. Ayer.” Rumphius says that the entire intertwined canes are made into cablés for anchors, OBSERVATIONS.—À very distinct species because of its leaflets usually pointing different ways and being grouped in many fascicles; by the leaf-rachis being armed beneath with long straight deflexed spines, which often are ternate and divaricate: and by the glomerulate male flowers which form a very short subscorpioid HM Ü. viminalis) —- BECCARI. MONOGRAPH OF THE GENUS CALAMUS. 205 at each spathel The typical C. viminalis (as I have already: said) is the Javan plant named C. litoralis by Blume, which however is hardly distinguishable by constant characters from the var. fasciculatus, which name I have assigned to -the ancient C. fasciculatus of Roxburgh, common in many parts of India and in Cochin-China, m In some Javan specimens, as, for instance, in those of Zollinger No. 2881 (Herb. Boiss. and Herb. Deless,) and in No. 2652 (in Herb, DeCand.) as many as 20-24 of the uppermost leaflets are regularly alternate in one plane, but in other specimens, also from Java, al the leaflets are distinctly fascicled and pointing different ways, and only 4-5 are regularly set at the apex. The spathels and the involucre of the male. spikelets are perhaps a little smaller in the Javan than in the Indian plant. C. viminalis was first made known by Rumphius, and was figured in vol. v of the Herbarium Amboinense, plate 55, f. 2, under the name Palmijuncus viminalis, * Rotang Java." The explanatory description of that plate is however worthless, being a ^ mixtum compositum” derived from heterogeneous elements, but the plate itself is so highly characteristic on account of the peculiar armature of the leaf- rachis formed by long straight deflexed spines, as to leave no doubt as to its identi- fication. Willdenow (1799) in the Species Plantarum first applied the name of C. viminalis to Palmijuncus viminalis, but he was wrong in considering all the figures of the plate 55 in vol v of the Herbarium Amboinense as belonging to a single species, while figure 1 in that plate represents another species—the Palmijuncus verus latifolius ( C. pisicarpus Bl,); but by the diagnosis of C. viminalis “aculeis . . , . . frondium distantibus reflexis," it is easy to see that Willdenow had applied that name only to the species represented in figure 2 of the said plate 55. i | Blume ( Rumphia iii, p. 45-46) has well established that the name of C viminalis Willd., ought to be applied to the above mentioned f. 2, plate 55, but ia this plate Blume has not recognised his own C. litoralis, which is certainly synonymous with C. viminalis. Furthermore, Blume (l. c., p. 46) believes that the f. 9 of the same plate 55 cannot represent a Javan species of Calamus, as Rumphius writes when speaking of the Javan Calamus that he had not seen the leaves, and from this passage Blume infers that the Calamus figured in plate 55, f. 2, must be one from Buru, included by Rumphius under the comprehensive name of Palmijuncus viminalis But I have to point out that Rumphius, when speaking of the “Dragon Blood,” shows he had had much correspondence with Javan people about Rotangs, and it is very probable that figure 2 of plate 55 had been drawn after the chapter on the Palmijuncus viminalis had been already written. But this is of little or no importance. What is certain is that Willdenow has is his €. viminahs on plate 55, fig. 2, vol v of the Herbarium