pay Sy : _ gC Say ; a Keeper EMER GENRES Se : Ta MAT ss . . Pa SS a GE as sep ss aE re Fe ps ! op oS ee PU ot” ae ne “ ner a = Le tT Es as 3 ee ane Fat 7 ine SP One OR oan Pon Ce eek eee ADP ny ie ee NEEES eee EN Th aaa. eee ali . he ; hc ‘ a ¢ i > eat SS ae a Pre TN eT aad ° ets , oper i : a Sot Sw, ‘ . : Mi 7 Ath TAXONOMICAL REVISIONS a ae mW 7) he ( 4: ax 4 "we 2a hi is st +f 83 1 on : cy Psy er Lie | Ve : iad ME ek nk te ee kc OM 2 J eel ‘ : a é PASH ir Y 7 ; 4 eivists - aa PP eres 8 be. 7 ” REPUBLIK INDONESIA REPUBLIC OF INDONESIA LEMBAGA ILMU PENGETAHUAN INDONESIA (L.LP.L) INDONESIAN INSTITUTE OF SCIENCES FLORA MALESIANA BEING AN ILLUSTRATED SYSTEMATIC ACCOUNT OF THE MALESIAN FLORA / INCLUDING KEYS FOR DETERMINATION / DIAGNOSTIC DESCRIPTIONS | REFERENCES TO THE LITERATURE|/|SYNONYMY]/AND DISTRIBUTION / AND NOTES ON THE ECOLOGY OF ITS WILD AND COMMONLY CULTIVATED PLANTS PUBLISHED UNDER THE AUSPICES OF LEMBAGA BIOLOGI NASIONAL BOTANIC GARDENS OF INDONESIA / BOGOR / JAVA AND OF THE RIJKSHERBARIUM / LEYDEN / NETHERLANDS PREPARED ON ANINTERNATIONAL CO-OPERATIVE BASIS UNDER THE SUPERVISION OF SEVERAL DIRECTORS OF BOTANIC GARDENS/ KEEPERS OF HERBARIA AND VARIOUS PROMINENT BOTANISTS FOR THE PROMOTION OF BOTANICAL SCIENCE AND THE CULTURAL ADVANCEMENT OF THE PEOPLES OF SOUTH-EASTERN ASIA TO THE SOUTHWEST PACIFIC REGION SERIES I VOLUME 9 SPERMATOPHYTA GENERAL EDITOR: Dr C; G. G. J. VAN SITEENIS DIRECTOR OF THE FOUNDATION ‘FLORA MALESIANA’ PUBLISHED BY: MARTINUS NIJHOFF / DR W. JUNK PUBLISHERS THE HAGUE / BOSTON / LONDON WONT — iN Oks) Distributors: for the United States and Canada Kluwer Boston, Inc. 190 Old Derby Street Hingham, MA 02043 USA for all other countries Kluwer Academic Publishers Group Distribution Center P.O. Box 322 3300 AH Dordrecht The Netherlands Library of Congress Cataloging in Publication Data (Revised for volume 9, part 3) Main entry under title: Flora Malesiana. At head of title: Republic of Indonesia, Ministry of Agriculture. Issued in pts. Includes bibliographical references. 1. Botany--Malay Archipelago--Collected works. 2. Phytogeography--Collected works. 3. Plant collectors--Malay Archipelago--Collected works. I. Steenis, Cornelis Gijsbert Gerrit Jan van, ed. II. Indonesia. Departemen Pertanian. III. Series. QK366.F56 582 .09598 (2p le ISBN 90-247-2780-4 (this volume) Publication dates Part 1 27 Dec. 1979 Part 2 23 July 1982 Part 3 28 Jan. 1983 Copyright © 1983 by Martinus Nijhoff Publishers, The Hague. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, mechanical, photocopying, recording, or otherwise without the prior written permission of the publisher, Martinus Nijhoff Publishers, P.O. Box 566, 2501 CN The Hague, The Netherlands. PRINTED IN THE NETHERLANDS CONTENTS ERROR ene Se bea CNG ia «6 aa «1a oleae ale ante d sr BSE bas ar aie (3) UICTIINI Eo as ee a a wid ea C28OCE GE eek ss ee (5) Dedication by R.E.G. Pichi Sermolli & C.G.G.J. van Steenis ...................- (7) aia RISENTIS ORINIT SIDED ete aa. cee os 2 2's? Jas 3S P a eke ae ere Pee bee ee ee (45) TAXONOMICAL REVISIONS in alphabetical sequence aeaemnimeene —T ENY WY, Be; ENDENENSOOEN eet sin ca ae as gs eta a i cre 1 eyperaceac—IP by: 3-H. Kern .&,H:-P..Nooteboom. 22825 2e> ene eee ss ee ee 107 Pcrocm pacene by PS. AShtonis... 33.9 2n;s ke ee ee oe eee 237 mehecne—F By: 1:P. Jessop). cia. 2 nsec ton oan. Ree eee eee 189 ADDENDA to volumes 4—9 Addenda, corrigenda et emendanda by C.G.G.J. van Steenis ef al. ............... 553 INDEX Index to scientific plant names by M.J. van Steenis-Kruseman & E.E. van Nieuwkoop 575 1906 5 a tele, F Vtuseo, thd. Karmann. 14 Jd Dovewhn 1919. C uff nice Pas : O = ; 5, : Oren to Atunento + neatorel per une prtitihe, pratire Lhe halla AMM» COR ma Bre. perience bla AN Aapore. te fra Be turk A “Boren A proves per esta, d.. tyadare Alt. fore prof file dn Oaths Parne auche 1 prdesre. br feperiiia. dire. te t1 Opts Pirie 2 a escron purde ee a ore Arimaudan._ de Ae pulla A erte tidy avyeuuts gleth ptgalrve A bile (oT | 5 4, ahtte_ A ian refprthas whe Yt Bin at feuipn Coffs pus t¢elrgen IS ae Crk Liu cor Al__ fiatl~ AS Ay atc Aids atect fies >) Affe Alti~-w : DV hettar, fe Dedicated to the memory of ODOARDO BECCARI DEDICATION A dedication to ODOARDO BECCARY, the greatest botanist ever to study in Malesia, is long overdue. Although best known as a plant taxonomist, his versatile genius extended far beyond the basic field of this branch of Botany, his wide interest leading him to investigate the laws of evolution, the interrelations between plants and animals, the connection between vegetation and environ- ment, plant distribution, the cultivated and useful plants of Malesia and many other problems of plant life. But, even if he devoted his studies to plants, in the depth of his mind he was primarily a naturalist, and in his long, lonely and dangerous explorations in Malesia he was attracted to all aspects of nature and human life, assembling, besides plants, an incredibly large number of collec- tions and an invaluable wealth of drawings and observations in zoology, anthropology and ethnol- ogy. He was indeed a naturalist, and one of the greatest of his time; but never in his mind were the knowledge and beauty of Nature disjoined, and, as he was a true and complete naturalist, he was at the same time a poet and an artist. His Nelle foreste di Borneo, Viaggi e ricerche di un naturalista (1902), excellently translated into English (in a somewhat abbreviated form) by Prof. E. GicLiotiand revised and edited by F.H.H. GUILLEMARD as Wanderings in the great forests of Borneo (1904), is a treasure in tropical botany; it is in fact an unrivalled introduction to tropical plant life and animals, man included. It is a most readable book touching on all sorts of topics and we advise it to be studied by all young people whose ambition it is to devote their life to tropical research. In the last years of his life, BECCARI was rearranging his diaries, notes and observations of the expeditions to eastern Malesia with the intention of publishing a second book on his explorations, but very unfortunately death did not allow him to carry out his wish. He left only a revised copy of his diaries and field notes which formed the bulk of the book Nuova Guinea, Selebes e Moluc- che, published posthumously by his son NELLO BEccaRiin 1924. It is neither well known nor duly appreciated outside Italy, since no translation has been published. Undoubtedly it lacks the glam- our and freshness of the previous book, being devoid of the original and acute observations deriv- ed from his field research, which are largely profuse in his Nelle foreste di Borneo, but it offers a good and fascinating description of his adventurous travels with a wealth of interesting remarks, and it is an invaluable documentation of the natural features of those almost unexplored coun- tries, particularly of the characteristics and customs of their inhabitants. This synthesis, however, is only part of his oeuvre and before entering on his achievements, let us first look at his life and the development of his ideas and ideals. BECCARI’s early youth was ill fated. He was born in Florence, in his father’s home in the Via dei Benci at the corner of Borgo dei Greci, on November 16, 1843. His mother, ANTONIETTA MI- Nuccl, from Radda in Chianti in Tuscany, died soon after his birth, and his father GruSEPPE BEC- CARI, from an ancient family native of Rimini (Romagna), died in 1849 when ODOARDO was Six years old; he was brought up by his maternal uncle Minuccio Minuccl. In April 1853 he entered the Collegio ‘Ferdinando’ in Lucca, where his love for botany was nurtured by the Vice Rector and Prefect of Studies, the Abbé IGNAzio MezzetTtT1! and by his Professor of Botany in the Lyceum of Lucca and Director of the Botanic Garden, CESARE Biccui. The latter, aware of the talent of his pupil and perhaps foreseeing his glorious future, in 1860 dedicated to him a new spe- cies, Tulipa beccariana’, the first of the numerous plants and animals to be named in his honour. Beccarr’s first collections date back to 1856, when he was a student of the College of Lucca and still only 13 years old. During his stay there he assembled a herbarium, which was still in exis- (1) In his honour, in 1871, BEccARI named a new genus of Annonaceae Mezzettia. (2) Tulipa beccariana Biccut, Agg. Fl. Lucch. (1860) 21, nom. nud.; 1 Giardini 8 (1861) 50, t. 2. (7) FLORA MALESIANA tence at the beginning of the present century. A search for it in Lucca as a separate herbarium was unsuccessful, but several specimens with labels headed ‘Erbario Beccari’ are present in the herbarium of IGNAZIO MEZZETTI, now kept in the Lyceum Machiavelli of Lucca, where BECCARI attended secondary school until July 1861. Whether these specimens are part of the separate BEc- CARI herbarium included in MEzzeTTI’s herbarium or duplicates of it, is uncertain. Other plants collected in the period in which BEccarI was a student in Lucca are kept in WeBp’s herbarium in Florence. In August 1861 he published his first paper and in the autumn of the same year, when he was 18 years old, BECCARIcommenced his studies in the Faculty of Natural Science at the University of Pisa. At first, perhaps under the influence of Biccut, he devoted his attention to the Cryp- togams and already in 1862 BECCARIs name appears together with those of several eminent botan- ists of the time, among the collectors of the ‘Erbario Crittogamico Italiano’, the classical exsiccata with printed labels, founded in 1858 by GIUSEPPE DE Notaris of Genoa. In the University of Pisa, BECCARI distinguished himself so much in botany, that the celebrated botanist PlETRO SAvi made him an assistant to the chair of botany in January 1863, while he was still an undergraduate. Dissatisfied with the conservatism of SAvi, however, he gave up his assis- tantship and moved to the University of Bologna from where he took his degree in Natural Science on July 1, 1864, by disputing his thesis on the structure of the lichen Arnoldia cyathodes Mass. [ = Plectopsora cyathodes (Mass.) KORBER] with the famous professor in botany ANTONIO BERTO- LONI. Before his graduation, BEccaARthad already planned a long journey to far away regions, when in June 1864, in the laboratory of Prof. GIOVANNI CAPELLINI, geologist at the University of Bo- logna, he met Marquis Giacomo Doria, a young, impassioned naturalist, later patron and Maece- nas of science, and founder of the Museo Civico di Storia Naturale at Genoa which bears his name. The two young men, united by the same enthusiasm for scientific exploration in unknown countries, soon fraternized. Shortly after his graduation, BECCARI visited Doria at Genoa, where they decided to undertake a long exploration together and, counselled also by the celebrated British naturalist JOHN BALL, they chose the Kingdom of Sarawak, in Borneo, as the destination of their enterprise. As part of his preparations, BECCARI spent the period from February to April 1865 in the great British centres of botany, the British Museum in London and Kew Gardens, ob- viously to acquaint himself roughly with the plants of Borneo. He met the HooKERs, CHARLES Darwin and Sir JAMES BROOKE, the Rajah of Sarawak, who assured him of the assistance of his nephew, the Tuan-muda, Sir CHARLES BROOKE, then governing the territory in his absence. Thus prepared, young OpoarRpDo, at the age of 22, commenced his studies on the flora of the Malesian tropics, which was to become his main life occupation, and in which he would rise to the greatest heights as a scientific explorer, naturalist, and botanist. He sailed from Southampton on April 4, 1865, and met Doria and his own brother, GIOVANNI BATTISTA BECCARI (who was on his way to Japan), at Alexandria. From there they travelled by train to Suez and by boat to Aden and then to Ceylon, where they spent a fortnight. There BECcARI visited the famous Botanic Gar- dens at Peradeniya and climbed Mt Petrotallagalla, where he made his first personal acquaintance with the tropical flora and started collecting. Via Penang island and Singapore, the voyagers arriv- ed on June 19, 1865 at Kuching, capital of Sarawak, which they had selected as their base of opera- tions. At the beginning they were guests of the Tuan-muda, Sir CHARLES BROOKE; later they settled in a house of their own with servants, and also bought a small boat (‘sampan’) for their excursions in the forests along the river. Soon BEccARIand Doria took up their botanical and zoological col- lecting in the dense and primitive forests which at that time surrounded Kuching. BECCARI was anxious to know the mountains and to collect intensively; thus he undertook the construction of a (8) Dedication big hut in the forest of Gunong Mattang at an altitude of about 300 m with the intention of making it the base for the explorations and collections in the primitive forests of Borneo. However, after some months the health of Doria deteriorated to such a point that at the beginning of March 1866 he was forced to return to Italy. Thus, BEccARI, having accompanied his friend to Singapore, re- mained alone to carry out the programme which he had planned with him and had hoped to realize together. At the beginning of April he moved to his house in the forest of Gunong Mattang which he called ‘Vallombrosa’, after the great monastery hidden in the dense mountain forest of Prato- magno, east of Florence. Together with a Chinese cook and four Malesian boys he spent nearly all the remaining months of 1866 there, except for four excursions and a period in October-No- vember at Kuching to pack his collections. On one of these excursions to Gunong Poe he discov- ered a new species of Rafflesia, the famous parasitic plant with vegetative parts extremely reduced and a gigantic flower, the largest in the plant kingdom, apparently arising directly from the stem of the host liana. The largest flower of the new species, Rafflesia tuan-mudae, so named in honour of the Tuan-muda of Sarawak, Sir CHARLES BROOKE, attained about 56 cm in diameter. In Nelle foreste di Borneo BECCARI gives many details of his hut at Mattang. He had cleverly designed it for drying plants and preparing zoological specimens rather than for lodging; soon it became an active and efficient laboratory, full of all sorts of products of nature. In his book he described his primitive life there as very happy and fully suited to his temperament. With only a cotton coat, trousers and a Chinese straw hat, mostly bare-footed, he carefully explored the sur- rounding primary forest, assembling marvellous collections of plants and animals. Back at his hut, he devoted many hours to arranging his collection, making drawings and descriptions and recording those notes and observations which later became the basis of his fascinating book Nelle foreste di Borneo. At the beginning of 1867 BEccaRt abandoned the hut at Mattang and spent the first two months at Kuching collecting in the surroundings, but chiefly arranging and packing his large collections. In March 1867 he again undertook his adventurous wanderings with the intention of visiting the interior of Sarawak. One of his trips from mid-March to the last days of May was devoted to the exploration of Batang-Lupar and the lakes of Kapuas with the main purpose of hunting orang- utan. He assembled there one of the best collections of these animals (skin, skeletons, heads and skulls, and even a foetus) and a wealth of observations which allowed him to express the opinion that the hominids did not originate in dense forest, like that of Borneo, and that the orang-utan, particularly well adapted to an arboreous environment, would be, not an ancestor, but a collateral of man. In his opinion, the hominids were derived from forerunners, allied to the great anthro- poids of tropical Africa, with an anatomical conformation, particularly of the limbs, more suited to evolve towards a biped gait and an erect habit and they had their origin in more open vegetation, like that of some regions of tropical Africa, where we find the greatest number of large mammals with rapid locomotion. Recent research in south-western Ethiopia seems to support this hypo- thesis. From August 12 to September 14, 1867, BEccarr collected in the district of Bintulu and in the country of the Kayan. From there, he was looking forward to organizing an expedition to the inte- rior regions of Sarawak, which at that time were still nearly unexplored and hardly visited by Europeans; but his project found every possible difficulty and obstacle. Despite them, without guide or interpreter, but with only four men and a small boat, he set out on September 15 from Bintulu on his journey through the interior of Sarawak along the basins of the main rivers Bintulu, Redjang and Batang Lupar and their tributaries, across the ridges of hills and mountains which represent the watershed between them. The journey was made mostly sailing up, or down, the rivers in various native paddle boats obtained from time to time from the natives, but also on foot (9) FLORA MALESIANA to overcome some impassable rapids or to cross the ridge between two adjacent basins; often he was forced to walk with difficulty in the stream beds, or to proceed slowly with a compass through the dense forests; more than once he was in real danger, even near the end of his travels when, having lost his compass, without food, in an unhabited region, he got lost for two days in a dense forest. On November 20, 1867, BEccaARI arrived at Kuching where he concluded this long, hard and risky enterprise. In the first two years of his stay in Borneo BEccarr’s health remained excellent, but in the last months it had been deteriorating. Already in June 1867 he had suffered the first attack of malaria and later many others followed. Furthermore, in July of the same year he had observed the first symptoms of elephantiasis on his right ankle. After the expedition to the interior of Sarawak he spent two months in Kuching arranging and packing the collections he had made. He had planned another long journey crossing the inland of Borneo from Kuching to Pontianak; but in January his health worsened, and being unable to subdue the high fever which had troubled him for some days, Beccariwas forced to undertake his homeward-bound voyage. He left Kuching on January 29, 1868 and arrived in Italy on March 2, after explorations in Borneo which had lasted almost three years. In Florence BEccARIwas the guest of his old friend from the College of Lucca, EmiLio MARCUC- ci, who had taken up the profession of architect but had not given up his love of botany, and who greatly assisted BECCaRI in that period in recovering his health. The house was located in Borgo Tegolaio 48, very close to the Museum of Physics and Natural History, where BEccarti had prob- ably assembled his collections. The house soon became a meeting place for young lovers of natural history including LEvreR and Sommrer. Soon after his return from Borneo G. Doria and R. GEsTRO, from the Civic Museum of Natural History of Genoa, were also his guests for several days, evidently to be informed, in detail, about the large zoological collections he had assembled in Borneo. At that time BECCARI was very busy sorting out and working on his collections; he also made agreements with collaborating specialists to study particular groups such as seagrasses (ASCHER- SON 1871), pteridophytes (CESATI 1876), mosses (HAMPE 1872), lichens (VON KREMPELHUBER 1875) and hepatics (DE NoTaris 1876), efc. (see Appendix 4); he probably also started distributing dupli- cate specimens of his Bornean plants. However, together with his technical work, he carried on with the study of his collection. In March 1869, Beccari started, at his own expense, the publication of a new periodical, the Nuovo Giornale Botanico Italiano, which was intended as a replacement of the Giornale Botanico Italiano founded in 1844 by F. PARLATORE, but interrupted in 1852. BEccaARI edited three volumes of the new periodical (1869, 1870, 1871) with the help of his friend Marcuccl, to whom he dedi- cated the new genus Marcuccia (Annonaceae) as a sign of gratitude for the help received in editing these volumes, particularly during his travels in Ethiopia in 1870. His first accounts of Bornean plants appeared in early volumes of his journal and many other papers dealing with his collections written by himself and other botanists were published in subsequent volumes. However, in spring 1871, when preparing for his expedition to New Guinea, BECCARI became aware of the difficulties of editing a journal when abroad making long expeditions in distant regions and handed the man- agement of the Nuovo Giornale Botanico Italiano over to T. CARUEL, who edited it until the end of 1893, when the journal became the official publication of the Italian Botanical Society, which it continues to be. Fascinated by his primitive life in Borneo, BECCARI was not satisfied with city life. Probably he had already developed the idea of undertaking a second journey to Malesia, when he received an offer to join an Italian expedition to Ethiopia. He sailed on February 14, 1870, from Genoa and (10) Dedication together with the zoologist Marquis ORAZIO ANTINORI and Prof. Arturo IssEL, geologist at the University of Genoa, visited the Bay of Assab and later, on behalf of the Italian Geographical Society, the country of Bogos. There he assembled a rich collection of plants (315 species of sper- matophytes and pteridophytes and 289 species of mosses, algae, fungi and lichens) enumerated and partly described in MARTELLI’ Florula Bogosensis (1886). He came back to Italy on October 20, 1870. Soon after his return to Florence, BECCARI materialized his project to visit Malesia again and after careful preparations, training himself in geodetics, astronomy and meteorology, he set out on November 24, 1871 from Genoa for the island of New Guinea, accompanied by Count Luici MarRIA D’ALBERTIS, an Italian nobleman who was passionately fond of hunting and natural history. Their first visit was to West Java where they stayed for some time in the Botanic Gardens at Bogor. The young director, Dr. RUDOLPH SCHEFFER, must have facilitated his exploring for a few days on Mt Gedeh, with its Tjibodas mountain garden, and adjacent primary forest on Mts Pangerango and Megamendong. Further stops were made at Flores and Timor in the Lesser Sunda Islands, and the islands of Banda and Ambon, where they arrived on March 7, 1872, and enjoyed the kind and helpful hospitality of Captain P.F. Kraatand his wife, the Italian lady AMALIA Ma- LAN. After a short journey to Buru and Ceram for information they came back to Ambon, where they organized the expedition to western New Guinea renting a small schooner, the ‘Bu- rung-Laut’, of 25 tons with a crew of eight men. On March 21, 1872, BEccARI and D’ALBERTIS sailed from Ambon to New Guinea, and having touched the islands of Geser and Goram reached Kapaor on the west coast of the mainland. Later, on April 30, they arrived at the small island of Sorong where they dismissed the Burung-Laut and rented a hut in which they established their first scientific station, collecting chiefly along the Ramoi river. They remained on Sorong island till July 15 when, having left their collections there with one of their men as keeper, they sailed in a ramshackle indigenous sailing boat with a crew of eleven Papuas to Dorei and later to Andai, where they arrived on August 7 after a long voyage, full of adventures. BeccaRI and D’ALBERTIS established their home and headquarters for the exploration of the region in a large Papuan hut built on high palafittes near the Andai river surrounded by the forest, with a splendid view of both the sea and the Arfak Mts. While D’ALBERTIS made a trip in the mountains, Beccarti collected intensively in the vicinity of Andai, but on September 28 he moved to Putat on the lower slopes of the Arfak Mts with the intention of exploring the higher regions. Unfortunately on October 9, he was informed that D’ALBERTISs had fallen seriously ill and he was compelled to return to Andai. The poor health of D’ ALBERTIS necessitated that the travellers return to Ambon, but only on November 2 was BEccariable to find an indigenous boat to reach Sorong and later Ambon. But in Sorong, unexpectedly, they found a schooner sent from Ambon in search of them. After recovering, safe and sound, the collections left there four months before, they sail- ed to Ambon, where they arrived on December 5, 1872. There BECCARI and D’ALBERTIS were greatly surprised to find the Italian Royal Corvette ‘Vettor Pisani’ on which D’ ALBERTIS obtained a passage, leaving his friend alone. Thus, BEccari concluded his first expedition to New Guinea, during which, despite all sorts of difficulties and serious health troubles, he had assembled a col- lection of about 700 species of plants and a rich amount of zoologica!, ethnographical and mine- ralogical specimens. BECCARI remained in Ambon for about two months to arrange and pack the collections. There, as a guest of Captain KraaLand his wife, he soon regained his health and prepared a trip to the Aru and Kei islands. Beccari departed from Ambon on February 8, 1873, having obtained a passage on a Dutch (11) FLORA MALESIANA Government steamer. Stricken with smallpox en route, BECcARI nonetheless reached the Aru is- lands on February 22, and with his base on Wokam, he collected plants and animals and made a topographical survey of the islands (see Appendix 1, C: Maps). On July 6, he moved to the Kei islands in a big local sailing boat, a Bughis prahu, on which Beccari was the guest of its Chinese master. But the boat suffered shipwreck on the east coast of Grand Kei. Fortunately he could save all his collections and collecting equipment. BEccARtI found the flora of these islands unexpectedly poor and after visiting Small Kei as well, he sailed on October 4 to Ambon with four men in a small indigenous sailing boat of only 4 tons, bought at Dulan. Despite the premonitions of the natives, the risky voyage of about 350 miles was successful and on October 23 he reached Ambon, where he stayed for two weeks, partly to arrange his collections, but chiefly to recover his strength, being again a guest of his friends, the KRAALs. On November 5, 1873, BEccaRIsailed by the steamer ‘Koning Willem III’ towards the West Mo- luccas (Buru and Ternate), proceeding via North Celebes towards Southwest Celebes, where he disembarked at Makassar on November 18. He stayed in the region for nearly three months until February 6, 1874. From there, as a paying passenger on an old Chinese boat of about 40 tons, similar to a prahu, he went to the larger islands south and southeast of Celebes (Kabaena and Muna) and to Kendari on the southeast coast of Celebes where he arrived on February 23. He re- mained in this district for six months to collect and make topographical surveys, but the collec- tions did not increase very much because the flora was not particularly interesting and because the region was plagued with pirates on the sea, and head-hunters on land. He chiefly collected in- land at Lepo-Lepo. Here he was informed that a Dutch vessel was looking for him at Kendari. It was the Escort vessel ‘Sumatra’ of the Royal Dutch Navy, which had been sent from Makassar in search of him, since it was rumoured that he was in danger from the pirates which infested the sea of Kendari. BEccarthad already decided to leave Kendari and accepted with pleasure the kind offer of the Captain of the vessel to take him and his men aboard to Makassar. Having packed his collections he sailed from Kendari on August 10 and arrived at Makassar after a voyage of five days. In these last months Beccarr’s funds had been running out, but early in 1874 he had already written from Makassar to his friend G. Dortain Genoa for financial help to carry out his project of a second expedition to New Guinea. When he was back at Makassar on August 29 he received the joyfull news that his friend had convinced the authorities of Genoa to contribute 15,000 lira towards a new, second expedition to the great island. Aware that the season was not suitable for sailing to New Guinea, BEccartsoon left Makassar by the same steamer ‘Koning Willem III’ on which he had travelled from Ambon to Makassar some months before. He proceeded to Bali, Surabaja, Semarang, and through the interior of Java to Bogor, to recuperate and to sort out his collections. There, he also spent some days at Tjibodas and on Mt Pangerango collecting. Unwearied, BEccarileft Jakarta on October 15, 1874, and via Surabaja, Makassar, the island of Bima and Timor in the Lesser Sunda Islands he arrived at Ter- nate island in the Moluccas on November 11. He remained there about 20 days and assembled rich botanical and zoological collections in the primitive forest near the hut (named by him ‘Para- disino’) which the Dutch Resident had built for him on the slopes of the volcano. BEcCARI intended to organize his travel to New Guinea from Ternate, but soon he realized that this was impossible, and on December 4 he left the island by the mail-steamer arriving at Ambon three days later. There he prepared for his new expedition financed by the Province and the Municipality of Genoa to West New Guinea, his old hunting grounds of 1872. He hired for his voyage the brig-schooner ‘Deli’ with a crew of 10, and accompanied by 8 men and a young boy for collecting plants and animals. (12) Dedication He sailed on January 22, 1875, and arrived at Sorong Island on February 1, establishing his base in the schooner and making trips to Ramoi, Dorei Hum, Mt Morait and venturing inland from Has as far as a river, the War Samson, not then recorded on the maps. Together with plants he amassed a rich collection of birds. On March 5 he left Sorong and after a visit of some days to Waigeu Island chiefly hunting for birds, he proceeded to Dorei and soon to Warbusi and Momi on the west coast of Geelvink Bay mainly with the aim of obtaining some specimens of cassowa- ries. Later, in April, always in the ‘Deli’, he went to the islands of the Bay which he had not pre- viously visited, spending nearly twenty days on Japen, a week on the uninhabited Mios Num, twenty days on Schouten Island and a week on a small island, Pulo Manim, near Mafor. On June 2 he arrived at Dorei where he found the Italian Corvette ‘Vettor Pisani’, and where he received the warmest welcome. Having arranged his collections he prepared the expedition to the Arfak Mts; on June 16 he started from Andai towards the mountains establishing his exploration base at Hatam (1500 m) in the centre of the mountain group. On June 23 he climbed one of the summits of the Arfak Mts (about 2000 m). He had planned to remain two months in the mountains, but on July 12 he was forced to cut short his exploration and to come down to the coast because of beri-beri among the crew of the schooner: two men had already died and the others were seriously ill. Thus BEccaRI realized that there was no choice; he had to go back to Ternate: his second ex- pedition to New Guinea was nearing its end; his dream of climbing and exploring the highest re- gions of the Arfak Mts had vanished for a second time. On July 18 he left Mansinam, near Andai, where the ‘Deli’ was riding at anchor, and on August 4, 1875, he arrived at Ternate, but in the meantime the beri-beri had killed most of the crew. He remained at Ternate three months ar- ranging his collections, notes and observations. The scientific results of the second expedition to New Guinea were very important. Even if the botanical specimens were not particularly numerous, the zoological collections were very plenti- ful, especially the skins of birds which surpassed 2000 in number, and included a set of birds of paradise which still remains one of the best of its kind. No less abundant were the ethnological collections consisting of every sort of object in use by natives. Also an important set of approxi- mately 200 Papuan skulls enriched the anthropological collections. But the expedition was also very fruitful because of his untiring activity as a naturalist and explorer in making notes of every- thing that attracted his attention and in studying all the aspects of the regions he visited. During the expedition he had also made various topographic surveys which later allowed the geographer Gurpo Cora (see Appendix 1, C: Maps) to draw maps of some regions of New Guinea, and he had assembled a great wealth of botanical, zoological, ethnological and anthropological observa- tions which are profusely reported in his letters published by E.H. GiGLioti, and in his book Nuova Guinea, Selebes e Molucche. Full of interest are the observations on the characteristics and origin of the Papuans, and on the life of birds, particularly those on the bower-birds of paradise, Amblyornis inornata, and its ‘capanne e giardini’, which are carefully and at the same time poeti- cally described by him in a paper full of interesting scientific and philosophical considerations, pervaded with a deep-rooted love and admiration of Nature. Learning that a Dutch expedition to New Guinea was being prepared, with the vessel ‘Soeraba- ja’, with the aim of performing a bathymetric survey, BECCARIreceived permission to accompany this. It lasted from November 11, 1875 till January 29, 1876, visiting Dorei, the Bay of Wanda- men, the islands of Roon and Krudu, the Bay of Humboldt, the island Arimosa, Awek (Japen I.), Dorei, Waigeu, Misool, the Bay of MacCluer, the Bay of Gouns, the island of Geser (off Southeast Ceram), and Ambon. From there he returned on the mail steamer to Ternate. He stayed there about a month to arrange and pack his latest collections and to ship them to Italy. This third expedition to New Guinea had not yielded results as far as BEccaARr’s botanical and zoological in- (13) FLORA MALESIANA terests were concerned, but it had allowed him to assemble many ethnological and anthropological notes, and to improve his topographic surveys. On March 12, 1876, Beccart sailed from Ternate to Java, on the first stage of his homeward voyage. He arrived, unexpectedly, at Florence on June 18, 1876, after about four years of bold and glorious exploration. On his return BECcARI was received with great honour. On July 14, 1876, the Municipality of Florence bestowed the freedom of the city on him; some scientific societies, such as the Zoological Society in London, and the Italian Anthropological Society, elected him an honorary member. Other scientific associations, such as the Italian Geographical Society and the Tuscan Society of Horticulture as well as the Faculty of Science of the Royal Institute of Advanced Studies of Flor- ence awarded him a gold medal. But he was not affected by these honours and devoted his time to his collections and to his friends in Florence and Genoa. However, the glamour of exploration and the call of the wild were too strongly in his nature and after a year BECCARI made one further long voyage to the Malesian islands. He and Captain Count Enrico A. D’ALBERTIS, a cousin of his former companion, set out from Genoa on October 14, 1877, on a trip, properly intended more for pleasure than for science, to Australia, en route travelling through India from Bombay, Lahore, Delhi, Benares, Lucknow, to Calcutta, touching Singapore and Kuching (December 1877), meeting in Australia FERDINAND VON MUELLER, and proceeding to Tasmania and New Zealand. On the return voyage he parted from D’ALBERTIS in Singapore and proceeded to Jakarta and then to Bogor where he spent two weeks, preparing a collecting trip in Central West Sumatra. Sailing from Jakarta on 28 May 1878, he arrived in early June via Padang and Padang Pandjang at Mt Singalang, a primary-forest-clad, long-extinct volcano of nearly 2900 m height. Here he had a hut built, as before in Sarawak and Ternate, which he made his headquarters. The hut, named by him ‘Bellavista’, was placed above the limit of cultivation and on the lower fringe of the primi- tive forest, at an altitude of about 1700 m. He remained there from June 12 to early August, mak- ing rich collections on the flanks and on the top of the volcano. Later he set his base in a house in the village of Ajer Mantcior at the base of Mt Singalang till September 20. After a short stay in Padang to arrange his collections, he undertook a journey on October 4 in the provinces and on October 22 he sailed from Padang to Bangkok where he arrived on November 10. During his travels in West Sumatra (see map of his itineraries in BECCARI 1930) he assembled large botanical and zoological collections; the largest were made on Mt Singalang, the harvest of plants running to a thousand numbers in all. Amongst them were the famous Rafflesia arnoldii and the then un- known, largest, erect aroid in the world, Amorphophallus titanum, a really colossal herbaceous plant, the tuber being up to 53 cm in diameter, the inflorescence more than 1.5 m high, the lamina of leaf covering a surface of about 15 m in circumference and the petiole attaining about 29 cm diameter at the base. From Bangkok he began his homeward journey to Italy, arriving in Florence on December 28, 1878, thus concluding the last of his fascinating explorations in Malesia. Reviewing the results of his botanical activities in the six years exploration in the field through almost the whole of Malesia, it is evident that BEccarr’s exploration in Sarawak was the most fruitful and thorough, with the huge number of over 4000 collections in two and a half years. The great virtue and value of his collections can only be properly estimated if one takes into considera- tion that Beccari collected species rather than specimens, and that he seldom collected a species twice. Each species was studied, dissected and annotated on the spot and mostly carried flowers and fruit. The Sumatran collection again was rather large, about 1000 numbers in five months travel, especially when one considers that BECcCaARr’s interests were wide; in Sumatra he also dedi- cated time to the study of agriculture, forest products and fruit trees, as he had done in Borneo. (14) Dedication The amount of these collections clearly contrasts with less than 1000 numbers in the three years spent in the Moluccas, Celebes and New Guinea (see Appendix 4), although really there are more, as most collections from the Kei and Aru islands and Kendari in Southeast Celebes are unnum- bered and unlisted. As the flora of at least New Guinea is not less rich in proportion to that of Borneo or Sumatra, the reasons for this contrast can only be explained by Beccarr’s activity in the field. Firstly, we must consider that he was a thorough collector and disliked gathering occa- sional or incomplete specimens; he preferred always to stop some days to collect systematically in a place which he considered botanically interesting, rather than to gather here and there along his path, en route, when moving from one place to another in his long expeditions. In a word, he preferred to collect intensively rather than extensively. In Borneo and in Sumatra where he had huts for drying, labelling and drawing his specimens (‘Vallombrosa’ and ‘Bellavista’ respectively) his collections were more numerous than in New Guinea, Celebes and Moluccas when he frequent- ly moved his collecting base, sometimes being forced to do so because threatened by native head-hunters, or by pirates. Finally, we must also bear in mind that the second expedition to New Guinea was made thanks to the financial support of the Province and Municipality of Genoa, se- cured on the warmest recommendation of Giacomo Doria, his friend and zoologist of the expedi- tion to Borneo. BEccaARI knew that he longed to enlarge the zoological collections of the Civic Museum of Natural History which he had founded in 1867, and, in his profound honesty, he felt bound to assemble large zoological collections for the Genoa Museum and to put zoology before botany. Indeed, the zoological collections were very rich in quality and in quantity, while the bo- tanical collections were not particularly numerous. Another reason for this contrast can also be found in Beccarr’s health which was worse in eastern Malesia than in Borneo and Sumatra. In 1878 Beccari was still only 35 years old, but had accumulated an unrivalled, immense amount of material, great scientific-botanical experience and knowledge of the Malesian flora, in fact had proved himself the greatest explorer of his time. He would prove himself also to be the greatest botanist in the elaboration of his results, surpassing BLUME in the width of the field he covered, including plant-geography, ecology and biology. Having concluded his explorations in Malesia, BEccart devoted the rest of his life to the study of his collections and of palms, except for an unhappy experience as Director of the Botanical Col- lections and Garden of the Royal Museum of Physics and Natural History of Florence in 1878— 1879, and a short journey to Ethiopia soon after. The vicissitudes of BEccaRr’s life as Director of the Botanical Collections and Garden cannot be understood without knowledge of some of the events in the history of the Florence Museum and the sale of BEccarr’s collections. The Royal Imperial Museum of Physics and Natural History in Florence was founded in 1775 by PreTRo LEOPOLDO Di LorENA, Grand Duke of Tuscany, and was installed in a building in Via Romana, not far from the Palazzo Pitti, the palace of the Grand Duke, at the base of the great and famous Giardino di Boboli. Part of this was soon designated as the Botanical Garden of the Museum. Thanks to the great interest of the Grand Duke in Natural Science, the scientific collections were greatly increased in the years thereafter and the Museum was subdivided into various sections (Cabinets) and in 1789 an astronomical observatory was also installed there. From then on all the Museum complex was usually named ‘La Specola’ by the Flo- rentines. The botanical section consisted of the so-called Botanical Collections (herbaria, carpo- logical collections, vegetable products, wax models and fossil plants) and the Botanical Garden. After various events dominated by the historical course of Tuscany in the first half of the 19th century, FiLippo PARLATORE was appointed in 1842 director of the Collections and Garden. He greatly contributed to the growth of the herbaria and to the organization of the department. In 1854 Pump BARKER WEBB died in Paris and bequeathed his invaluable herbarium and library to (15) FLORA MALESIANA the Florence Museum, together with an annual income of 6945.58 lira (derived from the sale of a palace in Paris) and known as Wesp’s Legacy, for their maintenance and increase. In 1859 the Government of Tuscany established the Istituto di Studi Superiori Pratici e di Perfezionamento in Firenze, which had its centre near the Monastery of San Marco; and the Museum of Physics and Natural History with its collections and the botanical garden, although situated on the op- posite side of the river Arno, became part of that Institute as the seat of the Faculty of Science. However, the Museum continued to have a director of its own, and in 1868 PARLATORE was ap- pointed to that office. Very unfortunately, the Institute of Advanced Studies, which only became the University of Florence in 1923, had no Rector responsible for the scientific and didactic activi- ty, but only an Administrative Board which determined the course of events in the Museum in the following years without an adequate knowledge of the problems and needs of scientific re- search. In 1860, after various vicissitudes, the Giardino dei Semplici, founded in 1545 and one of the most ancient in the world, became state property, and, in 1869, together with the adjoining buildings (originally the stables of the Grand Duke) was assigned to the Institute of the Advanced Studies being situated near the centre of the Institute at San Marco. In 1872 the Italian Govern- ment, the Province and the Commune of Florence signed a convention for the enlargement of the Institute of Advanced Studies, and the Board of the Institute decided to move some of the Cabi- nets of the Museum to the centre of the Institute. The latter also foresaw the removal of the Botan- ical Collections and Garden of the Museum from La Specola to the Giardino dei Semplici and pertinent buildings at San Marco, in order to have the botanical department nearer to the Institute, and to maintain only one garden, the Giardino dei Semplici, by far more famous than that of the Museum. Thus, during 1877 and 1880 the Cabinets of Chemistry, Physics, Geology and Mineralo- gy were shifted from the Museum to San Marco. Only in 1879 did the Commune of Florence ac- tually hand over the Giardino dei Semplici to the Institute of Advanced Studies. The latter decided that the Botanical Collections and Garden ought soon to be moved to San Marco. However, they were conveyed there only several years later owing to the opposition of several botanists, among whom BEccarI, and other personalities, who were against the removal of the botanical collec- tions, as they considered it, for several reasons, to be a great error. Fitippo PARLATORE died on September 9, 1877. He was the last Director of the Museum of Phys- ics and Natural History in Florence as the Institute had decided that the Dean of the Faculty was to hold the directorship of the Museum. However, the office of Director of the Botanical Collec- tions and Garden was vacant, and according to general opinion BECcaRI was the best qualified and most worthy successor to the work of PARLATORE, who had so greatly enlarged the herbarium and library and raised them to the level of the greatest museums in the world. But BEccaRr’s ap- pointment was strongly opposed by the Dean of the Science Faculty and the Board of the Institute of Advanced Studies, particularly because BECCARI was firmly convinced that the Director of the Botanical Collections and Garden should have no hand in teaching. On October 14, 1877, hardly more than a month after PARLATORE’s death, BECCARI undertook his travels with E. D’ALBERTIS, sketched above, and at that time no resolution had been taken. Only on March 26, 1878, while BECCARI was journeying in Australia, was he, in spite of the opposition, appointed Director of the Botanical Collections and Garden of the Royal Museum of Physics and Natural History of Florence, with the duty of supervising the practical phytographic research of the students. BECcA- Ri returned to Florence from the exploration of Sumatra on December 28, 1878, and soon took up the office of Director. In the preceding years BEccaARI had organized, worked and studied on his own in the field of botany, and his brilliant achievements were naturally a one-man show. In the field he had to make his own decisions, and learned to do so immediately. He had no rivals and had always very subor- (16) Dedication dinate personnel whom he could command. That was very different from the situation in which he was now placed as a Director, with a graded staff accustomed to some privileges acquired dur- ing the long directorship of PARLATORE. On the other hand Beccari was indefatigable and tidy in his work and he required everybody to be active and precise in carrying out his duty. He wanted to infuse new life into the operations of the botanical collections and garden, introducing methods that differed from the traditional ones. Very soon his reforms became unpopular among several of the staff, who felt that he lacked respect for the memory of his predecessor; this state of affairs acerbated the hostility towards him of the Institute of Advanced Studies. But such hostility was not something to scare him. It is clear, of course, that BECCARI, with his enormous drive and ambition, proved by his unique exploration and study of the tropical floras, thinking big, botanically and otherwise, wanted to raise the Florentine centre into an institute which could compete with the leading world herbaria. For this there was excellent opportunity as the Florence botanical collections had already acquired the enormous and (still) most important herbarium of PHiLIp BARKER WEBB, bristling with types of old collections, with funds attached for its maintenance, and further the important herbaria of CESALPINO, MICHELI, TARGIONI, and PARLATORE. To these could now be added his own numer- ous collections from the East, with its enormous mass of duplicates for the further enrichment of the Florence herbarium by exchange. There was, therefore, every reason and opportunity to fulfill his ambition. When Beccaritook on the directorship, he was aware that the botanical collections were under threat of removal from the Museum to the buildings adjoining the Giardino dei Semplici near the centre of the Institute of Advanced Studies. At first he was rather in favour of the project since he thought that the Institute had large funds for the construction of new buildings and that they would be better and more suitable for the collections than those of the Museum. But when he dis- covered that the funds of the Institute were scarce, and the buildings were the old stables of the Grand Duke of Tuscany, very humid, unsuitable for both the herbarium and library, and far worse than those of the Museum, he became a most obstinate and relentless opponent to the removal of the collections. His hostility was the primary cause of a wide gulf between him and the Institute Board. On the other hand Beccari began to understand that his ambition of raising the Florence Botanical Collections to the level of the other great European herbaria would be dif- ficult to realize. But this controversy was not the reason for his resignation as Director, at least not the main one. Indeed Beccarr’s resignation was for a different reason. On one of his visits to Java during his travels he had been requested by the Dutch East Indies Government to sell his collections to the Bogor (Buitenzorg) Herbarium for the cash payment of a considerable amount of money and his appointment as botanical explorer in the Garden, or a life annuity of 5000 lira. The offer was alluring, but BEccaRI wished his collections to remain in Italy and to spend the rest of his life in Florence, attending to their study. However, his own estate was seriously compromised owing to the expenses for his long expeditions, and before his depar- ture for the last journey to Australia, New Zealand and Sumatra (1877—1878) BEccARI undertook negotiations with the Florence Institute of Advanced Studies for the sale of his Malesian collec- tions. The Institute asked Marquis G. DortA, the Director of the Civic Museum of Natural History for an appraisal of BEccaRr’s botanical collections, which were estimated at 75,065 lira. On the basis of this valuation and considering the offer of the Dutch East Indies Government, the Insti- tute of Advanced Studies offered to buy all his botanical collections from Malesia against an annu- ity of 5000 lira for the rest of his life. BEccaRt accepted this offer but on the condition that the collections were entrusted to the Museum of Physics and Natural History of Florence and that he was entitled to have them at his disposal during his lifetime; in exchange he would assume (17) FLORA MALESIANA responsibility for their study and conservation. During his journey, in May 1878, when he was in Batavia prior to his expedition to West Sumatra, BECCARI was informed that the Board of the Institute of Advanced Studies had accepted his conditions and had officially approved the pur- chase of his collections. Consequently he refused the offer of the Dutch East Indies Govern- ment. When Beccaricame back to Florence and took up the directorship of the Botanical Collections and Garden of the Museum, he ought to have accepted the contract but learnt that the Board of the Institute of Advanced Studies intended to use the money from the Wess legacy for the pay- ment of his life annuity. BECCARI was greatly disappointed on hearing this decision, since he had thought that his life annuity would be paid by different Institute funds: he disliked the idea that the WEBB collections were to be deprived of nearly all their endowment until his own death. Fur- thermore, he knew that the WEBB legacy was the main source of income of the botanical depart- ment of the Museum and without it, his ambition to make the Florence herbarium one of the greatest in the world and a leading centre of tropical botany could not be accomplished. He under- stood too that under these conditions, the sale of his collections was incompatible with the duty of his office as Director. Indeed, he got a personal benefit from the sale, while as Director of the Collections and Garden it was his duty to avoid that these were deprived of a large amount of money necessary for their maintenance and increase during his lifetime. Thus he made every effort to persuade the Board of the Institute of Advanced Studies to use different funds to purchase his collections, but without result. He did not underestimate the hostility that the Institute had shown him since the beginning, and particularly recently, and when he was invited to sign the contract, he clearly understood that he was regarded by them as a troublemaker, and that the decision to pay the price of his collections with the WEBB legacy was merely an expedient devised by the In- stitute in order to compel him either to lose his prestige as Director or to resign from his office. Indeed, if BEccarthad sold his collections and kept his office he would have lost his prestige as a man and as Director, having put his personal interest before his duty. But the Board of the In- stitute knew that BECCARI was a man of honour and that it would achieve its aim: his resigna- tion. Then, as a last attempt, he tried to find some way in which, without going back on his word, he could withdraw from the compromise of the sale, but without success. Thus, on July 26, 1879, Beccari resigned as Director of the Botanical Collections and Garden of the Florence Museum. Only later, on October 31, 1879, as a private citizen, did he sign the contract for the sale of his own collections. The fight had been lost, but his honour was saved! Soon after the end of this unhappy and painful experience, on November 16, 1879, BECCARI left for Ethiopia to stay with his old friend and benefactor, Marquis Giacomo Doria, as members of an Italian expedition to the Assab Bay on the Red Sea which he had already visited in 1870. They also spent some days collecting in Aden and returned to Florence on February 26, 1880. After his return from Ethiopia, BEccARI resumed the study of his collections, which were lo- cated in a few small rooms on the top floor of the Museum of Natural History. In those modest and secluded rooms, alone, like a hermit, without any assistant or help, but together with his rich collections, he worked until his death. There, he wrote his famous works on Malesian plants and on palm taxonomy, and made the splendid drawings and photographs which adorn his publications. The first months there, however, were unfortunately rather hard for him. After his resignation, in November 1880 TEoDoRO CARUEL was appointed Director of the Botanical Collections and Garden of the Museum. He was soon instructed by the Institute of Advanced Studies to study the advisability of removing the Botanic Collections and Garden from the Museum to the Giardino (18) Dedication dei Semplici and adjoining buildings, and eventually to prepare a project for such a removal which had already been decided, but not realized, before PARLATORE’s death. BECCARI was aware that CARUEL between 1866 and 1871 had been Director of the Giardino dei Semplici and that he was in favour of the removal and was preparing the pertinent project. BEccArt had already expressed his resolute opposition to the removal, chiefly because the buildings near the Giardino dei Semplici were unsuitable for the collections and library as they were very humid, smaller and worse than those of the Museum, but also because library, herbaria and garden were well settled in the Museum and there was no need to remove them, and in doing so waste a large amount of money which could have been used for their maintenance and growth. Besides, he was strongly convinced that the great botanical collections and library at the Museum, as a centre of taxonomic research, had to be kept distinct from the centre of teaching and research on anatomy and physiology at the Giardino dei Semplici. Thus, in 1880 and 1881 Beccari tried everything, with letters and ar- ticles in various Italian newspapers, to convince the Faculty of Science, the Board of the Institute of Advanced Studies, and public opinion that the removal of the botanical collections and garden would be a great and irreparable mistake. He carried out a referendum against the removal of the herbaria and library among botanists in Italy and abroad. This was spread far and wide and was discussed in many Italian and foreign publications. Numerous botanists from every part of the world, and among them the most eminent taxonomists of the time, declared themselves against the removal. In 1881 Beccari published a paper in which the reasons for the protest against the removal of the botanical department from the Museum and the result of the pertinent referendum were given. In the same year CARUEL published his study for carrying into effect the project of the removal of the Botanical Collections and Garden. Despite the opinions of many and outstanding botanists expressed in the referendum, the Board of the Institute of Advanced Studies decided on the removal of the Botanical Collections, Library and Garden from the Museum of Natural History, in Via Romana, to the Giardino dei Semplici and adjoining buildings near San Marco, on the op- posite side of the river Arno. However, BEccartdid not give in, and he continued to publish other articles and papers against the removal until 1903. Even if his campaign did not gain its aim, it greatly contributed to further resolutions of the Institute of Advanced Studies which decided to enlarge and improve the buildings annexed to the Giardino dei Semplici and later to reserve for Botany the part of them originally intended for the Zoology department, which remained at the Museum, at La Specola, where it still is today. These deliberations greatly delayed the removal of the Botanical Collections, though the living plants of the Garden of the Museum were all moved to the Giardino dei Semplici during 1883. CARUEL continued to give his botanical lectures at the Museum until his retirement, in 1896, but the following year, his successor, O. MATTIROLO, under- took his teaching in the building near the Giardino dei Semplici. When in 1900 he moved to Turin, P. BAccaRINI succeeded him as Director. At that time, the Library and the Botanical Collections were still located in the Museum. In 1901, when the removal was close at hand, BEccaARI was requested to inform the Institute of Advanced Studies in which rooms of the new botanical building at the Giardino dei Semplici he wished to have his Malesian collections deposited. He disdainfully replied that he wished his collections to remain at the Museum in Via Romana in agreement with the contract of their sale, adding that, if the Institute had decided to move them to the new buildings, he would not follow them and would give up their study. His Malesian collections remained in the same rooms at the Museum until the end of his life, but, in 1905, in spite of further protest and particularly after controversy with P. BACCARINI, the removal of the Library and the Botanical Collections (includ- ing all the herbaria) from the Museum of Natural History to the new Botanical Institute near the (19) FLORA MALESIANA ancient Giardino dei Semplici, was brought to a conclusion. However, let us resume the course of BEccart’s life after his sad experience as Director of the Botanical Collections and Garden of the Museum in 1878—1879, and his hard fight against their removal started in 1881. These regrettable events marked a turning-point in BECCARI’s career. He realized that he had lost the chance of making the Florence Herbarium one of the leading centres for research in plant taxonomy, and decided to retire to private life, devoting himself entirely to taxonomic research, chiefly to elaborate his own Malesian collections for which he had gathered a wealth of field observations and drawings. On January 23, 1882, BECCARI married NELLA GORETTI DE FLAMIN], from a noble family of Ca- sentino, in the high valley of the Arno. They had four sons: NELLo, Dino, Baccio and RENZO. The eldest, his devoted son NELLO, became a professor of Comparative Anatomy at Florence Uni- versity and took great pains in editing some posthumous papers by the father, among them the book Nuova Guinea, Selebes e Molucche based on the original diaries of his father’s explorations in eastern Malesia from 1871 to 1876. He also encouraged U. MARTELLI and R.E.G. Picut SER- MOLLI to revise and edit some works on palm taxonomy which had been left unfinished by his father. The years immediately following BECCARI’s marriage, entirely devoted to his family and to study, were peaceful and fruitful. He set up his home in a villa inherited from his father, the me- diaeval Castello del Bisarno, near Ripoli in the immediate vicinity of Florence, and he lived there until his death. According to information obtained from his nephew and from letters to his friends, we know that he also had another house in the city of Florence where he and his family spent the week-days, particularly in winter. We also know that he used, as in the years before his marriage, to spend several weeks, particularly in the summer holidays and during the grape-har- vest at Radda in Chianti, on the old country estate of his mother’s family. He was very fond of country life and, following the tradition of the old families of the region, he was particularly inter- ested in wine-making, in which he attained great experience. He was one of the first producers, together with Baron B. Ricasott, of that typical wine, well-known in Italy and abroad as ‘Chianti, Gallo nero’. Nevertheless, during his holidays he did not stop his research, even if he did not work so actively as in the Florence Museum, where his collections were housed. However, after a few years his life was troubled by another sad event. In 1877 he had under- taken the publication of a great work, Malesia, mainly with the intention of embodying in it the results of the studies dealing with his own collections from the Malesian Archipelago; in addition, other papers or abstracts of works published elsewhere on plants of that region were also to be included. The first two volumes were printed in Genoa and BECCARI was greatly helped in editing them by his faithful friend R. Gestro, the Director of the Civic Museum of Natural History of Genoa, particularly during his last journey to the East. However, BECcARI undoubtedly corrected the proofs of all the instalments of Malesia and also those of fascicle 3 of volume 1, issued when he was in West Sumatra. This is proved by a letter to Gestro from Buitenzorg, now Bogor (dated May 2, 1878), which accompanied the corrected proofs of that fascicle. BECCARI published the first two fascicles of volume 1 at his own expense, but later Malesia became a publication of the Florence Institute of Advanced Studies. However, despite its great interest, the sale of this work was obviously limited, the text being written entirely in Italian. Consequently the Institute of Ad- vanced Studies decided that it was not worth continuing its publication and suddenly, in 1887, stopped all contributions to it while fascicle 3 of volume 3 was not yet complete. Actually, in the cover of fascicle 3 we find a note which informs us that the publication of Malesia is ended and explains the reasons for it. However, BECCARI wished to publish at least the text pertinent to the drawings of the account on Bombacaceae already issued in fascicle 3, but as far as possible, also (20) Dedication other papers and drawings ready for the press. Hence, he was compelled to beg in Italy and abroad for funds necessary to publish the last two fascicles of volume 3 of Malesia. Fortunately, the Min- ister of Education, PAOLO BosELLI, and the BENTHAM Trust in England, where he was highly es- teemed, allowed him the necessary financial support for bringing volume 3 of Malesia to a close. The last issue appeared in March 1890. Great was BEccarRt’s disappointment at the unhappy conclusion of the publication of Malesia, not only because the resolution of the Institute of Advanced Studies represented a slight to him and to his work, but also because he had lost a safe and certain means of publication for the results of the study of his collections. He was so much upset by this event that he even thought of visiting Malesia again. Actually, in the letter to Gestro(April 4, 1890) which accompanied the last fascicle (‘ultimo definitivo’) of Malesia he asked his friend for information on the departures from Genoa to Batavia and about the liners. Anyhow, the end of Malesia was another turning-point in his life: it marked the beginning of a long period of inactivity, after which he never resumed the study of his own Malesian collections. Having concluded the studies already undertaken, he published no scientific papers from 1893 to 1902, except some articles and letters protesting against the removal of the Botanical Collec- tions of the Florence Museum, described above, and the temporary closing of the herbaria and library in connection with this removal. In these years he was on the point of giving up his botanical activity entirely and none of his colleagues and friends, not even his devoted pupil UGo- LINO MaRTELLI, were able to induce him to resume his research. However, another person was to have the credit for reviving in him the enthusiasm for the country where he had spent the most fruitful period of his youth. After his explorations in Sarawak Beccarthad kept alive his friendship with the Rajah and the Ranee of Sarawak. They liked to spend part of the year in the surroundings of Genoa. It is difficult to say whether it was by chance or with the definite intention of helping BECcaRI to overcome his scientific inactivity, but in the early days of May 1897 the Ranee visited Florence and met BECCARI. A woman of great culture and sensibility, deeply fond of her kingdom of Sarawak, Lady MARGa- RET BROOKE succeeded in convincing BECCARI to write a book on his fascinating explorations in Borneo. He soon began his work; the Ranee kindly assisted him in the preparation of the book, particularly in providing him with the illustrations. Several of them, in fact, area selection from many fine photographs taken by the Ranee herself in Sarawak; these were assembled in a great album, still kept in the Florence Botanical Museum, which she presented to BECCARIon June 10, 1897. The preparation of the book, Nelle foreste di Borneo, led BECcaRI to recover a certain interest for scientific work and it was not difficult for Prof. OrREsTE MATTIROLO, the Director of the Botanical Department of the Museum, to complete the Ranee’s enterprise and convince him to resume botanical research after the publication of his book. However, BEccarthad already real- ized that the removal of the library and the herbaria from the Museum to the Giardino dei Semplici was close at hand, and that without them the elaboration of his own Malesian collections would be rather difficult. Consequently, he did not resume their study, interrupted in 1890 after the pub- lication of Malesia was stopped, and decided to devote his entire botanical activity to a single group. This decision was neither easy to take nor satisfying for him, but represented the best solu- tion for carrying out his research without a big library and a great herbarium at his disposal at any moment of the day. The selection of the group was easy, as the Palms had intrigued him since his first visit to Malesia, and he had already done some research on them. On the other hand he knew that this group offered him a taxonomically nearly unexplored, big field. Thus, BECCARI started again on his studies with renewed enthusiasm, spending the whole day at the Museum in his (21) FLORA MALESIANA small rooms where he had concentrated everything necessary for his work, including the great camera he had designed himself specially for making the marvellous photographs which are repro- duced in the plates of his truly monumental works on Palms. In the years that followed he had no difficulty in publishing his writings thanks to the great esteem he enjoyed abroad and the friendship of several Italian botanists. Actually, the publication of his chief work, Asiatic palms, was made possible by Sir GEorRGE KING, the Director of the Botanic Garden of Calcutta, and other important papers were published by his close friend, U. MARTELLI, in the periodical Webbia, which the latter had founded in honour of PHit1p BARKER WEBB. Some interesting works appeared also in L’Agricoltura Coloniale, a journal edited by the Istituto Agricolo Coloniale, the founda- tion of which was promoted by Beccarrand other Italian personalities. Several other papers on palm specimens from all over the world entrusted by their collectors to him for determination, were issued in various periodicals and books. Beccariled this last period of his life completely secluded from Italian academic life and nearly forgotten by most Italian botanists, but he was always overwhelmed by the sympathy and esteem of foreign botanists. In this period, perhaps more than before, he enjoyed the affection of his old and devoted friends and particularly of UGOLINO MARTELLI, his only pupil, a very keen botanist himself, well known for his basic works on the great family of Pandanaceae, whose study he had undertaken on BEccarr’s advice. This period, entirely devoted to his family and the palm studies, was serene and creative. In the last years of his active and eventful life he assembled the materials for a book on his explorations of eastern Malesia. He had already sorted out a final copy of his diaries and he had also begun to prepare the illustrations for his book, but unexpectedly death prevented him from accomplish- ing this last performance. He died peacefully in the evening of the 25th of October 1920, in Florence, at the age of 77. ODOARDO BECCARI was a great explorer but at the same time a very clever, many-sided, careful collector. Indeed he did incredible work in the field. In his long and lonely explorations in Malesia, BECCARI was attracted by all aspects of nature and human life and assembled an enormous wealth of botanical, zoological, ethnological and anthropological collections. Not only the number of the specimens makes his collections really invaluable, but also the fact that these are often accom- panied by notes and descriptions and sometimes by splendid and detailed drawings made in the field. The botanical collections from Malesia are kept in the Herbarium and Museum of the Florence University. They amount to more than 21,000 sheets, about 2400 flasks of material in alcohol, about 800 carpological specimens and more than 200 wood samples with the pertinent voucher specimens (see further information in VAN STEENIS’s Thesaurus Beccarianus). Many collection numbers are represented in BEccaART’s herbarium by more than one sheet. To these specimens we must add many duplicates which were distributed to the most important herbaria, among which those of the British Museum, Kew, Paris, Geneva, Leningrad, Berlin, Leiden, Vienna, Munich, Stockholm, and Bogor. The zoological collections consist of several thousand specimens belonging to a very high num- ber of species, many of which were described as new. Those in the higher groups are represented by skins, but sometimes also by skeletons, skulls and even heads or other parts of the body pre- served in alcohol. They are kept in the Civic Museum of Natural History of Genoa, which bears the name of its founder and Maecenas, Giacomo Doria, the companion of BEccARIon the expedi- tions to Borneo and the Red Sea. BEccaRI paid attention to special groups and assembled several (22) Dedication collections of particular importance such as that of the great ape, the orang-utan (48 specimens including a foetus), the set of birds of paradise, one of the best of its kind, and the collections of fishes, spiders, coleoptera, and ants. These collections have supplied a very rich harvest of study for many specialists. Indeed 205 works devoted to the study of BEccaArRI’s zoological collections had been published by 1920 in the Annali del Museo Civico di Storia Naturale of Genoa alone (see GEsTRO’s biography of BECCARI). Beccari also collected rich and interesting ethnological collections of great value and beauty. They are kept in the Anthropological and Ethnological Museum of Florence and some of them are exhibited in a hall of that Museum. A set of ethnological collections was sent to the Italian Geographical Society in Rome. Some of the idols, weapons, implements, ornaments, clothing, etc. were described and illustrated in BEccarr’s books Nelle foreste di Borneo and Nuova Guinea, Selebes e Molucche for the purpose of trying to establish, with the aid of the anthropological fea- tures, the origin of some of the peoples of Malesia. The value of these collections was greatly in- creased by the detailed information on the customs of those peoples which he described in his books and in the letters to his friends published by E.H. GiGLioL1, G. Cora and the Italian Geo- graphical Society. The anthropological collections are kept in the Anthropological and Ethnological Museum of Florence. They consist of skulls and a few skeletons of natives of Malesia. The most important is undoubtedly the set of some 200 skulls assembled in Korido in the island of Schouten in north- western New Guinea. These collections were the base of the first craniological investigations on Papuans, by P. MANTEGAZzZzA and E. REGALIA. BEccaRr’s activity in the field also extended to the geographical features of the districts he visited. During the preparation of his expeditions, he had trained himself in geodetics and topo- graphy and he had also invented a new instrument (Nuovo orizzonte artificiale. Rivista Marittima 6, 1873, 198—200, f. 1—5) for topographic surveys. During his travels he also made a topographic survey of several territories, particularly of the northwestern parts of New Guinea. These surveys later allowed G. Cora to prepare the maps he published in Cosmos (see Appendix 1, C: Maps). He also discovered a great river, War Samson, in northwestern New Guinea, near Sorong. Even though Beccarthad succeeded in assembling such enormous and invaluable collections, his fame is mainly due to his scientific work condensed into more than 150 publications, some of which consist of monographs of basic importance and those on palms especially still remain stan- dard works even today. BEcCaRI’ versatile mind allowed him to devote his attention to problems in different branches of natural science, but he carried out his activity chiefly in botany. In the first period, during which he made the great expeditions to Malesia, his botanical activity was essentially applied to the study of a part of his collections; the result was published mainly in the Nuovo Giornale Botanico Italia- no which he founded in 1869. Even if some of these papers were written here and there during his explorations, his scientific output in this period was necessarily small. With the end of his explorations in Malesia, the second period of BEccarr’s activity begins. Probably he was unsatisfied when he limited himself to the descriptions of new genera and species, and in this period he spread his field of research to the monographic or semi-monographic treat- ment of those families or genera from Malesia which had most attracted him. In this period BEc- CARI also undertook, at his own expense, the publication of a collection of botanical writings, Malesia, in order to facilitate the printing of his papers dealing with his Malesian plants and the reproduction of his splendid drawings. For this purpose he selected the quarto size. In Malesia he published some of his most outstanding works. The end of its publication in 1890 marks the end of this second period of his scientific activity and also the end of the study of his own collec- tions from Malesia. (23) FLORA MALESIANA After about a decade of complete scientific inactivity, due to the sad vicissitudes of his academic life and the end of Malesia, BEccaripublished Nelle foreste di Borneo in 1902. This year coincides with the resumption of his scientific studies and marks the beginning of the third period of his botanical research in which he devoted himself entirely to the study of the family of palms, becom- ing the best specialist who ever existed. In the following pages we want to go into more detail about the subjects contained in the mas- sive oeuvre of the Maestro. Confronted with the multitude of his activities we hope to weave this into a readable account, with a distinct feeling that our ability for writing falls short of the way in which Beccari could express himself, as testified by his ‘Wanderings’, which is still a thrilling guide for exploration in the tropics. Let us start with this work on Sarawak, the core of his main work in Malesia. He worked under favourable conditions, having ample equipment and time at his disposal, and the support of the Tuan-muda, CHARLES BROOKE. His big hut, called ‘Vallombrosa’ on Gunong Mattang, a hill west of Kuching, was used as a study centre. From there he made excursions and gradually familiarized himself with the very rich flora of the primary forest. He focussed attention on the big trees (Dipte- rocarpaceae, Bombacaceae, and others) as well as on the evasive tiny creatures of the saprophytic Triuridaceae and Burmanniaceae, the parasitic plants, the lianas and so forth, making beautiful and exemplary complete specimens in a skilled, professional way. This was, especially with un- wieldy plants such as palms and pandans, gingers and aroids, quite an effort, as every field botanist must be aware. Perseverance and patience fed by infinite interest must have induced him to take particular care with these groups. A special characteristic is that he knew his plants; hardly ever did he make two collections of the same species. As a scientific collector he was never equalled, and only approached by E.J.H. Corner and L.J. Brass. What a contrast with most other collectors who, even today, stick to the disgusting grab-as-grab-can way of collecting on hurried cross-country trips, causing heavy dupli- cation and absence of vital field notes. How Beccarimanaged all this at the age of 22, with only a few months training in tropical form knowledge at Kew, can only be understood if we imagine him as an extraordinarily gifted person with an intense interest in botany; botany in the widest sense, because he was not satisfied only with the taxonomy of flowering plants, but collected for example also wood samples and crypto- gams of all major phyla. His horizon widened to collecting minerals and all sorts of animals, ob- servations on vegetation types, on edible and horticultural plants, and the way of life of his com- panions, the Dayak people; in short, he possessed the integrated interest of a born all-round natu- ralist, whose scope went far beyond the mere plant collecting and description in which he excelled. BECCARI assembled a great wealth of data on the geographical features of Borneo, on the matter of useful and horticultural plants, on fibres, rattan, bamboos, resins, camphor, getah percha yielding trees, and medicinal plants. He was aware of the primitive domestication of species of Durio (durian), Eugeissona (a palm), Artocarpus (breadfruit), bananas and species of Nephelium, which he learned from observation of Dayak life. BEccarRi briefly reported on these subjects in a short summary of his journey in Sarawak to the Italian Geographical Society (1868), and later he incorporated the complete data in some appendices to his book Nelle foreste di Borneo. Prob- ably, when gliding easily in canoes on the rivers or proceeding with difficulty on the mysterious peat of the swamp forests during his long journey in the interior of Sarawak, he ventured on ideas about the origin of coal in Borneo — in which he was correct. When hunting for orang-utan in Batang-Lupar or excavating in the limestone caves of Sarawak he made his first guesses about the origin of man, which he developed in Nelle foreste di Borneo. He surely must have had a very busy life in his ‘Vallombrosa’, because he could never have con- (24) Dedication ceived the ‘Wanderings’ thirty years later without having copious notes of his field observations and full diaries. He must also have started in the field to compose botanical papers and prelim- inary revisions; when still in Sarawak he published some small papers in Italy. BECCARI was much enchanted with Sarawak, the flora, the country, and the Dayak people, and he even conceived a project that the Italian government should purchase it as a crown colony. When Beccarireturned from Borneo to Florence he founded the Nuovo Giornale Botanico Ita- liano, in which he published freely some papers on spectacular Bornean plants; but before leaving for the expedition to New Guinea he handed over the journal to T. CARUEL, since it would be diffi- cult both to edit it and to explore in Malesia. When, loaded with further materials, manuscripts, field notes and drawings, he returned from New Guinea to Florence, he realized the difficulty of publishing the botanical results of his explorations in the Nuovo Giornale Botanico Italiano. Thus he decided to undertake a collection of writings he named Malesia, in which he concentrated near- ly all his works of that period. The first two instalments of it were published during his one year interval, spent in Florence, between the last expedition to New Guinea and the journey to Austra- lia, New Zealand and Sumatra, and the third one appeared during the last mentioned journey. He undoubtedly must have worked very hard during that period, but he was able to manage it thanks to his efficient organization en route and the help of his friend R. GesTroin Genoa, where Malesia was printed. BeccaRI published in Malesia several works which are important for the subject in itself, but also various original considerations about some particular subjects, such as evolutionary pro- cesses, dispersal of seeds, geographical distribution, efc. are dealt with in them. First should be mentioned the extensive work on the ant plants devoted to the study of the symbiosis between plants and ants, which occupies the entire second volume. In it, BECCARI gives us his interesting views on the evolution and the common origin of plants and animals, discussing concepts still valid and topical today. Likewise of great interest are his considerations on the origin of the insectivor- ous plants and the distribution of plants in the Malesian archipelago, dealing especially with the Nepenthaceae. Other important works are the monograph on the genus Phoenix and the account of the palm genus Pritchardia in which he resumes his considerations on the dispersal of seeds and fruits and the origin of the flora of the Pacific islands. The three volumes of Malesia contain also a number of monographic or semi-monographic revisions of families and genera from Malesia, e.g., Icacinaceae, Menispermaceae, Nepenthes, Bombacaceae, Triuridaceae, Burmanniaceae, etc., and also a first survey of the palms of New Guinea. We can conclude that Malesia was intended to embody the botany of Malesia as a repository. It must have aroused great interest in the botanical world, containing novelties of fascinating plants with marvellous drawings made by Beccarthimself, a ‘must’ for every botanical institute. The use of the Italian language, even for monographic contributions from non-Italian collabora- tors, e.g., on Araceae by ENGLER, was certainly an obstacle to a wide sale, and the edition was subvented from Italian sources and the third and last volume could only appear thanks to the aid of the BENTHAM Trustees. Whether Beccartever intended or hoped to achieve an ultimate incorporation of all the mono- graphs of Malesian plant families, that is, an attempt towards a true Flora Malesiana, remains uncertain. The fact that he had certainly intentionally explored all areas of Malesia, except the Philippines, and that the first instalments of Malesia contained several monographic treatises, may support this idea. What is certain is that he fully realized that he could never dream of accom- plishing all this himself. Accordingly he freely entrusted many groups to colleagues in Italy and to his many correspondents abroad as appears from the bibliography in Appendix 4. Through the distributed duplicates BECCARI’s material went to various herbaria and later came (25) FLORA MALESIANA into the hands of specialists, but much of his original collection, embodied in the original Herba- rium Beccarianum, has not been examined by specialists. This original material is separately shelv- ed in Florence, 400 bundles in 33 cupboards; to it belong a card system with field data and a cover with drawings; there is a rather large number of unicates or collections of which no duplicates were distributed. In 1951 vAN STEENIS very roughly sampled a number of families and found that BEc- CARI Often had indicated and annotated genera in sched. as new, which were later based on other material, e.g., Koordersiodendron ENGL. (1898), Clavistylus J.J.S. (1910), Neosepicaea DIELS (1922), Octamyrtus Diets (1922), Haplolobus H.J. Lam (1931), Kjellbergiodendron BURRET (1936), Macadamia hillebrandii STEEN. (1952), Eriandra v. ROYEN & STEEN. (1952), Whiteoden- dron STEEN. (1952). It is a pity that in the past five decades too little initiative has been taken by the curators of the Florence Herbarium to attract and induce specialists to study the original BEccaRi collections in Florence. We suppose that it is due to the understaffing of this great Herbarium. Especially the Herbarium Beccarianum is not a reliquiae, not a closed chapter, but truly a thesaurus, still con- taining unknown botanical treasures. As this is not always realized by specialists, we urge them to borrow material of their speciality from this century-old, inexhaustible source. We sincerely hope that a revival of interest in the Her- barium Beccarianum is welcome to the future curators of the Florence Herbarium. Its possession brings with it the scientific obligation of using it, not just in honour of the Maestro, but mainly for the benefit of scientific botany and as a contribution to the fame of the Florence centre. After publication of Nelle foreste di Borneo in 1902, BEccaARI decided to concentrate, for the rest of his life, on the study of one large family on which he possessed more field knowledge than anyone, before or since, namely the palms. His first contribution to their knowledge dates from 1871, with a provisional account of those of Borneo. Old love never dies! In 1877 he had ac- counted for the palms of New Guinea, in 1885 for those cultivated in the Botanic Gardens at Bo- gor, but in about 1890 he spread his wings towards those beyond Malesia, the Indian empire, Indo- China, and later to Madagascar, Africa, the Pacific islands, etc. Everybody entrusted him with palm material and from this emanated a massive knowledge of Asiatic palms, embodied partly in the Records of the Botanical Survey of India, in HooKker’s Flora of India, partly later in Web- bia, founded by his old pupil and friend U. MarTELLI, but largely in the sumptuous volumes of the Annals of the Royal Botanic Garden, Calcutta, which also included those of Malesia. For the large folio plates of these massive plants BECCARI designed a special large camera with suitable accessories in order to achieve excellent illustrations!. He devoted his attention mainly to the tax- onomy of palms, but he also studied the cultivated species in some works which appeared in L’Agricoltura Coloniale, edited by that Institute once named Istituto Agricolo Coloniale, now Istituto Agronomico per |’Oltremare of which he had solicited the foundation in 1903. Of this big work a large number of unpublished manuscripts appeared in print after his death, through the untiring devotion of his pupil, friend, and colleague MARTELLI, who must be given a tribute of honour for his singularly unselfish efforts. The last of the manuscripts on palms which BECCARI left unfinished, that of the subfamily Arecoideae, was completed and published by PicHiI SERMOL- L1in 1955. We should also refer here to Moore’s important and competent evaluation of BECCA- RI’s massive contribution to the knowledge of the fascinating palm family. (1) The large camera and other microphotographic cameras designed by BEccarRI are described by LuIGI PAMPALONI, Apparecchio fotografico universale per laboratorio biologico ideato dal Dottor Beccari. Rend. Congr. Bot. Naz. Palermo (1903) 164—168, cum fig., and Gli apparecchi microfotografici del Dott. O. Becca- ri. Bull. Soc. Fotogr. Ital. 14 (1902) 129—145, fig. 1—7. (26) Dedication Dealing with his botanical activity we cannot silently pass over his descriptive work. Also in this BEccari excelled and showed that he had a remarkable insight into affinities. It appears that his new genera were always placed in the proper plant family and, moreover, that hardly ever were new species proposed by him reduced later, stamping him as a most accurate taxonomist. Indeed he was a taxonomist, but BEccARIShowed his sharp intelligence in other branches of botany often including his considerations in taxonomical papers. Describing the details of Gnetum led him to considerations about the ancestry of the flowering plants from the Gymnosperms. The plant geo- graphy of the palms led him to hypothetical ideas about former landbridges and sunken conti- nents. His gatherings in Sumatra led him to consider the affinities of its flora with those of South- east Asia and Java, concluding that the flora of volcanic ranges must be much younger than that of the more ancient and more stable Sunda lands. Other observations deal with the dissemination by earthworms; the double dispersal, anemochorous and zoochorous, of the plants of the periodic swamp forest provided with floating fruits and succulent seeds; pollination by pigeons; the various colours of flowers of the forest plants, and some others on physiology and ecology. BEccaRi also left traces of his versatile genius in various writings (papers and letters to his friends) which lie outside botany but must be mentioned briefly to understand how great he was as a naturalist. Particularly interesting are the letters to E.H. Gicuiot1and G. Cora in which he disclosed his views on the origin of the peoples of Malesia, in particular of the Papua-Mafor which he regarded as derived from a crossing of aboriginals, perhaps descended from Negritos and Hin- du peoples. Other interesting observations are those on the connections between mosquitos and malaria which he was one of the first to suppose, those on the agent of the bee pest which he sus- pected to be due to a protozoon later discovered in America, those on the connections between flies and cholera and numerous other observations particularly on the customs of animals. It was in Borneo that BeccarRi perceived the true value of evolution and was primarily fascinat- ed by the importance of adaptation to environmental conditions. But only later did his views on the processes of evolution take shape in his mind. It is a fact that the prolonged stays of gifted naturalists in the tropical wilderness, when their minds are set free from daily minutiae and domes- tics and solely occupied with the bewildering structural wealth of tropical plants and animals, allows their minds to open to new, big ideas and syntheses, generating philosophical thought. For this, one has only to think of von HUMBOLDT, JUNGHUHN, WALLACE, DARWIN, and CorRNER. To this, BECCARI, with his eager mind and astute power of observation, was no exception. BECCARI was used to going back from the facts to the causes, and his views on the evolutionary processes, which arose from the observations he made in nature, were consolidating in his mind in the course of time. Thus we find his views sketched in some papers and later resumed in others, whenever he had the chance to develop them on the basis of particular new observations. He did not supply us, or perhaps he did not want to supply us, with a synthesis of his views on evolution in an ad hoc publication, perhaps out of humility, since he disliked giving the impression that he was able to explain the laws of evolution, or perhaps out of honesty, because he perceived that his ideas had made their way into his mind by intuition and reasoning, without adequate investiga- tion. BECCARI was undoubtedly an evolutionist, but he was one in a very original manner. The first foundation of his theory of ‘plasmation’ was explained in his paper (1876) on the huts and gardens of Amblyornis inornata, the small bower-bird of paradise which builds a pretty hut with, in front, a lovely garden of soft moss on which it scatters flowers in shining colours changing them when they wither. This theory was resumed in the introduction to his work on ant plants (1884) and was later developed in his paper (1889) on the flowering of Amorphophallus titanum, the gigantic Ara- cea, and was summarized in his book, Nelle foreste di Borneo. (27) FLORA MALESIANA According to this theory the evolutionary processes of living beings took place, beginning with the most ancient geological times, fundamentally in two different epochs: a first epoch of plasma- tion, and a second epoch of conservative heredity, displaying in the course of time the two funda- mental rules of variability and fixity. In the first epoch, the plasmative strength, unhindered by heredity, may have given free play to the variability and to the adaptation stimulated by environ- mental conditions. In that epoch, the organism may have been liable to yield to the stimulus of external factors and more subject to modelling itself to them, the modifications occurring with the greatest of ease and even quite suddenly without the offspring necessarily being like the par- ents. This epoch may have been a period of youth for living beings in which each individual was allowed to modify itself in conformity with its needs, or rather even according to its wishes, its vanities, its whims. This epoch of plasmation, with a maximum of variability and a minimum of fixity, may have been followed by an epoch of conservative heredity, characterized by that strength which aims at the conservation of the acquired characters and owing to which the individuals belonging to a spe- cies transmit to their descendants the characteristics they have inherited from their ancestors. Thus the plasmative epoch may have been replaced by an epoch with a minimum of variability and a maximum of fixity. The strength of the conservative heredity becoming stronger in the course of time may consequently have weakened the faculty to vary, perhaps even to cancel it entirely, thus impressing the stamp of fixity on all living beings. He recognized the great influence of environmental factors on the plasmation of living beings and pointed out several cases of correlation between the morphology of some apparatuses and the environmental factors. For instance, he was the first to correlate the life form of flood-resistant plants with the environment and was struck by their similarity in leaf-shape: his ‘stenophyllous plants’, now called ‘rheophytes’. A still more important correlation amply studied by him was that of the symbiosis between plants and ants, the ‘piante ospitatrici’, or formicarian plants, to which he devoted a very large and detailed account (1884). However, he clearly and repeatedly recog- nized that plasmation could also be stimulated by an interior strength, by the wish of having some particular functions facilitated, such as defence, pollination, seed dispersal, etc. But BEccARtalso supposed that plasmation was even influenced, particularly in the animal kingdom, by a psychic push stimulated by the beauty of the environment, as could have happened in the birds of paradise desirous of imitating with their feathers the glowing colours of the aurora and dusk of the tropics, which they greet from the highest trees of the forest with very lively dances. Though recognizing that the extant being cannot, as a rule, undergo modifications because of the environment, BECCARI admitted however that even today some changes can take place in the species on account of a cross between individuals of different species or of the sudden appearance of hereditary modifications of various, even if unknown, origin. He admitted that these could be induced by new poisonous substances or by new enzymes arising in the substratum; thus he fore- saw the existence of mutations and mutagenic substances. BECCARI was a man of great intelligence, versatility and intuition, who united an exceptional personality and liberality with uncommon integrity and strength of mind. He was an indefatigable worker, who devoted hours and hours to his research, without a moment of rest. But he did not work out of a wish to be praised; prizes and honours did not interest him. He loved his research studies since he was zealous of the beauty and perfection of nature and only happy when he could entirely devote himself to investigating its manifold and marvellous aspects. For the same reason he liked to draw plants and animals and very few excelled him in scientific drawing. Beccari had an austere and inflexible character, but he was neither obstinate nor autocratic. His temperament was based on a keen sense of duty. Also in private life, although deeply fond of (28) Dedication his wife, sons and friends, every action was characterized by a clear austerity, and he was greatly beloved but at the same time he always inspired a certain awe. But this austerity concealed a great goodness of heart. He had also a great sense of justice and was a very courageous and stalwart man. These gifts united with the wish to inspire respect and love rather than fear, were greatly esteemed by native people and on his dangerous expeditions he never suffered injury from anybody. Undoubtedly he must have encountered serious danger, but he faced them with resolution, and he recounts them with indifference and without boasting. BeEccar! is described as a proud, almost misanthropic spirit and perhaps he was not an easy character and we understand why some regarded him as a troublesome person. From his youth, moulded in his long and solitary explorations in Malesia, he was set apart, destined to travel and to work alone, and he was not afraid of solitude; on the contrary, in his ripe age he found refuge in it, the better to devote himself to his studies and serve his single purpose and sole end: the science of nature. R.E.G. Picut SERMOLLI & C.G.G.J. VAN STEENIS (29) FLORA MALESIANA Appendix 1 — Bibliography of Odoardo Beccari A — Books and papers (excluding reviews of books) 1861. Escursione botanica. — L’Araldo Cattolico, Lucca, anno XVIII, 14 agosto 1861, n. 33: 264 (not seen). 1862. Illustrazione dell’ Arnoldia cyathodes Massal. — Comment. Soc. Crittog. Ital. 1: 128—130, tewle 1868. Descrizione di tre nuove specie di piante Bornensi. — Atti Soc. Ital. Sc. Nat. 11: 197—198. — Cenno di un viaggio a Borneo. — Boll. Soc. Geogr. Ital. 1: 193-214. 1869. Illustrazione di nuove specie di piante Bornensi. (Balanophoreae, Rafflesiaceae). — Nuovo Giorn. Bot. Ital. 1: 65—91, t. 2—5S. —— Varieta e notizie. — Ibid.: 158—160. — Lamenti del Redattore. Rivista bibliografica. — Ibid.: 222—224. 1870. Illustrazione di nuove specie di piante Bornensi. (Aristolochiaceae). — Nuovo Giorn. Bot. Ital. 2: 5—8, t. 1. — Nota di una nuova specie del genere Stenomeris. — Ibid.: 8—12, t. 2. — Nota sul Trichopodium zeylanicum Thw. — Ibid.: 13-19, t. 3. — Nota sull’embrione delle Dioscoreaceae. — Ibid.: 149-155, t. 4. —— Disepalum coronatum nuova specie di Anonacea bornense. — Ibid.: 155—156, t. 5. —— (O. Antinori & A. Issel). Relazione sommaria del viaggio nel Mar Rosso dei Signori Antinori, Beccari e Issel. — Boll. Soc. Geogr. Ital. 5(2): 43-60. 1871. Descrizione di due nuove specie di Hydnora d’ Abissinia. — Nuovo Giorn. Bot. Ital. 3: 5—7. — Petrosavia. Nuovo genere di piante parassite della famiglia delle Melanthaceae. — Ibid.: (Sle ieeale — Note sopra alcune palme Bornensi. — Ibid.: 11—30. — Le Hydrocotyle d’Europa. — Ibid.: 102. —— Sui generi Bihania ed Eusideroxylon. — Ibid.: 102—103. — Cattedra di Botanica nell’Universita di Torino. — Ibid.: 103-104. —— Illustrazione di nuove 0 rare specie di piante Bornensi. (Anonaceae). — Ibid.: 177—193, t. 2-7. 1874. Descrizione di una nuova specie del genere Myrmecodia della famiglia delle Rubiaceae. — Nuovo Giorn. Bot. Ital. 6: 195—197, t. 6. 1875. Osservazioni supra alcune Rafflesiaceae. — Nuovo Giorn. Bot. Ital. 7: 70—75. 1876. Le Capanne ed i Giardini dell’ Amblyornis inornata. — Ann. Mus. Civ. St. Nat. Genova 9: 382—400, t. 8. —— Lettera del Dr. A.B. Meyer al Mse. G. Doria (con nota di O. Beccari). — Ann. Mus. Civ. St. Nat. Genova 8: 333-334. 1877. Relazione intorno alla Baia di Assab. — Cosmos 4: 230—232. — Della organogenia dei fiori feminei del Gnetum gnemon L. — Nuovo Giorn. Bot. Ital. 9: 91—100, t. 7 (reprinted in Amer. Journ. Sc. 13: 469—471). —— Sulla Cardiopteris lobata Wall. — Nuovo Giorn. Bot. Ital. 9: 100—108, t. 8. — Die Flora der Aru-Inseln. — Ausland 50: 759-760. —— Sul nuovo genere Scorodocarpus e sul genere Ximenia L. della famiglia delle Olacineae. — Ibid: 273=279 ts 11. (30) Dedication — Della disseminazione delle palme. — Bull. R. Soc. Tosc. Ortic. 2: 167—173. —— II sagu della Nuova Guinea, Metroxylon rumphii Mart. — Ibid.: 247—249. — Catalogue of the plants of the Fly River (New Guinea) collected by De Albertis, 1877. — Journal of Melbourne Logbook 30 Jan. 1877 (not seen). — Le specie di palme raccolte alla Nuova Guinea da O. Beccari e dal medesimo adesso descritte, con note sulle specie dei paesi circonvicini. — Malesia 1: 7—96, t. 1 & 2. — Nuove osservazioni sulle palme della Nuova Guinea. — Ibid.: 97—102. — Studio monografico sopra le piante della famiglia della Icacineae e delle Menispermaceae sin qui scoperte nella Malesia e nella Nuova Guinea. — Ibid.: 103—165, t. 3—8. 1877—78. Piante nuove o rare dell’Arcipelago Malese e della Nuova Guinea, raccolte, descritte ed illustrate da O. Beccari. — Leguminosae-Caesalpinieae, Palmae, Chailletiaceae, Gymno- spermae papuanae, Coniferae, Gnetaceae, Cycadaceae, Violaceae, Magnoliaceae, Monimia- ceae, Araliaceae, Ericaceae, Vacciniaceae, Nepenthaceae, Corsia, Burmanniaceae. — Ibid.: 1 (1877) 167—192; contd. (1878) 193—254, t. 9-15. 1878. Sulle piante raccolte alla Nuova Guinea dal Sig. L.M. D’Albertis durante l’anno 1877, con descrizione di tre nuove specie di Icacineae. — Ibid.: 1 (1878) 255—256; contd. (1883) 257—258 (incl., pp. 257—258: Su di un nuovo genere della famiglia delle Olacineae). — II Conophallus titanum Beccari. — Bull. R. Soc. Tosc. Ortic. 3: 290—293, f. 32. 1879. Le Nepenthes. — Bull. R. Soc. Tosc. Ortic. 4: 13-19. — La pit piccola delle Araceae: Microcasia pygmaea Becc. — Ibid.: 179-181, cum fig. 1880. Notes on the plants collected by Sig. L.M. D’Albertis in New Guinea. — In: L.M. D’Alber- tis, New Guinea: what I did and what I saw. 2: 391—395. —— Catalogue of the plants of the Fly River, 1877. — Ibid.: 396—400. — Sul viaggio di O. Beccari in Sumatra. Giugno—Ottobre 1878. — Boll. Soc. Geogr. Ital. ser. 2, 5 (1880) 300—302. — Lettera di O. Beccari al Preside della Facolta di Scienze naturali del R. Istituto di Studi Supe- riori di Firenze. Dicembre 1880 (Published by U. Martelli, Webbia 5 (1921) 329—330). 1881. R. Museo di Fisica e Storia Naturale. L’Orto botanico. — Giornale ‘La Nazione’ 6 Gennaio 1881. —— (& G. Doria) Viaggio ad Assab nel Mar Rosso, dei signori G. Doria ed O. Beccari con il R. Avviso ‘Esploratore’ dal 16 Novembre 1879 al 26 Febbraio 1880. — Ann. Mus. Civ. St. Nat. Genova 16: 523—524. — A proposito del progetto di remozione delle collezioni botaniche dal Museo di Storia Natura- le di Firenze (not seen). — Sull’abbandono del museo e del giardino botanico della Specola a Firenze. Protesta dei culto- ri della botanica in Firenze colle adesioni e le osservazioni dei botanici italiani ed esteri. — Firen- ze, Tip. Sborgi, in 8°, 72 pp. — Cenni sopra la Flora di Assab. — Bull. R. Soc. Tosc. Ortic. 6: 108—110. —— Beitrage zur Pflanzengeographie des malayischen Archipels. — Bot. Jahrb. 1: 25—40 (con- densed version in German of Malesia 1: 214—238). 1883. Su di un nuovo genere della famiglia delle Olacineae. — Malesia 1: 257—258 (included in: Sulle piante raccolte alla Nuova Guinea dal Sig. L.M. D’Albertis durante l’anno 1877, con de- scrizione di tre nuove specie di Icacineae. See 1878). 1884. Piante ospitatrici, ossia piante formicarie della Malesia e della Papuasia descritte ed illustra- te da O. Beccari. — Ibid.: 2: 5—128, f. 1—9, t. 1-25; contd. (1885) 129—212, f. 10—15, t. 26—54; contd. (1886) 213—284, f. 16—17, t. 55—65. —— Veratronia malayana Miq. — Bull. R. Soc. Tosc. Ortic. 9: 70—72, t. 4. (31) FLORA MALESIANA — II The in Italia. — Boll. Notiz. Agrar. Minist. Agric. 6: 279—281; also in L’ Agricolt. meridio- nale 7: 116-118, and La Natura, Milano, 1: 267—268. — II colera e le mosche. — La Natura 1884(33): 81-83. 1885. Orto Botanico di Firenze. — Giornale il ‘Corriere’, 14 Maggio 1885. — Reliquiae Schefferianae. Illustrazione di alcune palme viventi nel Giardino Botanico di Bui- tenzorg. — Ann. Jard. Bot. Buitenzorg 2: 77—171, t. 1—14. —— Cyrtosperma (Alocasia Hort.) johnstonii Becc. — Bull. R. Soc. Tosc. Ortic. 10: 5—7. —— Plantes a fourmis de l’Archipel Indo-Malais et de la Nouvelle Guinée. (Compte rendu de Mr. E. Levier). — Archiv. Ital. Biol. 6: 305—341. 1886. Rivista delle specie del genere Nepenthes. — Malesia 3: 1—15, t. 1—3. —— Rivista delle felci e licopodiacee di Borneo e della Nuova Guinea, enumerate o descritte dal Barone V. Cesati nella memoria che porta il titolo: ‘‘Felci e specie nei gruppi affini raccolte a Borneo dal Sig. O. Beccari’’ e dell’altra: ‘‘Prospetto delle felci raccolte dal Sig. O.B. nella Poli- nesia’’. — Ibid.: 16—55, t. 4—S. — Nota sopra alcune felci raccolte dal Sig. J.E. Teysmann all’isola di Sumba o Sandal-wood ed in Timor. — Ibid.: 56—57. — Nuovi studi sulle Palme Asiatiche. — Ibid.: 58—149, t. 6—11. — Sulla Quercus robur Linn. Nota di botanica legale. Perizia nell’interesse della causa Le Pen- nec e Monetti contro Vitali, Picard, Charles e Comp.i. — Firenze, Tip. Nicolai, in 4°, 44 pp. 1887. Turgescenza dei petali di Magnolia Yulan. — Malpighia 1: 420. 1887—88. Le palme incluse nel genere Cocos Linn.: Studio preliminare. — Malpighia 1 (1887) 343-354, 441-454, t. 9; 2 (1888) 85—95, 147-156. 1888. Nuove specie di palme recentemente scoperte alla Nuova Guinea. — Nuovo Giorn. Bot. Ital. 20: 177-180. 1889. Nuove palme asiatiche. — Malesia 3: 169—200. — Le Bombaceae Malesi descritte ed illustrate da O. Beccari. — Ibid.: 201—280, f. 1—16, t. 12—36 (plates 12—35 were published before the text in vol. 3, no. 3, August 1887). — Fioritura dell’Amorphophallus titanum Becc. — Bull. R. Soc. Tosc. Ortic. 14: 250—253, 266—278, t. 8 (Reprint with separate pagination: 17 pp., 3 plates). — Palmae. — In: K. Schumann & M. Hollrung, Die Flora von Kaiser Wilhelms Land (Beiheft zu den Nachrichten tiber Kaiser Wilhelms Land und den Bismarck-Archipel): 15—17. 1890. Le palme del genere Pritchardia. — Malesia 3: 281—317, t. 37—38 (the plates were published before the text in vol. 3, no. 4, September 1889). — Le Triuridaceae della Malesia. — Ibid.: 318—344, t. 39—42 (the plates were published before the text in vol. 3, no. 4, September 1889). —— Rivista monografica delle specie del genere Phoenix Linn. — Ibid.: 345—416, f. 17, t. 43—44. 1891. Letter to Prof. G. Papasogli. — In: G. Papasogli, La nitrobenzina usata come insetticida. — Agric. Toscana 9: ?—? [repr. 1—6]. 1892—93. (& J.D. Hooker) Palmae. — In: J.D. Hooker, The Flora of British India 6: 402—448, 449-483. 1893. Palmae. — In: G.F. Scott Elliot, New and little known Madagascar plants, collected and enumerated by G.F. Scott Elliot, M.A., B.Sc., F.L.S. — J. Linn. Soc. London, Bot. 29: 61—62. 1901. Lettera al Presidente della Societa Botanica Italiana. — Bull. Soc. Bot. Ital. (1901) 119—124 (Dated March 23, 1901). —— Per il Giardino Botanico. — Giornale ‘La Nazione’, 12 Maggio 1901. — Trasloco delle collezioni botaniche del Museo di Firenze. Lettera al Presidente della Societa Botanica Italiana. — Bull. Soc. Bot. Ital. (1901) 202—210. (32) Dedication — Sul trasloco delle collezioni botaniche della Malesia dal Museo di Storia Naturale in Via Ro- mana ne’ nuovi locali presso il Giardino de’ Semplici. — Corrispondenza del D. Odoardo Bec- cari coll’Istituto di Studi Superiori. — Firenze, Tipogr. S. Landi, 4 pp. (Dated July 2, 1901). 1902. Nelle Foreste di Borneo. Viaggi e ricerche di un naturalista. — Firenze, xvi+667 pp., 81 fig., 4 maps. —— Systematic enumeration of the species of Calamus and Daemonorops with descriptions of the new ones. — Rec. Bot. Surv. India 2: 197—230. 1903. L’Istituto di Studi Superiori di Firenze, la chiusura del Museo Botanico e le sue peripezie. Osservazioni e Critiche. — Rocca San Casciano, Licinio Cappelli, 16 pp., 3 fig. — Per un Istituto Agricolo Botanico Coloniale. — Giornale ‘Fieramosca’, 21 Marzo 1903 (not seen). — L’avvenire agricolo dell’Eritrea. — ‘La Tribuna’, Roma, | Giugno 1903 (not seen). 1904. Palmae. — In: J. Perkins, Fragmenta Florae Philippinae 1: 45—48. — Wanderings in the great forests of Borneo: travels and researches of a naturalist in Sarawak (transl. by E.H. Giglioli, rev. and ed. by F.H.H. Guillemard). — Constable, London, xxiv + 424 pp., 61 fig., 3 maps. 1905. Le palme del genere Trachycarpus. — Webbia 1: 41—68, f. 1—16. — Note anatomiche sul frutto dei Trachycarpus. — Ibid.: 68—72, f. 17—18. — Notizie sul Nannorhops ritchieana H. Wendl. — Ibid.: 72—73; also Bull. R. Soc. Tosc. Ortic. 30: 325—326. — Palme nuove papuane. — Ibid.: 281—313, f. 1-8. — Le palme delle Isole Filippine. — Ibid.: 315—359. — Palmae. — In: K. Schumann & C. Lauterbach, Nachtrage zur Flora der Deutschen Schutzge- biete in der Stidsee: 60—61. 1906. Palmarum madagascariensium synopsis. — Bot. Jahrb. 38, Beibl. 87: 1—41, f. 1. 1907. Le Palme Americane della tribu delle Corypheae. — Webbia 2: 1—343. — Notes on Philippine palms. I. — Philip. J. Sc. 2: Bot. 219—240. 1908. Asiatic palms — Lepidocaryeae. Part I: The species of Calamus. — Ann. R. Bot. Gard. Calc. 11 [Fol.]: Letter-press I-VI, 1—518, t. 1—2; Plates t. 1-236 (1—231+25A, 104A, 172A, 188A, 191A). — Le palme ‘Dum’ od ‘Hyphaene’ piu specialmente quelle dell’ Affrica italiana. — Agric. Co- lon. 2: 137-183, f. 1—2, t. 1-3. —— Palmae novae antillanae. — In: Fedde’s Repert. Bot. Syst. 6: 94—96. — The palms of the Batanes and Babuyanes Islands. — Philip. J. Sc. 3: Bot. 339-342. 1909. New or little known Philippine palms. — Elmer, Leafl. Philip. Bot. 2: 639-650. — Notes on Philippine palms. II. — Philip. J. Sc. 4: Bot. 601—639, t. 30—31. —— Palmae. — Nova Guinea 8: 203—222, t. 50—S1. 1910. Glaziova treubiana, nouvelle espece de Cocoinée, avec observations sur le genre Cocos. — Ann. Jard. Bot. Buitenzorg ser. 2, suppl. 3: 791—806, 2 fig., t. 31. — Le palme del genere Raphia. — Agric. Colon. 4: 137—170, f. 1—3, t. 1—6 (reprinted with the original pagination, as an issue of the series ‘Biblioteca Agraria Coloniale’). —— Studio monografico del genere Raphia. — Webbia 3: 37—130, f. 1—8, t. 1. —— Palme australasiche nuove 0 poco note. — Ibid.: 131—165, f. 1—6. —— Descrizione di una nuova specie di Trachycarpus. — Ibid.: 187—190. — Palmae dell’Indo-China. — Ibid.: 191—245. — Contributo alla conoscenza delle Lepidocaryeae affricane. — Ibid.: 247—294. — La Copernicia cerifera in Riviera ed una nuova specie di Livistona. — Ibid.: 295—305, f. 1, thy. (33) FLORA MALESIANA 1911. Asiatic palms — Lepidocaryeae. Part II: The species of Daemonorops. — Ann. R. Gard. Calc. 12(1) [Fol.]: Letter-press [I—II], 1—237, t. I—II; Plates t. 1—109. — Classification des palmiers d’Indo-Chine. — Bull. Mus. Hist. Nat. Paris 17: 148—160. — Le palme che producono fibre di Piassava nel Madagascar. — Agr. Colon. 5: 320—324, f. A-D. — The palms of the island of Polillo. — Philip. J. Sc. 6: Bot. 229—230. 1912. Palmae. — In: H. Winkler, Beitrage zur Kenntnis der Flora und Pflanzengeographie von Borneo. II. — Bot. Jahrb. 48: 89—93. — Palmae. — In: I. Urban, Symbolae Antillanae 7: 170-172. 1912—13. The palms indigenous to Cuba with appended remarks on the general structure of the trunk of a palm. — Pomona Coll. J. Econ. Bot. 2 (1912) 253—276, f. 109—118; 351—377, f. 144—153; 3 (1913) 391—417, f. 154-172. 1912—14. Palme del Madagascar. — Firenze, Istituto Micrografico Italiano, [Fol.], [i—iv] +62 pp., 50 fig., 50 tav. 1913. Contributi alla conoscenza delle palme. — Webbia 4: 143—240, f. 1—17. — Manipolo di palme nuove Polinesiane conservate nell’erbario di Kew. — Firenze, M. Ricci, October 1913 (preprinted with the same pagination from Webbia 4: 253—291, f. 1—31, 1914). — Studio sui ‘Borassus’ e descrizione di un genere nuovo Asiatico di ‘Borasseae’. — Firenze, M. Ricci, November 1913 (preprinted with the same pagination from Webbia 4: 293—385, f. 32-42, 1914). 1914. Asiatic palms — Lepidocaryeae. Supplement to Part I: The species of Calamus. — Ann. R. Bot. Gard. Calc. 11 (App.) [Fol.]: Letter-press i—viii, 1—142; Plates t. 1—83. — Contributo alla conoscenza delle palme a olio (Elaeis guineensis). — Agric. Colon. 8: 5—37, t. 1—5; 108-118, t. 7-14; 201—212, t. 24-26; 255—270, t. 40—41 (reprinted with separate pagi- nation: 80 pp., 18 tav., as an issue of the series ‘Biblioteca Agraria Coloniale’). — Neue Palmen Mikronesiens. — In: G. Volkens, Beitrage zur Flora von Mikronesien. — Bot. Jahrb. 52: 4. — Neue Palmen Papuasiens. — In: C. Lauterbach, Beitrage zur Flora von Papuasien iv: 26B. — Bot. Jahrb. 52: 23—39. — Palmae. — In: K. Rechinger, Botanische und zoologische Ergebnisse einer wissenschaftli- chen Forschungsreise nach den Samoainseln, dem Neuguinea-Archipel und den Salomonsinseln von Marz bis Dezember, 1905. V. Teil. — Denkschr. Kaiserl. Akad. Wiss. Math.-Naturwiss. Kl. 89 (‘1913’) 502—S09, f. 5—12. 1916. Palmae. — In: J.F. Rock, Palmyra Island with a description of its flora. — Coll. Hawaii Publ. Bull. 4: 44—48, t. 17—18 (reprinted with the title ‘Palmae. Cocos nucifera L. forma pal- myrensis. Characteristics of the Coconuts of the Palmyra Islands’ and different pagination: p. 1—5, t. 18-19). —— Il genere Cocos e le palme affini. — Agric. Colon. 10: 435—471, t. 1—4; 489-532, t. 5—13; 585—623, t. 14—15 (reprinted with separate pagination: 128 pp., 15 tav., as issue of the series ‘Biblioteca Agraria Coloniale’). — Palmae. — In: E. Chiovenda, Resultati scientifici della Missione Stefanini-Paoli nella Soma- lia Italiana. Volume I. Le collezioni botaniche. Appendice: Le raccolte di Mangano, Scassellati, Mazzocchi e Provenzale in Somalia. — Firenze, Pubbl. R. Ist. Stud. Super. Mus. ed Erb. Co- lon. p. 176—177, 230. 1917. The origin and dispersal of Cocos nucifera. — Philip. J. Sc. 12: Bot. 27—43 (reprinted in Principes 7: 57—69, 1963). — A new species of Calamus from Amboina. — Ibid.: 81. (34) Dedication — Ona new South Polynesian palm, with notes on the genus Rhopalostylis Wendl. et Drude. — Trans. & Proc. New Zeal. Inst. Bot. 49: 47—50. — Palmae. — In: L.S. Gibbs, Dutch N.W. New Guinea. A contribution to the phytogeography and flora of the Arfak Mountains, &c. p. 91—98, 200. — Pelagodoxa henriana Becc. — In: D. Bois, Pelagodoxa henriana. Palmier nouveau des Isles Marquises. — Rev. Hort. 89: 302—304, f. 76-79. 1918—1921. Asiatic palms — Lepidocaryeae. Part III. The species of the genera Ceratolobus, Ca- lospatha, Plectocomia, Plectocomiopsis, Myrialepis, Zalacca, Pigafetta, Korthalsia, Metroxy- lon, Eugeissona. — Ann. R. Bot. Gard. Calc. 12(2) [Fol.]: Letter-press [I—II], 1-231, t. i—vi; Plates t. 1-120 (1—118+50A, 63A). 1919. Palms of the Philippine Islands, collected and distributed by A.D.E. Elmer. — Elmer, Leafl. Philip. Bot. 8: 2997—3067. —— The palms of the Philippine Islands. — Philip. J. Sc. 14: 295—362, t. 1—3. 1920. Palmae novae antillanae. II. — In: Fedde, Repert. Spec. Nov. Regni Veg. 16: 436—437. — Recensione delle palme del vecchio mondo appartenenti alla tribu delle Corypheae con de- scrizione delle specie e varieta nuove che vi appartengono. — Firenze, M. Ricci, 10 December 1920 (preprinted with the same pagination from Webbia 5: 5—70, 1921). — Le palme della Nuova Caledonia. — Firenze, M. Ricci, 128 pp., 13 tav., 10 December 1920 (preprinted with separate pagination from Webbia 5: 71—197, t. 1—13, 1921). 1921. Nelle foreste di Borneo. Viaggi e ricerche di un naturalista. Ed. 2. — Firenze, xvi+ 469 pp., 72 tav., 11 fig., 4 maps, 1 portr. (published posthumously by Beccari’s son Nello Beccari). 1923. Neue Palmen Papuasiens II. — In: C. Lauterbach, Beitrage zur Flora von Papuasien. X. — Bot. Jahrb. 58: 441—462. — Note botaniche e botanico industriali bornensi. — Webbia 5: 451—581, f. 1—6 (reprint of Appendix, Nelle foreste di Borneo, 1902). 1924. Neue Palmen Mikronesiens. — In: L. Diels, Beitrage zur Flora von Mikronesien und Poly- nesien. III. — Bot. Jahrb. 59: 11—16. — Nuova Guinea, Selebes e Molucche; diarii di viaggio ordinati dal figlio Nello Beccari, con introduzione e note del Prof. Luigi Buscalioni. — Firenze, Soc. Anon. Edit. ‘La Voce’, XXXViii +468 pp., 53 fig., 22 tav., 1 map (Diaries published by Beccari’s son Nello Beccari). — Palme della trib’ Borasseae. Ed. U. Martelli, G. Passeri, Firenze, 56 pp., 18 fig., 45 tav. 1927. Odoardo Beccari nel Mar Rosso e tra i Bogos. Frammenti di diario inediti, trascritti e ordi- nati dal figlio prof. Nello Beccari. — Boll. R. Soc. Geogr. Ital. ser. 6, 4: 625—646. 1930. Odoardo Beccari in Sumatra e la scoperta dell’ ‘Amorphophallus Titanum’. (Frammenti di diario inediti ed ordinati). — Boll. R. Soc. Geogr. Ital. ser. 6, 7: 569—595, f. 1—6, 1 map. (The author of the paper is given as Nello Beccari). 1933. Asiatic palms — Corypheae. The species of the genera Corypha, Nannorhops, Sabal, Co- pernicia, Serenoa, Brahea, Acoelorhaphe, Washingtonia, Pritchardia, Erythea, Livistona, Li- cuala, Pritchardiopsis, Phlolidocarpus, Teysmannia, Rhaphis, Chamaerops, Trachycarpus, Rhaphidophyllum, Trithrinax, Acanthorhiza, Hemithrinax, Thrinax, Coccothrinax, Cryso- phyla. (Posthumous work by Dr. Odoardo Beccari. Revised and edited by Prof. Ugolino Mar- telli). — Ann. R. Bot. Gard. Calc. 13: Letter-press [Qu.] [i—x], 1—356; Plates [Fol.] t. 1-102 (1—99+ 59 bis, 68 bis, 89 bis), (1931’). 1934. Generi, specie e varieta nuove di palme gerontogee della trib Arecaceae lasciate inedite dal Dr. Od. Beccari ed ordinate a cura di U. Martelli. — Atti Soc. Tosc. Sc. Nat. Pisa. Mem. 44: 114-176. [repr. pp. 1—65]. 1935. (By U. Martelli from Beccari manuscripts). I generi e le specie delle palme gerontogee della (35) FLORA MALESIANA tribu delle Arecaceae. Esposizione geografica secondo la monografia inedita del Dr. O. Beccari. — Nuovo Giorn. Bot. Ital. ser. 2, 41: 693—723 (‘1934’). —— (By U. Martelli from Beccari manuscripts). La sinonimia delle palme gerontogee della tribu delle Areceae. — Ibid. 42: 17-88. 1955. (& R.E.G. Pichi Sermolli) Subfamiliae Arecoidearum palmae gerontogeae. Tribuum et ge- nerum conspectus. — Webbia 11: 1—187, f. 1—47. B — Letters by Odoardo Beccari During his travels BECCARI wrote several letters to his friends in Italy, chiefly to G. Doria, E.H. GIGLIOLI, G. CorA, T. SALVADORI, R. GESTRO, and O. ANTINORI. They contain a wealth of very interesting observations and comments on the botanical, zoological, ethnological and other natu- ralistic aspects of Malesia. These letters or fragments of them were published in various Italian periodicals, usually accompanied by information and comments on BEccarr’s scientific discover- ies, and on the itineraries and the main events of his adventurous travels. The bibliographic citations of the papers in which these letters are published are given below together with an indication of the name of the friend to whom the letter was addressed, and the date and place in which it was written. In order to facilitate and render more systematic the consul- tation of these letters, they are quoted according to the periodicals in which they were published. Letters published in the Nuova Antologia BEcCARI’s travels in Malesia, Assab and the country of Bogos between 1865 and 1876 were de- scribed by Enrico H. GIGLIOLI in various instalments published in the Nuova Antologia with the general title of ‘Odoardo Beccari ed i suoi viaggi’. They were also reprinted, with independent pagination, and assembled in a special book (Firenze, Le Monnier, 309 pp., 9 fig., 2 small maps, 1872—76) with the title ‘I viaggi del Dott. Beccari da Firenze tracciati e commentati’. There Gi- GLIOLI published several letters or fragments of letters by BECCARI to his friends and also some passages from his original diaries. All of them are quoted verbatim between GIGLIOLI’s descrip- tions and comments. Borneo. 1865—1868. — Nuova Antologia 21 (1872) 119—160 (Passages of the original diaries). Samhara e Bogor. 1870. — Ibid.: 22 (1873) 658—668. Malesia, Molucche e Papuasia. 1871—72—73. — Ibid.: 22 (1873) 668—709. (Letters to G. Doria, from the Red Sea, December 10—11, 1871; to E.H. Giglioli, from Batavia, February 8; to G. Doria, from Makassar, February 23; Ceram, March 13; Ambon, March 21; Kapaor, April 21, with a small map; to E.H. Giglioli, from Sorong, May 3; to G. Doria, from Sorong, June 21, 1872). Papuasia: Sorong-Mansinam-Andai (Monti Arfak). (Giugno 1872 al Gennaio 1873). — Ibid.: 23 (1873) 194—225, 2 fig. (Letters to G. Doria, from Andai, August 30; Andai, September 8—9, 1872; Ambon, January 2; Ambon, February 2—7, 1873). Papuasia — Le isole Aru e Kei (Febbraio—Settembre 1873). — Ibid.: 24 (1873) 835—866, 1 map; 25 (1874) 163—192, 1 fig. (Letters to C. Correnti, from Ambon, January 3; to O. Antinori, from Ambon, January 3; Ambon, February 5; to G. Doria, from Dobbo (Aru Is.), February 24; Wo- kan (Aru Is.), March 10; Wokan, July 3—4; Tual (Kei Is.), August 27, 1873). I. Macassar — Kandari (Celebes). II. I Papua (Dicembre 1873, Giugno 1874). — Ibid.: 27 (1874) 420—463, 5 fig. (Letters to E.H. Giglioli, from Makassar, December 4; to R. Gestro, from Ma- (36) Dedication kassar, December 1; to G. Doria, from Makassar, December 4; Makassar, December 26, 1873; Makassar, January 15; to T. Salvadori, from Makassar, January 14; to R. Gestro, from Makas- sar, January 15; to G. Doria, from Kandary, April 23, and May 2 and 11, 1874). Celebes, Giava, Ternate, Amboina (Giugno 1874—Gennaio 1875). — Ibid.: ser. 2, 2 (1876) 802—822. (Letters to G. Doria, from Kandari, June 4; Makassar, August 30; to O. Antinori, from Makassar, August 28; to T. Salvadori, from Makassar, August 30; to Baron Podesta, the Mayor of Genoa, from Makassar, end August; to G. Doria, from Ambon, December 7, 1874; to E.H. Giglioli, from Ambon, January 7; to ??, from Ambon, January 15, 1875). Seconda esplorazione della Nuova Guinea. La baia di Geelvink (Febbraio—Novembre 1875). — Ibid.: ser. 2, 3 (1876) 147-163. (Letters to G. Doria, from Dorei, June 5; Andai, June 15; Hatam, June 21; to T. Salvadori, from Ternate, August 4, 1875). Terza esplorazione della Nuova Guinea. La baia di Humboldt (Novembre 1875—Marzo 1876). — Ibid.: ser. 2, 3 (1876) 333—363, 1 fig. (Letters to G. Cora, from Ternate, March 11; to E.H. Giglioli, from Ternate, March 6, 1876). Letters published in the Bollettino della Societa Geografica Italiana Estratti e frammenti di lettere dirette al marchese Giacomo Doria dal naturalista botanico Odoar- do Beccari, durante il suo viaggio alla Nuova Guinea. — Boll. Soc. Geogr. Ital. 8 (1872) 148—151. (Letters to G. Doria, from Wahaai, March 13; Ambon, March 21; Sorong, May 3, 1872). Odoardo Beccari nella Nuova Guinea e nelle isole Aru. — Ibid.: 9 (1873) 145—158. (Letters to G. Doria, from Andai, August 30, 1872; to C. Correnti, from Ambon, January 3; to O. Antino- ri, from Ambon, January 4; to G. Doria, from Ambon, February 2, 5, 7; to O. Antinori, from Ambon, February 5; to G. Doria, from Dobbo, Aru Is., February 24; Wokan, Aru Is., March 10, 1873). Lettera di O. Beccari al marchese Doria. — Ibid. 10: 4—5 (1873) 66—69. (Letter from Wokan, Aru Is., July 3, 1873). Frammento di lettera di O. Beccari a Doria da Makassar, il 18 Novembre 1873. — Ibid.: 10: 6 (1873) 38. Odoardo Beccari alle isole Key. — Ibid.: 87-89. (Letter to G. Doria, from Tual (Small Kei), Au- gust 27, 1873). Ultime notizie di O. Beccari. — Ibid.: 11 (1874) 78-81. (Letters to R. Gestro, from Makassar, December 1; to G. Doria, from Makassar, December 4; to O. Antinori, from Makassar, Decem- ber 7; to G. Doria, from Makassar, December 26, 1873). Corrispondenze del dott. Odoardo Beccari. — Ibid.: 276—282. (Letters to T. Salvadori, from Ma- kassar, January 14; to R. Gestro, from Makassar, January 15; to G. Doria, from Makassar, January 15, 1874). Lettere di Odoardo Beccari. — Ibid.: 480—488, carta della parte sud-est di Celebes. (Letters to G. Doria, from Kandari, April 23 and June 4, 1874). Nota sui Papua e sulla Nuova Guinea. — Ibid.: 652—659. (Letter to O. Antinori, from Makassar, August 28, 1874). Lettere di Odoardo Beccari. — Ibid.: 660—667. (Letters to the Mayor of Genoa, from Makassar, without dates; to T. Salvadori, from Makassar, August 30; to G. Doria, from Kandari, June 4, 1874). Lettera di O. Beccari. — Ibid.: 12 (1875) 117—122. (Letter to G. Doria, from Ambon, January 4, 1875). (37) FLORA MALESIANA La Nuova Guinea Olandese. — Ibid.: ser. 2. 1 (1876) 5SO—557. (Letter to G. Cora, from Ternate, March 11, 1876). Letters published in Cosmos The editor of the journal Cosmos of Turin, Gurpo Cora, gave ample information on BECCARIs travels in Malesia. He published several letters by BEccarito his friends in Italy, which are listed below. Furthermore, Cora gave various reports on the different stages of BECCARI’s journeys. The references to the latter are given in Appendix 3 dealing with BEccarrs itineraries. Recenti spedizioni alla Nuova Guinea. Spedizione di Beccari e D’Albertis: 1. Da Singapore ad Amboina. — Cosmos | (1873—74) 11—15. (Letters to G. Doria, from Ba- tavia, February 8; Makassar, February 23, Wahai(N. Ceram), March 13; Amboina, March 2 1872): 2. Da Amboina a Sorong. — Ibid.: 15—20. (Letters to G. Doria, from Kapaor, April 21; So- rong, May 3 and June 21, 1872). Recenti spedizioni alla Nuova Guinea. Esplorazioni di Odoardo Beccari: 1. Ricerche geografiche nella Nuova Guinea. — Cosmos 2 (1874—75) 7—9. (Letter to G. Cora, from Makassar, December 1, 1873). 2. Note sulle Isole Kei. — Ibid.: 9-10. (Letter to G. Cora, from Makassar, December 1, 1873). 3. Da Makassar a Kandari. — Ibid.: 92—96. (Letters to G. Doria, from Kandari, May 2 and 11, 1874). 4. Escursioni intorno a Kandari. Ritorno a Makassar. — Ibid.: 203—207. (Letters to G. Doria, from Kandari, June 4; Makassar, August 30, 1874). 5. Notizie sull’ornitologia di Celebes. — Ibid.: 207—208. (Letter to T. Salvadori, from Makas- sar, August 30, 1874). 6. Appunti etnografici sui Papua. — Ibid.: 400—404. (Letter to O. Antinori, from Makassar, August 28, 1874). 7. Soggiorno a Ternate. Da Ternate ad Amboina. Preparativi pel terzo viaggio alla Nuova Guinea. — Cosmos 3 (1875—76) 83—88. (Letters to G. Doria, from Amboina, January 4, 8, 9, 1875). 8. Da Amboina a Dorei, per Soron e Wakkaré. Scoperta del fiume Wa Samson. Esplorazione della baia di Geelvink, determinazione della sua vera ampiezza. — Ibid.: 88—92. (Letter to G. Doria, from Dorei, June 5, 1875). 9. Esplorazione dei Monti Arfak. Ritorno a Ternate per la via di Salvatti, Batanta, Koffiao. — Ibid.: 92—95. (Letters to G. Doria, from Andai, June 15; Hatam, June 21; to T. Salva- dori, from Ternate, August 4, 1875). 10. Viaggio a bordo del trasporto olandese ‘Soerabaja’. Itinerario progettato. Da Ternate a Dorei, per Salvatti. Visita ad Ansus. Lavori idrografici. — Ibid.: 220—221. (Letters to G. Doria, from Ternate, November 7; Dorei, November 26, 28, 1875). 11. Viaggio a bordo del trasporto olandese ‘Soerabaja’. Da Dorei alla baia di Humboldt per la baia Vandamen, l’isola Run, il sud di Jobi, le foci dell’ Ambermo. — Ibid.: 349-352. (Letter to G. Cora, from Ternate, March 11, 1876). 12. Saggio statistico sulla Nuova Guinea Olandese. Popolazione, Commercio, Climatologia, Nomenclatura. — Ibid.: 352—360. (Letter to G. Cora, from Ternate, March 11, 1876). 13. Viaggio a bordo del trasporto olandese ‘Soerabaja’. La Baia di Humboldt edi suoi abitanti. (38) Dedication La Baia del Disinganno, il Vulcano Ciclope e le isole Arimoa. — Ibid.: 364—372. (Letter to E.H. Giglioli, from Ternate, March 6, 1876). 14. Viaggio a bordo del trasporto olandese ‘Soerabaja’. Dalla Baia di Humboldt a Ternate pel nord di Jobi, Dorei, Waigheu, Misol, il golfo Mac Cluer, la baia Gouns, Ghesser, Amboi- na. — Ibid.: 372—374. (Letter to E.H. Giglioli, from Ternate, March 6, 1876). 15. Questioni etnologiche sui Papua. — Ibid.: 375—379. (Letter to E.H. Giglioli, from Ternate, March 6, 1876). Letters published in various journals Il viaggio di O. Beccari alla Nuova Guinea. — Nuov. Giorn. Bot. Ital. 4 (1872) 208—212. (Letters to a friend of Florence, from Wahai, N. Ceram, March 13; Ambon, March 21, 1872). Il viaggio di O. Beccari alla Nuova Guinea. — Ibid.: 291—294. (Letters to G. Doria, from Kapaor, April 21; Sorong, June 21, 1872). Lettera di O. Beccari dalle isole Aru. — Ibid. 5 (1873) 330. (Letter to [G. Doria], July 3—4, 1873). Brano di lettera di O. Beccari da Makassar in data 1° Dicembre 1873. — Ibid. 6 (1874) 205—206. Una pianta maravigliosa. — Bull. R. Soc. Tosc. Ortic. 3 (1878) 270—271. (Letter to R. Corsi Sal- viati, from Sumatra, without date. Comment by E.O. Fenzi). Lettera ornitologica di O. Beccari intorno agli uccelli osservati durante un recente viaggio alla Nuova Guinea. — Ann. Mus. Civ. St. Nat. Genova 7 (1875) 704—720. (Letter to T. Salvadori, from Ternate, August 4, 1875. Introduction by T. Salvadori). Lettera del Prof. Odoardo Beccari a Giacomo Doria. — Ibid. 13 (1878) 451—455. (From Kajt Ta- nam, Sumatra, September 8, 1878). C — Maps In the period of preparation for his travels BEccarRI carefully trained himself also in geodetics and topography. He also invented a new instrument (Nuovo orizzonte artificiale) for topographic survey. During his explorations BEccarI devoted great attention to the topography of the places he vis- ited and his surveys allowed him to draw some maps which greatly contributed to the delimitation of the coasts of certain areas of the Malesian Archipelago. Some of these maps were published by BEccartr himself, others were utilized by G. Cora, to- gether with surveys of other explorers, to elaborate some of the maps published in his periodical Cosmos. BeEccarr’s itineraries are traced on all maps listed below. Carta originale del viaggio di Beccari e d’Albertis nella Nuova Guinea ovest (Aprile 1872), costrut- ta e disegnata da Guido Cora. Scala 1:700.000. — Cosmos (G. Cora) 1 (1873-74) Tav. I. — Explanatory notes in Cora G. Recenti spedizioni alla Nuova Guinea. Memoria sulla Tavola I. — Ibid.: 22—24. Carta originale del viaggio di O. Beccari nel Sud-est di Celebes (Maggio— Agosto 1874), costrutta e disegnata da Guido Cora. Scala 1:1.200.000. — Ibid. 2 (1874-75) Tav. V. — Explanatory notes in Cora G. Viaggio di O. Beccari nel Sud-est di Celebes. Note sulla Tavola V. — Ibid.: 201-202. Carta originale della Nuova Guinea N.O. e delle isole Salvatti, Batanta, William, ecc. secondo i rilievi originali di Lovera, Cerruti, Beccari e le esplorazioni anteriori costrutta e disegnata da (39) FLORA MALESIANA Guido Cora. Scala 1:1.000.000. — Cosmos (G. Cora) 3 (1875) Tav. II. — Explanatory notes in Cora G. Recenti spedizioni alla Nuova Guinea. Note sulle Tavole II e III. — Ibid.: 81—83. Carta originale della Baia di Geelvink e del littorale N.O. della Nuova Guinea secondo i rilievi di Odoardo Beccari, 1875 del Geelvink, di Duperrey, Dumont D’Urville, ecc. costrutta e dise- gnata da Guido Cora. Scala 1:1.800.000. — Including an inset map: Tracciato comparativo della Baia di Geelvink secondo la carta dell’ Amm. Ingl. N° 2759a ed i rilievi di O. Beccari. Scala 1:4.800.000. — Ibid.: Tav. III. — Explanatory notes in Cora G. Recenti spedizioni alla Nuova Guinea. Note sulle Tavole II e III. — Ibid.: 81—83. Carta originale della Nuova Guinea Nord dai Monti Arfak alla Baia d’ Humboldt secondo i rilievi di Odoardo Beccari, 1875—76, della nave oland. ‘Soerabaja’, cap. Swaan e le esplorazioni ante- riori, costrutta e disegnata da Guido Cora. Scala 1:2.000.000. — Including an inset map: I Mon- ti Arfak. Scala 1:800.000. — Ibid. 3 (1876) Tav. X. — Explanatory notes in Cora G. Recenti spedizioni alla Nuova Guinea. Note sulla Tavola X. — Ibid.: 347-349. Piano della Baia d’ Humboldt (Telokh Lintciu) secondo i rilievi della nave oland.e ‘Etna’ 1858 e le ricerche di O. Beccari, 1875, di Guido Cora. Scala 1:70.000. — Ibid.: Tav. XI. Carta originale della Provincia di Sarawak, compilata sopra vari documenti e secondo le osserva- zioni dell’autore. — In O. Beccari, Nelle foreste di Borneo (1902) fig. 36 (p. 187). Abbozzo di carta (originale) del fiume Bintulu e suoi affluenti. — Ibid.: fig. 58 (p. 351). Carta del sistema idrografico e delle attuali divisioni politiche di Borneo, compilata sopra i docu- menti pili recenti e le osservazioni dell’autore. — Ibid.: fig. 66 (p. 407). Carta orginale degli itinerari dell’autore in Sarawak. — Ibid.: fig. 75 (facing page 504). Carta speciale della Nuova Guinea Ovest cogli itinerari di O. Beccari e L.M. D’Albertis (1872—1876) costrutta e disegnata da Guido Cora. — Including an inset map: Carta originale del viaggio di O. Beccari nel Sud Est di Celebes. 1874. Disegnata da G. Cora. — In O. Beccari, Nuova Guinea, Selebes e Molucche (1924) at the end of the book. Carta delle Isole Aru secondo i rilievi di Odoardo Beccari. Luglio 1873. — In O. Beccari, Lettera di O. Beccari al Marchese Doria. — Boll. Soc. Geogr. Ital. 10: 4—5 (1873) 64—65, at the end of the paper. Carta della Parte Sud-est delle Celebes secondo la relazione di O. Beccari e traccia del suo viaggio da Makassar a Kandari, Febbraio 1874. — Ibid.: 11 (1874) 480—488, at the end of the paper. Appendix 2 — Biographies of Odoardo Beccari BALDASSERONI, V. & D. CaRAzzi. L’opera biologica di Odoardo Beccari. Rassegna Sci. Biol. 3 (1921) 84-88. BARGAGLI PETRUCCI, G. L’ opera biologica di Odoardo Beccari. Pubbl. Ist. Stud. Sup. Firenze, Sez. Sci. Fis. Nat. In memoria di Odoardo Beccari, pp. 5—16. 1921. BeccarI, N. Enciclopedia Italiana di Scienze, Lettere ed Arti. Milano, Ist. G. Treccani, 6 (1930) 462. —— Brief Obituary, Itineraries and Bibliography — Manuscript compiled by Beccari’s son Nello for Flora Malesiana, Oct. 1947 (in Library Rijksherbarium, Leiden). BeEGuINoT, A. Boll. R. Soc. Geogr. Ital. ser. 5, 12 (1923) 194—209, portr. — Rivista Biol. 3 (1921) 118—121, portr. BERTACCHI, C. Geografi ed esploratori italiani contemporanei. Milano, De Agostini Ed. 1929, pp. 368—372, portr. BurkiLL, I.H. & J.C. MouLton. Journ. Str. Br. R. Asiat. Soc. 83 (1921) 166-173. (40) Dedication BUSCALIONI, L. La vita e l’opera di Odoardo Beccari. In: O. Beccari, Nuova Guinea, Selebes e Molucche. Firenze, 1924, pp. xxiii—Xxxviii. CavaRA, F. Bull. Soc. Afr. Ital. 40 (1921) 46—S51. CHIOVENDA, E. Notizie biografiche. Odoardo Beccari. Annuario R. Ist. Stud. Sup. Firenze 1920-21 (1921) 155—156. — Nuovo Giorn. Bot. Ital. ser. 2, 28 (1921) 5—35, portr., bibl. (in historical sequence). Cora, G. Recenti spedizioni alla Nuova Guinea. I. Spedizione di Beccari e d’Albertis. (Studi e viaggi di Beccari dalla sua adolescenza al 1872). Cosmos (G. Cora) 1 (1873-74) 8-11. Dori, G. I naturalisti italiani alla Nuova Guinea e specialmente delle loro scoperte zoologiche. Boll. Soc. Geogr. Ital. ser. 2, 3 (1878) 154-169. Gestro, R. Ann. Mus. Civ. St. Nat. Genova, ser. 3, 9 (1921) 242—297, 11 fig. (10 portr.). GUBERNATIS, A. DE. Dizionario biografico degli scrittori contemporanei. Firenze, 1879, 1: 117-118, portr. — Dictionnaire international des écrivains du jour. Florence, 1888, 1: 220—221. — Dictionnaire international des écrivains du monde latin. Suppl. et Ind. Florence, 1906, 30*. Jackson, B.D. (‘B.D.J.’). Proc. Linn. Soc. London 134 (1922) 37—40. LEONE E. DE & M. A.ippI CAPPELLETTI. Dizionario biografico degli Italiani. Roma, 1965, 7: 440-444. Lo Priore, G. Ber. Deut. Bot. Ges. 39 (1922). Gen. Vers. Heft (56)—(87), portr. bibl. (alphabeti- cal according to titles). MarRTELLI, U. Webbia 5 (1921) 295-353, portr., 3 maps, bibl. (alphabetical according to titles). Moore Jr., H.E. Odoardo Beccari (1843—1920). Principes 25 (1981) 29—35, portr., bibl. (on palms). NissEN, CL. Die botanische Buchillustration. 2 (1951) 11 (on his drawings). PALADINI, C. La morte di Odoardo Beccari. ‘Il Nuovo Giornale’ 27 Ottobre 1920. PAMPANINI, R. Agricoltura Coloniale 14 (1920) 449—453. Po.iaccl, G. Atti Ist. Bot. Univ. Pavia. ser. 6, 6 (1935), i—xiii, bibl. (in historical sequence). Puccion1, N. L’opera etnografica ed antropologica di Odoardo Beccari. Pubbl. Ist. Stud. Sup. Firenze, Sez. Sci. Fis. Nat. In memoria di Odoardo Beccari. pp. 17—26, 1921. Ropotico, F. Naturalisti esploratori dell’Ottocento Italiano. Firenze, Le Monnier, 1967, pp. 191—222, t. 4 (with reprint of original passages). Roster, G. Bull. R. Soc. Tosc. Ortic. 46 (1921) 33—36, portr. Saccarbo, P.A. La Botanica in Italia. Materiali per la storia di questa scienza. Mem. Ist. Veneto Sci. Lett. Arti. 25:4 (1895) 25; 26:6 (1901) 16. S.A.S. Kew Bull. (1920) 369—370. STEENIS, C.G.G.J. vAN. Thesaurus Beccarianus. Webbia 8 (1952) 427—436. STEENIS-KRUSEMAN, M.J. vAN. Flora Malesiana I, 1 (1950) 43—46, portr. —— Flora Malesiana I, 8 (1974) xiii (brief addition). Wittrock, V.B. Catalogus illustratus Iconothecae Botanicae Horti Bergiani Stockholmiensis. Acta Horti Berg. 3:2 (1903) 160, t. 31 (portr.); 3:3 (1905) 175. (41) FLORA MALESIANA Appendix 3 — Accounts of Odoardo Beccari’s itineraries in Malesia and Ethiopia (see also Letters and Maps) Beccarl, N. Brief Obituary, Itineraries and Bibliography. — Manuscript compiled by Beccari’s son Nello for Flora Malesiana, Oct. 1947. (in Library Rijksherbarium, Leiden). Burtt, B.L. Beccari’s ascent of ‘Mount Poi’. Flora Males. Bull. 19 (1964) 1131—1132. (Pointing out that Mt Poi (Poe, Pueh) in Southwest Sarawak, mentioned in the ‘Wanderings’, a famous type locality, is not what is nowadays on maps indicated as Mt Poi, but a more southeasterly peak in the Poi Range, now known as Gunung Berumput, or Gunung Rumput). Cora, G. Spedizione italiana alla Nuova Guinea. Roma, Stab. Civelli, 1872, 39 pp. — Recenti spedizioni alla Nuova Guinea. — Cosmos (G. Cora) 1 (1873-74) 7-8. — Recenti spedizioni alla Nuova Guinea. Spedizione di Beccari e D’Albertis. — Ibid.: 8-11, 141-143. — Recenti spedizioni alla Nuova Guinea. Odoardo Beccari. — Ibid.: 215-218, 265; 2 (1874-75) 2—4, 86; 3 (1875-76) 75—76. — Viaggio di O. Beccari nel Sud-est di Celebes. — Ibid.: 2 (1874—75) 200—202, t. V (map). — Recenti spedizioni alla Nuova Guinea. Secondo viaggio della ‘Vettor Pisani’. — Ibid.: 3 (1875—76) 77-78. — Recenti spedizioni alla Nuova Guinea. Quarto viaggio di O. Beccari alla Nuova Guinea (1875—1876) — Ibid.: 217. — Recenti spedizioni alla Nuova Guinea. I Monti Arfak. — Ibid.: 217-218. — Carta speciale della Nuova Guinea Ovest cogl’itinerari di O. Beccari e L.M. D’Albertis (1872—1876) costrutta e disegnata da Guido Cora. — Ann. Mus. Civ. St. Nat. Genova 12 (1878). — (see also Appendix 1 — Letters by O. Beccari). Recenti spedizioni alla Nuova Guinea. — Cosmos 1 (1873—74) 11—15, 15—20; 2 (1874-75) 7-9, 9-10, 92—96, 203—207, 207—208, 400—404; 3 (1875-76) 83-88, 88-92, 92-95, 220-221, 349-352, 352-360, 364-372, 372—374, 375-379. Gestro, R. Ricordo biografico di Giacomo Doria. Ann. Mus. Civ. St. Nat. Genova, ser. 3, 10 (1921) 1—78, portr. (information on Beccari’s travels). GiGuio1!, E.H. (see also Appendix 1 — Letters by O. Beccari). Odoardo Beccari ed i suoi viaggi. — Nuova Antologia 21 (1872) 119-160; 22 (1873) 658—668; 668—709; 23 (1873) 194—225; 24 (1873) 835—866; 25 (1874) 163-192; 27 (1874) 420—463; ser. 2, 2 (1876) 802—822; 3 (1876) 147-163, 333-363. Maaistris, L.F. DE. Biografie di Geografi e di Esploratori contemporanei. IV. Giacomo Doria. Novara, Ist. Geogr. De Agostini, 1917, 18 pp., portr. (information on Beccari’s travels). MaRTELLI, U. Odoardo Beccari. Webbia 5 (1921) 295—353, 3 maps. STEENIS-KRUSEMAN, M.J. vAN. Flora Malesiana I, 1 (1950) 43—45. Vink, W. Nova Guinea, Bot. n. 22 (1965) 479—481, f. 6. (on itinerary in New Guinea). (42) Dedication Appendix 4 — Studies based on Odoardo Beccari’s botanical collections (incomplete) ARCANGELI, G. L’Amorphophallus titanum Beccari, illustrato da G. Arcangeli. — Nuovo Giorn. Bot. Ital. 11 (1879) 217—223, cum fig. ASCHERSON, P. Plantae phanerogamae marinae, quas Cl. Eduardus Beccari in Archipelago Indico annis 1866 et 1867, et in Mari Rubro anno 1870 collegit, enumeratae. — Nuovo Giorn. Bot. Ital. 3 (1871) 299-302. BAGLIETTO, F. Lichenes in regione Bogos Abissiniae septentrionalis lecti ab O. Beccari; illustravit F. Baglietto. — Nuovo Giorn. Bot. Ital. 7 (1875) 239-254. BaILLon, H. Stirpes exoticae novae. — Adansonia 11 (1873—76) and 12 (1876—79). — Referring to several Bornean plants collected by Beccari. BAKER, J.G. On a collection of ferns made by Dr. Beccari in western Sumatra. — Journ. Bot. 18 (1880) 209-217. BARGAGLI-PETRUCCI, G. Sulla struttura dei legnami raccolti in Borneo dal dott. O. Beccari. — Malpighia 17 (1903) 280—371, t. 4—15. CESATI, V. Felci e specie nei gruppi affini raccolte a Borneo dal Signor Odoardo Beccari. — Atti Accad. Sci. Fis. Mat. Napoli 7:8 (1876) 1—37, t. 1-4. —— Prospetto delle Felci raccolte dal Sign. O. Beccari nella Polinesia, durante il suo secondo viaggio d’esplorazione in quei mari. — Rend. Accad. Sci. Fis. Mat. Napoli 16 (1877) 23—31. — Mycetum in itinere Borneensi lectorum a cl. Od. Beccari enumeratio. — Atti Accad. Sci. Fis. Mat. Napoli 8:3 (1879) 1—28, t. 1—4. CHRISTENSEN, C. Revision of the Bornean and New Guinean ferns collected by O. Beccari and described by V. Cesati & J.G. Baker. — Dansk Bot. Arkiv 9 (1937) 33-52. DuBarpD, M. Description de quelques types nouveaux ou peu connus de Sapotacées (Illipées) d’aprés les documents de L. Pierre. — Bull. Mus. Hist. Nat. Paris 14 (1908) 405—409. ENGLER, A. Araceae, specialmente Bornensi e Papuane raccolte da O. Beccari. — Bull. R. Soc. Tosc. Ortic. 4 (1879) 265—271; 295—302. (with an introduction, pp. 265—266, by O. Beccari). —— Araceae della Malesia e della Papuasia raccolte da Beccari, determinate ed illustrate. — Malesia 1 (1883) 259—304, t. 16—28. GEHEEB, A. Weitere Beitrage zur Moosflora von Neu Guinea. I. Uber die Laubmoose, welche Dr. O. Beccari in den Jahren 1872—73 und 1875 auf Neu Guinea, besonders dem Arfak-Gebirge, sammelte. II. Uber einige Moose von westlichen Borneo. — Bibl. Bot. Heft 44 (1898) 29 pp., Qetave — Musci frondosi in monte Pangerango insulae Javae a Dr. O. Beccari annis 1872 et 1874 lecti. — Rev. Bryol. 21 (1894) 81-85. Hampg, E. Musci frondosi in insulis Ceylon et Borneo a Dr. Od. Beccari lecti. — Nuovo Giorn. Bot. Ital. 4 (1872) 273-291. Hem, F. Diptérocarpacées de Borneo. — Bull. Mens. Soc. Linn. Paris 120 (1891) 954—958; 122 (1891) 970—976. — Deux Richetia nouveaux. — Bull. Mens. Soc. Linn. Paris 123 (1891) 979-981. IRMSCHER, E. Neue Begoniaceen von O. Beccari in Malesien gesammelt. — Webbia 9 (1954) 469—509, f. 1-8. KRANZLIN, F. On orchids collected by Beccari, in Pflanzenreich; corrections on it by R. Schlechter in Fedde, Repert. 9 (1911) 286—287. KREMPELHUBER, A. VON. Lichenes foliicoli quos legit O. Beccari annis 1866—1867 in insula Bor- neo. — Miinchen, 1874. (43) FLORA MALESIANA —— Lichenes quos legit O. Beccari in insulis Borneo et Singapore annis 1866 et 1867. — Nuovo Giorn. Bot. Ital. 7 (1875) 5—67, t. 1-2. MaRTELLI, U. Le Composte raccolte dal Dottore O. Beccari nell’ Arcipelago Malese e nella Papua- sia. — Nuovo Giorn. Bot. Ital. 15 (1883) 281—305. — Le Dilleniaceae Malesi e Papuane delle collezioni Beccari. — Malesia 3 (1886) 150—167. — Florula Bogosensis. Enumerazione delle piante dei Bogos raccolte dal Dott. O. Beccari nell’anno 1870, con descrizione delle specie nuovo 0 poco note. — Firenze, 1886. vii+ 169 pp., 1 tav. — Contribuzione alla flora di Massaua. — Nuovo Giorn. Bot. Ital. 20 (1888) 359—371. MERRILL, E.D. A brief survey of the present status of Bornean botany. — Webbia 7 (1950) 309-324. Notaris, C. DE. Epatiche di Borneo raccolte dal Dre O. Beccari nel Ragiato di Sarawak durante gli anni 1865—66—67. — Torino, Stamperia Reale Paravia, 1874, 44 pp., 35 tav. (preprinted from Mem. Acad. Sci. Torino 18 (1876) 267—308, t. 1—35; reprinted also in Nuovo Giorn. Bot. Ital. 8 (1876) 217-251). PASSERINI, G. Funghi raccolti in Abissinia dal Signor O. Beccari. — Nuovo Giorn. Bot. Ital. 7 (1875) 180—192, t. 4—S. PicHi SERMOLLI, R. Rapporti tra parassita ed ospite nella Rafflesia tuan-mudae Becc. e Cissus sp. — Nuovo Giorn. Bot. Ital. ser. 2, 44 (1937) 385—421, t. 8—10, f. 1-16. REICHENBACH, H.G. Odoardo Beccari novitiae orchidaceae papuanae describuntur. — Bot. Centr. Bl. 28 (1886) 343—346. SCHLECHTER, R. Orchidaceae novae Beccarianae. — Notizbl. Berl. Dahl. 8 (1921) 14—20. SCHUMANN, K. Sterculiaceae Beccarianae. — Bot. Jahrb. 24 (1897), Beibl. 58: 14—21. Soitms-LAUBACH, H. Uber die von Beccari auf seiner Reise nach Celebes und Neu Guinea gesam- melten Pandanaceae. — Ann. Jard. Bot. Buitenzorg 3 (1883) 89—104, t. 16. STEENIS, C.G.G.J. vAN. Thesaurus Beccarianus (including ‘Malaysian sea-grasses collected by Beccari’ and ‘Some records of Malaysian plants’). Webbia 8 (1952) 427—436. — & M.J. vAN STEENIS-KRUSEMAN. Numberlists of Beccari: Sumatra, Borneo, New Guinea. 1951—1953. Typed copies in Leiden, Bogor, Florence and the Arnold Arboretum, the Leiden copy kept up to date by the first author. Piante Sumatrane, PS 1—979, Piante Borneensi, PB 1—4069, Piante Papuane, PP 1—986. These are his collecting numbers; they do not correspond with the herbarium ‘sheet numbers’ as given in the ‘Herbarium Beccarianum’ at Florence. STEENIS-KRUSEMAN, M.J. VAN. Flora Malesiana I, 5 (1958) cclvii. — Explaining that Beccari later renumbered his collections originally named ‘Plantae Beccarianae’ in three series, each starting with number 1, viz the PS series (Piante Sumatrane), PB (Piante Borneensi) and PP (Piante Pa- puane). VENTURI, G. DE. Muschi raccolti dal Signor Odoardo Beccari nella terra dei Bogos in Abissinia. — Nuovo Giorn. Bot. Ital. 4 (1872) 7—22. ZANARDINI, J. Phycearum indicarum pugillus, a cl. Eduardo Beccari ad Borneum, Sincapore et Ceylanum annis MDCCCLXV—VI-—VII collectarum. — Mem. R. Ist. Veneto Sci. Lett. Arti 17 (1872) 109—170, t. 1-12. (44) ABBREVIATIONS AND SIGNS acc. = according Ak. Bis. = Aklan Bisaya (Philip. language) Alf. Cel. = Alfurese Celebes (language) alt. = altitude Anat. = Anatomy Ap. = Apayao (Philip. language) app. =appendix, appendices appr. = approximate Apr. = April Arch. = Archipelago atl. = atlas auct. div. =auctores diversi; various authors auct(t). mal. = auctores malayenses; authors dealing with Malesian flora auct(t). plur. = auctores plures; several authors Aug. = August Bag. = Bagobo (Philip. language) basionym = original name of the type specimen; its epithet remains permanently attached to the taxon which is typified by it provided it is of the same rank. Bg. = Buginese (language) Bik. = Bikol (Philip. language) Bil. = Bila-an (Philip. language) Bill. = Billiton Bis. = Bisaya (Philip. language) Bon. = Bontok (Philip. language) Born. = Borneo Bt= Bukit; mountain Bug. = Buginese (language) Buk. = Bukidnon (Philip. language) c. =circiter; about C. Bis. = Cebu Bisaya (Philip. language) cf. =confer; compare Chab. = Chabecano (Philip. language) citations = see references cm = centimetre c.n. =see comb. nov. comb. nov. =combinatio nova; new combination CS = cross-section or transversal section of an organ c.s. =cum suis; with collaboration cum fig. =including the figure cur. = curante; edited by D (after a vernacular name) = Dutch Daj. = Dyak (language) d.b.h. =diameter at breast height D.E.I. = Dutch East Indies descr. added behind a reference=means that this contains a valid description diam. = diameter Distr. (as an item) = Distribution Distr. (with a geographical name) = District ditto =the same, see do Div. = Division, or Divide div. = diversus (masc.); various do = ditto (Ital.); the same Dum. = Dumagat (Philip. language) dupl. = duplicate E=east (after degrees: eastern longitude) E (after a vernacular name) = English Ecol. = Ecology ed. = edited; edition; editor e.g. =exempli gratia; for example elab. = elaboravit; revised em(end). = emendavit; emended em(erg). ed. =emergency edition Engl. = English etc., &c. =et cetera; and (the) other things ex auctt. =ex auctores; according to authors excl. = exclusus (masc.); excluding, exclusive of ex descr. =known to the author only from the de- scription f. (before a plant name) = forma; form f. (after a personal name) = filius; the son f. (in citations) = figure fam. = family Feb(r). = February fide =according to fig. = figure fl. =flore, floret (floruit); (with) flower, flowering For. Serv. = Forest Service fr. =fructu, fructescit; (with) fruit, fruiting Fr. (after a vernacular name) = French G.= Gunung (Malay); mountain Gad. = Gaddang (Philip. language) gen. = genus; genus genus delendum = genus to be rejected Germ. = German geront. =Old World haud = not, not at all holotype=the specimen on which the original de- scription was actually based or so designated by the original author homonym =a name which duplicates the name of an earlier described taxon (of the same rank) but which is based on a different type species or type specimen; all later homonyms are nomenclaturally illegitimate, unless conserved I. = Island ib(id). =ibidem; the same, in the same place Ibn. = Ibanag (Philip. language) ic. =icon, icones; plate, plates ic. inedit.=icon ineditum, icones inedita; inedited plate(s) id. =idem; the same i.e. =id est; that is If. =Ifugao (Philip. language) Ig. =Igorot (Philip. language) Ilg. =Ilongot (Philip. language) Ilk. = Iloko (Philip. language) in adnot. =in adnotatione; in note, in annotation incl. = inclusus (masc.); including, inclusive(ly) indet. = indetermined Indr. = Indragiri (in Central Sumatra) inedit. =ineditus (masc.); inedited in herb. =in herbario; in the herbarium in litt. =in litteris; communicated by letter in sched. =in schedula; on a herbarium sheet in sicc. =in sicco; in a dried state in syn. =in synonymis; in synonymy Is. = Islands Is. (after a vernacular name) =Isinai (Philip. lan- guage) Ism. = Isamal (Philip. language) isotype=a duplicate of the holotype; in arboreous plants isotypes have often been collected from a single tree, shrub, or liana from which the holotype was also derived Iv. =Ivatan (Philip. language) J(av). = Javanese (language) Jan. = January Jr = Junior Klg. = Kalinga (Philip. language) Kul. =Kulaman (Philip. language) Kuy. = Kuyonon (Philip. language) Lamp. = Lampong Districts (in S. Sumatra) (45) FLORA MALESIANA Lan. = Lanao (Philip. language) lang. = language l.c. =loco citato; compare reference lectotype=the specimen selected a posteriori from the authentic elements on which the taxon was based when no holotype was designated or when the holotype is lost livr. =livraison, part IKeex= ex (plurs) LS = longitudinal or lengthwise section of an organ m= metre M = Malay (language) Mag. = Magindando (Philip. language) Mak. = Makassar, Macassar (in SW. Celebes) Mal. = Malay(an) Mal. Pen. = Malay Peninsula Mand. = Mandaya (Philip. language) Mang. = Mangyan (Philip. language) Mar. = March Mbo = Manobo (Philip. language) Md. = Madurese (language) Minangk. = Minangkabau (a Sumatran language) min. part. =pro minore parte; for the smaller part mm = milimetre Mng. = Mangguangan (Philip. language) Morph. = Morphology ms(c), MS(S) = manuscript(s) Mt(s) = Mount(ains) n. =numero; number N=North (after degrees: northern latitude); or New (e.g. in N. Guinea) NE. = northeast nec=not neerl. = Netherlands, Netherlands edition Neg. = Negrito (Philip. language) N.E.I. = Netherlands East Indies neotype=the specimen designated to serve as no- menclatural type when no authentic specimens have existed or when they have been lost; a neotype retains its status as the new type as long as no auth- entic elements are recovered and as long as it can be shown to be satisfactory in accordance with the original description or figure of the taxon N.G. = New Guinea N.I. = Netherlands Indies no =numero; number nom. = nomen; name (only)=nomen nudum nom. al.=nomen aliorum; name used by other authors nom. alt(ern).=nomen_ alternativum; name nom. cons(erv).=nomen conservandum, nomina conservanda; generic name(s) conserved by the In- ternational Rules of Botanical Nomenclature nom. fam. cons.=nomen familiarum conservan- dum; conserved family name nom. gen. cons. =see nomen conservandum nom. gen. cons. prop.=nomen genericum conser- vandum propositum; generic name proposed for conservation nom. _ illeg(it). =nomen name nom. leg(it). =nomen legitimum; legitimate name nom. nov. =nomen novum; new name nom, nud. = nomen nudum; name published without description and without reference to previous pub- lications alternative illegitimum; illegitimate (46) nom. rej(ic.) = nomen rejiciendum; name rejected by the International Rules of Botanical Nomenclature nom. seminudum=a name which is provided with some unessential notes or details which cannot be considered to represent a sufficient description which is, according to the International Rules of Botanical Nomenclature, compulsory for valid publication of the name of a taxon nom. subnudum = nomen seminudum nom. superfl. =a name superfluous when it was pub- lished; in most cases it is a name based on the same type as an other earlier specific name non followed by author’s name and year, not placed in parentheses, and put at the end of a citation = means that this author has published the same name mentioned in the citation independently. These names (combinations) are therefore homo- nyms. Compare 56b line 5—4 from bottom. The same can happen with generic names. (non followed by abbreviation of author’s name) be- fore a reference (citation) headed by an other author’s name = means that the second author has misinterpreted the taxon of the first author. Compare p. 419a under species 47 the synonym H. celebica. DIELS misapplied the name H. celebica as earlier described by BURCK. non al. =non aliorum; not of other authors non vidi=not seen by the author nov. =nova (femin.); new (species, variety, efc.) Nov. = November n.s. =new series n. sp. =nova species; new species n. (sp.) prov.=nomen (specificum) provisorium; provisional new (specific) name n.v. =non vidi; not seen NW. =northwest Oct. = October op.cit. = opere citato; in the work cited Pp. =pagina; page P.=Pulau, Pulu (in Malay); Island Pal(emb.) = Palembang Pamp. = Pampangan (Philip. language) Pang. = Pangasinan (Philip. language) paratype = a specimen cited with the original descrip- tion other than the holotype part. alt. =for the other part P. Bis. = Panay Bisaya (Philip. language) P.I. = Philippine Islands pl. =plate plurim., = plurimus; most D.p. = pro parte; partly pr. max. p. = pro maxima parte; for the greater part pro=as far as is concerned prob. = probabiliter; probably prop. = propositus; proposed Prov. = Province pr.p. = pro parte; partly pt = part quae est =which is quoad basionym, syn., specimina, efc. =as far as the basionym, synonym(s), specimen(s), efc. are con- cerned references =see for abbreviations the list in vol. 5, pp. cxlv—clxv Res. = Residency or Reserve resp. = respective(ly) Abbreviations and signs S=south (after degrees: southern latitude) S (after a vernacular name) = Sundanese (language) Sbl. =Sambali (Philip. language) SE. = southeast sec. =secus; according to sect. =sectio; section sens. ampl. (ampliss.)=sensu amplo (amplissimo); in a wider sense, in the widest sense sens. lat. =sensu lato; in a wide sense sens. str. (Sstrictiss.)=sensu stricto (strictissimo); in the narrow sense, in the narrowest sense Sept. = September seq., seqq. =sequens, sequentia; the following ser. = series s.1. =sensu lato; in a wide sense S.-L. Bis. =Samar-Leyte Bisaya (Philip. language) Sml. = Samal (Philip. language) s.n.=sine numero; (specimen) without the collec- tor’s number Sp. =Spanish (language) sp(ec). = species; species specim. = specimen(s) sphalm. = sphalmate; by error, erroneous Spp. = species; species (plural) Sr = Senior S.S. =see sens. Str. ssp. = subspecies; subspecies s.str. =see sens. Str. stat. nov. =Status nova; proposed in a new rank Sub. = Subanum (Philip. language) subg(en). =subgenus; subgenus subsect. = subsectio; subsection subsp. = subspecies; subspecies Sul. = Sulu (Philip. language) Sum. E.C. = Sumatra East Coast Sum. W.C.=Sumatra West Coast Suppl. = Supplement SW. =southwest syn. =synonymum; synonym synonyms =the names of taxa which have been re- ferred to an earlier described taxon of the same rank and with which they have been united on taxonomical grounds or which are bound together nomenclaturally syntypes = the specimens used by the original author when no holotype was designed or more specimens were simultaneously designated as type t. = tabula; plate Tag. = Tagalog (Philip. language) Tagb. = Tagbanua (Philip. language) Tagk. = Tagaka-olo (Philip. language) Tapan. = Tapanuli (in NW. Sumatra) taxon=each entity throughout the hierarchic ranks of the plant kingdom which can be described and discriminated from other taxa of the same rank Taxon. = Taxonomy Tg =Tandjung (Malay); cape Ting. = Tinggian (Philip. language) Tir. = Tirurai (Philip. language) transl. = translated type = each taxon above the rank of a species is typi- fied by a type belonging to a lower rank, for in- stance a family by a genus, a genus in its turn by a species; a species or infraspecific taxon is typified by aspecimen. The name of a taxon is nomenclatu- rally permanently attached to its type; from this it cannot be inferred that the type always represents botanically the most typical or average structure found in the circumscription of the taxon. type specimen=the specimen or other element to which the name of a species or infraspecific taxon is (nomenclaturally) permanently attached; botan- ically a type specimen is a random specimen on which the name was based by description. There- fore, it does not need to represent the average or most typical representative of a population. See holotype, isotype, lectotype, syntype, paratype, and neotype typ. excl. = typo excluso; type excluded typ. incl. =typo incluso; type included typus =see type and type specimen var. = varietas; variety var. nov. =varietas nova; new variety Vern. = Vernacular vide = see viz. = videlicet; namely vol. = volume W =west (after degrees: western longitude) Yak. = Yakan (Philip. language) + =about ) = male (flower, efc.) 2 =female (flower, efc.) ( )=monoecious with unisexual flowers )= polygamous ) = polygamous c’= many > =more than (in size, number, efc.) < =less than (size, number, efc.) xX 2/5=2/5 of natural size Xx montana= means that the epithet montana is that of a hybrid (47) ~ i = = Wied tio) aegodti A eR es Ge ee a . i ae a ibhinard Pram ipueet Mire eam cove! «beh cae lei sseesey hs. Cota bots f@isad ims ai « agile. Miplhy, =e: | > in” CCE ere _ eA pair yrXs te ogi sea cage! ety rai igimper 7A) @ y putt e z a MA) aye Pan ' Pe ( mm iit < » were) e ¢ ’ one | a 1 oS rl vee el WR: Ce D . teeta ding ae he iets ott lee Te. : ie ee d a4 oie wale =ete Oe as | ro of i pepeteg be deen | + tie pata? » att ~ - wt olnker wv Nada iM : , : ‘i : vy 2 : A Aaithag at caxieg © wit fg . — oe | the iat y lear en he re ; H wip wry 2 Alina A ; ae ' = Tin As’ oe eee 4 ita a eves . : : . » he OE ees : . : ‘ : Daley, de na 7 ® é ' SS ae: soy ey o@ 414 ei &~—, Rs : +7 Be = fe 4 j ee iv ve : is: Al =i iit h —-— ees T me i re Wi .,. a (4 . tM er = Vo AYe lias) . Bits fi - tat — rt « eine @ a uPe 1 3 ; 4 fil ut Luisi.) 4 Sea Oinaey in ie Wage ow) Poghh stad) oma iene, | bawenmin Vie oll Hae ue ‘Mines on bela aoc " _ atl Mixt Pils @! Eiicwws! “am: cure we

44 ne eee 9. Gastonia OM Pedicellarticulatedibelow;themowel cme ae en eee een oe ae: 10. Polyscias 9. Leaves not pinnate. 11. Pedicel articulated below the flower. 12. Leaves digitately compound (or rarely unifoliolate) (Malay Peninsula, Sumatra, Java) 11. Macropanax 12. Leaves simple, not articulated with the petiole (West New Guinea) . . . . 12. Anakasia 11. Pedicel not articulated below the flower. 13. Leaf simple (or unifoliolate) or palmately lobed. 14. Leaf palmately lobed. BO VARY 2-COMEG: (yes a4 oc si o Sehst a os Ste S o 13. Brassaiopsis IS Ovary,.10=orsmore-celledvose-s.... oachaaet wud) Bile Soo Gickbs OR Seoue 14. Trevesia 14. Leaf simple (or unifoliolate). 16. Articulation present between petiole and blade. ........2.2.2.2.. Schefflera 16. No articulation between petiole and blade. AV ROVATYP2-COLCA) graye cers. oy Bn sk She Te Rs By GED tsuki, Sieies dates ess 13. Brassaiopsis ieeOvary’4-.or;more-celled o..) 0.546" ¢ os EA ee thes Cl NS at db es 15. Dendropanax 13. Leaf digitately compound. 18. Petiolules joined together by a web of tissue. . .........2.2.4.-. 14. Trevesia 18. No such web of tissue present. 19. Styles or stigmas 2. Boe Otyle DING teri Woes 6 We sb earane 6 Feb. oxen? ecaeey WA Moa aee” 16. Acanthopanax 20 estylestunited intola columnysure-ges eee lee) eRe ee nome) oe lee 13. Brassaiopsis ios Styles or stigmas more: than.2. oi Spill nls Shen sin eine Nelsen iene Schefflera 1. HARMSIOPANAX Wars. in E. & P. Nat. Pfl. Fam. Nachtr. 1 (1897) 166; HArRms, Bot. Jahrb. 56 (1921) 413; Hutcu. Gen. FI. Pl. 2 (1967) 62; PHiLipson, Blumea 21 (1973) 81. — Schubertia BL. Bijdr. (1826) 884, nom. illeg., non MirRB. 1812. — Horsfieldia BL. ex DC. Prod. 4 (1830) 87, non WILLD. 1805; Bru. in B. & H. Gen. Pl. 1 (1865) 937; BoerL. Hand. 1 (1890) 633; HARMs in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 60. — Fig. 1-3. Sparingly branched or single-trunked often monocarpic trees, up to 18 m. Stems stout, bearing terminal clusters of large, palmately lobed, often peltate, exstipulate leaves. Trunk, petioles, and sometimes the blades spiny. Flowers in very large, repeatedly branched, terminal panicles which develop after the leaves have fallen. Umbellules arranged racemosely on the ultimate branchlets, sessile or peduncled, each consisting of a few to many pedicelled flowers. Pedicels not jointed, subtended by a bract and bearing two subulate bracteoles. Flowers hermaphrodite or with hermaphrodite flowers on terminal and male flowers on basal branches. Calyx a , Rh SN a SS SH litte: y YY a ea a a Sana 2) REREeeah es Seen +) ,—,, NAN, . = ON * 4 r. i, o ayiNatal BP une > NY See Nea pa era SRO SN AS RY ACO Ke -—. Brea Ne vo Ss r ‘a rl (3 AR Sy a p = > =, SORUN AALS OS. Ueraaaia SS | aay, Sg AN oa, ie ete TA ba TR RS eS mame ie i MS [ser. I, vol. 91 FLORA MALESIANA 10 Seep Nd als ‘ Bs a lule, x 4, d. developing fruit, x 12 (a NGF 36901, b-d Puiipson 3483). Fig. 1. Harmsiopanax ingens PHILIPSON ssp. ingens. a. Leaf, x 4/., b. part of inflorescence, x 1/3, c. umbel- 1979] ARALIACEAE—I (Philipson) 11 minute rim. Petals.5, free, valvate with a broad base. Stamens 5, dorsifixed, versa- tile, introrse. Ovary inferior, narrowly obconic, densely bristly; cells 2; disk conical, deeply cleft between the two subulate styles. Fruit consisting of 2 dry mericarps, each 3-ribbed and bearing a persistent slightly hooked style. Distr. Malesia: 3 spp. from Java, the Lesser Sunda Is., Celebes, and New Guinea. Ecol. Montane and mossy forest and in regrowth on grassy hillsides. Notes. Harmsiopanax is a small structurally isolated genus confined to Malesia. The three species are uniform both in their vegetative and their reproductive features. It has long been recognized that some of the characters of this genus are anomalous within Araliaceae and a return to its earlier position within Umbelliferae would have something in its favour. The monocarpic habit is unknown elsewhere in Aralia- ceae, but is not uncommon in Umbelliferae. The character of the fruit, which splits into two dry mericarps, closely approaches the fruit structure of Umbelliferae, and the vascularization of the gynoecium is also characteristic of that family. However, the structure of the leaf-base, the woody habit, and the shape of the petals all incline towards Araliaceae. KEY TO THE SPECIES 1. Upper surface of leaves uniformly setulose. PemUuMIDEIUIESISESSIIC.. 5 6 es He 2 Wmbellules peduncled. . ..... .4 4. 1. Upper surface of leaves with many (or rarely few) larger spines among the setulose hairs 1. Harmsiopanax aculeatus (BL.) WARB. ex BOERL. Hand. 3 (1900) 88; Koorp. Exk. Fl. Java 2 (1912) 719; Atlas 4 (1916) f. 668 & 669; BakH. f. & OoststR. in Back. Bekn. Fl. Java (em. ed.) 7 (1946) fam. 159, p. 19; Back. & BAKH. f. Fl. Java 2 (1965) 171; STEEN. Mt. FI. Java (1972) pl. 3-2; PHILIPSON, Blumea 21 (1973) 82. — Schubertia aculeata BL. Bijdr. (1826) 885. — Horsfieldia aculeata (BL.) DC. Prod. 4 (1830) 87; BENN. PI. Jav. Rar. (1840) 123, t. 26; K. & V. Bijdr. 7 (1900) 57; BoeRL. Handl. 1 (1890) 647. — Horsfieldia peltata BTuH. in B. & H. Gen. Pl. 1 (1862) 937. — Fig. 2. Tree up to 4 m, with a slender spiny trunk. Young stems covered more or less densely with woolly hairs, bristles, and spines with bulbous bases, the spines enlarging on older stems. Leaves rounded, variable in size, often 60 cm or more in @, deeply palmately lobed, usually peltate in mature leaves, sinuses between the lobes broad or narrow, lobes 7-10, usually sharply and irregularly incised and toothed, apex acute, upper surface rather sparsely covered with evenly-spaced, appressed, sometimes branched hairs (denser on the main veins), underside densely clothed with a soft, woolly tomentum, often with some bristles on the main veins; petiole c. 60 cm, 1 cm @ at base, terete with clasping base, densely covered with woolley hairs, bristles, and some spines. Inflores- cence up to c. 70 cm long, main branches rather sparsely covered with a short tomentum and, when young, bearing numerous bracts similar to the leaves but smaller, not peltate, and often 3-lobed or entire; ultimate branchlets slender and often woolly-tomentose, bearing minute linear bracts which subtend the sessile umbellules. Umbellules si Yous 46: JeMve® “eLie se, im. a5 ea) Pw) ie, Se about 4 mm @ in flower, the broadly ovate outer bracts forming a more or less distinct involucre. Flowers hermaphrodite or male, either mixed in an inflorescence, or separate, c. 10-15 per umbellule, each subtended by a lanceolate receptacular bract c. 2 mm long. Pedicel c. !/, mm long. Calyx rim fringed. Petals strap-shaped, c. 11/, mm long at anthesis. Filaments c. 2 mm; anthers c. 0.3 mm long, orbicular. Ovary covered with cilia which lengthen as the fruit ripens. Mericarps long-ciliate, crowned with the divergent styles. Distr. Malesia: Java, Lesser Sunda Is. (Bali, Lombok, Sumbawa, Flores, Timor), southern half of Celebes. There is a single KoRTHALS sheet in L, ticketed from Central Sumatra, but this is presum- ably wrongly localized. Ecol. Usually in rather dry, open localities, but also in forest, in secondary forest, also pioneering on rocks, in grasslands and on lava-streams, 300-1800 m. FI. fr. April-Nov. ScHMutTz found it in Flores flowering in October, but leaves had fallen. Vern. Java: djankurang, d. tjutjuk, djogloran- grang, S, gabus, garang, g. lanang, gungrang, udulan laki, J. Note. In this species lateral shoots usually appear below the infructescences so that the trees are not normally monocarpic. 2. Harmsiopanax harmsii K. Scu. in K. Sch. & Laut. Nachtr. (1905) 329; Harms, Bot. Jahrb. 56 (1921) 413. Tree up to 7 m, with a slender trunk. Young stem covered with bristles, hairs, and spines, older stems with smooth bark with small rounded lenti- cels and numerous spines. Monocarpic. Leaves 12 FLORA MALESIANA [ser. I, vol. 9! Fig. 2. Habit of Harmsiopanax aculeatus (BL.) WARB. ex BOERL. Coarse shrub on old lava-streams in E. Java (Mt Idjen) at c. 900 m altitude (Photogr. VAN STEENIS). rounded, up to 30 by 40 cm, deeply palmately lobed, cordate at base, lobes 5—9 with broad sinuses between them, margin unevenly and sharply den- tate, apex acute, upper surface densely covered with evenly spaced bristles of varying size (larger on the main veins), appressed and directed towards the leaf margin, often with woolly hairs inserted on their enlarged bases, the underside very densely woolly and with many bristles, usually bearing crisped hairs on their enlarged bases; petiole 50 cm, !/, cm @ at base, terete with clasping base, densely covered with bristles, woolly hairs, and spines. Panicle at first with numerous leaf-like bracts, the principal branches with some spines, rather sparsely covered with bristles and hairs, ultimate branches slender and tomentose, bearing linear bracts c. 4 mm long subtending peduncled umbellules; peduncles up to 5 mm, slender, tomen- tose, bearing 2 minute bracts. Umbellules spheri- cal, c. 4-5 mm @ in flower, outer bracts not form- 1979] ARALIACEAE—I (Philipson) 1) ing a distinct involucre. Flowers hermaphrodite, maturing in basipetal succession, the lower bracts of a branch either with sterile umbellules or lacking flowers; up to 60 in an umbellule, each subtended by a lanceolate ciliolate bract c. 1 mm long, and borne on a glabrous pedicel c. 1'/, mm long. Calyx rim fringed with many lacerate filaments. Petals ovate, c. 1 mm long. Filaments c. 1 mm; anthers c. '/, mm long. Ovary covered with cilia which lengthen as the fruit ripens. Mericarps with rounded ribs, long-ciliate, crowned by the diver- gent styles. Distr. Malesia: Papua New Guinea (Madang Distr., Western Highlands, Morobe Distr. & Central Distr.). Ecol. Forested hills, grassy slopes, and road- sides, 100-1800 m. Vern. Opme, Ganja, Mt Hagen, mafiong, Sattelberg, Morobe Distr. Note. Information about the habit is inade- quate. The stalked spherical umbellules are very distinctive. % 3 . 4 3 ® . 3. Harmsiopanax ingens PHILIPSON, Blumea 21 (1973) 84. ssp. ingens. — Fig. 1, 3. Unbranched tree up to 18 m witha thick or some- times slender trunk densely covered, except towards the base of mature specimens, with long, sharp, upwardly directed spines and marked with leaf-scars. Monocarpic. Leaves usually peltate, rounded, up to 1 m @, deeply palmately lobed, lobes usually with minor lobes and coarsely dentate, apex acute, upper surface bearing few to many long spines, especially on the midrib and principal veins between which the surface is often rugose and glabrous except for the remains of a tomentum of branched hairs, or with many bristles often with woolly hairs on their bases, the under- surface also with few to many long spines and usually clothed with a fawn or greyish woolly tomentum of branched hairs, or densely furnished with bristles usually with woolly hairs on their bases, or occasionally glabrous between the spines Fig. 3. Harmsiopanax ingens PHILIPSON. Left: apex of leafy stem; right: the large inflorescence (Photogr. FRODIN, Murmur Pass, 1971). 14 FLORA MALESIANA [ser. I, vol. 91 except for a few bristles; petiole up to 1 m and 3 cm @, terete with clasping base, covered with woolly hairs and bearing many spines. Panicle up to 5 m long and 5 m wide, leafless or with lobed bracts c. 10-20 cm long, principal branches spiny especially below, ultimate branches slender, tomentose, bearing linear bracts c. 1 cm long sub- tending peduncled or sessile umbellules; peduncles elongating as the fruit ripens, up to 4 mm, rather stout, tomentose, bearing 1 or 2 minute bracts. Umbellules bowl-shaped, c. 6-10 mm @ in flower, enlarging slightly in fruit, with an involucre of about 8 ovate bracts, 2-4 mm long and ciliolate distally. Flowers hermaphrodite, maturing in basipetal succession, terminal branches bearing maturing fruit while lower branches bear flowers or unopened buds; usually c. 12-16 (8-20) in an umbellule each subtended by an involucral bract or a narrower receptacular bract and borne on a glabrous pedicel 1-2 mm long. Calyx rim fringed with many lacerate filaments. Petals ovate, 1-2 mm long. Filaments 2-31/, mm; anthers 4/,—3/, mm long. Ovary covered with cilia which lengthen as the fruit ripens. Mericarps with rounded ribs, long- ciliate, crowned by the divergent styles. Distr. Malesia: New Guinea (NW. Irian and extending along the central mountains from the Orion Mts to the Owen Stanley Range, Murray Pass). Ecol. Montane and mossy forest and second- growth forest, 2000-3600 m, occasionally rather lower. Vern. Papua: mauku, Huli; Mandated Terr.: Sepik Distr., kKamul, Hindenburg Ra.; Western Highlands: murri, Hagen, tolsan, Minj, mauri, Melpa, mai, Mendi, kinogore, makua, makw, Enga; Eastern Highlands: kimu, Ka, ollu, Chimbu. Notes. A striking, single-trunked, monocarpic tree bearing immense inflorescences. The bark is described as grey brown and the wood white witha wide pith. The inflorescence has the appearance of bearing female flowers above and male flowers below, but this is evidently due to a basipetal sequence of anthesis. The terminal flowers have stamens when freshly opened and all those on lower branches bear styles. Apparently, the female organs of the lower branches are functional because branches from mature inflorescences bear fruit uniformly. Nevertheless, herbarium specimens cannot adequately represent such a large inflores- cence so that the possibility remains that some female-sterile flowers occur in this species. Variation occurs in both tomentum and in- florescence characters. For example, most speci- mens from West Irian have small umbellules and fewer leaf-spines. In the eastern part of the Eastern Highlands District a number of gatherings display a series of variations: the under-leaves give the appearance of being glabrous between bristles, the inflorescence branches bear small leafy bracts, the umbellules are sessile, with rather numerous {c. 18-21) small flowers subtended by rather broad bracts. Specimens from Mt Otto show all these features combined, but other specimens from this region diverge from the typical state in only some of these characters. No specimens of this subspecies are known from the Finisterre Range and only one from the Owen Stanley Range. ssp. moniliformis PHILIPSON, Blumea 21 (1973) 86. Umbellules disposed irregularly along the branches, singly or in small groups, with bare spaces intervening, sessile; flowers usually c. 20-30 per umbellule, floral parts smaller than in ssp. ingens; fruiting heads rather small (c. 5 mm @). Distr. Malesia: Papua New Guinea (districts bordering on the Huon Gulf). Vern. Morobe Distr.: mobian, Finschhafen. Note. This subspecies occurs at lower altitudes than is usual for ssp. ingens {1500-2000 m). No specimens of either subspecies have been collected from higher altitudes in the mountains north of the Markham River and the Huon Gulf. At higher altitudes in the Owen Stanley Range ssp. ingens is known from one gathering. The most south-easterly gathering at present known (CARR 13603) has a distinctive appearance due to the straight rigid in- florescence branches with small sessile umbellules. 2. ARALIDIUM Mia. Pl. Jungh. 3 (1855) 423; Fl. Ind. Bat. 1, 1 (1856) 762, t. 13; Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 25; Bru. in B. & H. Gen. Pl. 1 (1865) 936; HEmsL. in Hook. Ic. Pl. 16 (1886) t. 1549; Bogert. Handl. 1 (1890) 631; Harms in E. & P. Nat. Pf. Fam. 3, 8 (1894) 60; Hutcu. Gen. FI. Pl. 2 (1967) 59. — Fig. 4. Unarmed tree or shrub with simple, exstipulate pinnately lobed, irregularly incised, or entire leaves. Inflorescence a large panicle, with cymules of small flowers arranged racemosely on the branches. Pedicels articulated below the ovary. Dioecious. Male flowers: calyx 5; petals 5, imbricate; stamens 5, anthers dorsifixed. Female flowers: calyx and corolla similar to male; staminodes 5; ovary with 3 1979] ARALIACEAE—I (Philipson) 15 Fig. 4. Aralidium pinnatifidum (JUNGH. & DE VriesE) Mia. a. Habit, x 2/;, b. 3 flower and bud, c. 2 flower and bud, both x 8, d. fruit, seed, and CS, slightly enlarged (a VAN BALGOOY 2185, b CocKBURN FRI 8376, c SINCLAIR 9884, d fresh material). Drawn by W. R. PHILIPSON. 16 FLORA MALESIANA [ser. I, vol. 91 locules (2 abortive) and 1 ovule, styles 3-4, tapering from broad bases, stigmas terminal. Fruit drupe-like. Seed solitary, pendulous from a thickened funicle, 4-5-grooved; endosperm deeply ruminate. Distr. Monotypic. Peninsular Thailand and Malesia: Malay Peninsula, Sumatra, and Borneo. The record from Java by Mique (Ann. Mus. Bot. Lugd.-Bat. 1, 1863, 25) is erroneous (see K. & V. Bijdr. 7, 1900, 2). Ecol. Primary and secondary forest, from sea-level to c. 1250 m. Notes. The single species forms a genus with several unique features, namely the large, simple, charac- teristically pinnately lobed leaves, the absence of resin ducts, the diffuse panicles of male or female flowers, the 3-carpellate ovary with a single surviving loculus resulting in a single-seeded fruit, the dorsal raphe, the deeply ruminate endosperm, and the enlarged funicle. The genus is treated here as a member of the Araliaceae mainly as a matter of convenience. Sometimes it has been placed in the Cornaceae (e.g. RiDL. Fl. Mal. Pen. 1, 1922, 894, and Vicuter, Ann. Sc. Nat. Bot. 4, 1906, 171), and on full investigation it may well prove to be better placed in that family. The absence of resin ducts and the dorsal raphe strongly support a relationship with the Cornaceae and its immediate allies, though the absence of borders to the pits of the xylary fibres is characteristic of Araliaceae. Several genera formerly placed in the Cornaceae have now been elevated to the rank of family. If this course is followed then Aralidium should also be segregated. Many of the features of Aralidium approach those of Griselinia (segregated as Griseliniaceae) and possibly these two genera should be united as a single family. 1. Aralidium pinnatifidum (JUNGH. & DE VRIESE Mia. FI. Ind. Bat. 1, 1 (1856) 763; HEMsL. in Hook. Ic. Pl. 16 (1886) t. 1549; BoerL. Handl. 1 (1890) 631; Ripv. Fl. Mal. Pen. 1 (1922) 895; Puiieson, J. Bot. 78 (1940) 118. — Aralia pinnatifida JUNGH. & DE VRIESE, Ned. Kruidk. Arch. 1 (1846) 15; Ann. Sc. Nat. III, 6 (1846) 115. — A. dentatum Mia. Sum. (1861) 340. — A. integrifolium Herne in Fedde, Rep. 54 (1951) 245. — Fig. 4. Shrub or small tree up to c. 10 m, rarely reaching 20 m and 25 cm @, glabrous in its vegetative parts; buds enclosed in long sheathing leaf-bases. Leaves spaced with distinct internodes, usually c. 30 by 22 cm or more, + regularly pinnately incised, frequently as deep as the midrib, lobes oblong- acuminate and decurrent on the midrib, c. 2!/, cm wide or more, the lobing sometimes irregular, and occasionally the blade entire and broadly ovate (up to 25 by 20 cm) or rarely lanceolate, leaf mar- gin either entire or coarsely dentate, especially on the terminal lobe; petioles 5-12 cm, broadly channelled above, clasping the stem with a slightly dilated base, exstipulate. Panicles terminal, or occasionally in the upper axils, to 50 cm long, pendulous, puberulous; main bracts caducous, but the minute bracteoles often persisting until anthesis. Flowers numerous, small (buds c. 2!/, mm long), fragrant, creamy or red-tinged, ovary, calyx lobes and petals densely covered in a minute but coarse puberulence. Male flowers with the corolla persistent during anthesis, petals c. 1!/, mm long, strap-shaped, spreading, stamens c. 1 mm with flattened filaments and round anthers; stylopodium a succulent disk with a concave centre; styles absent, the ovary 11/, mm long, narrowly turbinate, without a loculus. Female flowers with the corolla caducous at anthesis together with the staminodes, styles divergent from their gibbous bases, ovary ovate with a single loculus (two abort early); ovule pendulous. Fruit usually obliquely ellipsoid, tapering to the apex and c. 3-4'/, cm long, but rarely subspherical, white when immature, ripening to purplish or black, juicy; exocarp fleshy, endocarp chartaceous. Seed broadly ellipsoid, 2—2'/, cm long, with the surface patterned with deep ruminations. Distr. Peninsular Thailand; in Malesia: Malay Peninsula (from Kedah southwards common; Singapore), throughout Sumatra (also in Simalur I.), Anambas Is. (Siantan) and throughout Borneo. Ecol. Frequent in evergreen primary rain- forest, also in open bamboo forest and secondary growths, from sea-level to c. 1250 m, in Borneo up to 1500-1800 m. FI. fr. Jan.—Dec. Uses. The only use, once mentioned, is from Brunei, as “leaves make good ghost medicine’’. Vern. Malay Peninsula: /émpédu buaya, (poko) balai, pungar, sahalat, sébalai tingal, sibilai, tébalai, M; Sumatra: (kayu) attarodan, Asahan, Batak lang., ségéntut, Gajo, médung, M, maneél silai, mannel dotan, sukun dotan, M, Simalur; Anambas Is.: ballok, M, Siantan; Borneo: daun tutchol antu, Brunei, Iban lang. Note. Entire leaves are not infrequent through- out the range of the species, so that the recognition of a second species using this character is not justified. Coarsely dentate leaf-margins were also employed as a specific character but are merely a minor variation. Some specimens from Mt Kina- balu have rather small globose fruits, but the typical form of fruit also occurs on that mountain. 1979] ARALIACEAE—I (Philipson) 17 3. ARALIA LinnE, Gen. Pl. ed. 5 (1754) 134; Sp. Pl. (1753) 273; DC. Prod. 4 (1830) 257; Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 6; BTH. in B. & H. Gen. Pl. 1 (1865) 936; BoeRL. Handl. 1 (1890) 629; Harms in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 56; STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 391; Hutcu. Gen. FI. Pl. 2 (1967) 63; STONE, Gard. Bull. Sing. 30 (1977) 134; PHiLipson, /.c. 97. — Acanthophora MERR. Philip. J. Sc. 13 (1918) Bot. 316, non LAMOUREUX, 1813 (Algae); STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 390. — Fig. 5, 6. Sparingly branched shrubs or small trees, or climbing, rarely (extra-Mal.) herbaceous, glabrous or hairy, often prickly. Leaves pinnate to tripinnate, usually with leaflets at the insertion of the lateral pinnae; leaflets serrate; petiole with a sheathing base. /nflorescence a terminal panicle; flowers sessile or pedicelled, with an articulation below the flower; calyx with 5—6 teeth; petals 5—6, imbricate; ovary 2-6-celled; styles 2-6 free or shortly connate below. Fruit a fleshy drupe; pyrenes cartilaginous, compressed; endosperm uniform. Distr. More than 30 spp. in North America (S. to Mexico) and East Asia, 6 spp. in Malesia: Sumatra, Malay Peninsula, Java, Lesser Sunda Is. (Sumba), Borneo, Celebes, Philippines, and West New Guinea. Ecol. Usually on scrubby hillsides and in secondary growth, often in ravines or near streams, or in thickets near or above the limit of tree-growth, at low altitude (100 m), but usually in the montane zone, up to 3000 m. Note. For a discussion of specific distinctions see VAN STEENIS, /.c. 391. Hut-Liw Li in Sargentia 2 (1942) 101, treated some species that extend into Malesia. MERRILL considered that the climbing habit and recurved spines of Acanthophora justified its separation as a distinct genus, but more recent authors have not agreed. KEY TO THE SPECIES 1. Flowers sessile (capitate) or very shortly pedicelled. 2. Flowers sessile, underside of leaf + densely tomentose, hairs of the branches and inflorescence + appressed and felted, bracts around the capitula enveloped in hairs 1. A. dasyphylla 2. Flowers short-pedicelled, underside of leaf sparsely tomentose, hairs of the branches and inflorescence + patent, bracts around the capitula less densely tomentose 2. A. javanica 1. Flowers +/,-11/, cm pedicelled (umbellate). 3. Climbing or scrambling liana, spines curved 3. Erect shrubs or small trees, spines straight. 4. Leaflets glaucous beneath, margins with few crenations. Fruit small (c. 3 mm long) 4. A. bipinnata 4. Leaflets green (or with fawn pubescence) beneath, margins serrate. Fruit rather larger (4-6 mm long). 5. Young parts and undersurface of leaves glabrous (but with small spines) 5. A. ferox 5. Young parts and undersurface of leaves pubescent Pe oR ee 2 eS ee te ce ee eA 1. Aralia dasyphylla Mia. FI. Ind. Bat. 1, 1 (1856) 751; Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 9, incl. var. strigosa Mia. et var. latifolia MiQ.; BOERL. Handl. 1 (1890) 646; K. & V. Bijdr. 7 (1900) 53; Koorp. Exk. Fl. Java 2 (1912) 718; Atlas 4 (1916) f. 673 A-K; Hut-Lin Lt, Sargentia 2 (1942) 20; BAKH. f. Blumea 6 (1947) 367, incl. var. urticifolia (BL. ex Mia.) BAKH. f.; BAKH. f. & OosTsTR. in Back. Bekn. Fl. Java (em. ed.) 7 (1948) fam. 159, p. 18; STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 391; NGoc-SANH Bul, Adansonia 4 (1964) 464; Back. & BAKH. f. Fl. Java 2 (1965) 170; STEEN. Mt. FI. Java (1972) pl. 3-1; Puitrrson, Gard. Bull. Sing. 30 (1977) 98; Y.-R. Lina, Acta Phytotax. Sin. 15 (1977) 86. — A. chinensis (non L.) BL. Bijdr. (1826) 870. — A. urticifolia BL. ex Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 9; BoerL. Handi. 1 (1890) 646; K. & V. Bijdr. 7 (1900) 55; Koorp. Atlas 4 18 FLORA MALESIANA [ser. I, vol. 91 @ (9 \, i) 2S 9 08 = od “N 2 @ re Z Ves) Fig. 5. Aralia bipinnata BLANCO. a. Upper branches of inflorescence, b. pinna, c. base of petiole, all x 2/., d. flower bud, e. fruit, both x 3 (JAcoss 7017). Drawn by W. R. PHILIPSON. 1979] ARALIACEAE—I (Philipson) 19 (1916) f. 673 L-N. — A. beccarii Riu. J. Mal. Br. R. As. Soc. 1 (1923) 64. — Fig. 6. Prickly shrub or small tree, often unbranched, to c. 5 m; young parts densely brown pubescent. Leaves forming large rosettes at the summit of the stems, c. 1 m long (or more), bi- or tripinnate, with a pair of leaflets (occasionally pinnate) at each division of the rachis, the petiole, rachis and lateral rachides prickly or unarmed, densely pubescent; leaflets subsessile or petiolule c. 5 mm long (or longer), usually densely pubescent on the lower surface, less dense above, ovate to oblong-ovate, c. 5—14(-18) by 3—5(—10) cm, base rounded to sub- cordate, apex acuminate, margin finely or some- times coarsely serrulate; petiole c. 40 cm, with an elongated sheathing base and a small ligule. Inflorescence a large terminal panicle, 70 cm or more long, densely brown pubescent, rachis bearing several secondary branches c. 30-40 cm long, with ultimate branches arranged racemosely, bracts ligulate, ending in heads of several sessile flowers, surrounded by an involucre of small usually densely pubescent bracts. Calyx with 5 usually obtuse teeth; petals 5, c. 11/, mm long, glabrous; stamens 5; ovary c. 2 mm long, glabrous, 5-celled; styles 5, slightly connate below. Fruit globose, c. 31/, mm @, ribbed when dry. Distr. Northwards to southern China; in Malesia: Sumatra, Malay Peninsula, West and Central Java. Ecol. Primary forest and secondary growths in deep ravines or open hillsides, from low altitude (c. 100 m) to 2500 m. Vern. Sumatra: kaju burle lasét, k. sépak- sipang, k. si marsuga-suga, k. sipang-sipang, sami- mpadan, M; Java: gorang, osangsing, J, pangang tjutjuk, S. Note. The capitulate flowers are characteristic (see also under A. javanica). The presence of this species in the Malay Peninsula has often been overlooked, though it extends into southern China. Its variability was discussed by VAN STEENIS (1948, Ec): 2. Aralia javanica Mia. Pl. Jungh. 3 (1855) 420; Fl. Ind. Bat. 1, 1 (1856) 749; Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 9; BoERL. Handl. 1 (1890) 646; K. & V. Bijdr. 7 (1900) 55; Koorp. Exk. Fl. Java 2 (1912) 718; Atlas 4 (1916) f. 670; Baku. f. & OoststrR. in Back. Bekn. Fl. Java (em. ed.) 7 (1948) fam. 159, p. 17; Back. & BAKH. f. Fl. Java 2 (1965) 170; Puiirson, Gard. Bull. Sing. 30 (1977) 98. A shrub or small tree, often unbranched, young parts covered with brown pubescence which per- sists on the stems and rachides of the inflorescence as + patent hairs. Leaves tufted at the ends of the branches, bipinnate with a pair of leaflets at the divisions of the rachis; leaflets variable in size, subsessile or the petiolules up to c. 2 cm, blade ovate to elliptic, up to 18 by 8 cm (usually smaller), both surfaces with sparse, short, appressed, bristly hairs, base cuneate to truncate, apex acuminate, margin finely and unevenly serrulate; petiole 20-30 cm. Inflorescence a large terminal panicle, rachis bearing several secondary branches c. 30-40 cm long, with the ultimate branches bearing heads (or subumbellules) of c. 10 flowers, sur- rounded by an involucre of small linear bracts. Calyx with 5 small teeth; petals 5; stamens 5; ovary c. 2 mm long, glabrous, 5-celled; styles erect at anthesis. Fruit ovoid, c. 5 mm long, ribbed when dry, with the persistent styles recurved. Distr. Malesia: West and Central Java (Mts Papandayan, Malabar, Diéng, Surakarta). Ecol. Mountain forests, 2000-3000 m. Note. This imperfectly known species may prove to be a form of the widespread A. dasyphylla, from which it appears to differ in the shortly pedicelled flowers, the sparser leaf-tomentum, the more patent hairs on the inflorescence branches and the bracts of the umbellules less thickly enveloped in hairs. 3. Aralia scandens (MERR.) HA, Nov. Sist. Vyssh. Rast. 11 (1974) 229; Stone, Gard. Bull. Sing. 30 (1977) 276, f. 1; Puiirson, /.c. 99. — Acantho- Phora scandens MeErRR. Philip. J. Sc. 13 (1918) Bot. 316; En. Philip. 3 (1923) 236; STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 390. — A. ferox (MiaQ.) Kina, J. As. Soc. Beng. 67, ii (1898) 45; Koorp. Minah. (1898) 498; MeErRR. En. Born. (1921) 458; Rip.. Fl. Mal. Pen. 1 (1922) 872; Masam. En. Phan. Born. (1942) 564. Prickly scandent shrub, glabrous, reaching a height of 10 m or more, stems c. 2!/, cm @. Leaves dispersed (c. 30 cm apart), up to 11/, m long, tri- or quadripinnate with a pair of leaflets at each division of the rachis, prickly on the petiole, rachides, and sometimes on the leaf veins; leaflets ovate to elliptic-ovate or ovate-lanceolate, petio- lules 3-10 mm, blade 5-14 by 2!/,-5 cm, base rounded or subcordate, apex acuminate, margins finely spinulose-denticulate; petiole to 35 cm, with an elongated sheathing base and a small ligule. Inflorescence a large terminal spiny panicle, the main rachis to c. 60 cm, bearing secondary branches singly or in whorls, up to 50 cm; ultimate branches 1-4 cm, subtended by lanceolate bracts, racemosely arranged, ending in umbellules; umbellules c. 10—20-flowered, pedicels slender, 10-12 mm, with lanceolate bracts 2-3 mm long, articulated below the flower. Calyx with 5-6 short acute teeth; petals 5-6, with a broad base, imbri- cate; stamens 5-6, filaments c. 4 mm long, anthers c. 1 mm long; ovary turbinate c. 2'!/, mm long, 5-6-celled, styles 5-6, free or only slightly connate below, at first erect. Fruit ellipsoidal, c. 5 mm long, purple to blue-black, deeply furrowed when dry, crowned by the persistent radiating styles. Distr. Malesia: Malay Peninsula (Perak, Selan- gor, Pahang), Sabah (Mt Kinabalu), Philippines [ser. I, vol. 9! FLORA MALESIANA 20 Fig. 6. Aralia dasyphylla Mig. Habit, Tjibodas, on slope of Mt Gedeh, W. Java, c. 1400 m altitude (Photogr. PHILIPson, 1973). 1979] ARALIACEAE—I (Philipson) 21 (Luzon, Panay, Catanduanes, Mindanao), Celebes (Menado, Buton I., S. Celebes). Ecol. Thickets on slopes and mountainsides, often near streams, or among secondary growths, 180-1550 m. Uses. In Mindanao is reported that scrapings of the bark are applied to wounds and a decoction of the boiled bark is drunk to releave internal pain. Vern. Philippines: simbar, Bag. Note. The only species with the habit of a liana, with spaced leaves, and recurved spines. The flowers are whitish or yellowish, and slightly fra- grant and are visited by numerous small bees. The fruit is purple and fleshy. cwangayan, Mindanao, 4. Aralia bipinnata BLANCO, FI. Filip. (1837) 222; Me_rkR. Sp. Blanc. (1918) 294; En. Philip. 3 (1923) 235; STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 392, incl. f. inermis STEEN.; PHILIPSON, Gard. Bull. Sing. 30 (1977) 99. — A. hypoleuca PREsL, Epim. (1851) 250; Mia. FI. Ind. Bat. 1, 1 (1856) 751; F.-VILL. Nov. App. (1880) 101; VmAL, Phan. Cuming. Philip. (1885) 117; Rev. Pl. Vasc. Filip. (1886) 144; Harms, Bot. Jahrb. 23 (1896) 18; MERR. Philip. J. Sc. 5 (1910) Bot. 369; En. Philip. 3 (1923) 235. — A. javanica (non Mi.) F.-VILL. Nov. App. (1880) 101. — A. glauca Merr. Philip. J. Sc. 2 (1907) Bot. 291; En. Philip. 3 (1923) 236. — A. apoensis ELMER, Leafl. Philip. Bot. 7 (1914) 2325; MERR. En Philip. 3 (1923) 235. — Fig. 5. A shrub or small, sparsely branched tree to 7 m, with prickly stems. Leaves to 1'/, m or more long, forming large crowns at the ends of the branches, bipinnate, with a pair of pinnae at each division of the rachis, with some prickles, especially on the petiole or unarmed, the rachis swollen and articu- lated at the nodes; leaflets sessile or with a short petiolule, ovate to lanceolate, usually 4-5 by 2-2'/, cm, but variable in size, apex acute or acumi- nate, base rounded to cordate, usually markedly oblique in lateral leaflets, margin conspicuously crenate, upper surface green, glabrous, lower sur- face glaucous, pubescent along the veins and some- times sparingly on the mesophyl!, sometimes only in the angles of the lower veins, or almost glabrous throughout, primary and secondary veins con- spicuous; petiole to 30 cm, base long sheathing and slightly ligulate. Inflorescence a large terminal panicle 30-70 cm long (or more), peduncle and also usually the main branches prickly, the whole either almost glabrous or pubescent; peduncle 5-18 cm long, stout; main rays c. 5-10, mostly clustered at the apex of the rachis, 25-65 cm long, bearing many short tertiary branches along their length; tertiary branches usually 5-10 cm long, ending in umbellules, and bearing a small number of lateral umbellules, or branches, minute lanceo- late bracts subtending the branches of the third or higher orders; umbellules with c. 20-30 radiating pedicels ; pedicels 5-10 mm. Calyx lobes 5, rounded, 1/, mm long; petals 5, 1'/, mm long; stamens 5; ovary 5-celled, styles subulate, free. Fruit spheroi- dal, c. 3 by 4 mm, strongly 5-ribbed when dry, persistent styles spreading. Distr. Taiwan; in Malesia: Philippines (Luzon, Leyte, Negros, Mindoro, Mindanao) and West New Guinea (Vogelkop Peninsula, possibly also in Swart Valley). Ecol. In rather open forests, ravines, and in thickets and secondary growths, (700-)1000- 2450 m. Vern. Philippines: badbaranai, C.Bis., dasanat, Neg., karugi, Buk., magkasau, Bis., mara-bauya, Bag., papang, Bon., sugsuga, Ig. Note. VAN STEENIS /.c. discussed the vari- ability in pubescence and the development of spines. 5. Aralia ferox Mia. Fl. Ind. Bat. 1, 1 (1856) 750; Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 9; BoeRL. Handl. 1 (1890) 629; K. & V. Bijdr. 7 (1900) 49; Koorop. Exk. Fl. Java 2 (1912) 717; Atlas 4 (1916) f. 671; Fl. Tjib. 2 (1923) 229; Baku. f. & OoststrR. in Back. Bekn. Fl. Java (em. ed.) 7 (1948) fam. 159, p. 18; STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 394; Back. & BAKH. f. Fl. Java 2 (1965) 170; PHILIPSON, Gard. Bull. Sing. 30 (1977) 99. — A. filicifolia R1pL. J. Fed. Mal. St. Mus. 8 (1917) 42, non C. Moore, 1876. Spiny shrub or small tree, usually unbranched, to c. 10 m. Leaves forming a large rosette at the summit of the stem, up to c. 1 m long, or shorter below the inflorescence, bi- or tripinnate, with a pair of leaflets (often pinnate) at each division of the rachis, prickly on the petiole, main rachis, and often on the lateral rachides; leaflets sessile or petiolules to c. 5 mm, ovate or ovate-oblong, usually c. 3 by 1°/, cm, but variable in size, base truncate to rounded or cuneate, apex acute acumi- nate, margin sharply serrate, both surfaces with small bristle-like spines, especially on the veins, sometimes with small spines on the underside of the midrib; petiole to c. 25 cm, with an elongated sheathing base and a small ligule. Inflorescence a large terminal panicle, 25-50 cm long, glabrous, the main rachis rather short, bearing a few lateral or a terminal cluster of branches c. 15-25 cm long; tertiary branches disposed singly or in subverticils, ending in umbellules and bearing a variable number of lateral umbellules. Flowers c. 10-12 per umbellule; pedicels c. 5-6 mm, articulated below the ovary; calyx a rim bearing 5 narrow or trian- gular teeth; petals and stamens 5; ovary turbinate c.2 mm long; styles 5, erect at first, soon spreading, scarcely united at the base. Fruit spheroidal, c. 6 by 5 mm, deeply furrowed when dry, the persistent styles radiating. Distr. Malesia: Central W. Sumatra (Mt Kerintji), W. Java (Mts Gedeh, Patuha and Tangkuban Prahu). 22 FLORA MALESIANA [ser. I, vol. 91 Ecol. In montane scrub and among scattered trees, 1900-2900 m. Vern. Pabong, pangang njirvan, panggang- tjérmé, S. forma nana STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 394, f. 1. Smaller, probably +/,-1 m, leaves tripinnate, 30 cm long, spiny all over; leaflets 4-13 by 2-7 mm, rachides of the ultimate pinnae winged. Distr. Malesia: Central W. Sumatra (Mt Talang). Ecol. Growing about 2500 m. Note. VAN STEENIS considered this interesting dwarf form to represent the extreme of a series in size variability. 6. Aralia montana BL. Bijdr. (1826) 870; Mia. FI. Ind. Bat. 1, 1 (1856) 750; Ann. Mus. Bot. Lugd.- Bat. 1 (1863) 9, incl. var. acutata MiQ.; BOERL. Handl. 1 (1890) 646; K. & V. Bijdr. 7 (1900) 51; Koorp. Exk. Fl. Java 2 (1912) 718; Atlas 4 (1916) f. 672; STEEN. Bull. Bot. Gard. Btzg III, 17 (1948) 391; BAKH. f. & OosTstrR. in Back. Bekn. FI. Java (em. ed.) 7 (1948) fam. 159, p. 17; BAKH. f. Blumea 6 (1950) 367, incl. var. crassifolia BAKH. f.; BACK. & BAKH. f. Fl. Java 2 (1965) 170; Puitirson, Gard. Bull. Sing. 30 (1977) 100. — A. bipinnata REINW. ex BL. Cat. (1823) 43, nomen; ex DE VRIESE, PI. Ind. Bat. Or. (1857) 84, nomen in synon. — Panax armatus WALL. [Cat. (1832) n. 4933, nomen] ex G. Don, Gen. Syst. 3 (1834) 386. — A. decomposita REINW. ex DE VRIESE, Pl. Ind. Bat. Or. (1857) 84, nom. illeg. altern. — A. armata (WALL.) SEEM. J. Bot. 6 (1868) 134; CLarkE, Fl. Br. Ind. 2 (1879) 723; KING, J. As. Soc. Beng. 67, ii (1898) 44; RIDL. Fl. Mal. Pen. 1 (1922) 873; Hur-Lin Lt, Sargentia 2 (1942) 106. — A. thomsonii SEEM. J. Bot. 6 (1868) 134; CLARKE, FI. Br. Ind. 2 (1879) 723; Kine, J. As. Soc. Beng. 67, ii (1898) 44; RipL. Fl. Mal. Pen. 1 (1922) 873; Hut-Lin Li, Sargentia 2 (1942) i» Shrub or small tree, frequently unbranched, with prickly stems, occasionally attaining a height of 12 m. Leaves to 1 m or more long, forming large crowns at the ends of the branches, bipinnate, with a pair of simple or occasionally pinnate leaflets at each division of the rachis, usually with some prickles, especially on the petiole, or unarmed, the rachis constricted at the joints; leaflets sessile or with a short petiolule, ovate, up to 14 by 7 cm, apex acute to acuminate, base truncate or rounded, oblique in lateral leaflets, margin sharply serrate, upper surface with the remains of a strigose tomen- tum, often + rugose, lower surface often with a + velvety tomentum, or with more harsh hairs + confined to the veins; petiole to 30 cm, its base sheathing and ligulate. Jnflorescence a large ter- minal panicle, peduncle and branches tomentose, prickles, if any, confined to the peduncle and main rachis, small usually persistent linear or ovate bracts c. 11/, cm long subtending the branches and also spaced along the peduncle; bracts of tertiary branches similar but smaller; secondary branches at intervals along the main rachis, c. 35 cm long, bearing numerous tertiary branches along their length; tertiary branches usually c. 6 cm long, ending in umbellules, and often bearing a number of lateral umbellules; umbellules with c. 20-30 radiating pedicels; pedicels usually 12-15 mm, occasionally shorter, pubescent. Flowers herm- aphrodite; calyx lobes 5, triangular or rounded; petals 5, c. 2 mm long; stamens 5; ovary 5-celled, glabrous; styles subulate, connate below, free and spreading above. Fruit spheroidal, up to c. 4 by 4 mm, strongly 5-ribbed when dry, surmounted by the reflexed styles. Distr. Malesia: Malay Peninsula, Sumatra, Java, Borneo (Sarawak, Sabah), Celebes, Lesser Sunda Is. (Sumba). Ecol. Primary and secondary forest, bamboo forest and low-lying moist ground, from near sea- level to 2600 m. Vern. Gorang, panggang tjutjuk, S; Malay Peninsula: poko dulang-dulang; Sarawak: tepa paluk. Notes. This species is considered to include all West Malesian examples with pedicelled flowers and pubescent leaves. This broad concept is contrary to former treatments which have recog- nized several species (A. thomsonii, A. armata). The alliance with A. chinensis L. and A. decaisneana HANCE is also very close. The application of names to this and other Javanese species has been very confused. The position is ably discussed by VALETON (in K. & V. Bijdr.) and by VAN STEENIS (/.c.). In most specimens the lower leaf surface and the pedicels are densely tomentose, but there are specimens in which the leaf is only sparsely hairy and the pedicels may be glabrous. A. armata appears to be within the range of variation of the complex although this plant is very spiny, its leaves and inflorescences always being provided with numerous short spines. The leaflets also are thinner and smoother and, like the pedicels, are less densely pubescent (see NGoc-SANH Bul, Adan- sonia 9, 1969, 461). However, A. armata (if distinct) has been collected only very rarely in the Malay Peninsula and only in the extreme north. The most aberrant specimens are those with glabrous umbel- lules which are mostly from Sumatra but also from Java: they may indicate that A. foliolosa SEEM. should also be included in this complex. Excluded Aralia capitulata JUNGH. & DE VRIESE, Ned. Kruidk. Arch. 1 (1846) 17; Ann. Sc. Nat. III, 7 (1846) 116 is, cf. SLEUMER, FI. Males. I, 7 (1971) 24 = Gomphandra capitulata (JUNGH. & DE VRIESE) BECC. (Icacinaceae). 1979] ARALIACEAE—I (Philipson) 23 Fig. 7. Delarbrea collina VIEL. a. Habit, x */s, b. flower bud, c. flower and ditto in LS, d. petal, e. stamen, f. flower after anthesis, x 13, g. fruit and ditto in CS, x 3 (ae, g RIDSDALE NGF 36736, f SOEKMA $.n.). Drawn by HELENE MULDER. 24 FLORA MALESIANA [ser. I, vol. 92 4. DELARBREA VIEILL. Bull. Soc. Linn. Norm. 9 (1865) 342, 393; Bru. in B. & H. Gen. Pl. 1 (1865) 935; BRITTEN in Forbes, Nat. Wand. (1885) 506 (see also p. 354); HARMs in E. & P. Nat. Pfi. Fam. 3, 8 (1894) 61; in K. Sch. & Laut. Fl. Schutzgeb. (1900) 485; Hutcu. Gen. FI. Pl. 2 (1967) 63. — Fig. 7. Glabrous unarmed shrubs or small trees. Leaves large, imparipinnate, with a stipular sheath; leaflets alternate or opposite, entire or indistinctly dentate. Flowers in umbellules grouped in large terminal panicles; pedicels articulated below the flower. Calyx lobes 5. Petals 5, imbricate, obovate, narrowed towards the base. Stamens 5, filaments stout, anthers dorsifixed. Ovary inferior, 2-celled, disk fleshy, obconic, crowned by two erect styles with clavate stigmas. Fruit ovoid, crowned by the small calyx lobes and the recurved style arms (which eventually fall); exocarp thin, fleshy, with peripheral oil vesicles; endocarp papery; endosperm with shallow longitudinal grooves not ruminate. Distr. Queensland, Melanesia and East Malesia, 3 or 4 spp., from the Lesser Sunda Is. eastwards to New Guinea (also New Britain), Queensland, Solomons, New Caledonia, and New Hebrides. In Malesia 1 sp. Ecol. Lowland to montane forest. Note. The corolla is distinctive, the petals being strongly imbricate and narrowed towards their insertion. The fruit also has a characteristic appearance, since the calyx and stylopodium, although persistent, do not enlarge as in most other araliads. The fruit, therefore, is a smooth ellipsoid berry without a prominent rim around the apex. 1. Delarbrea collina VIEILL. Bull. Soc. Bot. Norm. 9 (1865) 342; PuiLipson, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 18. — D. sp. HEMSL. Rep. Challenger, Bot. 1, pt 3 (1885) 155. — D. paradoxa (non VIEILL.) BRITTEN in Forbes, Nat. Wand. (1885) 506. — D. lauterbachii HARMs in K. Sch. & Laut. Fl. Schutzgeb. (1900) 485. Sparsely branched shrub to 5 m high, with the multijugate leaves clustered at the ends of the branches. Leaves c. 70-100 by 30-40 cm; rachis not articulated; leaflets alternate or in pairs, c. 7 on each side; petiolules c. 1 cm long; lamina c. 17-20 by 4-6!/, cm, lanceolate, ovate, oblong or elliptic, gradually tapered to an acute apex, base truncate, rounded or cuneate, usually oblique, margin entire; petiole c. 17-20 cm, terete, lenticellate, with a heavily lenticellate clasping base with mem- branous margins. Inflorescence a terminal panicle of umbellules, rachis up to 60 cm long, bearing well-spaced secondary branches 6-25 cm long, bracts caducous; tertiary branches c. 2-8 cm long, terminating in a circlet of broadly ovate bracts (mostly caducous) surrounding the umbellules, sometimes with smaller (male) lateral umbellules; umbellules c. 2 cm @ at anthesis, with c. 30-40 flowers. Pedicels c. 5 mm (elongating to 10-15 mm in fruit), pustulate. Calyx lobes 5, obtuse, united below into a tube. Petals 5, c. 11/, by 3/, mm, keeled within. Stamens 5, 1 mm long. Ovary sometimes prominently ribbed when dry, c. 2 mm long; disk and styles c. 1 mm high at anthesis. Fruit 16 by 10 mm, purplish black when mature. Distr. Solomon Is. to New Caledonia and Queensland; in Malesia: Lesser Sunda Is. (Timor, Wetar, Babar), Moluccas (Tenimber Is., Banda), New Guinea (Aru Is., Kar Kar Is., Madang, New Britain). Fig. 8. Ecol. Rain-forest, from sea-level to 1000 m. Vern. Don, Madang. Notes. The most wide-ranging of any Malesian araliad. It was collected in Malesia by ForBEs in Fig. 8. Range of Delarbrea collina VIEILL., localities in Malesia dotted. 1979] ARALIACEAE—I (Philipson) 25 Timor in 1882, when it was incorrectly identified as the New Caledonian species D. paradoxa VIEILL. Eight years earlier it had been collected in the Aru Is. during the Challenger Expedition (HEMSLEY, /.c.) and also been referred as close to D. paradoxa. Nearly twenty years later LAUTER- BACH collected it in the Moluccas, when HARMS described it as a new species. The statement by HaArMs that the genus was known previously only from New Caledonia cannot be reconciled with his note in the Pflanzenfamilien in which he recorded BRITTEN’s report of it in Timor. Solomon Islands collections were identified as D. collina VIEILL. by PHILIPSON in 1951. The arrangement of the um- bellules differs in the panicles of D. collina and D. paradoxa. All the material from Malesia con- forms to the characters of D. collina. The plant is evidently rare, few collections having been made in spite of its wide distribution. It was formerly cultivated in the Botanic Garden at Bogor until about 1958, having been introduced from Banda. 5. PENTAPANAX SEEM. J. Bot. 2 (1868) 294; HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 55; Koorp. Bull. Jard. Bot. Btzg III, 1 (1919) 181; Hurcu. Gen. FI. Pl. 2 (1967) 63; NGoc-SANH Bul, Adansonia 9 (1969) 389; PHILIPsoN, Austrobaileya 1 (1977) 23. — Fig. 9. Trees or shrubs, often scandent, unarmed. Leaves imparipinnate, exstipulate, glabrous. Flowers in racemes or umbels, which are arranged in panicles or com- pound umbels, pedicels articulated below the ovary. Calyx 5. Petals 5, imbricate in the bud. Stamens 5. Ovary inferior, 5-celled, disk + conical, surmounted by the styles which are united their whole length or become free down to half their length. Fruit globose; exocarp leathery, enclosing crustaceous pyrenes. Seeds compressed, endosperm smooth. Distr. About 14 spp. in India, Thailand, Vietnam, Ceylon, Burma, southern China, Taiwan, in Malesia: 1 sp. locally in E. Java. South American species formerly included are best excluded, and the 2 Queensland spp. are now referred to Polyscias. Ecol. Forest and scrub. 1. Pentapanax elegans Koorpb. Bull. Jard. Bot. Btzg III, 1 (1919) 182, pl. 16 & 17; Baku. f. & Ooststr. in Back. Bekn. Fl. Java (em. ed.) 7 (1948) fam. 159, p. 17; BAcK. & BAKH. f. Fl. Java 2 (1965) 169. — Fig. 9. var. elegans. Epiphytic scrambler or terrestrial shrub up to 10 m, with unarmed branches, leaf and flower buds separate, enclosed in + persistent imbricated cataphylls. Leaves disposed along the branches; petiole c. 6-10 cm, flattened above, base scarcely dilated, and sometimes minutely fimbriated, articu- lated with the rachis, and the rachis articulated with the petiolules, articulations minutely fimbriate, rachis to 5 cm, petiolules of lateral leaflets up to 5 mm, of terminal leaflet to 20 mm, leaflets 5 or fewer, ovate to oblong-elliptic, the lateral some- times oblique, up to 10 by 7 cm, usually c. 5 by 2'/, cm, apex acute, base rounded or cuneate, margin entire or with subulate teeth, glaucous beneath. Inflorescence terminal with persistent cataphylls at the base of the main axis, umbels solitary or 1-4 smaller (apparently male) lateral umbels arising from the axils of minute bracts on the rachis; rachis 8-16 cm, slender, glabrous; terminal umbel 3-6 cm @, many-flowered; pedicels 1/,-3 cm, filiform, glabrous, with minute brac- teoles surrounding the articulation below the flower. Calyx lobes ligulate, obtuse, c. 1 mm long; petals triangular to ligulate c. 2 mm long; filaments yellow, 3 mm, anthers purple, 1/, mm long. Ovary broadly obconic, surmounted by a stylar column, 2 mm long. Fruit globose, 3-4 mm @, disk broadly conical, crowned by the persistent calyx lobes and an awl-shaped stylar column which may divide at apex. Distr. Malesia: E. Java (Mts Ardjuno and Tengger), the variety in Thailand. Ecol. Uncommon, in light forest or scrub, including Casuarina junghuhniana forest, 1700- 2600 m. Note. A remarkably isolated species of a genus otherwise unrepresented in Malesia. var. pubescens Koorb. Bull. Jard. Bot. Btzg III, 1 (1919) 183 (as var. puberula in f. 17); NGoc-SANH Bul, Adansonia 9 (1969) 389. 26 FLORA MALESIANA [ser. I, vol. 91 WY © & A ‘ % iY . aN 6 as ae Fig. 9. Pentapanax elegans Koorv. a. Habit, x 1/, b. flower bud, c. flower in anthesis, d. ovary in CS, enlarged (VAN STEENIS 10879). Drawn by P. PRENDERGAST. 1979] ARALIACEAE—I (Philipson) 27 Peduncle and pedicels tomentose. Note. Since all specimens from Mts Tengger Distr. Thailand; in Malesia: E. Java (Mt Jang). and Ardjuno lack pubescence, the retention of the Ecol. In Casuarina junghuhniana forest, scat- variety appears justified. However, more collections tered, 1900-2300 m. are required from all localities. 6. MACKINLAYA F.v.M. Fragm. 4 (1864) 119; Harms in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 62; Bot. Jahrb. 56 (1921) 413; PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 3; Hutcu. Gen. FI. Pl. 2 (1967) 65. — Anomopanax Harms [in Dalla Torre & Harms, Gen. Siph. (1903) 364, nomen;] Ann. Jard. Bot. Btzg 19 (1904) 13; in K. Sch. & Laut. Nachtr. (1905) 332; in E. & P. Nat. Pfl. Fam. Nachtr. 3 (1908) 255; Bot. Jahrb. 56 (1921) 414; Hutcu. Gen. FI. Pl. 2 (1967) 59. — Fig. 10. Glabrous unarmed shrubs, often unbranched (sympodial). Leaves with a petiole having a dilated sheath encircling the stem and (in dried material) a constriction at the apex, and with a leaf-blade either unifoliolate or digitately compound, the central leaflet, or the three central leaflets, sometimes digitately lobed or compound. Inflorescence terminal (but sympodium often continued by axillary branching), the peduncle bearing umbellately arranged branches which terminate either in umbel- lules or in cymes. Flowers male or hermaphrodite, the male flowers either in distinct inflorescences or towards the periphery of mixed inflorescences. Pedicel articulated below the flower. Calyx lobes 5—6, triangular or lanceolate. Petals 5-6, narrowed below into a distinct claw, and above into a long incurved process. Stamens 5-6; anthers subglobose. Ovary inferior, with two uni-ovulate cells. Disk prominent, with a crenulate margin. Styles 2, subulate, free, recurved in fruit. Fruit strongly compressed, 2-seeded (or one aborted), with a longitudinal furrow between the seeds; exocarp leathery, endocarp cartilaginous. Endosperm smooth. Distr. 5 spp., Solomon Is., Queensland, and in Malesia (3 spp.): Philippines, Celebes and throughout New Guinea. Ecol. Understorey of rain-forest and montane forest, or epiphytic. Also in secondary growth. Notes. The leaves are palmately divided or they may be reduced to a single leaflet, especially on the upper branches of M. schlechteri. The central leaflet, or the three central leaflets, are either lobed or com- pound, a character rarely found in other genera of this family. Another foliar character rare in the family is the insertion of the leaf-sheath round the whole circumference of the stem. This character occurs also in a section of Polyscias, but is more typical of Umbelliferae. The narrow base of the petal is also very rare in the family, but is characteristic of Umbelliferae. The constantly 2-celled ovary is also typical of Umbel- liferae, but other characters of the fruit appear to justify the retention of these plants in the Araliaceae. Reasons for regarding Anomopanax as congeneric with Mackinlaya are given by PHILIPSON (i.c.). KEY TO THE SPECIES 1. Ultimate branches of the inflorescence in irregularcymes ............ 1. M. celebica 1. Ultimate branches of the inflorescence in umbellules. 22 Primary, fays of the inflorescence many (30-50) < <= sss). 2 2 = eee ee 2. M. radiata > Prumary rays.of the inflorescence ‘¢.,15 ordewer 2) a0-e 4 ee eee 3. M. schlechteri 1. Mackinlaya celebica (HARMS) PHILIPSON, Bull. (1904) 15. — Anomopanax warburgii Harms, /.c. Br. Mus. Nat. Hist. Bot. 1 (1951) 8. — Anomopanax 15. — M. amplifolia Hemst. Kew Bull. (1909) celebicus Harms, Ann. Jard. Bot. Btzg 19 (1904) 260; Harms, Bot. Jahrb. 56 (1920) 413. — Anomo- 14; Ic. Bog. 2 (1906) t. 176 & 177. — Anomopanax panax arfakensis Gripes, Arfak (1917) 163. — philippinensis Harms, Ann. Jard. Bot. Btzg 19 Anomopanax digitata Merr. Philip. J. Sc. 17 28 FLORA MALESIANA [ser. I, vol. 92 Fig. 10. Mackinlaya celebica (HARMS) PHILIPSON. a. Part of inflorescence, x ?/., b. leaf, c. leaflet, xis d. flower cluster, e. flower, x 7, f. petal, x 13, g. fruit, x 5/; (BRAss 28056). Drawn by HELENE MULDER. 1979] (1920) 301. — Polyscias cibaria WHITE & FRANCIS ex LANE-POOLE, For. Res. (1925) 129, descr. angl. minim. — Anomopanax variifolius C. T. WHITE, J. Arn. Arb. 10 (1929) 256. — M. digitata (MERR.) PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 7. — M. warburgii (HARMS) PuHiuipson, /.c. 8. — Fig. 10. Shrub or small sparsely branched tree to 6 m. Leaves very variable in size and complexity; petiole up to 52 cm, 1'/, cm wide, terete, striate, with a membranous base ensheathing the stem; leaflets 5 (rarely 3) or the central petiolule (or the central 3 petiolules) frequently dividing to bear three, or more rarely 5 leaflets; lateral petioles short (c. 1-2 cm), the three central longer (up to 12 cm); lamina elliptic or ovate, up to 48 by 22 cm, base abruptly attenuated into the petiolule or subcordate, apex acuminate or gradually narrowed, acute, margin entire, denticulate or coarsely serrate, especially towards the apex, membranous, lamina of the terminal leaflet (and less frequently of the central three leafiets) sometimes deeply 3-lobed or with 3-5 separate leaflets, of which the lateral are markedly oblique at the base. Inflorescence a terminal compound umbel, very variable in size, either entirely of male flowers or with male and hermaphrodite fiowers, often overtopped by sym- podial growth; peduncle terete, striate, stout, up to 30(-45) cm, 3/, cm 2, bearing lanceolate bracts below the rays; primary rays c. 9-18, 10-20 cm, striate, with distal small linear bracts; secondary tays about 5-10, 3-6 cm, dividing again (often repeatedly) either umbellately or in an irregular cymose manner, the central ray frequently more strongly developed. Calyx lobes 5, triangular, c. 1 mm long. Petals 5, obovate c. 11/, mm long. Filaments c. 11/, mm, anthers small. Ovary obconic, 1-2 mm long, narrowly turbinate in male flowers, ovoid and quickly swelling in female flowers. Fruit up to 2!/, by 3 cm, compressed, rotund, constricted in the mid-axis, the two halves frequently unevenly developed. Distr. Solomon Is.; in Malesia: New Guinea (incl. New Britain and Aru Is.), Celebes, and Central & S. Philippines. Ecol. Rain-forest, open hill forest, and montane forest, also in secondary growths, from sea-level to 1450 m. Uses. LANE-POOLE (/.c.) recorded that at Mt Obree leaves and flowers are cooked with coconut oil and put in armlets in dances. Vern. Philippines: binlaon, C.Bis., pararau, Bag., tagima, Sub., lumot-lumot, Mindanao; New Guinea: bugini, wale, yam bonga, Sepik Distr., lak-lak, W. Highlands, po’undo, S. Highlands, Papua, nere, Central Distr., Papua, narona, New Britain. Notes. The leaves have a strong parsley-like odour. The flowers are creamy white, and the fruits blue to purple with a glaucous bloom. Salt is said to be obtained from the ashes of the leaves. ARALIACEAE—I (Philipson) 29 Although collected frequently in the Philippines and New Guinea, this species is unrecorded for the Moluccas. There is considerable variation in the size of the leaves and of the inflorescence. A few New Guinea specimens are intermediate in charac- ter between this and the equally common M. schlechteri, and are interpreted as hybrids. 2. Mackinlaya radiata PHiipson, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 6. Slender shrub to 5 m. Petiole c. 20 cm, terete and finely striate, base ensheathing the stem, mem- branous. Leaflets 5, or the central petiolule bearing three leaflets, the two lateral petiolules short (1- 11/, cm), the three central longer (6-8 cm, or the midpetiolule to 11 cm); lamina of the lateral leaflets elliptic or ovate, up to 20 by 12 cm, base abruptly attenuated into the petiolule, apex gradually narrowed, acute, margin entire or minutely denticulate towards the apex, mem- branous; lamina of the central leaflet similar or deeply 3-lobed or with 3 separate leaflets of which the lateral are strongly oblique at the base. Jn- florescence a terminal compound umbel; peduncle terete, striate, stout, from 20 cm to considerably longer, 4-6 mm @, bearing lanceolate bracts below the rays; primary rays numerous (c. 30-50), 9-18 cm, slender, striate, with distal minute, linear, caducous bracts; secondary rays (pedicels) numerous (35-130), filiform, 1-2 cm; outer flowers male, central hermaphrodite. Calyx lobes 5, narrowly triangular, c. '/, mm long. Petals 5, obovate, c. 1 mm long. Ovary narrowly obconic in male flowers, ovoid in hermaphrodite flowers, c. 0.7 mm long. Fruit (immature) ovate, com- pressed. Distr. Malesia: New Guinea (NW. Irian; Sepik Distr.). Ecol. Montane rain-forest and mossy forest, 900-1200 m. Vern. Apiyetimber, Sepik Distr. Note. Flowers creamy white. 3. Mackinlaya schlechteri (HARMS) PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 7. — Anomo- panax schlechteri Harms in K. Sch. & Laut. Nachtr. (1905) 332, t. 13. — Anomopanax versteegii Harms, Nova Guinea 8 (1910) 276. — M. versteegii (HARMS) PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 7. — M. brassii Puitipson, I.c. 6. — M. klossii Putrrson, I.c. 6. — M. subulata PHILIPSON, /.c. 7. Slender shrub to 6 m. Leaves 1-3{rarely 4-)folio- late. Petiole usually less than 10 cm, but occasion- ally longer (to 20 cm), especially in compound leaves, terete and finely striate, base ensheathing the stem, membranous. Lamina elliptic obovate, or oblong, occasionally irregularly lobed, up to 28 by 12 cm but usually considerably smaller, base cuneate or truncate (of lateral leaflets often 30 FLORA MALESIANA [ser. I, vol. 9? On nw y ch W 3 4 iF (\ fd ke IZ e iy “1 ‘ae PF { D i : Yale Y 50 0 aS aigtZ XV D . 0 Soy aoa 4: At Wy * Wa a ‘ my * sf 2 eo" ys \ » Art 1A / rY bes eh | Y *. v QQ § SIG GS pr } \ WW SS} | g 5 % ° Z | £ NN \ f Fig. 11. Osmoxylon novoguineense (ScuerF.) Becc. a. Leaf half, b. base of petiole, both x 2/;, c. inflores- cence, x */;, d. flower and CS of ovary, x 4, e. false fruit and ditto in CS (CRAVEN & SCHODDE 789). Drawn by W. R. PHILIPSON. 1979] ARALIACEAE—I (Philipson) 31 oblique), apex shortly acuminate, acute, margin entire or dentate towards the apex, membranous or chartaceous. Inflorescence a terminal compound umbel, often overtopped by a lateral branch at its base; peduncle terete, striate, c. 10-20 cm, bearing small lanceolate bracts below the rays; primary rays 6-15, c. 4-7 cm, striate, with minute distal caducous bracts; secondary rays (pedicels) c. 10-20, filiform or rather rigid, usually 5-12 mm; male flowers towards the outside of the umbellules. Calyx lobes 5, triangular to subulate, '/,-1 mm long. Petals 5, obovate. Ovary narrowly turbinate, in male flowers obconic or ovoid, c. */,; mm long in hermaphrodite. Fruit large, 15 by 22 mm, com- pressed, rotund, constricted above and below on the central axis; styles persistent, recurved. Distr. Malesia: New Guinea (along the Central Ranges, from the Star Mts east to Meyamya), also in New Britain. Ecol. Rain-forest and montane forest, 600— 2300 m. Uses. The cut stem exudes a viscous sap which is an irritant. The leaves are aromatic. The plant is reported to be poisonous and to have a number of medicinal uses. The boiled leaves are eaten to reduce fever and to relieve ‘korima’. Pieces of leaf placed in a cavity relieve toothache. The leaves are wrapped around taro at planting to encourage growth. Vern. Dako, Wissel Lakes, kolobang, kulbang, Sepik Distr., auke, kenata, muklofo, E. Highlands, narona, New Britain. Notes. The flowers are white and the ripe fruit mauve to purple with a glaucous bloom. A large number of collections made in recent years throughout New Guinea all have regularly compound umbels with the flowers borne on branches of the third degree in the form of strict umbels. In two of the earliest gatherings (SCHLECH- TER 14365 and VERSTEEG 1419) the third degree branches frequently divide again either umbellately or cymosely. These two specimens were described as species by HARMS. PHILIPSON later (1951) kept the forms with regular umbellules separate (de- scribing three species). All these five entities are now considered conspecific, the SCHLECHTER and VERSTEEG specimens being regarded as rare anomalies in a widespread and abundant species. It is possible that the VERSTEEG plant is a hybrid with M. celebica. 7. OSMOXYLON Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 5; BTH. in B. & H. Gen. Pl. 1 (1865) 944; Becc. Malesia | (1878) 193; BoeRL. Ann. Jard. Bot. Btzg 6 (1887) 123; O. K. Rev. Gen. Pl. 1 (1891) 645; Harms in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 32; Bot. Jahrb. 56 (1920) 384; Hutcu. Gen. FI. Pl. 2 (1967) 73; PHILIPSON, Blumea 23 (1976) 99. — Eschweileria Zipp. ex BOERL. Ann. Jard. Bot. Btzg 6 (1887) 112, non Eschweilera Mart. 1828; Handl. 1 (1890) 640. — Pseudosandalum O. K. Rev. Gen. Pl. 1 (1891) 271 (‘Pseudosantalum’), nom. illeg. — Boerlagiodendron HARMS in E. & P. Nat. Pfi. Fam. 3, 8 (1894) 31; in K. Sch. & Laut. Fl. Schutzgeb. (1900) 484; Bot. Jahrb. 56 (1920) 377; Hutcu. Gen. FI. Pl. 2 (1967) 72. — Fig. 11, 13-16. Unarmed, glabrous or tomentose shrubs or trees. Leaves palmately lobed or simple, rarely digitately compound; stipules forming a ligule, and the base of the petiole furnished with one to several spiral or transversal crests or collars (very rarely absent). Inflorescence a terminal compound umbel; peduncle short; primary rays each terminating into three branches; the central branch bearing a head or umbellule of almost always sterile bacciform flowers (‘pseudo-fruits’); the two lateral branches each bearing a head or umbellule of hermaphrodite flowers. Calyx an obsolete rim or 0. Corolla with few to many lobes above, tubular below. Stamens 4-30, filaments thick, anthers oblong, exserted. Ovary inferior, not articulated with the pedicel, cells 1-many; disk flat with a central raised boss bearing the pustulate stigmas. Fruit subglobose (ribbed when dry); exocarp fleshy, endocarp crustaceous. Seeds compressed, endosperm smooth or wrinkled. Distr. About 50 spp., of which 40 occur in Malesia, extending from Borneo and the Philippines east- wards through Celebes and Moluccas to New Guinea, the remainder lying further north and east in Taiwan, Micronesia, Melanesia and the New Hebrides. Fig. 12. a2 FLORA MALESIANA [ser. I, vol. 9! Ecol. Mainly understorey trees in primary rain-forest, also in second growth forest, usually at low altitudes, especially in shaded situations and near rivers, 15. O. borneense a characteristic rheophyte. Notes. The foliage, inflorescence and flowers of this genus are all unique within the family. The base of the petiole often bears a spiral crest, or this may form a simple collar; the blade varies from simple to elaborately compound; the inflorescence is composed of trifid rays, the central branch bearing sterile bacciform flowers; the corolla is tubular. The central bacciform flowers (pseudo-fruits) are sterile, except in 12. O. yatesii, in which apparently mature seed was once found. By exception a specimen of 40. O. /uzoniense had apparently fertile flowers on the central branches of the inflorescence. KEY TO THE SPECIES 1. Leaves simple, without lobes. Dweetioleybase without crestsi(dNewaGuinea)) sas. carsueeneee ec ese ee 1. O. miquelii 2. Petiole base with 1 or more crests. Ss Petioleibaseawithiseveralcrests (Philippines)ia = —p ence eee Geen 8. O. oblongifolium 3. Petiole base with a single collar-like crest. 4. Fertile flowers pedicelled (Philippines). 5. Leaf obovate to oblanceolate, petiole 3cmorless........... . 2. O. dinagatense SS elWeatdelliptics petiole icmiormmoressaewacs 200s ees cl cence ee ree 3. O. simplicifolium 4. Fertile flowers sessile (or subsessile). 6s Weafibroadly obovate (Solomons) masseuse) | ee ae ete 4. O. spathipedunculatum 6. Leaf lanceolate or narrowly obovate. deePseudo-truits;pedicelled’ (Moluccas) Ream .ct suche cant eee ee 5. O. articulatum 7. Pseudo-fruits sessile. $=inflorescenceses60:cmyG (Moluccas), 20 net se eee een een 6. O. umbelliferum 82 Inflorescencee: 30%em) @ (New ireland) ag. e405 eee cee eee nee 7. O. lanceolatum 1. Leaves lobed or digitately compound. 9. Leaves digitately compound. 10 Ovary 10=16-celled\((New Guinea) = 0. 4 616 see ee ee 13. O. geelvinkianum 10. Ovary 4—5-celled. 11. Leaflets lobed (Philippines). 12. Primary rays of intiorescence'c; 1Ojor fewer... ... se) ee ee 8. O. oblongifolium 40. Leaf palmately lobed. 41. Petiolar crests long-pectinate (Celebes) . . ......-..... 36. O. celebicum 41. Petiolar crests fimbriate, entire, or undulate. 42. Articulation of lateral branches of umbels close to the base (Talaud Is.) 37. O. talaudense 42. Articulation of lateral branches of umbels near the middle. 43. Inflorescence over 20 cm @ (New Guinea) ......... 31. O. sessiliflorum 43. Inflorescence under 20 cm @ (Philippines). 44. Leaf usually 3-lobed. Inflorescence rays delicate, indistinctly setose to glabrous 39. O. trilobatum 44. Leaf usually 5—7-lobed. Inflorescence rays sturdy, markedly setose 40. O. luzoniense 1. Osmoxylon miquelii BorrLt. Ann. Jard. Bot. Sparsely branched tree, 15 m. Leaves glabrous, Btzg 6 (1887) 125, t. 16; Harms, Bot. Jahrb. 56 simple, subcoriaceous; stipules small forming a (1920) 384; Puitipson, Blumea 23 (1976) 103. — __ bicuspid ligule; petioles long (to 19 cm), swollen O. amboinense Mia. Ann. Mus. Bot. Lugd.-Bat. 1 distally; blade oblong-elliptic, 22-36 by 9-12 cm, (1863) 6, p.p.; Becc. Malesia 1 (1878) 194, p.p.— subrounded at base and apex or mucronulate, Gastonia simplicifolia Zier. ex SEEM. J. Bot. 3 midrib prominent below, secondary veins arched- (1865) 75, nomen in synon.; ex BoERL. Ann. Jard. ascending and uniting, c. 1-2 cm apart, margin Bot. Btzg 6 (1887) 125, nom. inval. in synon. —_ entire to undulate. Umbel terminal, sessile, with Pseudosandalum miquelii (BOERL.) O. K. Rev. Gen. many (28-32) radiating rigid, angular, trifid Pl. 1 (1891) 271. branches c. 7 cm long to first joint. Central 34 FLORA MALESIANA [ser. I, vol. 91 =o * \ x. DS a Ce i =|o pee N aye 1 —— — —— — Fig. 12. Species density of Osmoxylon Mia. in Malesia; above the hyphen the number of endemic species, below it the non-endemics. branches unknown. Lateral branches c. 5 cm long, articulate near the base. Flowers 20—30, sessile on the expanded ends of the inflorescence branches. Corolla and stamens unknown. Drupes crowded, subrotund, c. 4 mm @ (dry), c. 8-10 ribbed when dry, crowned by a semiglobose entire stigma, 8-10-celled. Seeds with slightly ruminate endo- sperm. Distr. Malesia: West New Guinea. Only known from the type (coll. ZIPPELIUs). 2. Osmoxylon dinagatense (MERR.) PHILIPSON, Blumea 23 (1976) 103. — Boerlagiodendron dinagatense MeERR. Philip. J. Sc. 17 (1920) 301; Me_rr. En. Philip. 3 (1923) 222. Glabrous shrub, c. 2 m. Leaves crowded at the ends of the branches; petiole 2-3 cm, channelled above, with a small triangular base, bearing a short stipular ligule (2-3 mm long) and extending around the base of the petiole as a single narrow collar; blade obovate to oblanceolate, to 23 by 8'/, cm; base narrowed into the petiole, apex rounded with a very short apiculum, margin slightly revolute, entire to obscurely undulate with minute teeth, coriaceous. Inflorescence a terminal compound umbel; peduncle c. 1!/, cm, with 1-few bracts (reduced leaf-bases) with small triangular bracts (3 mm long) among the primary rays; primary rays about 15, 8-10 mm long, flattened, bearing opposite bracts (2 mm long) at the apex, each bearing three branches; central branch c. 4 mm long, bearing a head of sessile, bacciform flowers c. 2!/, mm @; lateral branches 1'/,-2 cm long with two opposite small bracts about the middle and ending in an involucre of minute rounded bracts around a terminal umbellule of c. 7-10 flowers; pedicels 11/,-2'/, mm long. Calyx a minute rim. Corolla and stamens unknown. Ovary 3-4-celled. Ripe fruit unknown. Distr. Malesia: Philippines (Dinagat I.). Note. A species clearly demonstrating the con- generity of Osmoxylon and Boerlagiodendron. 3. Osmoxylon simplicifolium (ELMER) PHILIPSON, Blumea 23 (1976) 103. — Boerlagiodendron simplicifolia ELMER, Leafi. Philip. Bot. 7 (1914) 2329; MERR. En. Philip. 3 (1923) 224. Glabrous shrub, to 5 m, with numerous /eaves clustered near the ends of the branches; petiole 5-7 cm, 2 mm wide, terete, with a small clasping base, an inconspicuous stipular ligule, and a single broad disk-like crest around the lower part of the petiole; blade simple, elliptic, base broadly cuneate, apex acute to apiculate, to 20 by 6!/, cm, coriaceous, margin thickened, coarsely dentate, midrib prominent, principal nerves c. 8-10 mm apart. Inflorescence a terminal compound umbel, spherical, c. 7 cm @; peduncle 11/,-2 cm; primary rays c. 25-30, c. 1 cm long with two small obtuse bracts at the apex, ending in three branches; the central branch c. 6 mm long, bearing a subglobose umbel of c. 10-12 sterile bacciform flowers (2 mm @), 3-4 mm pedicelled; lateral branches c. 2—2!/,cm long, articulated about the middle, terminating in an umbel of c. 10-20 flowers; pedicels c. 2!/, mm. Calyx rim obsolete. Corolla 3—4-lobed, tubular below, 2 mm long. Stamens 3-4, exserted, 3 mm long. Ovary subcylindric, 24-celled, 1 mm long. Drupe spherical, c. 5mm @ (dry), 2-4-ribbed when dry; surface of endosperm shallowly wrinkled. Distr. Malesia: Philippines (Mindanao: Agusan Prov., Cabadbaran). Ecol. On wind-swept ridge at 1750 m, on moss- covered soil with stones. Vern. Bolauanon, Mbo. Note. A wide-spreading shrub. Bark thick, yellowish, becoming grey. Wood soft, yellowish. Twigs repeatedly branched, the leafy portion sub- erect, leaves mostly ascending, rigidly coriaceous. Inflorescence branches green. Flowers orange, odourless. Berries becoming purple-black. 4. Osmoxylon spathipedunculatum (PHILIPSON) PHILIPSON, Blumea 23 (1976) 103. — Meryta spathipedunculata PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 12. Glabrous tree, to 20 m, with spreading branches. Leaves crowded at the ends of the branches; petiole to 14 cm with a small clasping base bearing a stipular ligule and a rim-like collar around the base of the petiole; blade obovate 17-30 by 10-15 cm, attenuate at the base, apex obtuse, margin entire, midrib prominent, secondary veins arched and uniting, 1!/,-2 cm apart. Inflorescence a terminal compound umbel; primary rays c. 12, stout, com- pressed 9-17 cm long, bearing three branches at the apex; central branches and sterile flowers unf- known; the two lateral branches 9-14 cm long with an articulation c. 1-2 cm from base, bearing helmet-shaped bracts which fall to reveal the terminal head of c. 12 flowers sessile on an ex- panded receptacle with an involucral rimc.1cm @. Calyx rim 0. Corolla split into 5 lobes above, tubular below. Stamens 5. Ovary subcylindr:c, + 1979] ARALIACEAE—I (Philipson) 3D 9-celled; disk raised in the centre to the pustulate stigmas. Drupes in a spherical head, globose, c. 12 mm @, c. 9-ribbed when dry, crowned by the prominent, persistent, confluent stigmas. Distr. Solomon Islands (Bougainville and Guadalcanal). Ecol. Rain-forest, 800-1200 m. 5. Osmoxylon articulatum PHILIPSON, Blumea 23 (1976) 103. Tree with stout branches, glabrous. Leaves well spaced towards the ends of the branches; petiole 10-13 cm, 2 mm broad, narrowly channelled above, with a small triangular base, bearing a short stipu- lar ligule (c. 2 mm long) and extending around the base of the petiole as a collar; blade obovate, to 27 by 9 cm, base narrowly cuneate, apex rounded or acute and shortly apiculate, margin thickened, remotely dentate towards the apex, midrib promi- nent, principal lateral veins c. 11/,-11/, cm apart. Inflorescence a terminal compound umbel, almost sessile, saucer-shaped bracts caducous; primary rays c. 24, 8-11 cm long, flattened, c. 4 mm broad, bearing three branches at the apex; central branch 4-6'/, cm long, the apex expanded and bearing an umbel of c. 10 sterile bacciform flowers c. 6 by 6 mm (when dry) apparently 1-celled, c. 9-14 mm pedicelled; the two lateral branches c. 7 cm long at anthesis with an articulation c. 8-10 mm above the base, bearing helmet-shaped bracts which fall to reveal the terminal head of c. 15—18 flowers, sessile on an expanded receptacle with an involucral rim c.6mm @. Calyx rim 0. Corolla split into c. 4 lobes above, tubular below, c. 2!/, mm long. Stamens 5, exserted. Ovary subcylindric, c. 11/, mm long, 7-8-celled, disk with a pustulate central stigmatic boss. Drupes in a spherical head c. 2 cm @ (when dry), strongly 7—8-ribbed (when dry), c. 8 mm @; stigmas persistent, prominent. Distr. Malesia: Moluccas (Halmaheira: Ake Mumar to upper reaches of the Kakatua-matawe). Fr. Sept. Note. This plant is evidently similar to 6. O. umbelliferum described by RUMPHIUS. However, a number of differences make it unlikely that it is the same species. The diameter of the inflorescence of the Halmaheira plant is only about half that given by RuMPHIUs; the lateral rays of the inflores- cence are distinctly articulated near the base, a feature now shown in Rumpuius’ figure; and the sterile bacciform flowers are long-pedicelled, where- as RuMPHIus described and figured his as borne in capitula. 6. Osmoxylon umbelliferum (LAMK) MEerR. Int. Rumph. (1917) 406; Puitirson, Blumea 23 (1976) 104. — Pseudo-Sandalum amboinense RUMPH. Herb. Amb. 2: 54, t. 12. — Aralia umbellifera LAMK, Encycl. 1 (1783) 225. — Hedera umbelli- ferum (LAMx) DC. Prod. 4 (1830) 262. — Gilibertia saururoides DC. l.c. 256. — Gastonia saururoides Roxs. [Hort. Beng. (1814) 90, nomen;] FI. Ind. ed. Carey 2 (1832) 408 (‘sasuroides’). — O. amboi- nense Miq. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 6, p.p. — Pseudosandalum umbelliferum (LAMK) O. K. Rev. Gen. Pl. 1 (1891) 271. According to RUMPHIUs: Tree with stout trunk, the branches marked with prominent round leaf- scars. Leaves clustered at the ends of the branches, glabrous; petioles long; blades simple, lanceolate (30-36 by 10-12 cm), base rounded, apex acute, margin dentate. Flowers in large spreading umbels, the radiating branches tripartite, c. 30 cm long, each ending in a capitulum. Distr. Malesia: Moluccas. Infrequent on Am- bon, but said to be more numerous in Ceram and the Sula Islands. Only known from Rumputus’ excellent plate and description; not yet re- collected in Ambon. Ecol. Evidently in forest in the hills of Ambon, and also planted at the time of RUMPHIUS. Vern. Sasuru, Leytimor, tonokuku, Hitu. Notes. Valued for the perfume of its wood and foliage. Since this plant is known only from a description and a figure, some uncertainties remain as to its specific characters. In the description it is stated that the young leaves possess a few small teeth of which some signs remain on the older leaves. It is not clear whether the teeth are best developed on distinctive juvenile foliage, or whether the newly expanded normal foliage is intended. The leaves in the figure have prominent teeth, but as they are not shown associated with the inflorescence, they may be from a juvenile shoot. The description of the size of the flower buds is confusing, and it seems likely that sterile pseudo-fruits were mistaken for flower buds. Nevertheless, most characters of the plant are adequately portrayed and there can be no doubt that this species is distinct from the other simple-leaved species with a single, collar-like, petiolar crest. 7. Osmoxylon lanceolatum PHILIPSON, Blumea 23 (1976) 104. — Fig. 13. Small tree with few branches, up to 16 m, glabrous. Many /eaves clustered towards the ends of the branches; petiole 8-15 cm, terete, with a small triangular base, bearing a short stipular ligule (c. 2mm long) and extending around the base of the petiole as a collar; blade oblanceolate, to 33 by 7'/, cm, base narrowly cuneate, apex acute or slightly apiculate, margin entire, midrib promi- nent, lateral veins arched ascending, c. 2-3 cm apart. Inflorescence a terminal compound umbel, almost sessile, saucer-shaped; bracts caducous; primary rays c. 15, c. 10 cm long, flattened, c. 4-5 mm broad, bearing three branches at the apex; central branch 5-6 cm long, the apex expanded and bearing a spherical head of c. 8-12 sessile, sterile, bacciform flowers c. 5 by 5 mm (when dry), 1-2-celled; the two lateral branches 36 FLORA MALESIANA [ser. I, vol. 92 Fig. 13. Osmoxylon lanceolatum PutILipson. Above: habit of inflorescence and leaves; below, left: twig showing collar-like crests at the base of the petioles; below, right: the trifid branches of the inflorescence (New Ireland, SANDs 795). 1979] ARALIACEAE—I (Philipson) a7 c. 5 cm long at anthesis, with an articulation immediately above the base, bearing two helmet- shaped bracts which fall to reveal the terminal head of c. 8-10 flowers, sessile on an expanded receptacle with an involucral rim c. 7 mm @. Calyx rim 0. Corolla known only in bud, c. 2!/, mm long. Stamens c. 5. Ovary gibbous, c. 1'/, mm high, 4-celled. Fruit unknown. Distr. Malesia: New Ireland (Namatanai Subdistr., Danfu R. area, inland from Manga). Ecol. Understorey tree in ridge top forest on limestone, 750-850 m. Note. The bark is pale grey, + smooth with fine cracks. The twigs and cut branches are strongly aromatic. The wood is soft and dark straw- coloured. The central branches of the inflorescence rays are held + horizontally or depressed and come to maturity before the lateral branches which are held erect. 8. Osmoxylon oblongifolium PHILipson, Blumea 23 (1976) 105. Shrub c. 2 m, glabrous when mature, setulose on young parts. Leaves clustered at the ends of the branches; petiole to 16 cm, channelled above, 4 mm broad, with a clasping base prolonged upwards as a stipular ligule 11/,-2 cm long, and with entire or fimbriate crests encircling the lower part of the petiole; blade simple, oblong-ovate, occasionally with a small triangular lobe on each side below the middle, to 46 by 17(-24) cm, base rounded to truncate, apex shortly acuminate, margin serrate, midrib prominent, principal lateral veins c. 3-4 cm apart (at broadest part of leaf). Inflorescence a terminal compound umbel, hemispherical, c. 13 cm @; peduncle 3-4 cm, c. 6 mm wide, bearing fimbriate, lanceolate bracts (to 2 cm long) along its length and around and among the primary rays; primary rays c. 20, rather short and stout (16-20 by 3-4 mm) with small opposite caducous bracts at apex, each ray ending in three branches; the central branch c. 4 mm long; sterile flowers un- known; the two lateral branches 31/,-4 cm long, articulated below the middle, terminating in a head of c. 20-30 sessile or subsessile flowers. Flowers unknown. Fruits crowded in a spherical head, drupes c. 5mm @ (when dry) on pedicels c. 1 mm, 4-ribbed, pyrenes 4, cartilaginous; endosperm with faint reticulate ridging. Distr. Malesia; Philippines (Samar). Ecol. In dipterocarp forest, along creek bank, at 200 m. 9. Osmoxylon catanduanense (MERR.) PHILIPSON, Blumea 23 (1976) 105. — Boerlagiodendron catanduanense MERR. Philip. J. Sc. 13 (1918) Bot. 318; En. Philip. 3 (1923) 222. Shrub c. | m, glabrous except for parts of the inflorescence. Leaves clustered at the ends of the branches; petiole c. 30 cm, terete, 3 mm wide, base with a short ligule, and inconspicuous recurved crests; blade digitately compound, leaflets 7, chartaceous to subcoriaceous; petiolules 4-5 cm, the lateral shorter; leaflets lanceolate, mid-leaflet to 26 cm long, base cuneate, apex + caudate; irregularly lyrately lobed, the sinuses reaching to within c. 8 mm of the midrib, lobes 2-4 cm long, patent, margin slightly thickened, entire or obscurely dentate. Inflorescence a terminal compound umbel, c. 10 cm @, peduncle stout, bearing ovate bracts c. 1 mm long; primary rays c. 7-10, 2-21/, cm long, to 2 mm wide, minutely pubescent, with 2 broadly ovate bracts 4-5 mm long at apex; central branch 3 mm or less, bearing a head of sterile flowers; lateral branches 2!/, cm long, bearing 2 broad bracts near the middle, and ending in a spherical head of c. 15 sessile flowers. Calyx an obsolete rim. Petals and stamens un- known. Ovary 4-celled. Fruit ovoid, 4-ridged when dry, 7 by 5 mm. Distr. Malesia: Philippines (Catanduanes). Ecol. On forested slopes, at c. 350 m. Note. Together with 10. O. caudatum and 11. O. heterophyllum this species forms a small group with leaves composed of lyrate leaflets. Although the foliage is similar (though not identical), the inflorescences are distinctive. In O. heterophyllum the rays are short, resulting in a compact compound umbel; in O. caudatum there are few rays (10 or fewer); in O. catanduanense there are many, relatively long rays, resulting in a large, diffuse compound umbel. Since O. catanduanense and O. caudatum are known only from the type collec- tions, the range of variation of these species is not known, but the inflorescence differences justify the retention of all three species. 10. Osmoxylon caudatum (MERR.) PHILIPSON, Blumea 23 (1976) 105. — Boerlagiodendron caudatum MeErR. Philip. J. Sc. 14 (1919) 440; En. Philip. 3 (1923) 222. Erect shrub, c. 2 m, becoming glabrous. Leaves clustered towards the ends of the branches; petiole c. 45 cm, terete, striate, 4-5 mm wide, base with a stipular ligule c. 11/, cm long, and 2-3 fimbriate or pectinate crests; blade digitately compound, leaflets S—9, subcoriaceous; petiolules of the central leaflets to 5 cm long, lateral leaflets + sessile; leaflets elliptic-lanceolate, mid-leaflet to c. 25 cm long, base decurrent on the slightly winged petio- lule, apex caudate-acuminate, the larger leaflets with 1-3 pairs of lyrate lobes reaching almost to the midrib, lobes ascending, margin slightly revolute with often prominent usually incurved teeth. Inflorescence a terminal compound umbel, c. 18 cm @; peduncle stout, c. 2 cm, bearing few ligulate bracts c. 11/,-2 cm long, rough, with short setae on the back; primary rays c. 25, c. 4-5 cm long, glabrous and striate, subtended by ligulate bracts, similar opposite bracts at the apex, c. 1 cm long; central branch 1-1'/, cm long, glabrous, ending in a whorl of obtuse rough coriaceous bracts 38 FLORA MALESIANA [ser. I, vol. 91 c. 3 mm long and a head (c. 1 cm @) of sterile bacciform flowers c. 3 mm @, on pedicels 3-8 mm; lateral branches 4-5 cm long, with opposite fimbriate bracts near the middle, bearing an ellipsoid head of c. 20 sessile flowers, each flower subtended and + enclosed by an ovate fimbriate bract 3-4 mm long. Calyx an obsolete rim. Petals and stamens unknown. Ovary 2-2!/, mm long, turbinate, obscurely angled, 4-celled. Fruit unknown. Distr. Malesia: Philippines (Luzon: Ilocos Norte Prov., Mt Palimlin). Ecol. On forested slopes near the summit, at c. 1000 m. Note. Apparently never re-collected. For a dis- cussion of distinctive features see under 9. O. catanduanense. 11. Osmoxylon heterophyllum (MERR.) PHILIPSON, Blumea 23 (1976) 106. — Boerlagiodendron heterophyllum Merr. Philip. J. Sc. 9 (1914) Bot. 329; En. Philip. 3 (1923) 223. Erect tree, c. 5 m, glabrous. Leaves clustered at the ends of the branches; petiole to 25 cm, base clasping, prolonged as a stipular ligule to 21/, cm long, and bearing several pectinate crests (becom- ing recurved when the bristles may be obscured or shed); blade to 35 cm long, digitately com- pound (or some leaflets imperfectly separated); leaflets 3-7, unequal in size, oblong-ovate to broadly oblong-oblanceolate, acuminate, lyrately lobed and irregularly dentate, the base gradually narrowed to the petiolule; petiolule up to 7 cm. Inflorescence a dense terminal compound umbel; peduncle stout, c. 2 cm, bearing many lanceolate bracts 2-3 cm long; primary rays c. 30, 10-15 mm long, subtended by lanceolate c. 2!/, cm long bracts, sometimes with bristles on the back, and bearing similar opposite terminal bracts 8-10 mm long, each ending in three branches; central branch short (not seen fully developed) terminating in an umbellule of c. 15 sterile bacciform flowers (c. 3 mm long), pedicels 3-4 mm; lateral branches 8-10 mm long (? fully developed), articulation present ending in heads (c. 1 cm @) of numerous flowers. Calyx rim obsolete. Corolla lobes 4, 2 mm long. Stamens 4. Ovary 4-celled. Fruit unknown. Distr. Malesia: Philippines (Samar, Biliran and Mindanao). Ecol. Primary forest, under shade near creek, 100-550 m. Vern. Arafas, Bis., kayuang, Mbo, magusayag, C.Bis. Note. The description is partially based on the original publication as I have seen only immature inflorescences. For a discussion of distinctive features, see under 9. O. catanduanense. 12. Osmoxylon yatesii (MERR.) PHILIPSON, Blumea 23 (1976) 106. — Boerlagiodendron yatesii MERR. Philip. J. Sc. 13 (1918) Bot. 44; En. Philip. 3 (1923) 225% Shrub, 1 m, glabrous, except for the inflores- cence. Leaves clustered towards the ends of the stout branches; petiole to 38 cm, channelled above, 5 mm @, base with a stipular ligule 1 cm long, and 1-2 inconspicuous non-fimbriate crests around the back of the petiole; blade digitately compound, leaflets 5-7; petiolule 2-7 cm (the lateral shorter); blade elliptic-oblong to ovate, mid-leaflet to 23 by 8 cm, base gradually tapered, apex acuminate- caudate, margin dentate or somewhat undulate. Inflorescence a terminal compound umbel, sub- sessile or peduncle stout 1-3 cm, bearing few ovate bracts 1 cm long; primary rays 5-10, tomentose, c. 3 cm long and 3 mm wide, subtended by ovate bracts 6 mm long, similar opposite bracts at apex; central branch 2-3 mm long, pubescent, ending in a whorl of blunt bracts (3 mm long) and an umbellule of c. 10 sterile flowers 4 mm @, 2-4- celled, pedicels 3-8 mm; lateral branches 2—21/, cm long, pubescent, with small opposite bracts about the middle, bearing a head of 10-15 sessile flowers, subtended by ovate ciliate bracts. Calyx rim obsolete, sometimes fimbriate. Corolla 4—5- lobed, tubular below, 11/, mm long in bud. Stamens 4-5. Ovary 4-celled. Young fruit (MERRILL, /.c.) shortly pedicelled (2-3 mm). Distr. Malesia: Philippines (Luzon and Catan- duanes). Ecol. In rain-forest and mossy forest, from low altitude to 1250 m. Vern. Magalayag, Dinagat. Note. The leaves are unlike any other Osmoxy- lon, resembling those of Macropanax or Schefflera. The flowers are described as yellow, and the fruit black. The only instance of a pseudo-fruit con- taining apparently normal seeds occurred in this species. 13. Osmoxylon geelvinkianum BeEcc. Malesia 1 (1878) 196; PHILipson, Blumea 23 (1976) 106. — Eschweileria geelvinkiana (BECC.) BOERL. Ann. Jard. Bot. Btzg 6 (1886) 120. — Trevesia geelvin- kiana (BEcc.) O. K. Rev. Gen. PI. 1 (1891) 272. — Boerlagiodendron geelvinkianum (BECC.) HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 32. — Eschweiler(i)a elegans RIDL. Trans. Linn. Soc. II, 9 (1916) 63. — Boerlagiodendron elegans (RIDL.) Hars, Bot. Jahrb. 56 (1920) 380. — Boerlagioden- dron stenolobum HARMS, /.c. 382, f. 1 kt. Glabrous shrub with few slender branches, up to 3 m high. Leaves clustered near the ends of the branches; petiole up to 22 cm, usually shorter, 2-3!/, mm wide, narrowly channelled above, with a sheathing base prolonged as a membranous stipular ligule up to 4 cm or longer and with fimbriate or + entire crests encircling the lower part of the petiole; blade up to 30 cm @ (usually 20 cm or less) very deeply 5—11-lobed, or with distinct digitately arranged leaflets, the lobes or 1979) ARALIACEAE—I (Philipson) 39 leaflets linear-lanceolate to lanceolate-obovate, entire or irregularly pinnatifid with narrow finely- tapering lobes, base gradually narrowed, apex narrowly caudate, margin serrate; leaves below the inflorescence sometimes reduced to a single leaflet. Inflorescence terminal hemispherical, c. 12-20 cm @; peduncle 1 cm or less, with caducous lanceolate bracts mostly clustered below the primary rays, 1-2 cm long; primary rays rather few, spaced, 2'/.-6 cm long, slender, bearing two caducous lanceolate bracts at the apex, up to 1'/, cm long, each ray ending in three branches; the central branch c. 4-6 mm long bearing a whorl of lanceo- late caducous bracts and an umbel of c. 7-12 sterile, globose or ovoid bacciform flowers (c. 7 mm @ when dry) with pedicels c. 5 mm long and 6-9- celled; the two lateral branches c. 3-4 cm long at anthesis, articulated about the middle, terminating in a small head of 10-20 sessile or subsessile flowers. Calyx rim obsolete; corolla splitting into c. 4 irregular lobes above, tubular below, c. 21/, mm long. Stamens 10-14, exserted, 3 mm long, anthers small. Ovary cylindric, c. 2 mm long, 10—16-celled; disk with a central raised boss formed by the pustulate stigmas. Fruit globose, fleshy (ribbed when dry), c. 10 mm @. Distr. Malesia: New Guinea (Irian Jaya, to Sepik and Fly R. areas). Ecol. Primary forest, along creeks and river banks, flood-resistant, from near sea-level to 850 m. Vern. Amamutapu, Kamora, korinki, Orne, ida’ pforpforsami, Kutubu. Note. The narrow leaf segments, almost or quite separated at their bases, are characteristic, even though variable in outline. The flowers are described as orange or reddish, and the soft fleshy fruits as dark purple, dark blue, or black. 14. Osmoxylon lineare (MERR.) PHILIPSON, Blumea 23 (1976) 106. — Boerlagiodendron lineare MERR. Philip. J. Sc. 3 (1908) Bot. 253; En. Philip. 3 (1923) 2233 Glabrous, erect shrub, c. 3 m. Leaves crowded near the ends of the branches; petiole 4-6 cm, the base with a small stipular ligule (5 mm long) and a few fimbriate crests; blade to 20 cm @, digitately compound with 4-7 leaflets (or very deeply divided into as many lobes); leaflets linear- lanceolate, c. 1—1!/, cm wide, the base decurrent on the winged petiolule, apex attenuated, margin thickened, denticulate especially above. Jnflores- cence a terminal compound umbel; peduncle short (c. 1 cm), bracteate; primary rays c. 10, 2-23/, cm long, each ending in three branches; the central branch 4-5 mm long bearing a spherical head of numerous fimbriate bracts (sterile flowers fallen); the lateral branches 3-3!/, cm long with a pair of minute fimbriate bracts about the middle, ending in a capitulum of c. 20 sessile flowers subtended by small fimbriate bracts, c. 7 mm @. Calyx rim minute. Corolla and stamens unknown. Ovary 5-celled. Fruit ovoid c. 3 mm long, 5-ridged (when dry). Distr. Malesia: Philippines (Luzon). Note. Apparently never re-collected. The original specimen is without field information. The foliage is similar to that of 15. O. horneense but with shorter petioles and fewer leaflets. 15. Osmoxylon borneense SEEM. J. Bot. 6 (1868) 141; PuiLirson, Blumea 23 (1976) 107. — O. helleborinum Becc. Malesia 1 (1878) 198. — Eschweileria helleborina (BECC.) BOERL. Ann. Jard. Bot. Btzg 6 (1887) t. 13. — Trevesia helleborina (BECC) 2O Ke Rev Gen! Ply 1.113891)? 272. — Boerlagiodendron helleborinum (BEcc.) HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 31. — Boerlagio- dendron borneense (SEEM.) MERR. En. Born. (1921) 456. Glabrous, spreading shrub, up to 3 m. Leaves clustered near the ends of the branches; petiole up to 24 cm, narrowly channelled above, 2-3 (or 5) mm wide, with a sheathing base prolonged as a mem- branous stipular ligule up to 2!/, cm long, and usually with fimbriate, + entire, or more rarely long-setose crests encircling the lower part of the petiole; blade up to 20 cm @ digitately compound (or the bases of the leaflets joined by a very short web of tissue); leaflets 9-13, linear-lanceolate to lanceolate, gradually narrowed to the base and apex, up to 20 by 3 cm, usually much narrower, margin serrate, principle veins numerous, c. 5— 10 mm apart. Inflorescence terminal, hemispherical, c. 6-13 cm @; peduncle 1-2 cm or shorter, with caducous lanceolate-ovate entire or fimbriate bracts mostly clustered below the primary rays, 1-11/, cm long; primary rays rather few (5-12), 2'/.-4 cm long, with 2 ovate bracts at the apex, c. 8 mm long, each ray ending in three branches; central branch c. 5-12 mm, bearing an umbel of c. 6-16 sterile, globose, bacciform flowers, c. 3— 5 mm @ (when dry) with pedicels 4-5 mm long and 5-celled; the two lateral branches 1'/,—3 cm long at anthesis, articulated about the middle, terminating in a small head of c. 20-25 sessile flowers with minute rounded bracts. Calyx rim obsolete; corolla splitting into few irregular lobes above, tubular below, c. 11/,-2 mm long; stamens 5-6, slightly exserted; ovary turbinate, angled, c. 1 mm long, 5-celled, disk with a central raised boss formed by the pustulate stigmas. Fruit globose, fleshy (ribbed when dry) c.5 mm @. Distr. Malesia: Borneo (Sarawak, Sabah and Kalimantan). Ecol. Characteristic of rocky river banks, not beyond flood-level, often in deep shade, from near sea-level to 950 m. Vern. Medong, Kayan, empasia abor, Iban, kayan, Tamang, koung, Kinabalu, bungor, Murut Bokan, salimpangaya, Murut Kalabakai. Notes. The leaves of some specimens of 13. O. geelvinkianum (New Guinea) resemble this 40 FLORA MALESIANA [ser. I, vol. 91 species closely. This species can be distinguished by the more compact and smaller fertile flowers, and by its leaflets being uniformly unlobed. Both species are characteristic of river banks, and O. borneense has a low spreading habit, with the branches often rooting, resulting in extensive patches of this low shrub. It is a characteristic rheophyte confined to below flood-level. The flowers are described as greenish white or cream and the inflorescence branches are frequently dark purple. 16. Osmoxylon pulcherrimum VIDAL ex F.-VILL. Novy. App. (1880) 102; Sinopsis Atlas (1883) 28, t. 55, f. B; PHmirson, Blumea 23 (1976) 107. — Eschweileria pulcherrima (VIDAL) BOERL. Ann. Jard. Bot. Btzg 6 (1887) 123. — Trevesia pulcher- rima (VIDAL) O. K. Rev. Gen. Pl. 1 (1891) 272. — Boerlagiodendron pulcherrimum (VIDAL) HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 32; MerrR. Philip. J. Sc. 3 (1908) Bot. 254; En. Philip. 3 (1923) 224. — Boerlagiodendron sibuyanense ELMER, Leafi. Philip. Bot. 7 (1914) 2328; Merr. En. Philip. 3 (1923) 224. Erect, sparsely branched tree, up to 10 m, glabrous when mature, except for the inflorescence. Leaves crowded at the ends of the branches; petiole to 1 m, channelled above, clasping base heavily lenticellate, prolonged as-a broad stipular ligule c. 2 cm long, usually with strong bristles on the back, and with strong long-pectinate crests encircling the base of the petiole; blade coriaceous, fan-shaped, c. 40 cm long, base broadly cuneate to truncate, palmately 7-11-lobed, lobes extending to within c. 12 cm from the base, lanceolate, coarsely serrate, sometimes irregularly lobulate, slightly narrowed towards the base, apex acute, sinuses rounded. Inflorescence a terminal compound umbel, c. 18 cm @; peduncle very short, bearing heavily setose bracts; primary rays 15-20, c. 34cm long, 3-4 mm broad, setulose, at the apex bearing opposite, ovate-lanceolate bracts 10-15 mm long, each ending in three branches; central branch c. 15-20 mm long, terminating in a globular head (c. 12 mm @) of c. 20-30, sessile, sterile, bacciform flowers (3-4 mm @) 3-celled, subtended by small ovate-lanceolate bracts; lateral branches c. 5!/, cm long (7'/, cm in fruit), with opposite bracts (c. 3-4 mm long) near the middle, terminating in a globose head of c. 40-50 sessile flowers, bracts between the flowers very small, setulose. Calyx rim obsolete. Corolla 4-lobed, tubular below, 2 mm long in bud. Stamens 4. Ovary 4-celled. Fruit globose c. 6-8 mm long, 4-ribbed (dry). Distr. Malesia: Philippines (Luzon, Mindoro and Sibuyan), recorded also from Formosa and Micronesia (Palau), cf. KANEHIRA, En. Micron. Pl. (1935) 384. Ecol. Damp primary forests, 225-800 m. Vern. Cf. MERRILL: paladukai, Bik., salapak, Neg.; cf. ELMER: palad-amok, Vis. Notes. The fan-shaped leaves with several nar- tow lobes and prominent main veins resemble those of 34. O. eminens but are less strikingly developed. The inflorescence is considerably smaller with the pseudo-fruits forming a compact head borne on a comparatively long peduncle. The heads of true flowers, and of the fruits, are considerably smaller than those of O. eminens. Although VIDAL’s material is no longer avail- able, the figure and description relate well to later collections. The specimens on which ELMER based his Boerlagiodendron sibuyanense have the lobes of the leaf rather simpler in outline than is usual, but the fragments of young inflorescence are quite typical of the taxon and the name is reduced to synonymy. 17. Osmoxylon soelaense PHILIPSON, Blumea 23 (1976) 108. A glabrous shrub or small tree. Petiole c. 32 cm, broadly channelled above, clasping base prolonged as a stipular ligule 11/, cm long, and bearing c. 3 fimbriate crests; blade c. 40 cm long, broadly cuneate at the base, palmately 7-lobed to within c. 12 cm from the base, lobes narrowly ovate to oblong-elliptic, slightly narrowed towards the rounded sinuses, apiculate, margin denticulate. Inflorescence a terminal compound umbel; peduncle c. 2 cm, bearing lanceolate bracts c. 2 cm long (similar smaller bracts subtend the primary rays); primary rays c. 1'/, cm long, 5 mm wide, flattened, bearing opposite, terminal, persistent bracts 10-12 mm long, with lenticels and branched bristles on the back, ending in three branches; central branch c. 10 mm long terminating in a head 1 cm @ of 10-15 sessile sterile bacciform flowers (4 by 3 mm when dry; 3-celled) surrounded by an involucre of ovate bracts (3 mm long) and with minute bracts interspersed; lateral branches 4'/,-5 cm long, bearing opposite ovate bracts (4 mm long) c. 8 mm above the base, terminating in a dense head c. 1 cm @ of 25-30 sessile flowers interspersed with inconspicuous obtuse fimbriate bracts. Calyx rim obsolete. Corolla 5-lobed above, tubular below. Stamens 5. Ovary 4~5-celled. Fruit unknown. Distr. Malesia: Moluccas (Sula Is.: Taliabu and Sulabesi). Note. Fora discussion of the distinctive features, see under 31. O. sessiliflorum. 18. Osmoxylon globulare PHiLirson, Blumea 23 (1976) 108. Shrub to 4 m, furfuraceous on the young parts. Petiole to 55 cm long, broadly channelled above, c. 1 cm wide, clasping base heavily lenticellate, prolonged as a stipular ligule 4 cm long sometimes scaly on the back, and bearing numerous irregular undulate crests on the base of the petiole often continued up the petiole, as rough fascicles of bristles as far as the blade; blade 45 cm long, base 1979] ARALIACEAE—I (Philipson) 4] cordate or emarginate, with some bristles under- neath, palmately 7-lobed to within c. 10-15 cm from the base, lobes narrowly ovate to oblong- elliptic, slightly narrowed towards the broadly rounded sinuses, apex acute, margin serrate. Inflorescence a terminal compound umbel, spheri- cal, c. 15 cm @; peduncle stout, 2-3 cm, bracts together with those among the primary rays caducous; primary rays 30-40, rigid only slightly flattened (subterete), 2-4 cm long, 2—2'/, mm wide, bearing opposite bract-scars at the apex, ending in three branches; central branch 8-10 mm long, terminating in a head c. 13 mm @ of c. 20 sub- sessile sterile bacciform flowers (5 by 4 mm when dry, 2-3-celled); pedicels to 2 mm interspersed with minute bracts; lateral branches 3-4'/, cm long, with the scars of opposite bracts near the middle, terminating in a dense head 1-11/, cm @ (in bud) of 20-30 sessile flowers interspersed with inconspicuous bracts. Calyx rim obsolete. Corolla 6-8-lobed above, tubular below, 2'/, mm long (in bud). Stamens 6-8. Ovary turbinate, obscurely ribbed, 5—8-celled (varying on the same plant). Fruit unknown. Distr. Malesia: Moluccas (Halmaheira, Moro- tai). Ecol. In forest from sea-level to 800 m. Said to be rare in Halmaheira but common in Morotai. Vern. Bungan-gutu, saha-sasate, Djailolo. Note. For a discussion of the distinctive features see under 31. O. sessilifiorum. 19. Osmoxylon kostermansii PHILIPSON, Blumea 23 (1976) 108. Glabrous, small tree, 8 m. Leaves clustered near the ends of the branches; petiole up to 35 cm, narrowly channelled above, c. 3 mm broad, witha sheathing base prolonged as a stipular ligule c.2 cm long, continued around the back of the leaf-base as a single wide crest with an entire recurved margin; blade up to 30 cm @, base cordate, deeply 5—7-lobed, lobes elliptic, slightly narrowed towards the sinuses and with a short acute apiculum, Margin minutely serrate and sometimes with small sub-lobes, sinuses rounded. Inflorescence a terminal compound umbel, hemispherical, c. 14 cm @ at anthesis; peduncle c. 1 cm long, bearing small lanceolate bracts (c. 3 mm long) below and among the numerous (c. 20-24) primary rays; primary rays 4-5 cm long and 1 mm broad, with opposite bracts (2 mm long) at the apex, each ending in three branches; the central branch 5S—6 mm long, bearing a spherical umbel of c. 20 small, sterile, bacciform flowers (2 mm @) on pedicels c. 5-7 mm long, 2-celled; the two lateral branches c. 31/,4 cm long, with no articulation or bracts except for a minute involucre around the terminal umbellule of c. 10-14 flowers; pedicels c. 2-3 mm. Calyx rim obsolete; corolla 4-lobed, 2 mm long in bud; stamens 4; ovary subcylindric, angled, c. 1 mm long, 5—7-celled, disk with a central stylar boss. Fruit spherical, c. 6 mm long, strongly ribbed when Distr. Malesia: Borneo (Kalimantan: Sang- kulirang Distr., Mt Medadam). Ecol. On limestone at 450 m. Note. The foliage is similar to that of 22. O. pal- matum, except for the distinctive petiolar crest. The inflorescence is also similar to O. palmatum, especially in the lack of an articulation on the rays below the umbellules. 20. Osmoxylon barbatum Becc. Malesia 1 (1878) 197; PHiLtipson, Blumea 23 (1976) 109. — Esch- weileria barbata (BECC.) BoERL. Ann. Jard. Bot. Btzg 6 (1886) 117. —Trevesia barbata (BEcc.) O. K. Rev. Gen. Pl. 1 (1891) 272. — Boerlagio- dendron barbatum (BEcc.) HARMS in E. & P. Nat. Pfi. Fam. 3, 8 (1894) 31. Small, glabrous tree. Leaves clustered near the ends of the branches; petiole to 32 cm, channelled above, 4-5 mm broad, with a sheathing base prolonged as a stipular ligule 2-3 cm long, and with several long setose crests encircling the lower part of the petiole; blade up to 43 cm @, base cordate to truncate; deeply 5-7-lobed; lobes elliptic-lanceolate, narrowed towards the sinuses and tapered to an acuminate apex, margin minutely serrate, sinuses broadly rounded. Inflorescence a terminal compound umbel, hemispherical, to 12 cm @; peduncle 1-2 cm; primary rays c. 20, 3—5 cm long, with three branches at the apex; the central branch c. 1 cm long, bearing a subglobose umbel of c. 15-20, sterile, bacciform flowers (3-4 mm @) on pedicels S—7 mm long, 2-4-celled; the two lateral branches about 4 cm long with no clear articulation but 1 or 2 obsolescent bracts, terminating in an umbellule of c. 15-20 flowers; pedicels c. 34 mm. Calyx rim obsolete; corolla 4-lobed above, tubular below, c. 4 mm long; stamens 7 or more, rarely fewer, exserted; ovary subcylindric c. 3 mm long, angled, 7-celled; disk with a central stigmatic boss. Fruit unknown. Distr. Malesia: SE. Moluccas (Key Is.). Note. Distinguished from the closely related 22. O. palmatum by the long-setose petiolar crests. For discrimination from 21. O. pfeilii see that species. 21. Osmoxylon pfeilii (WARB.) PHILIPSON, Blumea 23 (1976) 109. — Eschweileria pfeilii WARB. Bot. Jahrb. 13 (1891) 396. — Boerlagiodendron pfeilii (Wars.) Harms in E. & P. Nat. Pfil. Fam. 3, 8 (1894) 32. A tree developing a crown when mature, up to 16 m, glabrous when mature, young parts slightly setulose. Leaves in terminal clusters; petiole up to 60 cm, channelled above, c. 6 mm broad, with a sheathing base prolonged as a membranous stipular ligule 24 cm long, and with many pro- nounced, long-setose crests encircling the lower part of the petiole; blade up to 75 cm @, deeply 42 FLORA MALESIANA [ser. I, vol. 91 7-11-lobed, base cordate or emarginate; lobes lanceolate to narrowly elliptic-oblong, narrowed towards the sinuses and tapered to an attenuate apex, margin sharply and remotely serrate, sinuses broadly rounded. Inflorescence a terminal com- pound umbel, hemispherical, to 20 cm @; peduncle short (2-3 cm); primary rays numerous (c. 30-40), 4-5 cm long, glabrous, with obsolete bracts at the apex, each ending in three branches; central branch c. 10 mm long, bearing a subglobose umbel of c. 20, small, sterile, bacciform flowers (1—-11/, mm @) on pedicels c. 3-4 mm long, 2—5-celled; the two lateral branches c. 3-4 cm long, with no clear articulation but 1 or 2 obsolescent bracts, termi- nating in a head of c. 12-16 subsessile flowers (pedicel c. 1 mm, becoming longer in fruit) sur- rounded by an inconspicuous involucral rim. Calyx rim obsolete; corolla 5-lobed, 11/, mm long in bud; stamens 5; ovary subcylindric, angled, 11/, mm long at anthesis, 5—16-celled, disk with a central boss formed by the united pustulate stigmas. Fruit spherical, fleshy, c. 8 mm @, ribbed when dry, the stigmatic boss persistent and prominent; pyrenes cartilaginous. Distr. Malesia: Bismarck Archipelago (New Britain, Duke of York Group and New Ireland). Ecol. Primary rain-forest, from near sea-level to 600 m. Vern. Sare, sasare, sare a lauvolau, New Britain, Pomio; a ibalur, New Ireland. Notes. The bark is grey-brown and pustular, the wood straw-coloured and soft. The flowers are orange, and the ripe fruit dark red-violet. In the original description the ovary is recorded to possess 10-14 cells. However, some other specimens have as few as 5 cells in the ovary, but in other respects agree with specimens with the large number of seeds. Since the inflorescence, leaf-shape, and especially the nature of the petiolar crests, as well as the distribution, are all highly distinctive within the genus, all the specimens can be accepted as examples of ome species with a highly variable number of carpels. This species is very close to 20. O. barbatum of the Key Islands. The original diagnostic character of the number of cells in the ovary has been found to be unreliable. However, since the primary rays in the inflorescence are more numerous and the pedicels of the fertile and sterile flowers are shorter this geographically distinct species is maintained. 22. Osmoxylon palmatum (LAMK) PHILIPSON, comb. nov. — Folium polypi mas (et femina?) RUMPH. Herb. Amb. 4: 101, t. 43. — Aralia palmata LaMK, Encycl. 1 (1783) 224, type, non Lour. 1790, nec R. & S. 1820. — Trevesia moluccana Mia. FI. Ind. Bat. 1, 1 (1856) 748; Bonplandia 4 (1856) 137. — Trevesia zippeliana Mia. Ann. Mus. Bot. Lugd.- Bat. 1 (1863) 11. — Unjala bifida REINW. ex DE VRIESE, Pl. Ind. Or. (1867) 83, nomen in synon.; ex Boert. Ann. Jard. Bot. Btzg 6 (1887) 166, in synon. — O. moluccanum (MiqQ.) Becc. Malesia 1 (1878) 195; PHILIPSON, Blumea 23 (1976) 109. — O. zippelianum (MiQ.) BEcc. Malesia 1 (1878) 195. — Eschweileria palmata Zipp. ex BOERL. Ann. Jard. Bot. Btzg 6 (1887) 116, t. 14. — Boerlagio- dendron palmatum (Zipp. ex BOERL.) HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 31; Merr. Int. Rumph. (1917) 407. — Boerlagiodendron moluc- canum (Miq.) BAKH. f. & Ooststr. in Back. Bekn. Fl. Java (em. ed.) 7 (1948) fam. 159, p. 3; Blumea 6 (1950) 367; BACK. & BAKH. f. Fl. Java 2 (1965) 163. Small, glabrous tree to 15 m. Leaves clustered near the ends of the branches; petiole up to 40 cm, channelled above, 4-5 mm broad, with a sheathing base prolonged as a stipular ligule up to 5 cm long, and with several fimbriate crests encircling the lower part of the petiole; blade up to 45 cm @, base cordate, deeply S5S—9-lobed, lobes elliptic, slightly narrower towards the sinuses, acuminate, margin serrate and sometimes with small sub-lobes, sinuses rounded. Inflorescence a terminal, compound umbel, subspherical, to c. 20 cm @ at anthesis; peduncle c. 1-2 cm, with small caducous bracts below and among the numerous (20-60) primary rays; primary rays 3-5 cm long, 2-3 mm broad, with two opposite caducous bracts at the apex, each ending in three branches; central branch 6-10 mm long, bearing a spherical umbel of c. 20-30 small, sterile, bacciform, flowers (2-4 mm @) on pedicels 4-6 mm long, 4~7-celled; two lateral branches c. 2!/,-5 cm long, rigid and straight, with no articulation (indistinct scars of bracts may occur below the involucre), bearing a minute involucre around the terminal umbellule of c. 12-20 flowers; pedicels to c. 5 mm (occasionally flowers subsessile). Calyx rim obsolete. Corolla irregularly 4-5-lobed above, tubular below, c. 5 mm long. Stamens 6-9, exserted, filaments stout. Ovary subcylindric, angled, c. 2 mm long, 6-9-celled, disk with a raised central stigmatic boss. Fruit globose, fleshy, c. 10 mm @, strongly ribbed when dry. Distr. Malesia: Celebes (once, not localized) and Moluccas (Buru, Ceram, Ambon, Banda, Tenimber Is.). Also cultivated in the Bogor Botanic Garden. Ecol. An understorey tree in primary rain- forest. Uses. The leaves are used for culinary and medi- cinal purposes (against gonorrhoea). Vern. Daun gurita, pelenda darat, saha-saha, Moluccas, fumala-alas, Tenimber Is. Note. The spherical inflorescence is characteris- tic, having straight rigid rays with no articulation on the secondary branches, and the pseudo-fruits are well separated from the true flowers. MIQUEL distinguished Trevesia zippeliana because the collector noted that its ovary was 4-celled. Possibly this number related to the sterile flowers. In two gatherings from Ambon (Waai, TEYSMANN; G. Salhutu, BoERLAGE 179) the fertile flowers are 1979] ARALIACEAE—I (Philipson) 43 subsessile in heads, but otherwise conform to the characters of this species. The only record of this genus from the Tenimber Is. consists of leaves only, but their characters conform to this species. 23. Osmoxylon ramosii (MERR.) PHILIPSON, Blumea 23 (1976) 110. — Boerlagiodendron ramosii MERR. Philip. J. Sc. 11 (1916) Bot. 27; Me_rr. En. Philip. 3 (1923) 224. Erect, unbranched, or sparingly branched, 4 m high shrub, becoming glabrous. Petiole to 40 cm, the clasping base prolonged as a stipular ligule c. 2 cm long, and with few to several prominent recurved, obscurely fimbriate, crests surrounding the lower part of the petiole; blade to 30 cm long, base emarginate, palmately 3-7-lobed, lobes extending to within 3-8 cm from the base, oblong- ovate, margin serrate, sometimes lyrately lobulate, apex acuminate, sinuses broadly rounded. Jnflores- cence a terminal compound umbel, 10-15 cm @; peduncle stout, with lanceolate bristle-bearing bracts; primary rays c. 15, 2-3 cm long, 2-3 mm wide, flattened, subtended by lanceolate bracts c. 2 cm long, with bristles on the back and bearing similar opposite terminal bracts c. 1 cm long, each terminating in three branches; central branch c. 4mm long, slightly pubescent, terminating in an umbellule (c. 11/, cm @) of c. 10-15 sterile, bacci- form flowers 4-5 mm @, 2-celled, pedicels 5-8 mm long, subtended by caducous bracts; lateral branches 2!/, cm long (slightly longer as fruits form), without any articulation or bracts except for a caducous small involucre around the terminal head of c. 25-35 sessile or very short-pedicelled flowers, c. 1 cm 2, bracts among the flowers obscure. Calyx rim obsolete. Corolla 4-lobed above, tubular below. Stamens 4, exserted, filaments broad. Ovary subcylindric, obscurely angled, 4-celled. Fruit spherical 9 mm @, 4-ribbed when dry. Distr. Malesia: Philippines (Luzon). Ecol. On low-lying, wet ground, in forest, or on forested slopes, 700-800 m. Notes. The flat-topped inflorescence is c. 15 cm @ with much-reduced leaves below it. The in- florescence rays are dark purplish and the flowers orange-yellow. The bark is grey and the wood soft. This is the only species in the Philippines without opposite bracts on the lateral branches of the inflorescence rays. In this respect it resembles 22. O. palmatum and a few other species. 24. Osmoxylon novoguineense (SCHEFF.) BECC. Malesia 1 (1878) 197; PHiLipson, Blumea 23 (1976) 110. — Trevesia novo-guineensis SCHEFF. Ann. Jard. Bot. Btzg 1 (1876) 26. — Eschweileria novoguineensis (SCHEFF.) BOERL. ibid. 6 (1886) 118. — Boerlagiodendron novoguineense (SCHEFF.) Harms in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 31; Back. & BAKH. f. Fl. Java 2 (1965) 163. — Boerlagiodendron lauterbachii HARMS in K. Sch. & Laut. Fl. Schutzgeb. (1900) 484. — Fig. 11, 14. Fig. 14. Osmoxylon novoguineense (SCHEFF.) BECC. Petiolar base with ligule and collar-like crests (Photogr. PHILIPSON, Wantoat, 1968). 44 FLORA MALESIANA [ser. I, vol. 9! Tree or shrub, unbranched or sparingly branched, up to 16 m, the young parts rufous- furfuraceous, glabrescent. Large leaves forming terminal crowns; petiole up to 1 m, stout (1-2 cm broad), flattened above, with a sheathing base prolonged as a strong stipular ligule up to 7 cm long, and with fimbriate crests encircling the lower part of the petiole; blade up to 1.20 m @, with 5-7 strong ribs radiating from the top of the petiole, deeply lobed almost to the base of these ribs, the lobes in turn deeply lobed and incised, the central lobes especially being strongly pinnatisect or digitately tripartite, apices acute, margins serrate; upper leaves associated with inflorescences may be smaller, more simply lobed, or entire. Inflorescence terminal, a large compound umbel, bowl-shaped, up to 35 cm @; peduncle up to 10 cm, stout, with lanceolate caducous bracts (c. 4 cm long) below and among the numerous (c. 50-70) primary rays; primary rays c. 12-15 cm long at anthesis, c. 3 mm @, bearing two caducous bracts (1 cm long) at the apex, each ray ending in three branches; central branch c. 2 cm long, bearing an umbel of c. 20-40, sterile, bacciform flowers (c. 6 mm @ when dry) on pedicels c. 10 mm, and 2-6-celled; the two lateral branches c. 4-6 cm, with two opposite or sub- opposite bracts about the middle, terminating in a subspherical umbel 21/,-3 cm @ of 30-50 flowers on pedicels c. 8-10 mm long. Calyx rim obsolete, undulate. Petals with irregular erect lobes, tubular below. Stamens 6-10 exserted. Ovary turbinate somewhat angled; glabrous, 6—-14-celled; disk flat with a central double row of pustulate stigmas. Fruits on stiff radiating pedicels, ovoid or spherical, fleshy, ribbed when dry. Distr. Solomon Is.; in Malesia: throughout New Guinea and in the Bismarck Archipelago. Ecol. Primary and second-growth forest, from sea-level to 1600 m. Vern. Lebe, Mooi, teresakui, Manikiong, akriek, Biak, hoppung, Hottam, uger, Wagu, faliifalii, Tifal, ap gan dandam, aimaini, Mamig, ida’ pfopforsami, Kutubu, pulaka, Gazelle Peninsula. Notes. The foliage is similar to that of 30. O. boerlagei, but the pedicelled flowers of O. novo- guineense distinguish it readily from that species. The ripe fruits are usually ovoid, but in the Solomon Is. they are characteristically spheroidal, and this feature recurs in some specimens from the Bis- marck Archipelago and the adjacent coast of New Guinea. The fawn bark is pustulate with many lenticels. An orange exudate flows from the cut stems. The wood is soft and straw-coloured. The inflorescence branches are dark purple, the corolla and stamens usually deep red, and the ripe fruit shining purple or blue-black. 25. Osmoxylon teysmannii (BOERL.) PHILIPSON, Blumea 23 (1976) 111. — Eschweileria teysmannii Borrt. Ann. Jard. Bot. Btzg 6 (1887) 119. — Trevesia teysmannii (BOERL.) O. K. Rev. Gen. Pl. 1 (1891) 272. — Boerlagiodendron teysmannii (BOERL.) HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 31. A small, glabrous tree, 6 m. Leaves clustered at the ends of the branches; petiole to 40 cm, chan- nelled above, 4 mm broad, with a sheathing base prolonged as a stipular ligule 2-2'/, cm long, and with several fimbriate or entire crests on the lower part of the petiole; blade c. 30 cm @, membranous, cordate at the base, deeply 7-lobed, lobes elliptic, slightly narrowed to the broadly rounded sinuses, narrowed to a fine apiculum at the apex, margin finely serrulate. Inflorescence a terminal compound umbel, c. 10cm @; peduncle c. 1 cm, bearing ovate bracts (ligules of reduced leaves) and terminating in a cluster of bracts (c. 10 mm long) below and among the primary rays; primary rays c. 12-15, c. 3-4 cm long, with a pair of lanceolate bracts at the apex (c. 1 cm long); central branch c. 1 cm long, terminating in an umbellule of c. 5-8 sterile bacci- form flowers (c. 4 mm @ when dry) on pedicels 6-9 mm long interspersed with linear bracts 5 mm long; two lateral branches c. 3!/, cm long, articu- lated about the middle, terminating in an umbellule surrounded by caducous linear bracts (leaving a rim-like involucre); flowers c. 8-12, pedicels c. 2-3 mm. Calyx rim obsolete. Corolla 3 mm long, with 7-8 lobes above, tubular below. Stamens 7-8, filaments stout, anthers exserted. Ovary sub- cylindric, angled, c. 11/, mm long, 7—8-celled; disk flat, with a central stigmatic boss. Fruit unknown. Distr. Malesia: SW. Celebes (Tjamba, Kosali- Porema) and NW. Central Celebes (Palu-Parigi and Mt Nokilalaki). Ecol. In rain-forest, 800-1000 m., 26. Osmoxylon humile (ELMER) PHILIPSON, Blumea 23 (1976) 111. — Boerlagiodendron humilis ELMER, Leafl. Philip. Bot. 7 (1914) 2327; Merr. En. Philip. 3 (1923) 223. Erect, small, sparsely branched shrub, up to 11/, m. Leaves clustered at the ends of the branches; petiole to 25 cm, terete, base prolonged as a stipular ligule c. 1 cm long, and with c. 3 entire crests at the base; blade palmately 5-lobed, 24 cm long, base truncate or cordate, lobes reaching to within 3-6 cm from the base, elliptic, 4-6 cm wide, narrowed towards the broadly rounded sinuses, tapered to an acute apiculum, margin serrate in the upper part, the outer lobes with a lobule on the lower edge. Inflorescence a terminal compound umbel, 9 cm @, subtended by a few foliaceous bracts; peduncle stout, 2-3 cm, with furfuraceous, oblong bracts; primary branches crowded, numer- ous, 21/, cm, furfuraceous, flattened, striate with opposite minute bracts at the apex; central branch c. 3 mm, bearing an umbellule of sterile, bacciform flowers 31/, mm @, pedicels 4'/, mm; lateral branches 3-31/, cm, articulated about the middle, terminating in an umbellule of c. 15 flowers, bracts 1979] ARALIACEAE—I (Philipson) 45 inconspicuous, fimbriate, pedicels 11/,-2 mm. Calyx an obsolete rim. Corolla and stamens not known. Ovary 1'/,-2 mm, 4-celled, with a flat disk and a raised central stigmatic boss, 4-celled. Fruit 6 by 4 mm (dry) 4-ribbed; pyrenes crusta- ceous; endosperm rugose. Distr. Malesia: Philippines (Mindanao). Ecol. Damp fertile ground in dense forest, on south side of Baruring R., at 1000 m. Vern. Sardng-ka-mdno, Bag. 27. Osmoxylon pectinatum (MERR.) PHILIPSON, Blumea 23 (1976) 111. — Boerlagiodendron pectinatum MeErkR. Philip. J. Sc. 3 (1908) Bot. 253, 424; En. Philip. 3 (1923) 224; KANEHIRA, Form. Trees rev. ed. (1936) 520, f. 480; Hur-Lin Li, Woody FI. Taiwan (1963) 666, f. 273. Shrub or small glabrous tree up to 8 m. Leaves clustered at the ends of the branches; petiole to 18 cm, with a clasping base prolonged as a short acute stipular ligule, and with several basal crests fringed with 1-2 cm long bristles; blade to 25cm @, base truncate to broadly cuneate, palmately 5-7- lobed, lobes reaching to about the middle of the lamina, sinuses narrow-rounded, lobes oblong- elliptic, usually slightly narrowed below, obtuse to acute, margin thickened, coarsely dentate, coria- ceous. Inflorescence a terminal compound umbel; peduncle c. 1 cm, with small bracts (3 mm) below and among the primary rays; primary rays c. 25- 35, c. 2-3 cm long, with opposite ovate caducous bracts at the apex, each terminating in three branches; central branch c. 8-11 mm long, ending in an involucre of minute bracts (1 mm) surround- ing an umbellule of c. 15—20 ovoid sterile flowers (c. 3 mm long, 3-celled), pedicels 5-6 mm long; lateral branches c. 2!/, cm long at anthesis, with an articulation about the middle, ending in an umbel- lule c. 1 cm @ with minute fimbriate bracts; flowers c. 30, pedicels c. 11/, mm (elongating slightly in fruit). Calyx rim obsolete. Corolla lobes 4-5, tubular below, 2 mm long. Stamens 4-5. Ovary turbinate, 11/, mm long, 4—6-celled. Fruits globose, 5 by 5 mm (dry), 4—6-ribbed when dry. Distr. Taiwan (Botel Tobago and Lutao I., east off Taiwan proper); in Malesia: N. Philippines (Batan I.). Ecol. Forested slopes at 650 m. Vern. Narapan, Iv. 28. Osmoxylon masarangense PHILIPSON, Blumea 23 (1976) 111. Small tree, 5 m, the young parts setulose, becoming + glabrous. Leaves in terminal clusters; petiole c. 17 cm, rather narrow (1'/.-2 mm @), channelled above, with a small clasping base, prolonged as a stipular ligule, 1-11/, cm long, setulose on the back, and with a number of long- setulose crests encircling the lower part of the petiole; blade c. 18 by 22 cm, deeply 3—5-lobed (or below the inflorescence sometimes simple), the base truncate or emarginate, lobes oblong or elliptic, slightly narrowed to the broadly rounded sinuses, apex with a short apiculum, membranous, margin finely setulose-serrate, sinuses c. 6 cm frob- base of the blade. Inflorescence a terminal sum sessile compound umbel; primary rays c. 10, setulose, 1°/,-2 cm long, each ray ending in three branches; central branch 4-5 mm long, ending in an umbellule, pedicels 6 mm, sterile flowers not known; two lateral branches 2-2'/, cm long, 1 mm broad, with two bract scars about the middle but usually not opposite, terminating in an umbellule with c. 10 pedicels 3-5 mm long (in fruit). Flowers unknown. Fruit (when dry) ovoid, 6 by 4 mm, 5-seeded. Distr. Malesia: N. Celebes (Minahasa, Tomo- hon, Mt Masarang). Ecol. Secondary forest at edge of crater lake, at 1200 m. Note. This species is similar in aspect to the Philippine 39. O. trilobatum, but the petiolar crests are distinctive. 29. Osmoxylon insidiator BEcc. Malesia 1 (1878) 195; PHILIPSON, Blumea 23 (1976) 112. — O. carpo- phagarum BeEcc. Malesia 1 (1878) 196. — Esch- weileria insidiatrix (BECC.) BOERL. Ann. Jard. Bot. Btzg 6 (1886) 120. — Eschweileria carpophagarum (BEcc.) BOERL. /.c. 121, t. 15. — Trevesia insidiator (BEcc.) O. K. Rev. Gen. Pl. 1 (1891) 272. — Trevesia carpophagarum (BeEcc.) O. K. lc. — Boerlagiodendron insidiator (BECC.) HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 32. — Boerlagiodendron carpophagarum (BEcc.) Harms, /.c. — Boerlagio- dendron pachycephalum Harms, Nova Guinea 8 (1910) 271. Small tree to 12 m, young parts with uniform scurfy tomentum. Large /eaves forming terminal crowns; petiole up to 80 cm, stout (1-2 cm 9), broadly channelled above, with a sheathing base prolonged as a strong stipular ligule c. 9 cm long, and with moderately developed irregular (not fimbriate) crests encircling the lower part of the petiole; blade up to 85 cm @, with 5-7 strong ribs radiating from the top of the petiole, deeply lobed almost to the base of these ribs, lobes in turn deeply lobed and incised, the median often digitately tripartite, apices long acuminate, margin irregu- larly and remotely serrate, subglabrous when mature or showing remnants of the tomentum. Inflorescence a terminal hemispherical compound umbel, c. 15 cm high by 30 cm wide; peduncle short, stout (11/.-2 cm @), with lanceolate bracts 4-6 cm long below and among the numerous (15-20) primary rays; primary rays 6-10 cm long, c. 5-12 mm wide, rigid, bearing 2 lanceolate bracts (2-3 cm long) at the apex, each ray ending in three branches; central branch c. 11/, cm long, bearing an umbel of c. 30 sterile bacciform flowers (c. 5- 12 mm @ when dry), the pseudo-fruits and their pedicels + rufous tomentose, pedicels 10-12 mm, 46 FLORA MALESIANA [ser. I, vol. 92 and 6-celled, surrounded by an involucre of short ovate bracts (3-8 mm long); two lateral branches c. 6 cm at anthesis, rigid, slightly flattened, to 8 mm broad, bearing a pair of bracts (c. 11/, cm long) about the middle, terminating in a subglobose head 31/,4 cm @ of c. 30-40 sessile flowers, and surrounded by an involucre of ovate bracts c. 10-14 mm long. Calyx rim fimbriate. Petals irregularly 4-S-lobed, 7-8 mm long, connate below to form a fleshy tube, pubescent on the outer surface. Stamens c. 15-26, filaments strap- like, projecting beyond the corolla, anthers c.4mm long. Ovary shortly turbinate, 24 mm long, angled, furfuraceous, c. 13—25-celled; disk flat, with a central boss formed of the pustulate stigmas. Fruits in a compact spherical head, the individual drupes angled by mutual pressure, and bearing the persistent stigmas on the exposed face, c. 10-14 mm long, the numerous pyrenes compressed and flat; cartilaginous. Distr. Malesia: throughout New Guinea, but local; also Waigeo I. Ecol. Primary rain-forest and regrowths, frequently beside streams, from sea-level to 350 m. Vern. Angit, kangit, Waigeo, pennifogo, Orakawa, Papua. Notes. The bark is greyish brown, slightly fissured with many lenticels. The wood is soft and % ae xe white. The flowers are reddish-brown or purple, with orange-red filaments and the fruit purple. BEcCaRI provided a detailed description of the living plant, and noted that the fruits are eaten by various species of pigeon. Boerlagiodendron pachycephalum Harms has very strongly developed umbels and leaves, but apart from size, it does not differ from this species. Since a range in stature is shown by the several gatherings now available, the whole is best re- garded as a single species. Similarly, the very short pedicels of O. carpo- Phagarum, which BEccarI used to distinguish it from O. insidiator, can in fact be matched on several specimens of that species. 30. Osmoxylon boerlagei (WARB.) PHILIPSON, Blumea 23 (1976) 112. — Eschweileria boerlagei Wars. Bot. Jahrb. 13 (1891) 395. — Boerlagioden- dron warburgii HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 32, nom. illeg. superfl.; in K. Sch. & Laut. Fl. Schutzgeb. (1900) 484. — Boerlagiodendron boerlagei (WARB.) HARMS, Bot. Jahrb. 56 (1920) 382. — Fig. 15. Small to fairly large tree, unbranched or sparingly branched, up to 24 m, glabrous, at least when mature. Large Jeaves forming terminal crowns; petiole up to 1 m, stout (1-2 cm @), ” *e," Fig. 15. Osmoxylon boerlagei (WARB.) PHILIPSON. Showing that each ray ends into a central umbel of bacciform flowers and two lateral umbels with normal flowers (Photogr. PHILIPSON, Kassam Pass, E. New Guinea, 1968). 1979] ARALIACEAE—I (Philipson) 47 broadly channelled above, with a_ sheathing, heavily lenticellate base prolonged as a strong stipular ligule up to 7 cm long, and with lacerate crests encircling the lower part of the petiole; blade up to 1.15 m @, with 5—7 strong ribs radiating from the top of the petiole, deeply lobed almost to the base of these ribs, lobes in turn deeply lobed and incised, the central lobes especially being strongly pinnatisect or digitately tripartite, apices acute, margin undulate or indistinctly serrate. Jnflores- cence terminal (or overtopped by a lateral leafy branch), a large compound umbel, bowl-shaped with a slightly convex top, up to 60 cm @; peduncle c. 10 cm, stout, with lanceolate bracts below and among the very numerous radiating primary rays; outer primary rays c. 20 cm long at anthesis (elongating in fruit), inner rather shorter, woody, bearing two caducous bracts at the apex, each ray ending in three branches; central branch c. 4 cm, bearing an umbel of c. 20 sterile bacciform flowers (c. 8 mm @ when dry) with rigid pedicels c. 11/, cm long, and 5—6-celled; two lateral branches c. 9 cm long at anthesis, articulated about the middle, terminating in a button-like head of c. 20-30 sessile flowers and surrounded by ovate bracts which soon fall leaving a bowl-shaped involucre, c. 11/, cm @. Calyx rim obsolete. Petals c. 13, bud flat-topped, angled, minutely pubescent, apparently falling as a calyptra. Stamens 8-13. Ovary shortly turbinate, angled, glabrous, 10-14-celled; disk flat, with a central double row (or ellipse) of pustulate stigmas. Fruits spreading to form a + spherical head obscuring the involucre, each c. 9 by 7 mm (when dry) with prominent persistent stigmas. Distr. Malesia: throughout New Guinea. Ecol. Primary forest, and secondary growths on old cultivations, from near sea-level to 1800 m. Vern. Eunya, Gimi, apiatambay, Washkuk, ma-korr-korr, Jal, teresahui, Manikiong. Note. Bark yellow grey with shallow fissures and many pale corky pustules; wood straw- coloured, fibrous. The large terminal inflorescence is shallowly convex on top and is surrounded by several large leaves. Flowers reddish brown. Fruit purplish black and succulent. 31. Osmoxylon sessiliflorum (LAUT.) PHILIPSON, Blumea 23 (1976) 113. — Boerlagiodendron sessiliflorum Laut. Nova Guinea 8 (1910) 272. Small tree, up to 18 m, glabrous when mature, or tomentum persistent on the inflorescence. Leaves crowded at the ends of the branches; petiole to 60 cm, with the clasping base prolonged as a stipu- lar ligule up to 6 cm long, and with few to several strong or weak crests around the base of the petiole, margin undulate or fimbriate; blade to 50 cm long, base cordate, palmately 5—9-lobed, lobes extending to near the base, elliptic, coarsely serrate, often irregularly lobulate, apex acute, sinuses rounded. Inflorescence a terminal compound umbel; peduncle to c. 4 cm, bearing lanceolate bracts to 2/2 cm long, caducous or persistent, occasionally with some bristles on the back, primary rays c. 20-30, c. 9 cm long, pubescent, bearing opposite caducous or rarely persistent bracts at the apex; central branch variable in length (2-18 mm), bearing an umbellule or head 1-2'/, cm @ of sterile bacciform flowers (4-5 mm @), 4-8-celled, subtended by minute bracts, pedicels variable in length (5-18 mm); lateral branches 3-7 cm, articu- lated near the middle, terminating in a globose head of c. 20-30 sessile flowers (or pedicels 1'/, mm long), bracts between the flowers very small. Calyx rim obsolete. Corolla few- to many-lobed, tubular below, 1'/.4 mm long in bud. Stamens 6-17. Ovary 5—18-celled. Fruit a globose head of drupes; drupes c. 10 by 6 mm, obovoid, ribbed when dry. Distr. Malesia: throughout New Guinea. Ecol. Rain-forest, especially along the muddy banks of rivers, from sea-level to 100 m. Vern. Akriek, Biak, korinki, Orne, kwita-kwita, Milne Bay, sapi-ai, Jense, terrasahui, Manikiong. Notes. Unbranched or sparsely branched with crowns of large leaves. The bark is light brown and the wood cream. The inflorescence branches are purple, the flowers red, and the succulent ripe fruits black. The variation in the numbers of floral parts is considerable. Most specimens have more than 10 stamens and the same number of cells in the ovary, or more. Three gatherings have from 5—7 stamens and cells. These may possibly require to be segregated as a distinct species, but other evidence to support this course is lacking. An even more dis- tinctive gathering has central branches to 3 cm long with the pseudo-fruits on short pedicels (less than 5 mm) and flowers with 17 stamens and 25 cells in the ovary. These characters have not been included in the specific description as this specimen is only very tentatively referred to this species. This species forms an eastward extension of a complex of species, represented in the Moluccas by 37. O. talaudense, 17. O. soelaense and 18. O. glo- bulare. ‘Several similar species occur in the Philip- pines. They are characterized by dense spherical heads of flowers. Osmoxylon talaudense resembles some specimens of O. sessiliflorum rather closely, and the difficulty of preserving the characters of these large-leaved plants in an herbarium probably obscures several good diagnostic features. The most reliable charac- ter to distinguish these two species is the position of the articulation on the lateral branches of the inflorescence rays. In specimens from the Talaud Is. this is close to the base (below the apex of the central umbellule of pseudo-fruits) and the bracts are persistent, whereas in the New Guinea material it is near (or above) the middle, and is usually marked by two inconspicuous scars which fre- quently are not opposite. In both species the pseudo-fruits are pedicelled, whereas in the other 48 FLORA MALESIANA [ser. I, vol. 91 two Moluccan species the pseudo-fruits are sessile or subsessile forming spherical heads. In O. globu- lare (from Morotai and Halmaheira) the lateral branches are rigid and only slightly flattened with the articulation near the middle and the bracts caducous. In O. soelaense the articulation is much nearer the base, the bracts are persistent, and the branches are broader and much flattened. 32. Osmoxylon camiguinense (MERR.) PHILIPSON, Blumea 23 (1976) 113. — Boerlagiodendron camiguinense MERR. Philip. J. Sc. 3 (1908) Bot. 252; En. Philip. 3 (1923) 222. Shrub to 2 m, glabrous except for the inflores- cence. Leaves clustered at the ends of the branches; petiole to 18 cm, with a sheathing base prolonged as a stipular ligule 1 cm long, and with several entire crests encircling the lower part of the petiole; blade 20 by 22 cm, base truncate, 3—-5-lobed to about the middle, lobes oblong, scarcely nar- rowed towards the base, apex acuminate, sinuses broad, rounded, margin dentate, coriaceous. Inflorescence a terminal compound umbel; primary rays c. 2—2!/, cm, pubescent with opposite lanceo- late bracts (c. 12 mm long) at the apex; central branch c. 4 mm long, bearing a globose umbellule (c. 12 mm @) of sterile bacciform flowers (c. 2 mm @), pedicels c. 3 mm, subtended by numerous ligulate bracts; lateral branches c. 3 cm long, with opposite lanceolate bracts 4 mm long, ending in a head of c. 20-30 sessile flowers. Corolla and stamens not seen, described by MERRILL as 3-merous. Fruit globose 6 mm @ (dry), 3-seeded. Distr. Malesia: Philippines (Babuyan Is.: Camiguin I.). Ecol. On slopes in forest, at 500 m. 33. Osmoxylon fenicis (MERR.) PHILIPSON, Blumea 23 (1976) 114. — Boerlagiodendron fenicis MERR. Philip. J. Sc. 13 (1918) Bot. 44; En. Philip. 3 (1923) 223. — Boerlagiodendron tayabense MERR. Philip. J. Sc. 13 (1918) Bot. 45; En. Philip. 3 (1923) 224. Erect shrub or treelet a few m high, glabrous except for the inflorescence. Petiole to 45 cm, clasping base prolonged as a broad stipular ligule, 1!/,-3 cm long, several prominent long pectinate petiolar crests extending up the petiole as oblique groups of bristles; blade to 30 cm long, base cordate or truncate, deeply palmately 3—7-lobed, lobes extending to within c. 7 cm of the base, broadly elliptic to oblong, entire or with subsidiary lobes, somewhat narrowed towards the sinuses, apex abruptly apiculate, margin remotely denticu- late or serrate, sinuses very broadly rounded. Inflorescence a terminal, compound, furfuraceous umbel, hemispherical, c. 10-12 cm @; peduncle 2-3 cm, densely enclosed in broadly ovate bracts c. 2 cm long, bearing dense fascicles of strong bristles on their blades; primary rays 15-30, c. 21/,-3 cm long, 2'/, mm wide, pubescent, sub- tended by large bristly bracts c. 1'/, cm long, bearing opposite terminal usually bristly bracts 6-15 mm long, each ending in 3 branches; central branch c. 4-10 mm, pubescent, bearing a terminal umbellule of c. 10-20 sterile bacciform flowers 2-3 mm @, 2-celled, pedicels 3-4 mm tomentose sometimes with a ruff of hairs around the pseudo- fruits, subtended by early-caducous small fimbriate bracts; lateral branches 2!/,-3 cm long, with two opposite bracts (2-3 mm) about the middle, bearing a terminal head (8 mm @ without corollas) of c. 15-30 sessile flowers, subtended by incon- spicuous ovate tomentose bracts. Calyx an obsolete rim. Corolla in bud c. 1 mm long. Stamens 3. Ovary 3-celled. Fruiting head 10-12 mm @; drupes c. 6 mm long, crowded, sessile, 3-angled; seeds 3. Distr. Malesia: Philippines (Luzon). Ecol. In primary dipterocarp forest, on rocky slopes near streams, 300—400 m. Note. Closely allied to 40. O. luzoniense and 27. O. pectinatum, but the combination of long- fringed petiolar crests, pubescent inflorescence branches, excessively bristly bracts and 3-merous flowers is distinctive. 34. Osmoxylon eminens (BULL) PHILIPSON, Blumea 23 (1976) 114. — Trevesia eminens BULL, Cat. New Plants (1884) 17; Retail List (1885) 64, fig. — Boerlagiodendron mindanaense MerRR. Philip. J. Sc. 3 (1908) Bot. 154. — Boerlagiodendron eminens (BULL) Merr. En. Philip. 3 (1923) 223. Small tree, up to 10 m, with few stout branches. Leaves large, forming terminal crowns, glabrous when mature, young inflorescence densely fur- furaceous; petiole to 1 m, stout (to 2 cm @), flattened above, base clasping the stem, heavily lenticellate, prolonged as a stipular ligule 2 cm long, bicuspid, often with scales or bristles on the back, and bearing few to several entire, fimbriate or occasionally long setose crests; blade to 60 cm long, palmately 10—-19-lobed, base cordate, lobes reaching to near the base, lanceolate to oblong, up to 15 cm wide, in outline either strap-shaped or irregularly pinnatisect, or the central lobe occa- sionally distinctly tripartite, margin coarsely and irregularly dentate, apex acuminate. Inflorescence a terminal compound umbel c. 40 cm @; peduncle stout c. 6 cm, 1!/, cm wide, bearing many lanceo- late scaly bracts 3-6 cm long; primary rays numerous, rigid, flattened, 9-12 cm long, 6-10 mm broad, bearing opposite oblong scaly bracts (2-3 cm long) at the apex, each ending in three branches; central branch !/,—1 cm long, terminating in an umbellule (3-5 cm @) of c. 20-40 sterile bacciform flowers, 7 mm @, 2-3-celled, pedicels 1-2 cm long, surrounded by an involucre of small bracts (to 8 mm long); lateral branches c. 12 cm long, with opposite bracts (6-10 mm long) near the middle, terminating in a head of c. 50-60 sessile flowers, 1!/,-2 cm @ (with corollas), heads spherical at anthesis, ovoid after corollas absciss; 1979] ARALIACEAE—I (Philipson) 49 bracts between the flowers very small. Calyx rim obsolete. Corolla 5-6-lobed, tubular below, 4-5 mm long. Stamens 4-6, filaments 7 mm long, anthers 1'/, mm long. Ovary 2-3 mm long (at anthesis), 5—6-celled. Fruits crowded in dense ovoid heads 3-4 by 2!/.-3 cm, drupes c. 9 by 5 mm, 5-6-angled by mutual pressure, narrowed to the base, crowned by the persistent stigmatic boss; pyrenes crustaceous; endosperm wrinkled. Distr. Micronesia: Carolines; in Malesia: throughout the Philippines. Ecol. In primary forest from low altitude (100 m) to ridge forest and mossy forest at 950 m, often in shady ravines. Vern. Cf. MERRILL: apalong or apulong, Bis., bunglui-babde, pina-pina, Sul., mangunpulun, Bag., palad-ulot, S.L.Bis., ulo-ulo, C.Bis.; in addition: lolobongan, Lan. Notes. This is the most widespread and most frequently collected species in the Philippines. It is also the most striking. Its large, many-lobed, fan- shaped leaves and the strong inflorescences, with globular flower-heads and large clusters of pseudo- fruits are distinctive. Only 16. O. pulcherrimum resembles it somewhat in its leaf characters, but the central branches of the inflorescence rays of that species are much longer and its pseudo-fruits are sessile. The inflorescence branches are described as dull reddish brown, the flowers as light orange, and the fruits as indigo-black. 35. Osmoxylon serratifolium (ELMER) PHILIPSON, Blumea 23 (1976) 114. — Boerlagiodendron serratifolium ELMER, Leafl. Philip. Bot. 2 (1908) 505; Merr. En. Philip. 3 (1923) 224. Sparingly branched shrub to 5 m. Petiole to 50 cm long, channelled above, clasping base pro- longed as an obtuse stipular ligule, and with few narrow + fimbriate crests around the base of the petiole; blade to 50 cm long, base cordate, pal- mately lobed (up to 11 lobes), lobes extending to within about '/,; from the base, narrowly elliptic, margin serrate (or slightly lobulate), apex acumi- nate, sinuses narrowly rounded. Inflorescence a terminal compound umbel c. 30 cm @; peduncle stout, bracteate; primary rays 20-30, 4-5 cm long, 5-6 mm wide, flattened, subtended by lanceolate bracts 3-5 cm long, sometimes with bristles on the back, and bearing similar opposite terminal bracts 21/2 cm long, each terminating in three branches; central branch 12-15 mm long, terminating in an umbellule (3-4 cm @) of c. 20-25 sterile bacciform flowers 5-6 mm @, 3-4-celled, pedicels to 10 mm, interspersed with persistent small bracts; lateral branches c. 9 cm long, with opposite bracts (c. 6 mm long) 2-3 cm from the base, terminating in a spherical head of c. 30 sessile flowers c. 11/, cm @ (in bud), bracts between the flowers very small, obtuse, fimbriate. Calyx rim obsolete. Corolla 5-7-lobed above, tubular below, 5 mm long. Stamens 5-6, exserted, filaments 7 mm long, anthers 2 mm long. Ovary 3 mm long, 5-celled. Fruit unknown. Distr. Malesia: Philippines (Leyte, Camiguin, Panay). Ecol. ELMer noted that this species was rare in the low hills of Leyte. Notes. The flowers are orange-yellow (salmon), the fruits dark purple. The inflorescence is very similar to that of 34. O. eminens, but the leaves lack the many strong fan-like ribs of that species. 36. Osmoxylon celebicum PxHiLieson, Blumea 23 (1976) 115. — Boerlagiodendron celebicum HARMS ex Koorpb. Minah. (1898) 489, nomen. A small, sparsely branched tree, 6 m high. Large leaves forming terminal crowns; petiole 50 cm by 8 mm, flattened above, with a sheathing base prolonged as a strong stipular ligule 3 cm long with branched fibrous setae on the outer surface, and with several crests bearing similar setae on the lower part of the petiole; blade 50 cm @, base emarginate, deeply 9-lobed, lobes narrowly elliptic to lanceolate, narrowed towards the sinuses, apex broadly cuneate, margin minutely and remotely serrate, sinuses broadly rounded. Inflorescence a terminal compound subspherical umbel c. 20cm @; peduncle short, stout (15 mm wide) with large setose bracts (c. 4 cm long) below and among the primary rays; primary rays c. 15, c. 6 cm long, 5 mm broad, with a pair of large setose bracts (22 by 10 mm) at the apex, each ending in three branches; central branch c. 6 by 2 mm, terminating in an involucre of setose ovate bracts (c. 4 mm long) and an umbel of c. 20-30 sterile bacciform flowers (c. 3 mm @ when dry, 2-3-celled, on pedicels 6-10 mm long) interspersed with bracts covered with crisp reddish-brown setulae; the two lateral branches c. 2 cm long, with opposite setulose bracts (c. 1 cm long) below the middle, terminating in a spherical head of c. 30-40 sessile flowers each sub- tended by a reddish brown setulose cymbiform bract. Calyx rim obsolete. Corolla c. 2'/, mm long in bud (not seen in open condition). Stamens 5. Ovary subcylindric, c. 1 mm long in bud, 5-celled; disk with a central stigmatic boss. Fruit unknown. Distr. Malesia: Celebes (Minahasa, Manado). Ecol. On rich volcanic sand, at 10 m. Vern. Sinomaha. Note. The flower buds are orange and the fruits deep purple. 37. Osmoxylon talaudense PHILIPSON, Blumea 23 (1976) 115. Shrub or small tree, to 6 m, glabrous. Leaves at the ends of the stout branches; petiole to 60 cm, broadly channelled above, clasping base prolonged as a stipular ligule c. 2 cm long, and bearing 2-3 fimbriate crests; blade c. 50 cm long truncate to cordate at the base, palmately 7—-11-lobed to within 50 FLORA MALESIANA [ser. I, vol. 91 Fig. 16. Osmoxylon micranthum (HARMS) PHILIPSON. a. Habit, x 1/,, b. flower, c. false fruit and ditto in CS, x 15, d. CS of fruit, x 6 (a-c KANIs 1384, d PULLEN 428). Drawn by W. R. PHILIPSON. 1979] ARALIACEAE—I (Philipson) 51 + 1/4 of the base, lobes elliptic oblong slightly narrowed towards the rounded sinuses, apiculate, margin denticulate to undulate. Inflorescence a terminal compound umbel, peduncle 1-2 cm, bearing broad ovate bracts 2-3 cm long, with bristles on the back, (similar persistent bracts subtend the primary rays); primary rays c. 15, c. 3-4 cm long, 4 mm wide, flattened, bearing opposite terminal persistent bracts 11/,-2 cm long, sometimes with a few bristles on the back, ending in three branches; central branch 8-10 mm long, terminating in an umbellule 2 cm @ of 15-20 sterile bacciform flowers (4 mm @, 2-celled) surrounded by an involucre of obtuse bracts 1-2 mm long, pedicels 5—7 mm; lateral branches 4'/,-5 cm long, bearing opposite ovate persistent bracts (3-7 mm long) c. 5-10 mm above the base, terminating in a dense head c. 1 cm @ of 30-40 sessile flowers inter- spersed with inconspicuous obtuse bracts. Calyx rim obsolete. Corolla 5-lobed. Stamens 5. Ovary turbinate, 11/, mm long, 5-celled. Fruit in spherical heads 2 cm @ (when dry); drupes c. 9 by 6 mm, obovoid, 5-ribbed. Distr. Malesia: N. Moluccas (Talaud Is.: Karekelong and Salebabu). Ecol. Common in forest, besides streams, from near sea-level to 100 m. Vern. Laripatu, Talaud. Note. The flower is yellow-orange and the fruit dark purple. For a discussion of the distinctive features, see under 31. O. sessiliflorum. 38. Osmoxylon micranthum (HARMS) PHILIPSON, Blumea 23 (1976) 115. — Boerlagiodendron micranthum Harms, Bot. Jahrb. 56 (1920) 379. — Boerlagiodendron sayeri HARMS, /.c. 379, f. 1 a—j. — Eschweileria gawadensis BAKER f. J. Bot. 61 (1923) 22. — Boerlagiodendron tricolor PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 11. — Fig. 16. A sparsely branched shrub to 8 m, sometimes trailing or semi-scandent, young parts uniformly setulose, buds without cataphylls. Leaves in terminal clusters; petiole up to 30 cm, rather narrow (2-4 mm @), becoming sparsely setulose, channelled above, with a sheathing base prolonged as a membranous stipular ligule up to 3 cm long, and with a number of lacerate crests encircling the lower part of the petiole; blade deeply 3-5- or more rarely 7-lobed, or below the inflorescence sometimes simple, base cordate or emarginate, the central lobe up to 30 cm long, the lobes oblong, lanceolate or broadly elliptic, entire or irregularly lobed or incised, or with small sub-lobes, apices long cuspidate, acute, margin serrate, sinuses between the lobes broad and rounded, surfaces become sparsely setulose to subglabrous. /nflores- cence a terminal compound umbel, often appearing subterminal by growth of a leafy brafich at the base of the peduncle; peduncle short (1-2 cm), heavily setulose, occasionally with a flowering ray arising from the axils of bracts on or below the peduncle, bearing distally many lanceolate bracts 5-10 mm long; primary rays 12-18, 10-20 mm long, setulose, with two lanceolate bracts at the apex, each ray ending in three branches; the central branch very short (2-3 mm) bearing a subglobose umbel of many (c. 40) small sterile bacciform flowers (c. 1'/. by 1 mm) with filamentous pedicels c. 2 mm long, and 1-2 cells each with 1 abortive ovule; the two lateral branches c. 2 cm long, with two minute bracts about their middle, setulose, terminating in a head of c. 20 sessile flowers sur- rounded by an involucre of small rounded bracts. Calyx rim obsolete. Corolla 4(—5)-lobed, united below, c. 2 mm long. Stamens 4(—5), filaments ribbon-like elongating beyond the corolla tube at anthesis, 3-4 mm, anthers small. Ovary shortly subcylindric, c. 1 mm high, faintly angled, glabrous, 1-5-, usually 4-celled, disk fleshy, rising slightly to the central stigmas. Fruit an ellipsoid drupe with 1-5 cartilaginous pyrenes; seeds with smooth endosperm. Distr. Malesia: New Guinea (Irian Jaya: Idenburg R. to Milne Bay Distr.). Ecol. In primary forest from the foothills to the montane mossy forest, often in swampy or deeply shaded situations, 700-2400 m. Vern. Diande, Chimbu, kenata, Okapa. Note. The inflorescence branches are often red or purple, and the flowers either orange or reddish with yellow anthers. The ripe fruits are deep purple or black. The shape and size of the leaf can vary greatly, even on the same plant. The ovary usually has 4 cells, but plants with 3, 2 and 1 occur. Since these are alike in other respects they have been treated as a single species. Although the type of Boerlagiodendron tricolor has an ovary with 5 cells and is from much further west than other gather- ings, it is not considered to be specifically distinct. 39. Osmoxylon trilobatum (MERR.) PHILIPSON, Blumea 23 (1976) 116. — O. cumingii SEEM. J. Bot. 6 (1868) 141, nomen. — Boerlagiodendron trilo- batum Merk. Philip. J. Sc. 2 (1907) Bot. 289; En. Philip. 3 (1923) 224. Slender shrub or small tree up to 5 m, becoming glabrous except for slight pubescence on the inflorescence. Leaves clustered near the ends of the branches; petioles to 25 cm, c. 3 mm wide, clasping base prolonged as a broad stipular ligule c. 1 cm long, and with 2-3 entire, or obscurely fimbriate, often recurved crests surrounding the base of the petiole; blade 3- or occasionally 5-lobed (leaves below the inflorescence sometimes simple), to 30 by 28 cm, base broadly cuneate, rounded or trun- cate (emarginate in 5-lobed leaves), lobes about 1/,*/, of the blade, narrowly or broadly oblong, often slightly narrowed below and sharply acumi- nate to caudate, margin serrate. Inflorescence a terminal compound spherical umbel, 7-15 cm 2, either rather compact or branches lax; peduncle 2-3 cm with broad ovate bracts; primary rays 8-20 52 FLORA MALESIANA [ser. I, vol. 9 or more, slightly pubescent, 11/.-4 cm long, sub- tended by ovate bracts 5-10 mm long, opposite ovate bracts at the apex, 1-3 mm long; central branch 11/,-6 mm long, pubescent, bearing an umbel (1—11/, cm @) of sterile bacciform flowers up to 5 mm @, 1-4-celled, pedicels 2-6 mm long, subtended by ovate bracts 1-3 mm long; lateral branches 1'/,-3 cm with opposite small bracts about the middle, bearing a terminal head, c. 1 cm @ of c. 8-20 flowers, surrounded by an involucre of small rounded pubescent bracts, pedicels c. 1 mm or less (up to 3 mm in fruit). Calyx rim obsolete. Corolla 4—5-lobed above, tubular below, 2-3 mm long. Stamens 4-5, exserted. Ovary subcylindric, 4-5-celled. Fruit a spherical drupe (when dry 4—5- ribbed, 7 by 5 mm). Distr. Malesia: widespread in the Philippines (Luzon to Mindanao). Ecol. In primary forest, frequently beside streams in damp ravines, from the lowland at 75 m to 750 m. Vern. Kamay-kamay, Tag., ayum, C.Bis. Note. Sparingly branched but wide-spreading slender shrub, with yellowish bark, at first heavily dotted with brown lenticels. The flowers are white and the ripe fruit smooth and purple. 40. Osmoxylon luzoniense (MERR.) PHILIPSON, Blumea 23 (1976) 116. — Boerlagiodendron luzoniense MerRR. Philip. J. Sc. 3 (1908) Bot. 252; En. Philip. 3 (1923) 223. — Boerlagiodendron clementis MERR. Philip. J. Sc. 3 (1908) Bot. 155; En. Philip. 3 (1923) 222. — Boerlagiodendron agusanense ELMER, Leafl. Philip. Bot. 7 (1914) 2330; MerrR. En. Philip. 3 (1923) 222. — Boer- lagiodendron diversifolium MERR. Philip. J. Sc. 10 (1915) Bot. 333; En. Philip. 3 (1923) 223. Erect, unbranched or sparsely branched shrub or tree to 8 m, becoming glabrous except for the inflorescence. Leaves clustered at the ends of the branches; petiole to 40 cm, with a clasping base prolonged as a short stipular ligule, and with several basal entire or shortly fimbriate crests; blade to 33 cm @, base truncate or cordate, deeply palmately 3—7-lobed, sinuses broad, rounded, lobes elliptic, usually narrowed below, entire or with subsidiary lobes (the central lobe especially often narrow below and strongly pinnately lobed), apex acute, margin coarsely serrate, coriaceous; upper- most leaves often reduced and simple. Inflorescence a terminal compound, subsessile umbel 10-15 cm @; primary rays c. 20-30, 2-3 cm long, subtended by lanceolate furfuraceous and + fimbriate bracts, furfuraceous villose or + hirsute, with opposite rounded or lanceolate hirsute bracts at the apex, each ending in three branches; central branch c. 2-8 mm long, hirsute, ending in an involucre of minute bracts (1 mm) surrounding a globose (2 cm @) umbellule of c. 15-20 sterile flowers (c. 6 by 6 mm, 2-3-celled), pedicels 2-3 mm, hirsute; lateral branches c. 21/,-31/, cm long at anthesis, with an articulation about the middle, + hirsute, ending in a globose head, 2 cm @ (with open corollas), with ovate tomentose obtuse bracts, c. 2 mm long; flowers 30-40, + sessile (rarely pediceis to 2 mm). Calyx rim obsolete. Corolla 4-5-lobed above, tubular below, 3'/.-4 mm long. Stamens 4-5, exserted, filament stout, 5 mm, anther 1 mm long. Ovary 4-S-celled. Fruit 6 by 5 mm (dry), strongly 4—-5-ribbed. Distr. Malesia: widespread in the Philippines (Luzon to Mindanao), also in N. Celebes. Ecol. In forests, often by streams and on ridge in mossy forest, 280-1650 m. Vern. Philippines: bolwang hi inalahan, If., iyangnok, Mbo, malakapayas,S.L.Bis., molonpolon, Buk., tafigan-tangan-batu, Buk., tachung, vafiganeg, Ig. Notes. Inflorescence yellow to red (salmon), fruits blue-black or purple. MERRILL did not liken his Boerlagiodendron diversifolium (from Mindanao) to this species, no doubt because he gave importance to the occur- rence of variable leaves and the 5-merous flowers. However, specimens from Luzon may possess simple leaves below the umbel, and both 4- and 5-merous flowers occur in both Luzon and Mindanao. The greater range of material now available establishes the identity of the two species. Similarly, no features seem to distinguish Boer- lagiodendron clementis, and B. agusanense though a greater range of collections would be desirable. The species is treated here in a broad sense. The inflorescence characters of most specimens are uniform, being hirsute and with the bracts at the apex of the primary rays obtuse and short. Some specimens (CURRAN 5088, ELMER 16762) have longer lanceolate bracts with some bristles on the back. The northernmost specimen, from Ilocos Norte, has finer and less hairy inflorescence rays (recalling 39. O. trilobatum), but the foliage agrees with this species. Leaf-shape is more variable, even on the same specimen. Other species which resemble O. luzoniense in some respects are: 32. O. camiguinense with broader, shallower lobing, a more delicate inflorescence, and _ tri-merous flowers; 27. O. pectinatum with glabrous inflores- cence branches and long-pectinate petiolar crests; and 26. O. humile with pedicelled flowers forming less dense heads. A specimen from Surigao Province (BS 83562) has a most interesting abnormal structure. The central branches of the inflorescence rays bear heads of apparently fertile flowers, with lobed corollas and exserted stamens. 41. Osmoxylon insigne (MIQ.) Becc. Malesia 1 (1878) 195; Puiipson, Blumea 23 (1976) 117. — Trevesia insignis Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 222. — Trevesia palmata var. insignis CLARKE, FI. Br. Ind. 2 (1879) 732, pro nomen. — Eschweileria insignis (M1Q.) BoERL. Ann. Jard. Bot. 1979] ARALIACEAE—I (Philipson) 53 Btzg 6 (1887) 122. —. Boerlagiodendron insigne (Miq.) Harms in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 31. A glabrous tree. Leaves large, palmately lobed; petiole to 45 cm, 8-10 mm wide, flattened above, with a sheathing base prolonged as a stipular ligule, numerous fimbriate crests around the base of the petiole, and irregular tufts of bristles along the whole length of the petiole; blade to 55 cm @, deeply 5S—7-lobed, the sinuses broadly rounded and c. 4-5 cm from the base of the blade, lobes pinnatilobed, with a narrow base and an attenuated apex, margins serrate. Inflorescence a terminal compound umbel c. 16 cm @; peduncle stout, with lanceolate bracts (11/,-2 cm long) subtending the primary rays; primary rays 25-30, c. 4 cm long, bearing 2 caducous lanceolate bracts (c. 1 cm long) at the apex, each ray ending in 3 branches; central branch c. 12-16 mm long, bearing an umbel (c. 2 cm @) of c. 20 sterile ovoid bacciform flowers (c. 4 mm long when dry) with pedicels c. 4-6 mm long, and 3-celled; two lateral branches c. 31/, cm long, with an articulation about the middle, termi- nating in an umbellule (c. 2 cm @) of c. 10-15 flowers on short stout pedicels c. 2 mm long, umbellules surrounded by a receptacular rim after caducous bracts have abscissed. Calyx rim minute. Petals c. 31/, mm long in bud (when dry), with 8-9 lobes above, tubular below. Stamens 8-9, with stout filaments. Ovary cylindric, 8—-9-celled; disk with a central double row of pustulate stigmas. Fruit unknown. Distr. Malesia: Moluccas (Batjan). Note. Tufts of bristles along the entire length of the petiole together with the pinnatifid lobes of the leaf are distinctive. SEEMANN (J. Bot. 4, 1866, 353) referred to 5-flowered umbels with 5-angled drupes, but this probably relates to the New Guinea specimen which he included under this name. Insufficiently known Boerlagiodendron ledermannii HARMS, Bot. Jahrb. 56 (1920) 383; PHILIPSON, Blumea 23 (1976) 117. — Type: LEDERMANN 12293. HARMS compared this species with Boerlagio- dendron geelvinkianum. The size of the foliage and flowers prevents it from being included within that species. If it represents a local species, it has not been re-collected since the original gathering of LEDERMANN in 1912. The type specimen, which was incomplete, was destroyed during the war. Boerlagiodendron monticola HaARMs in K. Sch. & Laut. Fl. Schutzgeb. Nachtr. (1905) 330; Puitip- SON, Blumea 23 (1976) 117. — Type: SCHLECHTER 14471. The incomplete type specimen, gathered by SCHLECHTER, was destroyed during the war. This species was evidently similar to 38. Osmoxylon micranthum, but the ovary was possibly 10-celled. I have tentatively identified RopBins 1644 as this species: it is close to O. micranthum but its ovary, with 8 cells, is outside the range of variation of that species and the pedicels are longer (in fruit). Its distribution (foothills of Adalbert Range) is not dissimilar to that of Boerlagiodendron monticola (Torricelli Mts). 8. ARTHROPHYLLUM BL. Bijdr. (1826) 878; DC. Prod. 4 (1830) 266; Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 27; Bru. in B. & H. Gen. PI. 1 (1865) 944; Harms in E. & P. Nat. Pfi. Fam. 3, 8 (1894) 54; Koorpb. Atlas 4 (1916) f. 675 & 676; HuTcH. Gen. FI. Pl. 2 (1967) 80; STONE, Gard. Bull. Sing. 30 (1977) 276; PHILIPSON, /.c. 299, f. 1-16; Adansonia 17 (1978) 329. — Mormoraphis JACK ex WALL. Cat. (1831) n. 4931, nomen. — Eremopanax BAILL. Adansonia 12 (1878) 158. — Fig. 17, 19-23. Unarmed, sparingly branched trees or shrubs. Leaves on vegetative shoots and lower leaves spirally arranged, imparipinnate, often crowded at the end of the branches, those on flowering branches often opposite, smaller, or reduced to a single leaflet; petiole terete; rachis articulated at the insertion of the pinnae and leaflets; leaflets entire; stipular sheath clasping, small, ligule a mere rim. Jnflores- cence consisting of compound umbels, either solitary and terminal or more commonly borne on a cluster of specialized leafy branches arising from the axils of the uppermost leaves; pedicels not articulated. Flowers bisexual. Calyx an undulate rim, sometimes with indistinct teeth, persistent. Petals 4-6, valvate in bud. Stamens 4-6, anthers curved, basifixed. Ovary turbinate, 1-celled; disk FLORA MALESIANA [ser. I, vol. 91 . 17. Habit of Arthrophyllum diversifolium Bu. (Photogr. HOOGLAND, Bogor). 1979] ARALIACEAE—I (Philipson) $5 fleshy, rising in the.centre to the + sessile capitate stigma. Fruit ovoid or spheroidal, often oblique; exocarp leathery; endocarp cartilaginous. Seed solitary, pendulous; endosperm deeply transversely ruminate. Distr. About 31 spp. (17 in Malesia) extending from the Nicobar Is. and Indo-China to the Philippines, New Guinea and New Caledonia. Fig. 18. Notes. The principal distinctive characters of this isolated genus are the single-celled ovary and the arrangement of the umbels on specialized lateral branches in the majority of the species. The present treatment, following that which I gave in 1977, /.c., remains tentative until widespread field studies can be undertaken. KEY TO THE SPECIES 1. Inflorescence becoming paniculate by the successive development of branches below the umbellules. Fig. 19a 1. A. proliferum 1. Inflorescence a compound umbel. 2. Inflorescence with four orders of branching. 3. Leaflets lanceolate. 4. Leaflets c. 4-7 cm long. Fig. 20f 4. Leaflets much longer 3. Leaflets broader. 5. Pedicels (at early anthesis) c. 10 mm long. 6. Peduncles of umbellules with bracts or their scars. Fig. 21a 6. Peduncles of umbellules without bracts or their scars. Fig.21b. . . .. . 2. A. ashtonii 3. A. angustifolium 4. A. ahernianum 5. A. engganoense 5. Pedicels (at early anthesis) c. 5 mm long, or shorter. i eaeeaves associated with. the umbels rotund. Fig. 2le.... . 2 0.2 26 ws ns 6. A. collinum 7. Leaves (or leaflets) associated with the umbels ovate or elliptic. 8. Leaves associated with the umbels ovate, + fleshy, with the lower surface smooth (lateral veins obscure). Fig. 21f 7. A. crassum 8. Leaves (or leaflets) associated with the umbels + elliptic, coriaceous or chartaceous, veins visible. 9. Young parts glabrous 9. Young parts with rufous tomentum. 10. Umbels at anthesis with numerous filamentous pedicels. Fig. 21c. 10. Umbels at anthesis with fewer stout pedicels. Fig. 21d Re E Spt GAG cuee cus 3m (sania date oN Sa ce 8. A. pacificum 9. A. diversifolium 10. A. macranthum 2. Inflorescence with three (or fewer) degrees of branching. 11. Leaflets membranaceous or chartaceous. 12. Mid-leaflets c.16cmlong. ....... 12. Mid-leaflets c. 8 cm long or shorter. C) eter te hela web Me. die lek baie) cel) ce) Mis ais, 11. A. papyraceum 13. Leaflets usually 5—7 (Fig. 20a). Primary inflorescence branches usually short (c. 3-6 cm) and without articulations Pernt? “Ope Whe Oe Be ake ao) 12. A. maingayi 13. Leaflets more numerous (Fig. 20d). Primary inflorescence branches longer (10-20 cm), with one or more leafy nodes 11. Leaflets coriaceous. 14, Leaflets c. 5 14. Leaflets more numerous. 15. Petals and stamens 6 15. Petals and stamens 4 or 5. 16. Leafiets 6 cm long, or longer. Fig. 20b . 16. Leaflets shorter. Fig. 20c phos Lek. jet itis. le amie 1. Arthrophyllum proliferum PHILIPSON, Gard. Bull. Sing. 30 (1977) 302, f. 3-4. — Fig. 19a-b. Medium-sized, glabrous tree. Leaves multijugate, up to 90 by 24 cm, of the flowering branches smaller with fewer pinnae or usually simple; petioles up to 28 cm, 5 mm @; petiolules 1-1'/, cm; leaflets obovate-oblong, c. 16 by 7 cm, chartaceous, margin slightly revolute, base broadly cuneate to truncate, es Ce oa ne CARTS ye Or MO. ey) mc Chet » (WES jel AY She IN CE Wire tS A) AA Fie We (iY AY BE Be AG ae AERERe CHhbhic ss Fig. 22. Arthrophyllum diversifolium BL. Diagrammatic sketch of habit, showing vegetative and flowering branches, 3/3, (Courtesy Gard. Bull. Sing. 30, 1977). the inflorescence-bearing branches and smaller rarely unifoliolate) leaves mostly in opposite pairs with fewer pinnae or unifoliolate; petiole up to and usually with flowering branches in the axils of 40 cm; petiolules '/,-11/, cm; leaflets ovate-oblong the upper leaves, ending in an umbel of secondary or elliptic, up to 24 by 11 cm (leaflets of bipinnate rays subtended by a whorl of pinnate or unifoliolate leaves usually c. 10 by 5 cm), + coriaceous or leaves; secondary rays up to c. 30 cm, bearing (especially in bipinnate leaves) somewhat mem- mainly simple leaves in opposite pairs with branaceous, margin slightly revolute, base truncate, flowering branches in their axils and terminating rounded, or cuneate, often oblique, apex shortly in compound umbellules each subtended by a acuminate, veins usually 5—7 pairs. Inflorescence a whorl of often caducous bracts; tertiary rays cluster of specialized leafy branches forming a (peduncles) c. 5 cm, articulated about the middle; terminal crown which abscisses after fruiting; umbellules with c. 30-40 flowers, c. 17-20 mm 2; main rays up to 150 cm, bearing pinnate (or more _ pedicels c. 20, c. 5 mm at anthesis (longer in fruit), 62 FLORA MALESIANA [ser. I, vol. 9! Fig. 23. Arthrophyllum diversifolium BL. Schematic drawing of a single flowering shoot, with four degrees of branching, « 1/; (Courtesy Gard. Bull. Sing. 30, 1977). 1979 ] with an involucre of minute caducous bracts. Petals 5, c. 2 mm long. Stamens 5, anthers curved. Ovary turbinate, often inconspicuous at anthesis; disk fleshy, rising in the centre to a sessile stigma. Fruit ellipsoidal, c. 9 by 7 mm; calyx and stylo- podium forming a conspicuous beak. Distr. Malesia: Sumatra, Malay Peninsula, Java, Borneo, Celebes. Ecol. In a wide variety of habitats, on dry sandy soil to swampy humus, in primary lowland and montane rain-forest and also in secondary forest, heath-forest and waste land, from sea-level to 1600 m. Uses. Concoctions of the root and bark are reported to have medicinal properties, including a remedy for syphilis, and the plant has stupifying and poisonous properties. Vern. Sumatra: antjaneudeung uding, bidju, bolu bolu, bulu, (kayu) abang-abang, k. attu turut, légung, Riouw Arch., mapang, miu, obang, potah, silanta, ténjam dakan, Banka, tocrah. Malay Peninsula: chindangan utan, jolok hantu, lupa dahan, (pokok) restong, ségan budahan, susun kélapa, tum bong ninyor, tusum perpah. Java: dajo, délahan, déleg, délek, djangkorrang, gombong, (kayu) gompang, kédrja, kidjangkurang, kingom- pang, ki ompong, kléntjang, krépang, langitj, malas bérdahan, pongporang, putjangan. Borneo: merje- meh, Sarawak, karadjungjung, ‘Kalimantan. Celebes: kKambabah, susangkangan. Notes. The very widespread A. diversifolium is variable in many characters, and may comprise a number of geographic subspecies, but no basis for this is apparent at present. Most individuals have the lower leaves simply imparipinnate, whereas others have bipinnate, or rarely tripinnate, leaves. The flowers and inflorescences of these forms appear to be identical, though rapid changes in the umbellules after flowering produce a deceptively distinctive appearance in specimens at different stages of development. Field experience over the . whole range of the species will be required to understand this interesting leaf-polymorphism. In treating all forms as one species I am partly influenced by the fact that most authors who have been familiar with the plants in Java (where both forms occur) have regarded the complex as a single species (the fact that some authors have recognized the variant from Mt Salak as a distinct species does not affect the problem of leaf-polymorphism). Apart from the strikingly different leaf forms just discussed, certain local variants may eventually be shown to justify specific rank. A form growing on Mt Salak (near Bogor) has often been regarded as distinct (see, for example, HOCHREUTINER, Candollea 2, 1925, 481, and BACKER & BAKHUIZEN VAN DEN BRINK f. FI. Java 2, 1965, 169). Indeed this form is the basis of the name 4A. diversifolium. I retain this name in preference to the other two names published simultaneously by BLUME because it has been most consistently adopted since ARALIACEAE—I (Philipson) 63 it was first used in this comprehensive sense by CLARKE, /.c. On the evidence available I do not consider the Salak plants any more distinctive than many other local variants. It might be considered that 5. A. engganoense is also no more than another such variant, but its facies is so marked that specific rank appears justified. It is possible that RIDLEY was correct in distin- guishing A. congestum, but the material is not good and appears inadequate to confirm specific status. Five collections from Brunei and a neighbouring district of Sarawak are all very alike and sufficiently distinct from both A. diversifolium and 7. A. crassum to suggest that they represent a separate taxon, but for the present they are tenta- tively retained as a form of A. diversifolium. Similarly, the two collections described by Rip.ey as A. rubiginosum and A. rufosepalum are based on collections which are not altogether typical of A. diversifolium, but which come closest to that species. In the absence of more supporting material, it is advisable not to retain them as species. The first of these names (A. rubiginosum) has been widely used in identifications of Bornean specimens, but the specimens concerned are either typical A. diversifolium or belong to the distinctive 7. A. crassum. Specimens from Mt Kinabalu described by Ripcey as A. havilandii have bipinnate leaves, and appear to conform well with A. diversifolium. This form was again collected on. Mt Kinabalu by CLEMENS and is also known from Sarawak. The smooth bark is whitish to greyish brown with pustulate lenticels; the wood is cream, with a colourless aromatic exudate. The flowers are yellowish with a sickly sweet scent. Seedlings have simple and trifoliolate leaves. 10. Arthrophyllum macranthum PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 18; Gard. Bull. Sing. 30 (1977) 308, f. 16. — A. diversifolium (non BL.) Harms, Bot. Jahrb. 56 (1920) 413. — Fig. 21d. Tree up to 25 m, sparsely branched with leaves crowded at the ends of the branches, all young parts with dense rufous tomentum which may persist on the flowers and inflorescence. Leaves multijugate, up to 100 by 30 cm; of the flower- bearing branches smaller with fewer pinnae, or simple; petioles stout, up to 40 cm, clasping base heavily lenticellate; petiolules '/,-2 cm; leaflets ovate to oblong, up to 16 by 8 cm, coriaceous, mar- gin revolute, base rounded, truncate, or cordate, very rarely cuneate, often oblique, apex obtuse or bluntly apiculate, principal veins arched-ascending, reticulations visible especially beneath (pinnae of leaves on the flowering branches usually elliptic with a cuneate base). Inflorescence a whorl of specialized leafy branches forming a terminal crown; main rays up to 60 cm, bearing pinnate leaves often in opposite pairs and with flowering 64 FLORA MALESIANA [ser. I, vol. 91 branches in the upper axils, and ending in a whorl of secondary rays subtended by a whorl of usually simple leaves (bracts); the secondary rays c. 10— 20 cm, bearing simple leaves, usually in opposite pairs, and terminating in compound umbels; umbellules (c. 2 cm) with c. 10-15 flowers; pedicels stout, 3-5 mm to 10 mm or more in fruit, sub- tended by minute bracts. Flower buds 5 mm or more long; calyx a rim or with 5 indistinct teeth. Petals 5, triangular, c. 4 mm long, fleshy. Stamens 5. Ovary obconical, c. 2'/, mm long; disk fleshy (hemispherical in living material, conical and furrowed when dry); stigma capitate, + sessile. Fruit ellipsoid, sometimes slightly oblique, c. 12 by 8 mm, the stylopodium forming a conical beak with the persistent stigma; exocarp fleshy, endo- carp cartilaginous. Distr. Malesia; New Guinea (from Irian Jaya to Milne Bay Distr. and the Bismarck Archipelago). Ecol. Usually a sub-canopy tree of rain-forest ranging from the lower montane zone to mossy subalpine woodland and scrubland, occasionally in second growth, usually above 1000 m (up to 2700 m), but occurring also on the coastal scarps of the Astrolabe Range. Vern. Maguva, W. Sepik, agagwa, agare, agugwa, alolo, angga, engga, pooli, tipilan, W. High- lands, arua, hagegoa, wonkurumeh, E. Highlands, kolom, S. Highlands. Note. Trees become very different in appearance when in flower or fruit: the spiral pinnate foliage leaves are surmounted by tufts of branches which end in inflorescences and bear much smaller leaves. The foliage leaves are fleshy, leathery and glossy. The ripe fruit is purple and shining. The bark is grey, at first smooth with many leaf-scars and lenticels, but small longitudinal fissures develop. The cut branches exude a brown latex and a scent of celery. The soft wood is white or straw-coloured. 11. Arthrophyllum papyraceum PHILIPSON, Gard. Bull. Sing. 30 (1977) 308. Shrub, rufous-tomentose on the young parts. Leaves alternate, imparipinnate; petiole 15—20 cm; leaflets c. 7, membranaceous, elliptic, c. 12-24 by 5-10 cm, base broadly cuneate, apex finely acuminate, margin slightly revolute. /nflorescence a terminal compound umbel; primary rays few (2), 2-3 cm long, 2 mm wide, without bracts (caducous), secondary rays few (3), c. 13-18 mm long, articu- lated about the middle, ending in an umbellule of c. 10-12 flowers; pedicels 2-3 mm, slightly fur- furaceous. Petals 5, c. 2 mm long in bud. Stamens 5, anthers curved. Ovary turbinate, glabrous, obscurely ribbed. Fruit unknown. Distr. Malesia: E. Sumatra (East Coast Res. near Aek Sordang), one collection. Ecol. Primary rain-forest. Note. Known from a single collection (with no duplicates) this species resembles 12. A. maingayi in its simple inflorescence, and the few pinnae of its foliage leaves. However, the large size of the leaflets precludes its inclusion in that species. 12. Arthrophyllum maingayi PHILIpson, Gard. Bull. Sing. 30 (1977) 309, f. 7. — A. pinnatum CLARKE, Fl. Br. Ind. 2 (1879) 734, excluding synonyms [see also SEEM. J. Bot. 4 (1866) 294]; RIDL. Fl. Mal. Pen. 1 (1922) 886. — Fig. 19c, 20a. Low shrub or slender tree, rarely as high as 10 m, rufous-tomentose on the very young parts, soon becoming glabrous. Leaves tufted at the ends of the branches, usually with 5—7 leaflets (but up to 15), up to 30 by 18 cm; petiole to 12 cm (usually shorter), 2 mm @; petiolules c. 0-10 mm; leaflets elliptic or elliptic-lanceolate (occasionally the lowermost pair of leaflets is replaced by pinnate leaf segments), c. 8 by 3!/, cm, rather thin, base cuneate or rounded, apex apiculate or caudate, margin slightly revolute, lateral veins faint and obscure; leaves below the flowering branches sometimes reduced to 3 or | leaflet(s). Inflorescence consisting of a number of primary branches radiating from the end of a leafy shoot (which forms a longer or shorter peduncle); primary branches, often rather few, usually 3-6 cm long and devoid of leaves except for a few terminal simple or trifoliolate leaves around the compound umbels, occasionally the branches bear pairs of opposite leaves when they may be up to 30 cm long; secondary rays c. 2-3 cm long; pedicels 4-10 cm. Petals 5, 11/,-2 mm long in bud. Stamens 5. Ovary turbinate, disk fleshy. Fruit spheroidal, c. 8 by 6 mm, calyx and stylopodium rather inconspicuous. Distr. Malesia: Central W. Sumatra (Mt Kerintji), throughout the Malay Peninsula (incl. Penang) and Borneo. Vern. Poko minta anak, Kedah, karon baru, Sumatra. Notes. The name ‘A. pinnatum’, misapplied by CLARKE, /.c., has been in general use for this species. However, this name was based on Panax pinnatum LAMK, which in turn was based on the Rumphian name ‘Scutellaria secunda’ ; this is quite a distinct plant (see under 6. Polyscias cumingiana). The three species A. maingayi, 16. A. montanum and 17. A. alternifolium are similar in having simpler inflorescences than 9. A. diversifolium. A. montanum can be distinguished from A. main- gayi by its more leathery leaves with more promi- nent nervation, and by the more woody and leafy flower-bearing primary branches of the inflores- cence. A. alternifolium is distinguished from both these species by its small, coriaceous leaflets with incon- spicuous nervation. Most specimens can be readily distinguished, but a few can be assigned to a species only doubtfully, usually because the material is inadequate. This is usually due to the junction between the vegetative (spiral phyllotactic) shoots and the flower-bearing branches being omitted. 1979 ] 13. Arthrophyllum kjellbergii PHriLIpsoN, Gard. Bull. Sing. 30 (1977) 309, f. 10. — Fig. 20d. Small tree, 10 m, branches c. 1 cm @, young parts covered by rufous tomentum. Leaves clustered towards the ends of the branches, multijugate, 25-30 by 16 cm; petioles c. 8 cm, 2 mm wide; petiolules c. 3-8 mm; leaflets elliptic or ovate- oblong, up to 8 by 3'/,; cm, chartaceous, margin minutely revolute, base rounded to cuneate often oblique, apex tapered to a blunt apiculum, or rounded and mucronate. Inflorescence a terminal cluster of c. 5—10 specialized branches (primary rays); primary rays c. 15-20 cm, 1'/,-2 mm wide, bearing near the middle an opposite pair of simple leaves or sometimes trifoliolate leaves with short flowering branches in their axils, and sometimes with a second pair higher up, and 2-3 similar leaves below the terminal cluster of secondary rays; secondary rays c. 12, c. 2'/,-3'/, cm long, each subtended by a small bract and bearing opposite caducous bracts near the middle, terminating in an umbellule of c. 12 flowers surrounded by an involucre of caducous bracts (1 mm long). Flowers known only in young bud. Fruit spheroidal, c. 5 by 4 mm, calyx and stylopodium prominent; pedicel 5-6 mm. Distr. Malesia: SE. Celebes (Kendari). Ecol. Primary rain-forest, 50-150 m. Note. The small leaves and relatively simple inflorescences are distinctive. 14. Arthrophyllum cenabrei Merr. Philip. J. Sc. 20 (1922) 417; En. Philip. 3 (1923) 235; PHILIPsoN, Gard. Bull. Sing. 30 (1977) 309. Glabrous tree, c. 10 m, ultimate branches c. 5mm @. Upper leaves pinnate, up to 10 cm long, leaflets mostly 5, sometimes 3, or the uppermost reduced to simple leaflets, the rachis and petiole c. 4 cm; leaflets mostly elliptic, 4'/,-6 by 2!/,-31/, cm, chartaceous to subcoriaceous, very shortly and obtusely acuminate, base acute, brownish oliva- ceous and slightly shining when dry, nerves 3-4 pairs, slender; petiolules 5-10 mm. Peduncles c. 4 cm, umbellately arranged at the tops of the branchlets, usually however with solitary inflorescences in the axils of the uppermost leaves, thus forming a somewhat leafy inflorescence. Fruits 5-8 in each umbel, ovoid, c. 7 mm @; pedicels 8-10 mm. Distr. Malesia: Philippines (Cebu; FB 28343, type, not seen). Ecol. On slopes at 600 m. Vern. Bingliu, C.Bis. Note. No specimen of this species has been located. The above description is taken from MERRILL’s original account. In placing this species in the key, it has been assumed that the inflorescence branching is relatively simple. 15. Arthrophyllum pulgarense ELmer, Leafl. Philip. Bot. 7 (1915) 2551; Merr. En. Philip. 3 (1923) 235; ARALIACEAE—I (Philipson) 65 PHILIPSON, Gard. Bull. Sing. 30 (1977) 311, f. 9. — Fig. 20c. Small tree, branches c. 1 cm @, young parts with red tomentum, becoming glabrous except on the ovaries. Leaves clustered towards the ends of the branches, leaflets c. 6 pairs, c. 22 by 8 cm; petioles c. 6 cm, 3 mm @; petiolules 5-6 mm; leaflets elliptic to rotund, c. 4 by 23/, cm, coriaceous, margin revolute, rounded to broadly cuneate, apex rounded or abruptly tapered to a short obtuse apiculum. Inflorescence a terminal cluster of specialized branches (primary rays); primary rays c. 6-10 cm, 3-4 mm @, bearing near the middle an opposite pair of simple rotund leaves, sometimes with flowering branches in their axils, and with a whorl of similar leaves below the terminal cluster of secondary rays; secondary rays c. 6-8, 2-4 cm long, articulated about the middle with scars of bracts or bearing a pair of small simple leaves terminating in an umbellule of c. 8-12 flowers; bracts caducous. Calyx with indistinct teeth. Petals 6, 2'/, mm long (in bud). Stamens 6, anthers curved. Ovary obconical, 2 mm long, furfuraceous. Fruit ellipsoid, c. 8 by 5 mm, the stylopodium forming a beak with stigma and calyx; exocarp fleshy. Distr. Pulgar). Ecol. Common in montane forest on Mt Pulgar. Note. The coriaceous, small, often rotund leaf- lets are characteristic. Malesia: Philippines (Palawan: Mt 16. Arthrophyllum montanum RIDL. J. Fed. Mal. St. Mus. 4 (1909) 24; Fl. Mal. Pen. 1 (1922) 886; PuHILIPsoN, Gard. Bull. Sing. 30 (1977) 311, f. 8. — A. nitidum RipL. J. Fed. Mal. St. Mus. 7 (1916) 42; Fl. Mal. Pen. 1 (1922) 886. — A. ovatum RIpDL. J. Fed. Mal. St. Mus. 7 (1916) 42; Fl. Mal. Pen. 1 (1922) 886. — Fig. 19d, 20b. Shrub or small tree to 6 m, unbranched or sparingly branched, rufous-tomentose on the young parts, becoming glabrous. Leaves tufted at the ends of the branches, multijugate, c. 30-55 by 12-22 cm; petiole 9-21 cm, 3 mm @; petiolules c. 10-15 mm; leaflets elliptic or oblong, 6-10 by 21/4 cm, coriaceous or chartaceous, base cuneate, apex with a short blunt apiculum, margin entire, revolute, the few principal lateral veins usually rather prominent; the leaves associated with the umbels usually unifoliolate, broadly ellip- tic to rotund, with a petiole to 4'/, cm. Inflores- cences on specialized leafy branches either in terminal clusters or axillary in the upper leaves; branches 10-30 cm, leaves mostly simple in opposite pairs, usually without flowering branches in their axils, branches ending in a whorl of simple leaves surrounding a compound umbel; primary rays c. 5-15(-20), 2-6 cm, with scars of caducous bracts; pedicels 4-8 mm. Petals (4-)5, 2 mm long in bud. Stamens (4-)5. Ovary turbinate; disk fleshy, stigma + sessile. Fruit spheroidal, c. 8 by 66 FLORA MALESIANA [ser. I, vol. 91 Fig. 24. Gastonia serratifolia (MiQ.) PHILIPSON. a. Habit, x */3, b. lower bud, c. flower, d. fruit, both x5 (a—c SCHMUTZ 3612, d VAN ROYEN 4090). Drawn by HELENE MULDER. 1979] ARALIACEAE—]I (Philipson) 67 5 mm, calyx and stylopodium rather prominent. Distr. Malesia: Malay Peninsula (Kedah to Selangor). Ecol. Primary forest ascending to the montane zone, and in second-growth, 250-1500 m. Note. Similar to 12. A. maingayi but dis- tinguished by the more leathery leaves and by the leafy inflorescence branches which are usually absent in A. maingayi. 17. Arthrophyllum alternifolium MAINGAY ex RIDL. Fl. Mal. Pen. 1 (1922) 886; PHILipson, Gard. Bull. Sing. 30 (1977) 311, f. 11. — A. pinnatum CLARKE, FI. Br. Ind. 2 (1879) 734, p.p., excel. basionym; Kina, J. As. Soc. Beng. 67, ii (1898) 59, p.p. — A. alternifolium MAINGAY ex CLARKE, FI. Br. Ind. 2 (1879) 734, nomen in synon. — Fig. 20e. Slender, sparingly branched shrub to 2 m, rufous-tomentose on the young parts, becoming glabrous. Leaves tufted at the ends of the branches, multijugate, c. 20-25(-30) by 9-12(-15) cm; petiole terete, c. 3-6(-9) cm, 2-3 mm @; petiolules c. 2 mm; leaflets ovate, elliptic or lanceolate, 31/,4(-6) by 1-2(-2'/,) cm, coriaceous, base cuneate, apex acuminate to caudate, obtuse, margin revolute, veins obscure; leaves associated with the umbels (if any) reduced, with fewer leaflets or unifoliolate, sometimes broadly ovate. Inflorescence usually a terminal compound umbel, occasionally a whorl of leafy branches (5-14 cm long) (leaves usually simple in opposite pairs), each ending in a compound umbel; peduncle 1'/,— 4'/, cm, with one or more usually caducous simple (or trifoliolate) leaves at the apex; primary rays c. 5, 3-4'/, cm, with scars of caducous leaves about the middle, each ending in an umbellule of c. 12-25 flowers, pedicels 5-8 mm. Calyx a rim or minutely 4-5-dentate. Petals 4-5, 2 mm long in bud. Stamens 4-5. Ovary turbinate; disk fleshy, stigma + sessile. Fruit spheroidal, c. 5 by 5 mm when dry, calyx and stylopodium small. Distr. Malesia: Malay Peninsula (Johore: Mt Ophir; Pahang, Selangor, and Malacca). Ecol. In shady montane forest, with Rhodo- dendron and Dacrydium, 900 m and above. Note. The small, coriaceous, often apiculate leaflets are characteristic. Although collected most frequently on Mt Ophir it occurs on other high ridges in southern Malaya. Excluded Arthrophyllum ceylanicum Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 27 (type in L), is according to VAN STEENIS, Rec. Trav. Bot. Néerl. 24 (1927) 819 = Oroxylum indicum (L.) Kurz (Bignoniaceae). Arthrophyllum reticulatum Bui. ex Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 27 (type in L), is according to MIQUEL (/.c. 318) and VAN STEENIS (vide supra) = Oroxylum indicum (L.) Kurz (Bignoniaceae). 9. GASTONIA Comm. ex LAMK, Encycl. 2 (1786) 610; Mig. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 5; HARMS in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 43; Hurcu. Gen. FI. Pl. 2 (1967) 68; PHILIPSON, Blumea 18 (1970) 491, 497, f. 1-10. — Tetraplasandra (non A. GRAY) Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 4; HARMs in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 29, p.p.; Nachtr. 2 (1900) 253. — Indokingia HEMSL. in Hook. Ic. Pl. (1906) t. 2805. — Peekeliopanax HARMS, Notizbl. Berl.-Dahl. 9 (1926) 478, fig. — Fig. 24, 26, 27. Trees unarmed with thick branches, glabrous or tomentose. Leaves large, im- paripinnate, exstipulate; rachis articulated; leaflets in pairs entire or crenate; petiole terete, with clasping base. Flowers in umbellules which are arranged race- mosely, or in verticils, on strong inflorescence branches; pedicels not articulated below the ovary. Calyx forming a continuous rim with an entire or indistinctly denticulate margin. Corolla of 5-13 free petals or calyptrate, fleshy, valvate. Stamens either equal in number to the petals or up to several times as many; filaments usually rather short and thick; anthers large, often irregularly lobed, dorsifixed. Ovary inferior, broadly obconic, cells 7-22; disk fleshy with stylar processes equal in number to the cells arising from its centre. Fruit a spherical berry (strongly ribbed when dry), with an indistinct calyx rim, a flattened disk, and a 68 FLORA MALESIANA [ser. I, vol. 9! prominent stylopodium bearing a ring of radiating stigmatic arms; exocarp fleshy, endocarp crustaceous. Endosperm with smooth surface. Distr. About 10 spp. in East Africa, Madagascar, the Seychelles and Mascarenes, Malesia, and the Solomon Is. Ecol. Primary and second-growth forest, or in open country, from sea-level to lower montane zone. Taxon. I have amply analyzed and discussed the affinities of Gastonia (Blumea 18, 1970, 497). I have come to the conclusion that within its alliances Gastonia is the only genus west of Samoa; in Polynesia there are three other closely allied genera. KEY TO THE SPECIES 1. Corolla with free petals. Ovary cells and style arms 6—9(-12). Whole plant glabrous. Leaflets usually entire or sparsely crenate. Main inflorescence branches borne along an elongated axis with caducous bracts; peduncles of the umbellules mostly aggregated into pseudo-whorls 1. G. serratifolia 1. Corolla calyptrate. Ovary cells and style arms 12-18(—22). Young parts with scurfy tomentum, per- sisting on the ovary and bracts. Leaflets strongly crenate. Main inflorescence branches borne sub- umbellately on a short axis with persistent bracts; peduncles of the umbellules scattered 1. Gastonia serratifolia (MIQ.) PHILIPSON, comb. nov. — Arthrophyllum serratifolium Mia. Sum. (1861) 341, type from Sibolga, leg. TEYSMANN (in U). — G. papuana Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 5; Harms, Bot. Jahrb. 56 (1921) 408; PHitipson, Blumea 18 (1970) 492, 500 f. 3. — Tetraplasandra paucidens Mia. Ann. Mus. Bot. Lugd.-Bat. 1 (1863) 4; Koorb. Minah. (1898) 488. — G. eupteronoides T. & B. Nat. Tijd. N. I. 25 (1863) 416. — Polyscias papuana (MiIQ.) SEEM. J. Bot. 3 (1865) 181. — Tetraplasandra koordersii Harms, Ann. Jard. Bot. Btzg 19 (1904) 12; Ic. Bog. 2 (1906) t. 178. — Tetraplasandra philippinen- sis Merr. Philip. J. Sc. 1 (1906) Suppl. 219; En. Philip. 3 (1923) 222. — G. winkleri Harms in Fedde, Rep. 15 (1917) 20. — Tetraplasandra solomonensis PHILIPSON, Bull. Br. Mus. Nat. Hist. Bot. 1 (1951) 11. — Fig. 24. Shrub or small tree, sometimes epiphytic, up to 27 m, with clear bole to 15 m, dbh 34 cm, crown sparsely branched with leaves crowded at the ends of the branches, glabrous. Leaves up to 80 by 20 cm; petiole c. 13 cm; leaflets c. 10 pairs, petiolules up to 1 cm; blade oblong, ovate or lanceolate, middle leaflets 8—-14(-18) by 2!/,- 31/,(-8) cm, chartaceous, entire and subrevolute, or with a few obscure, more rarely several promi- nent crenations, apex rounded and bluntly apicu- late or tapering and acute, base broadly cuneate, midrib prominent. Inflorescence terminal, glab- rous, with a stout primary axis 15-25 cm long, bearing scattered or clustered branches along its length and ending in an umbel of c. 10 branches; bracts caducous; secondary branches 30-40 cm, bearing numerous subverticillate peduncles along their length and ending in an umbel; peduncles c. 3 cm, somewhat longer in fruit; pedicels 1-21/, cm, forming umbellules of c. 10 flowers. Flower buds (when dry) c. 7 by 3 mm. Calyx rim 2. G. spectabilis undulate. Petals 5-9, slightly fleshy, fully separated. Stamens variable in number, (7—)14—55; filaments short; anthers broad and irregularly lobed, variable in size. Ovary glabrous, smoothly rounded below, slightly constricted below the calyx; cells 6-12, usually c. 9, disk with a prominent rim and, at anthesis, a central boss formed by closely appressed subulate style arms equal in number to the ovary cells. Fruit c. 9 by 7 mm (without stylopodium), the flattened stylopodium ending in a ring or double row of radiating subulate stigmatic arms, black when ripe, the fleshy exocarp enclosing com- pressed crustaceous pyrenes. Distr. Solomon Is.; in Malesia: Central W. Sumatra (Sibolga, Enggano I.), Malay Peninsula (Johore), Sunda Straits (islet Dwars in den Weg), West Java, Lesser Sunda Is. (Sumba, Timor, Wetar, Flores), W. Borneo (east of Pontianak, Karimata), N. Borneo (Sabah), Philippines Fig. 25. Distribution of Gastonia in Malesia. Known localities of G. serratifolia (M1Q.) PHILIP- SON: black dots; range of G. spectabilis (HARMS) PHILIPSON: broken line. 1979] ARALIACEAE—I (Philipson) 69 SEQ NTT RC Fig. 26. Gastonia spectabilis (HARMS) PHILIPSON, two inflorescences visible below the leaves (Photogr. PHILIPSON, Kassam Pass, E. New Guinea, 1968). 70 FLORA MALESIANA [ser. I, vol. 91 (Palawan, Balabac), Celebes (Lepo-Lepo, Luwuk, Minahassa), N. Moluccas (Talaud Is.), New Guinea (Vogelkop, NW. & SW. Irian, Schouten I., Waigeu I.). Fig. 25. Ecol. Primary and secondary forest, or in open country, usually at low altitude and often on the shore or sea-cliffs, but ascending to 1000 m. EyMa noted it to be a characteristic tree on Mt Tam- bunan, Luwuk, E. Celebes. Vern. Bajur talang ékoaho, M, Enggano, jarum, Sabah, raka, Sumba, kre, wangka, Flores, lampo pad, bungku, Celebes, bufigio, Palawan, lantora, Talaud Is., mansnongoree, mantsenongor, Schouten I., raauwrack, ara-orach, Vogelkop, Maibrat. Note. The entire or serrate nature of the leaf margins, the number of ovary cells, and especially the number of stamens are variable characters but show no discernable geographical segregation. The outer bark is described as light brown, with small oblong brittle scales. Inner bark, leaves and inflorescences with copious sticky juice. Wood soft white. The petals are variously described as purple, light green, and white. 2. Gastonia spectabilis (HARMS) PHILIPSON, Blumea 18 (1970) 494, pl. 1. — Peekeliopanax spectabilis Harms, Notizbl. Berl.-Dahl. 9 (1926) 478, fig.; PHILIPSON, Blumea 18 (1970) 500 f. 2. — G. bori- diana Harms, Bot. Jahrb. 69 (1938) 282. — Fig. 26, 27. Tree up to 40 m high, clear bole to 28 m, dbh 1.75 m, crown sparsely branched with the branches whorled or regularly forked and the leaves crowded at the ends of the branches, all young parts with scurfy indumentum more evident in dried material. Leaves up to 80 by 30 cm; petiole c. 15 cm, with some tomentum remaining at the joints; leaflets c. 11 pairs on petiolules c. 2-8 mm, oblong or elliptic, middle leaflets 10-15 by 4-6 cm, charta- ceous when dry, prominently crenate, apex nar- rowed to a short blunt apiculus, base rounded or truncate. Inflorescence in forks well below the leaves, when in bud covered with large scurfy cataphylls, and when mature with radiating branches forming clusters c. 130 cm @. Primary axis short (5-10 cm) with persistent bracts; secondary branches numerous, radiating, up to 65 cm, bearing small, persistent, scurfy bracts and peduncles scattered along their length and clustered in a terminal umbel; peduncles c. 1'/,-5 cm; pedicels 14/,-1'/, cm, forming umbellules of c. 5-12 flowers. Flower buds c. 8 by 5 mm when dry (fresh c. 12 by 8 mm). Calyx rim straight. Petals 6-12, very fleshy, incompletely separated (often splitting into c. 5 lobes). Stamens 25-66, often c. 35, filaments short; anthers broad and irregularly lobed, variable in size. Ovary with a short dense indumentum, smoothly rounded below, cells usually c. 16, very rarely fewer than 10, as many as 22; disk at anthesis with a prominent rim and a central boss formed by closely appressed subulate Fig. 27. Gastonia spectabilis (HARMS) PHILIPSON. Young tree, free bole 19 m, 47 cm @, crown 10 m, longest leaves 2 m, not yet flowering (Photogr. G. PEEKEL, New Ireland, Ugana, 1940). styles equal in number to the cells. Fruit c. 8 by 10 mm (dry and without stylopodium), the flattened disk with a prominent stylopodium ending in an elliptical ring of radiating subulate stigmatic arms; the fleshy exocarp enclosing compressed crusta- ceous pyrenes. Distr. Solomon Is.; in Malesia: New Guinea (Vogelkop, NW. Irian, Papua New Guinea), Bismarcks, and New Ireland. Fig. 25. Ecol. Primary and secondary rain-forest, and in cultivated areas, 200-2000 m. Vern. Amoriga, bekuak, djak, ntjier, tuju, Vogelkop, bohko, boinga, gabiel, jamwa, mestic, sikoto, tubat, tumbala, Madang Distr., waki, W. Highlands, aita, E. Highlands, kuhuh, Papua. Note. Possibly the largest araliad known. The regular habit of branching results from the vegeta- 1979] ARALIACEAE—I (Philipson) 71 Fig. 28. Polyscias nodosa (BL.) SEEM. a. Leaf and part of inflorescence, x */s, D. leaflet, x 2/3, c. branch of inflorescence, x 2/3, d. umbellule, e. flower bud and ditto in LS, x 8, f. ovary and ditto in CS, x 8, g. stamens (VERSTEEGH BW 3868). Drawn by HELENE MULDER. 7/p2 FLORA MALESIANA [ser. I, vol. 9! tive shoots springing in pairs or whorls from below terminal inflorescence buds. Several flushes of growth occur as an inflorescence bud matures, so that at anthesis the inflorescences are situated in forks well below the leafy crown, with a succession Bole without buttresses or with buttresses 1 m high and 2 m wide. Outer bark brown with prominent pustular lenticels and small shallow fissures. Exudate from cuts abundant, clear and aromatic. Wood soft. Flowers cream, stamens yellow. Ripe fruit dark red-brown. of younger inflorescence buds in higher forks. 10. POLYSCIAS J. R. & G. Forster, Char. Gen. (1776) 63, t. 32; DC. Prod. 4 (1830) 257; SEEM. J. Bot. 3 (1865) 179; Bru. in B. & H. Gen. PI. 1 (1876) 941; Harms in E. & P. Nat. Pfi. Fam. 3, 8 (1894) 43; Koorp. Atlas 4 (1916) f. 677-680; MerR. Int. Rumph. (1917) 408; Harms, Bot. Jahrb. 56 (1921) 409; Merr. En. Philip. 3 (1923) 233; Hutcu. Gen. FI. Pl. 2 (1967) 75; BERNARDI, Candollea 26 (1971) 13; PHILIPSON, Blumea 24 (1978) 169. — Eupteron Mia. Pl. Jungh. 3 (1855) 423; Fl. Ind. Bat. 1, 1 (1856) 762; Hutcu. Gen. FI. Pl. 2 (1967) 68. — Nothopanax Mig. P\. Jungh. 3 (1855) 425; Bonplandia 4 (1856) 139; Fl. Ind. Bat. 1, 1 (1856) 765; SEEM. Fl. Vit. (1866) 114; Merr. Int. Rumph. (1917) 409; En. Philip. 3 (1923) 233. — Irvingia F.v.M. Fragm. 5 (1865) 17, non Hook. f. 1860. — Kissodendron SEEM. J. Bot. 3 (1865) 201; ibid. 6 (1868) 129; Harms in E. & P. Nat. Pfl. Fam. 3, 8 (1894) 45; Bot. Jahrb. 56 (1921) 412; Hutcu. Gen. Fl. Pl. 2 (1967) 75. — Palmervandenbroekia Gisss, Arfak (1917) 162; Hutcu. Gen. Fl. Pl. 2 (1967) 75. — Gelibia Hutcu. Gen. FI. Pl. 2 (1967) 57. — Fig. 28, 30. Unarmed shrubs or trees, glabrous or furfuraceous. Leaves imparipinnate or 2-3-pinnate (rarely unifoliolate) with an elongated or short sheathing base; rachis articulated; petiole terete; leaflets in pairs, entire, crenate or dentate. Inflorescence terminal, often large, a panicle, corymb, or compound rarely simple umbel. Flowers in umbellules, capitula, or racemose; pedicels articulated below the ovary. Calyx a rim with an undulate or dentate margin. Petals 4-5(-8 or more), valvate. Stamens equal in number to the petals; anthers dorsifixed. Ovary inferior, 4-5(-8 or more)-celled; disk fleshy; styles either free and recurved (at least in fruit) or joined to form a beak-like stylopodium. Fruit a spherical or ovoid drupe, crowned by the persistent calyx rim and the styles or stylopodium; exocarp fleshy, endocarp chartaceous. Endosperm with an uneven surface, fissured, or rarely smooth. Distr. About 100 spp. throughout the tropics of the Old World (incl. Australia), and the Pacific Islands; in Malesia 23 spp. of which 3 adventive: rare in West (none native in Sumatra and Malaya), more common in East, the majority in New Guinea. Fig. 29. Ecol. Primary or second-growth forest, from sea-level to 2650 m. Note. Araliads with pinnate leaves and an articulated pedicel are here regarded as forming one genus, divided into several sections. These sections have formerly been segregated as genera, but I follow BERNARDI, /.c., in uniting them. The most distinctive is sect. Polyscias, characterized by an elongated leaf- sheath. Several of the species of this section are cultivated and have a number of cultivars. KEY TO THE SECTIONS 1. Leaf-sheath elongated, extending along the petiole for about '/, of its length (Spp. 1-7) 1. Sect. Polyscias 1. Leaf-sheath short or obsolete, restricted to the base of the petiole. 2. Style arms spreading, at least in fruit. 1979] 3. Flowers arranged racemosely (Sp.8)... . 3. Flowers arranged in umbellules or capitula (Spp. 9-14)... 2... . 2. Style arms erect, fused, forming a beak in fruit. 4. Inflorescence large, + as long as the leaves (Spp. 15-19) 4. Inflorescence much shorter than the leaves (Spp. 20-23). ARALIACEAE—I (Philipson) 13 . . . 4 Sect. Kissodendron 5. Sect. Palmervandenbroekia 1. Section Polyscias PHILIPSON, Blumea 24 (1978) 169. Aromatic, glabrous shrubs or small trees, often cultivated. Leaf-sheath elon- gated along the petiole for '/,'/, of its length. Styles spreading, at least in the fruit. Distr. Polynesia, Queensland, Malesia, and SE. Asia. KEY TO THE SPECIES femlbeaves 2—S-pinnate.s im cit,