She. * SS +} oe eres ae ate $? *, hte 3 un ; Library Arnold Arboretum IME FL] é UW Harvard University a . X ae-< 7, -~ « * . Som (Ct : 7 ’ - eee og SS: 5 _ 5, es —s -* 7 Ss 4 be ss . . oa pf ta 7 + * . z. > J 2 M a ae — F ; ‘ woe | . ~ ” a pr 7 set te | » oo « s No wih bs Nee Se oe as, ns 0 ee oe Pe Pa ad ry - A oud A 4 om - et ae i E = . Lat wee ~ i cas yon . ee > om — , Fe - * - ak * 4, (ads, : > Th a. - a iY - ) Pa ~ >a : a ee A as ae, 2 q ‘ ee ie ~~ - a ne . “q4.7 = * THE GARDENS’ BULLETIN STRAITS SETTLEMENTS Volume VIII October, 1934—October, 1935 To be purchased at the Botanic Gardens, Singapore Published by Authority. SINGAPORE: PRINTED AT THE GOVERNMENT PRINTING OrFice, SinGAPoRE, BY W. T. Cuerry, GoverNnMENT PRINTER. 1935 ane CONTENTS PAGE Part 1: 10th October, 1934 (pp. 1-68). SYMINGTON, C. F.: Notes on Malayan Dipterocar- paceae II (with plates nos. 1-10, & Index) as 1 ALSTON, A. H. G.: The genus Selaginella in the Malay Peninsula (with Index) ri 41 BAKER, J. A.: Notes on the Biology of " Macaranga spp. (with plates nos. 11-15) ; 63 Part 2: 26th January, 1935 (pp. 69-164). CARR, C. E.: Some Malayan Orchids V (with an Index) 69 MERRILL, E. oe Raditionk and Bawectisns to Ridley’ s Flora of the Malay Peninsula Sgt coke CORNER, E. J. H.: A Nectria Parasitic on a Liverwort: with Further Notes on Neotiella Crozalsiana 135 FURTADO, C. X.: Araceae alesis 145 FURTADO, C. X.: Palmae Matic ‘ena Wendlandiana Scheff. or Nenga a oy) Wendl.? i 159 Anonymous: Ob iaary: Walter Pom nc: AOE Part 3: 24th June, 1935 (pp. 165-264). CARR, C: E.: Two Collections of Orchids from British North Borneo Part 1 (with Index) A269 FURTADO, C. X.: Palmae Malesicae IJI—Notes on Some Malayan Calami (with Index to Collectors’ numbers) 241 SMITH, J. J.: A New Pridodeiiron nee Gunong Tahan fs 262 Part 4: 28th October, 1935 aide 265- ay. SYMINGTON, C. F.: Notes on Malayan Dipterocar- paceae III (with plates nos. 16-28, & Index) .. 265 HoLTTuM, R. E.: The Tree-ferns of the Malay Peninsula (with plates nos. 29-36, and Index to Collectors’ numbers) of 293 FURTADO, C. X.: Palmae Malesicae bee ren as described in Blanco’ s Flora de Filipinas rae A | FURTADO, C. X.: Palmae Malesicae V—Notes on Some Malayan Daemonorops (with plates nos. 37 & 38, and Index to Collectors’ numbers) 63909 Index to Volume VIII by Bi .. 369 @ ,? oe rocarpaceae II, by C.F, Syming- . Sees ee - ss : | i ee i a i F 2: tin the Malay" Penin ninéula, y ee ‘E et aon 7 Sap: ot x : 4 es asf SAUER vd Sr ew eee a» is . atc e ; + ; Ya a a ; ; J * ‘5 ol res, SE Nee Bat) J iy§ ethene 0 De) oer a THE GARDENS’ BULLETIN STRAITS SETTLEMENTS Svcl Vill 10th October 1994. Part 1 /NOTES ON MALAYAN DIPTEROCARPACEAE—IL. C. F. Symington, B.Sc. This paper continues the series commenced by my “Notes on Malavan Dipterocarpacee—lI’’ published by courtesy of the Director of Gardens, Straits Settlements in this Bulletin, Vol. VII, pt. 2 (1933). As previously, the species dealt with have been selected in no systematic sequence but are those concerned in various problems that have demanded attention from time to time. I have first discussed the genus Scaphula Parker. The study of this group involved some more or less intensive work upon all our species of Anisoptera, the results of which are thought to be worthy of publication. Next I have endeavoured to clarify the position with regard to the much confused species Hopea sangal, following my notes on which a description of its close ally H. semi- cuneata seems appropriate. Finally I have had the pleasure of examining the dipterocarps collected in Sarawak by Mr. Richards of the Oxford University Exploration Club. Two of these collections [Balanocarpus longiflorus (Brandis) Foxw. and Hopea laxa Sym.jare of peculiar interest as they raise the whole question of the status of the anomalous genus Balanocarpus. Atthough I do not feel that my study of the plants concerned in this group is sufficiently complete for me to give an authoritative and final treatise upon it, it seems desirable that a statement of my opinions should be given in justification of the publication of these two new binomials. As in my previous paper the concluding paragraph of my notes under each species occurring in the Malay Peninsula is addressed in particular to Malayan Forest Officers. I wish to take this opportunity of recording my grateful appreciation of the invaluable assistance I have received from the Director of the Royal Botanic Gardens, Kew, Dr. Burtt Davy of the Imperial Forestry Institute, Oxford, the Forest Botanist, Dehra Dun, the Curator of the 1 2 Herbarium, Calcutta, the Forest Botanist, Maymyo, Mr. Henderson of Singapore, Dr. van Slooten of Buitenzorg, the Curator of the Museum, Kuching, and the Acting Chief, Division of Botany of the Bureau of Science, Manila. I should like also to express my appreciation of the enthusiasm and interest of my colleagues in the Malayan Forest Service who are mainly responsible for the collection of recent herbarium material. Without such willing co-operation as has been accorded me, research of this nature would be impossible. Seaphula Parker in Fedd. Rep. 30: 325 (19382). {n uniting Vatica scaphula Dyer (=Hopea scaphula Roxb.) and Anisoptera glabra Kurz, Parker (l.c.) was of opinion that this interesting species could not be assigned to any of these genera and created for it the new genus Scaphula. As the recognition of this new genus would involve the transfer to another genus of two [A. scaphula (Roxb.) Pierre and A. laevis Ridl.] from among our six Malayan species of Anisoptera, which, as far as the forest officer or timber worker are concerned, form an essentially homogeneous group, it seems desirable to examine critically the evidence put forward in support of the new genus before making the transfer. Undoubtedly Scaphula is most closely allied to Anisoptera Korth. and the problem resolves itself into deciding whether A. scaphula and A. laevis should be removed to the new genus Scaphula or the generic conception of Anisoptera be amended to include them. Parker, fully aware of the close affinity of Scaphula and Antsoptera, has enumerated the differences between them as follows :— Scaphula Anisoptera 1. Calyx lobes | open imbricate in bud | | Stamens | 15-20, normally 15 those | 20-50 | opposite the sepals paired | and those opposite the petals solitary. Some or all | the stamens opposite the | petals may be paired with the additional stamens fully developed or more or less reduced* Connectives shortly mucronate long awned Style slender with small 8-lobed | stylopodium stout stigma passing gradually | and fleshy tipped into a stylopodium as in | with 3 short styles | Dipterocarpus without well-differ- entiated stigmas. * The presence of 20 stamens is evidently abnormal although not surprising. I have never met with more than 15 stamens in the flowers I have examined. bo he GO Gardens Bulletin, S.5. 3 One and 3 are not of great importance as generic characters. In the flowers of A. scaphula I find that the calyx lobes are actually slightly imbricate in bud. The - length of the connectival appendage is a very variable character within other genera of the Duipterocarpaceae although it is normally a reliable specific character. Occasional abnormal stamens with an awn-like appendage are to be found in A. scaphula. The number of stamens also is subject to great variation within some genera of this family including the genus Anisoptera (Scaphula excl.). In A. costata, A. marginata, and A. Curtisit the number ot stamens varies between 23 and 30 whereas in the flowers of A. megistocarpa examined by me there have been from 47-54 stamens. The main diagnostic character of Scaphula is the nature of the stylopodium. Reference to the figures (Pls. I—IV) will show how Anisoptera and Scaphula differ in this respect. In Anisoptera the ovary is surmounted by an erect, fleshy, sulcate stylopodium which is crowned with 3 {occasionally 4) short styles without well-differentiated stigmas. In Scaphula the ovary is surmounted by an enlarged flattened rugose disc. From this disc a slender columnar style rises rather abruptly and is terminated by three sessile stigmatic lobes. Scaphula differs from most species of Anisoptera also in being almost glabrous and in having subglobular flower buds. Against these differences must be considered the many characters that Anisoptera and Scaphula have in common and which are not shared by other genera in the family. The facies of A. scaphula and A. laevis is that of an Anisoptera. In form and bark characters and in the structure of the secondary xylem of the wood they agree with Anisoptera. The structure of the embryo and the germination of the seed is as in Anisoptera (vide Foxworthy, Mal. For. Rec. 10 p. 31) and they share the peculiar character of having young parts and the undersides of the leaves furnished with minute peltate scales (fugacious in A. scaphula). The fruits of Anisoptera and Scaphula are essentially the same and can be distinguished only by the persistent style and stylopodium or occasionally by persistent stamens. Whether or not Scaphula is worthy of generic rank must always be to some extent problematical but the writer is strongly opposed to the disruption of so characteristic and homogeneous a group as that formed by the described species of Anisoptera. He is of opinion that when the genus is next monographed slight alteration should be made in Vol. VII. (1934). 4 the generic definition of Anisoptera to include A. scaphula and A. laevis. Foxworthy (1. c. p. 93) has slightly altered the generic description of van Slooten [Bull. Jard. Bot. Buitenz. III, 8: 3 (1926) ] to embrace these species, but has — not allowed for their distinctive form of style and stylopo- dium nor for the fact that they are practically glabrous. The distinctive characters of A. scaphula and A. laevis should, however, be recognised by placing them in a section [Sect. Glabrae Heim, Recherches sur les Dipterocarpées, p. 38 (1892)] apart from the other known species of the genus Anisoptera. vAnisoptera scaphula (Roxb.) Pierre, Fl. For. Cochinch. sub t. 235 (1890) in obs.: Hopea scaphula Roxb. F1. Ind. 2: 611 (1832); Kurz, Prelim. Rep. For. Pegu, app. A: 19 et app. B: 29 (1875), et Fl. Brit. Burma 1: 121 & 547 (1877) : Anisoptera glabra Kurz in Journ. Asiat. Soc. Bengal 42, pt. 2: 61 (1873), Prelim. Rep. For. Pegu, app. A: 16 et app. B: 29 (1875), et FI. Brit. Burma 1: 112, (1877); Dyer in Hooke Brit. Ind. 1: 301 (1875) ; Brandis in Journ. Linn. Soe. 31: 41 (1895), et Ind. Trees 67, t. 29 (1911) ; Guérin ex Lecomte, F)}. Gen. l’indo-chine 1: 368 (1910); non Pierre, l.c. sub t. 235 et 253; non Ridl. in Agri. Bull. S. 5S. & F. M.-S. 1,. pt. 2: 60 (1901), et in Jourmeak. As. Soc. Str. Br. 54: 25 (1910): Vatica scaphula Dyer in l.c.,; King in Journ. As. Soc. Bengal 42, pt. 2: 127 (1893) in obs.; Brandis in Journ. Linn. Soc. 31: 182 (1895), et Ind. Trees 72 (1911): Anisoptera thurifera Ridl., Fi. Mal. Penins. 1: 219 (1922) ; Foxw. in Journ. Mal. Br. R. As. Soc. 5: 341 (1927) partim, Mal. For. Rec. 3: 78 (1927) partim (plate of leaves and fruit), et Mal. For. Rec. 10: 100 (1932), non Blume: Scaphula glabra Parker in Fedd. Rep. 30: 326 (1932) ; Cretzoiu in Acta Faun. Flor. Univers. Bot. 1, 9: 3 (1933) fig. lL & 2. Plate it The various descriptions published for this species under the names Vatica scaphula and Anisoptera glabra being hardly satisfactory, I have prepared the following with the help of our Peninsula collections and Burmese material kindly lent by Calcutta and Maymyo. Branchilets usually pale yellow-brown, glabrous, some- times covered with light brown scales. Leaves oblong, elliptic—oblong or obovate, gradually or abruptly acuminate, usually slightly cordate at base but may be rounded, truncate, or even sub-cuneate, usually about 12 cm. X 6 cm. but varying between very wide limits, glabrous, coriaceous, Gordens Bulletin, S.S. eo fl ee 2. 5 usually drying a light green-brown on both surfaces, the lower sometimes sparsely clothed with light-coloured scales ; midrib depressed above, elevate and yellow beneath; secondary nerves 15-24 pairs, usually at an angle of over 60° to the midrib, curved and anastomosing towards the margins, with frequent distinct intermediary nerves, prominent and usually yellow on both surfaces; fine reticu- lations evident on both surfaces, more prominent beneath; . petioles 2—4 cm., enlarged rugose and sulcate in upper half, frequently clothed with pale brown scales; stipules linear- lanceolate acuminate, 5 mm. long (may exceed 2 cm. in seedlings), densely clothed with brown scales, caducous. Panicles axillary and terminal, pendulous, racemose, up to Z0 cm. long, puberulous or densely tomentose towards the ends; ultimate branchlets distant, 1-5 cm. long, each bearing 2-7 distichous flowers about 5 mm. apart. Flowers globular in bud, slightly pointed at apex, 6 mm. long including the pedicel, foetid or sickly smelling; pedicel 1.5 mm., grey or tawny tomentose; calyx cup shallow. Sepals slightly imbricate in bud, tomentose on both surfaces, 2 large oblong obtuse, 3 smaller ovate acuminate. Petals slightly twisted in bud, reflexed when expanded, ovate-oblong, glabrous or sometimes sparsely puberulous along the margins, white, falling singly. Stamens 15 (15-20, Parker), pairs opposite the sepals, solitary opposite the petals; filaments flattened, shorter than the anthers; anthers somewhat hastate, posterior cells smaller than the anterior; appendage to connective mucronate or rarely extended into a short awn. Ovary deeply embedded in the receptacle and surmounted by a rugose, tomentose, disc-like or sub-conical stylopodium; style columnar, giabrous; stigma hardly enlarged, 3-lobed. fruit: stalk 2-8 mm. long, rugose, puberulous; calyx tube enclosing and adnate to the nut, globose when mature, about 1.25 cm. diam., more or less verrucose, glabrous; large calyx lobes usually linear spathulate, 14 cm. X 2.5 em. but very variable, somewhat verrucose; 3 small lobes, linear, acuminate or rounded at apex, up to 2.5 cm. long, also very variable; nut crowned with the enlarged tomentose disc-like stylopodium and erect style (a few stamens occasionally persist for a long time but have fallen before the fruit is mature). Embryo: cotyledons unequal; dorsal cot. occupy- ing the entire upper portion of the ovary, large; placental cot. lying below the dorsal, smaller; radicle medial and horizontal. Parker, in his notes on the genus Scaphula, has given an account of the early history of this interesting species which has been known in Burma variously as Vatica scaphula and Anisoptera glabra. As Parker points out, Vol. VIII. (1934). 6 although Kurz realised that these two species belonged to the same genus he did not realise that they are actually conspecific. Thus, although Kurz in 1877 proposed and Pierre in 1890 used the combination Anisoptera scaphula, it was not until 1932 that A. glabra Kurz was reduced to synonymy with V. scaphula Dyer by Parker. The last author, however, being (as were both King and Brandis) fully aware of the anomalous nature of the species, was not of opinion that it could remain in either of the genera to which it had previously been assigned. He recognised its close affinity to Anisoptera but considered that the floral characters were such as to warrant the creation of a new genus Scaphula. My reasons for retaining this species in the genus Anisoptera are given fully in my notes upon the genus Scaphula (p. 2). | In 1901 and again in 1910 Ridley recorded Anisoptera glabra from the Malay Peninsula but the collections examined by him were actually A. laevis Ridl., a mistake which he corrected in 1922 when publishing the description of his new species in the Flora of the Malay Peninsula. These collections, originally referred by Ridley to A. glabra, were examined by Foxworthy who, in 1911 (Phil. Journ. Sc. Bot. 6: 257), identified them with A. thurifera. Our earliest authentic collection of A. scaphula from the Peninsula is C. F. 453 (Kew No. 32) which was collected in Plus Forest Reserve in 1912 and is the specimen referred by Ridley to A. thurifera in his flora. This collection has fruit hardly mature and is in a condition in which it is difficult to distinguish these two species. Thus Foxworthy’s A. thurifera from the Malay Peninsula was actually A. laevis while Ridley’s A. thurifera was A. scaphula. In 1927 (Journ. Mal. Br. R. As. Soc. 5: 341 and Mal. For. Rec. 3: 78) Foxworthy included under A. thurifera collections of both A. laevis and A. scaphula. This is evident from the fact that he says that “A. thurifera Bl. is the commonest species’, but gives a plate of leaves and fruit based on C. F. 11685 which is A. scaphula. His former conception of the species is explained by Foxworthy in Mal. For. Rec. 10 (1932) but here, as in his previous two publications, he continues to record A. glabra Kurz as a synonym of A. thurifera (Bleo.) Bl. This is quite unjustified although sterile specimens of these species are undoubtedly rather alike. We may now say definitely that A. thurifera Bl. has never been collected in the Peninsula. Gardens Bulletin, S.S. _ Collections examined— BURMA: Kurz 116 & 2117 (Auth. spec. of A. glabra Kurz) ; King’s coll. 144 & 164; Brandis 114; Calcutta Nos. 885, 952, 966, 8089, 8358 & 8397; Maymyo Nos. 6505 (Field No. 2545), 7564 (Field No. 5062) & 14146 (Field No. 11892) ; For Dept. Tenas. Cire. 351; Dehra Dun Nos. 1188, 5023 & 389751. MALAY PENINSULA: KEDAH: C. F. 217385. PreRAKk: C. F: 453° (Kew No. 32), 10378, 30772 & 30807. KELANTAN: C. F. 32722. TRENGGANU: C. F. 10242. Pananc: C. ¥. 2399. SELANGOR: C. F. 9993, 11685, 14357, 22008, 22010 & 24962. NEGRI SEMBILAN: C. F. 631. Anisoptera scaphula is a large smocth-leaved form of mersawa not uncommon towards the north of the Malay Peninsula. It is identical with the tree known in Burma by the vernacular names of boilshara, taungsagaing, kaung humu, etc. and producing the timber known as “mascal wood”. The former identification of this tree with Anisoptera thurifera of the Philippines is shewn to be wrong. . Anisoptera levis Ridl., Fl. Mal. Penins. 1: 219 (1922); Brandis in Journ. Linn. Soc. 31: 41 (1895) (in obs. sub A. glabra); Foxw., Mail. For. Rec. 10: 101 (1932); Anisoptera glabra Ridl. in Agri. Bull. S. S. & F. M. S. 1, pt. 2: 60 (1901) et in Journ. R. As. Soc. Str. Br. 54: 25 (1910) ; non Kurz: Anisoptera thurifera Foxw. in Phil. Journ. Se. Bot. 6: 257 (1911) partim, quoad spec. mailay., in Journ. Mal. Br. R. As. Soc. 5: 34 (1927) partim et Mal. For. Rec. 3: 78 (1927) partim (plate of bole and bark); non Blume, non Ridl., Fl. Mal. Penins. l.c. Plate II. The following description is prepared as an aid to comparison of this species with its very close ally A. scaphula together with which it forms a group distinct from all other known species of the genus. Branchlets slightly sulcate towards the ends when dry, glabrous or more or less densely ‘clothed with reddish scaies. Leaves oblong to elliptic—oblong or lanceolate, gradually or abruptly acuminate, rounded or subcuneate at base, 8-10 cm. X 2.5—4.5 cm., coriaceous, when dry green-brown above, reddish beneath owing to the more or less dense covering of scales; midrib depressed above, elevate and reddish beneath ; secondary nerves 10-15 pairs at an angle of about Vol. VIII. (1934). a ees 8 50° to the midrib, curved and anastomosing towards the margins with occasional distinct intermediary nerves, prominent on both surfaces, elevate beneath; fine reticula- tions conspicuous particularly upon the lower surface; petioles 1.5-2.5 cm., enlarged and rugose in the upper half, usually clothed with minute reddish scales; stipules lanceo- late, densely squamate, caducous. Pamnicles axillary and terminal, pendulous, racemose, up to 18 cm. long, densely stellate fulvous tomentose towards the ends of the branchlets; ultimate branchlets distant, up to 2 cm. long, each bearing 2-8 closely set distichous flowers. Flowers globular in bud, pointed at the apex, 5 mm. long including the pedicel; pedicel rugose, 1.5 mm., fulvous tomentose; calyx cup shallow. Sepals valvate, 2 oblong obtuse, slightly larger than the remaining 3 which are deltoid acute, all fulvous tomentose on both surfaces. Petals slightly twisted in bud, reflexed when expanded, oblong, obtuse, glabrous on both surfaces or sparsely ciliate outside on the portion exposed in bud, pale yellow (Foxworthy). Stamens 15 (15-20, Foxworthy) in pairs opposite the sepals, solitary opposite the petals; filaments flattened, less than half the length of the anthers; anthers oblong, broader at the base, posterior celis smaller than the anterior; appendage to connective mucronate. Ovary deeply embedded in the receptacle, surmounted by the rugose tomentose disc-like stylopodium; style columnar, sulcate, glabrous; stigma slightly enlarged, obscurely 3-lobed (see Pl. Il). Fruit: stalk 2-3 mm. long, verrucose, sub-glabrous; calyx tube enclosing and adnate to the nut, globose when mature, 1—1.5 cm. diam., minutely verrucose, glabrous; 2 large calyx lobes linear spathulate, usually about 10 cm. X 1.5 cm., but may be up to 14 cm. long; 3 small lobes linear acuminate, 1-2.5 em. long; nut crowned by the enlarged tomentose disc-like stylopodium and persistent style. Embryo as in A. scaphula. The earliest collection of this species is probably Alvins (Cantley’s collector) 2160 from Merlimau in Malacca made in 1885, which is most likely the specimen Ridley had in mind when he recorded Anisoptera glabra as “mersawa merah” in 1901. Other early collections are Holmberg 841 (1891) and Derry 1166 (1892), also from Malacca, and Ridley 6886 (Kew No. 210) collected in the Botanic Gardens Singapore in 1895. The first was examined by Brandis who, in 1895, made some pertinent notes on the specimen and suggested that it might be a new species. In 1910 Ridley cited the four collections mentioned above, all of which were in fruit, under A. glabra, but in his Flora in 1922 he evidently realised that they represented Gardens Bulletin, S.S. eA 9 a distinct species, as Brandis had suggested, and described them under the name of A. laevis Ridl. Foxworthy’s early conception of this species is explained in my notes under A. scaphula. In Mal. For. Rec. 10 his conception is quite clear but C. F. 631, cited under A. laevis, is actually A. scaphula. Foxworthy (p. 102) remarks that it is difficult to distinguish this species from A. thurifera without flowering material. Personally I do not find it so. In fruit A. laevis is distinct by reason of its peculiar style and stylopodium while sterile specimens have smaller leaves with fewer nerves and are furnished with distinctive reddish scales on the under surface. I do, however, find sterile specimens of A. scaphula and A. thurifera almost indistin- guishable in certain conditions. Collections examined— MALAY PENINSULA: PERAK: C. F. 0103, 7050, 10394, 11020, 25536, 25602, 25823, 28560, 29915, 30660, 32262 & 33789. KELANTAN: C. F’. 30720. ; PAHANG: C. F. 4635, 4996, 5058, 5487, 5504, 10587, 15621, 26104, 28214 (?) & 28247. SELANGOR: C. F. 9645, 10939, 24025, 24033, 24064, 24455, 24704 & 27942. NEGRI SEMBILAN: C. F. 0638, 0647 & 23755. MALACCA: Holmberg 841; S’pore 2160 (Alvins) ; Derry 1166; C. F. 18276. SINGAPORE: S’pore 6886 (Kew No. 210) Ridley (auth. spec. of A. laevis Ridl.) & 2047. Antsoptera laevis is our most common form of smooth- leaved mersawa. Notes are made on early collections of the species and a botanical description is given for comparison with the closely related A. scaphula, a less common mersawa found mainly in the north of the Peninsula. eo Anisoptera costata Korth., Verh. Nat. Gesch. Ned. Overz., Bot. 67, tab. 6, fig. 1-9 (1839-’42) ; Ridl. in Agri. Bull. Ss. S. & F. M.S. 1: 60 (1901); in Journ. R. As. Soc. St. Br. 54: 25 (1910) partim, et Fl. Mai. Penins. 1: 218 (1922) partim; V. Sl. in Bull. Jard. Bot. Buitenz III, 8:7, fig. 1 (1926) ; Foxw. Mal. For. Rec. 3: 79 (1927) obs., et Mal. For. Rec. 10: 97 (1932). Plate III A. This species, described from Borneo by Korthals in 1839, was first collected in the Peninsula by Holmberg in Malacca in 1891. Ridley recorded this collection as A. costata in 1900 but in 1910 he included along with it his 6684 from the Gardens’ Jungle, Singapore. This latter Vol. VIIT. (1934). 10 collection is one of those upon which van Slooten founded his A. megistocarpa. This confusion of two species under the name of A. costata persisted in the Flora of the Malay Peninsula where the specimens cited from Batu Tiga (Holmberg) and Bang: (Mitchell, C. F. 1042) belong to A. costata and the remainder to A. megistocarpa. Van Slooten cleared up the confusion in 1926 when he described his A. megistocarpa but under A. costata cited only one collection (C. F. 1042) from the Malay Peninsula. Foxworthy (Mal. For. Rec. 10) also cites only the one collection, having also confused this species with A. megistocarpa. He records A. megistocarpa from Kuala Lumpur and Upper Perak on the evidence of collections (C. F. 13712 and 10409) which are actually A. costata. It will be seen from the citation of specimens that A. costata is by no means as rare a tree in the Peninsula as has been supposed. Under A. oblonga I have made some notes concerning the relationship of A. costata to A. oblonga and other closely allied species. | Collections examined— SIAM: C. F. 2993. MALAY PENINSULA: PERAK: C. F. 10409, 30310 & 30908. SELANGOR: C. F. 1042, 9992, 13712, 14840, 17073 & 32667. MALACCA: Holmberg 768. SUMATRA: F. R. 1., E. No. 399 & 546, T. No. 330 & 896. JAVA: F. R. 1, E. No. 1367. BORNEO: Korthals, G. Sakembang (Auth. spec. of A. costata Korth.). Anisoptera costata is a form of mersawa with rebtee larger leaves, yellow and hairy on the undersurface. It is found on low-lying land and is here shown to be more common than has been supposed. Reference is made to the former confusion of this species with A. megistocarpa. “ Anisoptera oblonga Dyer in Hook. f. Fl. Brit. Ind. 1: 301 (1874) ; Pierre, Fl. For. Cochinch. 15, t. 235 et 236 (1890) in obs.; Brandis in Journ. Linn. Soc. 31: 42 * (1895); Guérin ex Lecomte, Fl. Gen. l’indo-chine 1: 369 (1912): Shorea nervosa Kurz, For. Fl. Brit. Burma 1: 119 (1877) ; Pierre, l.c. t. 235 in obs. ; Brandis l.e.: ? Anisoptera cochinchinensis Lanessan, Pl. Utiles Col. Fr. 268 (1886); Pierre, lc. t. 2835 A et t. 253; Brandis, l.c.; Heim in Bot. Tidsskr. 25, 1: 44 (1902) ; Guérin, l.c. 367; Craib, Fl. Siam. Enum. 1: 139 (1925) ; Gardens Bulletin, S.S. 11 Lecomte, Bois d’indo-chine 115, fig. 15, 3, 4 & 5 (1926): ? Anisoptera glabra Pierre, l.c. t. 235 B, et Guerin, lc. 368, non Kurz: ? Anisoptera robusta Pierre, l.c. t. 236; Brandis, l.c.; Guérin, l.c. 368; Craib, l.c. 140: ? Anisoptera marginatoides Heim in l.c.; Guérin, l.c. 369 ; Craib, l.c. 139. Plate III B. Anisoptera oblonga was first collected by Griffith in Mergui and described by Dyer in 1872. In 1877 Kurz described his Shorea nervosa from a sterile collection from Tenasserim (not seen by me) which Pierre and Brandis later identified as A. oblonga Dyer. Up to the time of Foxworthy’s “Dipterocarpacez of the Malay Peninsula” (Mal. For. Rec. 10 (19382)) no collection suggestive of this species had been made in the Peninsula, but in 1933-34 three flowering collections were made in Kedah and Kelantan which exactly match Burmese collections of A. oblonga. Of the collections examined by me I consider the following to be A. oblonga Dyer without doubt— BURMA: Griffith 939 (Auth. spec. of A. oblonga Dyer) ; _ Maymyo 7513 (Field No. 5047), 9378 (Field No. 8055) & 9403 (Field No. 8058). _ MALAY PENINSULA: KEDAH: C. F. 21663 & 30133. KELANTAN: C. F. 29096 & 33284. After studying carefully the above collections and collections of A. costata, and after making numerous floral dissections, I am struck with the very close affinity of these species. Indeed I can find no satisfactory character upon which they can be separated, other than the nature of the epidermal: covering. All the collections of these species that I have examined have been furnished, on the under- sides of the leaves and to some extent on the branchlets and panicles, with three forms of trichome—1, peltate scales; 2, dense tufted stellate hairs; and 3, long simple hairs or stellate hairs with a few long arms (all gradations are found between 2 and 3). In A. costata usually all three forms of trichome are plentiful but in A. oblonga ‘3”’ are almost absent and “2” are sparsely distributed. This difference between A. oblonga and A. costata may be a local variation of a single species due to climate condition rather than a specific difference. However, it would not be wise to unite these species until more collections are available for study. I also consider that A. cochinchinensis Lanessan and A. robusta Pierre probably do not differ specifically from A. oblonga. Nor am I satisfied that A. glabra Pierre is distinct. From the evidence of Pierre’s description and Vol. VIII. (1934). 12 figure it is clear that his A. glabra is not A. glabra Kurz {[=A. scaphula (Roxb.) Pierre]. Pierre enumerates the diagnostic characters of A. cochinchinensis thus—“Spécifiquement, nous distinguons ?” A. cochinchinensis de |’ A. oblonga Dyer, par les fewilles munes Wun plus grand nombre de petites cétes; par des fieurs un peu plus grandes, contenant de 30 a 35 étamines au liew de 25; par les nervures des ailes du calice fructifére, également élevées sur les deux faces. Ses feuilles sont un peu plus oblongues que celles de Il A. costata, et moins souvent cordées, plus glabres; ses anthéres sont glabres, ainst que les ailes de son calice fructifére, d’ailleurs moins larges et moins longues’’. | All the characters mentioned by Pierre are subject to considerable variation within the species in Anisoptera. Even the number of stamens I have found to vary within wide limits in some of the species and I have counted 30 stamens in the flowers of one Burmese collection of typical A. oblonga. Pierre evidently felt that the evidence for specific separation of his A. cochinchinensis and A. glabra was slender because he adds—‘‘On ne saurait dissimuler le grand vapprochement de ces trois espéces (A. cochinchinensis, A. oblonga and A. costata), et méme de celle—ci et de lea suwante (A. glabra Pierre), qui se fait remarquer surtout par son calice fructifére, dont les petites ailes sont plus étroites et plus longues.” Concerning A. robusta, Pierre says—‘‘C’est avec beaucoup Whésitation que je distingue cette espece des A. glabra, A. oblonga et A. costata, trois espéces ayant une étroite affinité. De la premiére surtout, il est trés difficile de la séparer st lon considére que le fruit est seul connu et qwil est possible qu’elle soit A. oblonga. De toutes deux, elle differe par ses feuilles velues et par lextrémité arrondie de’ ses feuwilles. Del A. oblonga, elle se distingue encore par des fleurs plus grosses, des sépales plus longs et plus aigus. Dans la plante de Mergu, type de l A. oblonga, les plus grands sépales n’ont que deux mm. de longueur et sont pubérulents. Les pétales y ont 4 mm. Y% de longueur et les nervures sont au nombre de7. On y trouve 25 étamines. Les anthéres glabres longues @un mm. y sont terminées par un connectif long de 2 mm. Y%. Le style y est ovoide pubérulent. J’avais pensé aussi, avant davoir vu la plante de Korthals, que c’etait 1 A. costata, mais la les feuilles sont cordées acuminées, le nombre des petites cétes est de 20-24, les anthéres sont pubescentes et le style linéaire—oblong. Gardens Bulletin, S.S. 13 Del A. cochinchinensis on ne peut distinguer l A. robusta gue par les boutons plus allongés, par son inflorescence, ses feuilles velues et le nombre de ses étamines.” The only authentic specimen of A. robusta I have examined is Thorel 2696 which has flowers practically identical with those I have dissected on some of the sheets of A. costata. The predominance of the third form of trichome appears to be the only character that distin- guishes A. robusta Pierre from A. costata Korth., A. oblonga Dyer, A. cochinchinensis Lanessan, and A. glabra Pierre (non Kurz) all of which may be proved to be conspecific when more complete material is available. I have reason to suspect that A. marginatoides Heim which is represented by a single fruiting specimen from Chantaburi (not seen by me) may also have to be reduced to synonymy with the above. There is nothing in the description of the species to preclude such reduction. Under A. costata Korth. I have cited the collections examined that I consider typical of Korthal’s species and under A. oblonga Dyer those that I consider typical of Dyer’s form. The following collections probably do not differ specifically from the above but as they are hardly typical of either it seems desirable to keep them separate meanwhile— CAMBODIA: Foxworthy Nos. 4, 5 & 12. COCHIN CHINA: Pierre 1550 (auth. spec. of A. cochin- chinensis Pierre), Thorel 2696 (A. robusta Pierre). BURMA: Calcutta 964 & 8062; Bot. Survey Tenas. Circ. 328 (Maymyo 605 N/972). SIAM: For. Dept. Siam No. 871; S’pore 2708. MALAY PENINSULA: KELANTAN: C. F. 29122, 29082, 29969, 32725 & 33408. PAHANG: 5151, 28323 & 29969. Ansoptera oblonga is a form of mersawa occurring in Burma. It has recently been collected in Kedah and Kelantan. It is suggested that this tree may not differ specifically from A. costata and certain other forms described from Cochin China and Siam, but more collections are required to decide the point. “ Anisoptera marginata Korth. Verh. Nat. Gesch. Ned. Overz. Bez., Bot. 66, t. 6, fig. la-13a (1839-’42) ; Ridl., Fl. Mal. Penins. 1: 219 (1922) ; V. SI. in Buil. Jard. Bot. Buitenz. II], 8: 5 (1926) ; Foxw., Mal. For. Rec. 10: 96 (1932) : ? Anisoptera grandiflora Brandis in Journ. Linn. Soe. o1: 43, pl. 2, fig. 29 (1895); V. SIL. Le. 10, in nota. Plate IIT C. Vol. VIII. (1934). cee > iy 14 Van Slooten has given an excellent description of this species which is reproduced by Foxworthy in Mal. For. Rec. 10. In the flowers examined by me I find that the number of stamens varies from 25-28 and that occasionally there are 4 styles. Four styles have also been noticed in A. oblonga. Occasionally on dissection there appear to be more than 4 styles. This is due to the fact that each style is readily divisible into two, a condition found in all the species of Antsoptera I have examined (sect. Glabrae excepted). A. marginata was originally described by Korthals from Borneo and was first recorded from the Malay Peninsula by Ridley in 1922 on the evidence of a single collection (C. F. 3601) from Kuantan. In 1926 van Slooten cited only this collection and in 1932 Foxworthy added only one other from Ulu Rompin. It is now possible to add considerably to the citation of collections from the Malay Peninsula. It appears that this species is not uncommon in our swamp or semi-swamp forests near the coast. Van Slooten has suggested that A. grandiflora Brandis may prove to be a synonym of A. costata Korth. I am of opinion that it is only a large-leaved collection of A. marginata. This is, however, difficult to prove because these two species are essentially similar in flower structure, the best diagnostic character being the more numerous veined and more pubescent leaves of A. costata. Collections examined—— MALAY PENINSULA: PERAK: C. F. 27804, 27806 & 30753. SELANGOR: C. F. 23927, 23948, 24562, 27070, 29717, 29718, 29730, 29738, 30771 & 36029. PAHANG: C. F. 3601 & 15496. SINGAPORE: C. F. 30759. SUMATRA: F. R. I. Nos. 382, 385, 398 & 403; F. R. I. No. b.b. 5788. BANGKA: Teysmann sine no. JAVA: Cult. in Hort. Bog, [X-D-171. BORNEO: ? Haviland (Garai) 959 [auth. spec. of A. grandiflora Brandis]; Garai sine no. 26.2.92 (fruit— possibly from same tree as 959); F. R. I. No. b.b. 14895 & E. No. 5056. Anisoptera marginata is a form of mersawa found usually in semi-swamp jungle such as Utan Melintang or Telok. It was recently discovered in Sungei Buloh Reserve. “Anisoptera megistocarpa V. Sl. in Bull. Jard. Bot. Buitenz. Ill, 8: 12, fig. 2 (1926); Foxw., Mal. For. Rec. 10: 98 (1932): Anisoptera costata Ridl. in Journ. R. As. Gardens Bulletin, S.S. 15 Soc. Str. Br. 54: 25 (1910) partim, et Fl. Mal. Penins. 1: 218 (1922) partim; Burkill, Journ. Str. Br. R. As. Soe. 75: 43 (1917); non Korth. Plate IV A. Van Slooten and Foxworthy have given very complete descriptions of this species and van Slooten has published a very good figure of the branchlet, inflorescence and fruit. I should like to add the following notes—In examining the flowers of Ridl. 6684, Burkill 526, and C. F. 624 I find the number of stamens to vary between 47 and 54 (30-32 in published descriptions) and that the arrangement of the styles is essentially as in A. costata etc., i.e., there are three closely appressed styles with minute stigmas. On dissection each style is found to be notched at the apex and easily divisible into two (vide my notes under A. marginata). This species was collected by Ridley in the Botanie Gardens Jungle, Singapore as early as 1894. In 1910 and again in 1922 Ridley recorded these coilections along with authentic material of A. costata under the latter name. It was not until van Slooten described his A. megistocarpa in 1926 that the Peninsula collections of these two species were separated, but recently confusion has again occurred (vide my notes on A. costata). R We have at Kepong three. sterile collections (C. F. 28702, 30561 and 30610) from Brunei which look like small- leaved specimens of A. megistocarpa. It is _ probable, however, that they represent an undescribed species (vide also my notes on A. Curtisi?). Collections examined— MALAY PENINSULA: PERAK: C. F. 27337, 27821 & 310838. PAHANG: C. F. 29951. NEGRI SEMBILAN: C. F. 427, 624 & 669. MALACCA: C. F. 2091 & 9296. SINGAPORE: Ridley 6684 (cited as 6634 by V. SIL.) (ff. and ft.—auth. spec. of A. megistocarpa V. Sl.) ; S’pore Gdns. No. 526; S’pore sine no. (Burkill 1916) & sine no. (Md. Nur 1924). Anisoptera megistocarpa is a form of mersawa with a conspicuous dark crown due to the hairy red-brown under- surfaces of its leaves. It may be seen in Bikam, Pasir Panjang (N. S.), and the Malacca among other reserves, but is not plentiful. Anisoptera Curtisii Dyer ex King in Journ. As. Soc. Bengal 62, pt. 2: 100 (1893) ; Brandis et Gilg in Engler, Nat. Pflanzenfam. III, 6: 263, t. 122, fig. F. (1894) and’ 2nd edit. 21: 259, t. 112 fig. F. (1925); Brandis in Journ. Linn. Soc. 31: 42 (1895): Burkill, Journ. Str. Vol. VIIT. (1934). 16 Br. R. As. Soc. 81:-51 & 63 (1920) ; Merr., Bibl. Enum. Born. Pl. 400 (1921); Ridl. Fl. Mal. Penins. 1: 218 (1922); V. Sl. in Bull. Jard.. Bot. Buitenze tiles (1926) ; Foxw., Mal. For. Rec. 3: 79 (1927) in nota, et Mal. For. Rec. 10: 99 (1932) ; non Foxw. in Philip. Journ. Se. Bot. 6: 255, t. 41 (1911), l.eis: tere et ex Merr., Enum. Phil. Fl. Pl. 3: 92 (1923): Anisop- tera Curtisu var. latifolia King, }|.c. Plate IV B. The only description of this species based on both flowering and fruiting material is that of Foxworthy in Mal. For. Rec. 10. The following is an amplified description based upon my own experience of our collections. Branchlets rugulose, slightly flattened, densely clothed with a scabrous, ferruginous, fugacious, steilate tomentum. Leaves oblong or elliptic-oblong to obovate, usually rather abruptly tapering at both ends, apex acute or shortly acuminate, base rounded, obtuse to slightly cordate (rarely acute), usually about 10 cm. * 4 em. but may be much smaller or up to 14 cm. X 7 ecm. (var. latifolia King), glabrous, shining red-brown above when dry, beneath, when young, usually sparsely pubescent and pale yellow, later glabrescent and densely ochraceous lepidote; midrib deeply depressed above, somewhat purple beneath; secondary nerves 18-20 prs. conspicuously looped towards the margins, with occasional short intermediary veins, sometimes slightly depressed above, conspicuously raised and purplish beneath ; reticulations visible on both surfaces; petioles enlarged in the upper portion, 1—2.5 em., rugulose, scabrous, squamate; stipules linear oblong, sub-falcate, acute, 7 xX 2 mm.,, scabrous, caducous. Panicles lax, pendulous, racemose, up to 15 cm. long, rugose, scabrid; ultimate branchlets distant, up to 4 cm. long, each bearing 3-10 distichous flowers 2.5—5.0 mm. apart. Flowers acuminate in bud, 11.0 mm. including the pedicel; pedicel 3 mm., scabrid tomentose. Sepals slightly imbricate in bud, sub-equal in length, 2 ovate oblong obtuse, 3 ovate acuminate, all stellate tomentose outside, pubescent inside. Petals in bud slightly contorted, involute and reflexed at apex, lanceolate acuminate, glabrous except along one margin outside, “greenish white” (Haniff, S’pore 3692). Stamens 24-26 (usually 25) in two whorls; fila- ments longer than the anthers, slightly flattened, broader towards the base; anthers oblong, posterior cells much smaller than the anterior; appendage to connective filiform, more than 3 times as long as the anther. Ovary 2/8 inferior, tomentose, surmounted by a sub-globose, sub- suleate, tomentose stylopodium; styles 3 (occasionally 4), Gardens Bulietin, S.S. aes ~~. . iat. 17 short, glabrous, closely appressed in young flowers, each notched at apex and easily divisible into 2; stigmas minute. Fruit: stalk 4 mm. long, puberulous; calyx tube companu- late, adnate to 2/3 of the nut, about 1.2 cm. diam. and 1 cm. high, levigate or sub-verrucose puberulous; 2 large calyx lobes linear-spathulate, usually about 9 em. X 1.2 cm. but may be up to 12 cm. X 2 cm., minutely puberulous or glabrate; 3 small lobes linear acuminate, up to 1.5 cm., puberulous or glabrate; nut 2/3 enclosed in the calyx cup, globular, about 1 cm. diam., glabrate or puberulous, sur- mounted by the conical stylopodium and remnants of the styles (stamens frequently persist for a long time but have fallen before the fruit reaches maturity). The earliest collections of this species were made in the Larut hills by Kunstler in 1882 and in Penang by Curtis in 1885-92. Upon these collections King founded his species in 1893 using Dyer’s manuscript name. One of Curtis’s collections has broader leaves than the others, which fact induced King to form for it the variety latifolia. Subsequent authors, however, are agreed that varietal separation is not justified. Although it would be surprising to find A. Curtisii, which is so widely distributed on hills in the Malay Peninsula, strictly confined to this portion of the Malayan region, a wider distribution seems hardly sufficiently proven. The collection on the evidence of which Foxworthy (Mal. For. Rec. 10) records the species from Peninsular Siam, and Burmese collections under the name of A. Curtisii in the herbariums of Maymyo and Calcutta, are, I consider, related to A. oblonga (see citations on p. 13). From Sumatra van Slooten records the species on the evidence of three sterile collections which I have not seen. From Borneo Brandis recorded a collection by Burbidge, and van Slooten added several other collections from that island. Of these I have examined Haviland, sine no. (July 1905, Brunei) and bb. 8156 (ex herb. Bog.) and am of opinion that they represent an undescribed species with the facies of a small-leaved form of A. megistocarpa. Recent sterile collections from Brunei (C. F. 28702, 30561 and 30610) are from the same species. Foxworthy has recorded A. Curtisii from the Philippines but I am in agreement with van Slooten who questions his identification. I have examined Bur. Sci. Phil. 8985 cited by Foxworthy under A. Curtisii and I am of opinion that they are more correctly referable to A. brunnea Foxw. a species closely related to A. thurifera (Blco.) Bl. Although superficially A. Curtisii is, as van Slooten and Foxworthy have remarked, probably most liable to be Vol. VIII. (1934). B 18 contused with A. costata and A. marginata, its natural affinity, I venture to suggest, is rather with A. thurifera Bl. A. thurifera has almost twice as many stamens but the stylopodium and styles, as well as the other flower parts, are very alike; this may be seen by comparison of floral dis- sections given on plate IV B & C. Collections examined— MALAY PENINSULA: PENANG: Curtis 4428 (3 coll.) & sine no. Govt. Hill, 1893, fl. & ft. (Auth. specs. of A. Curtisu Dyer) ; Curtis 1400 (auth. spec. of A. Curtist var. latifolia King); Fox 78 & 85; Haniff, sine no. (1918); S’pore 1521, 3281, 3481, 3692, 3723, 3762 & 4566; C. F. 10845, 11662, 27781 & 28017. PERAK: Kunstler 3618 & 3706; C. F. 16585, 24635, 25475, 25488, 25546, 25601, 28559, 28563 & 32180. DINDINGS: C. F. 16591 & 27848. KELANTAN: C. F. 32717 & 33409. TRENGGANU: C. F. 26747, 26752 & 26761. SELANGOR: C. F. 16009, 22942, 23657, 24029 & 27136. NEGRI SEMBILAN: C. F. 742, 1192, 2103 & 18884. SINE Loc.: C. F. 11718. Anisoptera Curtisii is a form of mersawa with small bright yellow leaves found upon hills particularly in the northern portion of the Peninsula. It is very plentiful on Penang Island where it was first collected and is apparently known as rengkong. The plant is described and its probable distribution outside the Malay Peninsula is discussed. ‘Hopea sangal Korth. in Verh. Nat. Gesch. Bot. 75 (1839-’42) ; Blume, Mus. Bot. Lugd. Bat. 2: 34 (1852) ;. Mig. Fl. Ind. Bat. 1, pt. 2: 504 (1859) ; A.DC. in DC. Prod. 16, pt. 2: 635 (1868); V. Sl. ex Heyne, Nutt. Plant. Ned. Ind. 2nd edit.: 1111 (1927): Petalandra micrantha Hassk., Hort. Bog. Descr. 105 (1858) ; Mia... le. 505; A. DC. in Le.; King. in Journ, Asiat.. S00; Bengal 62, pt. 2: 126 (1893), in nota sub H. micrantha Hook. f.: Hopea fagifolia Miq., Fl. Ind. Bat. Suppl. 499: (1860); Scheff. in Nat. Tijdschr. Ned. Ind. 31: 351 (1870) ; Brandis in Journ. Linn. Soc. 31: 65 (1895) ; Beerl., Cat. Hort. Bog. 1 & 2: 103 (1899); Koord. & Val., Boomsoort. Java 5: 124 (1900); Sym. in Gdns. Bull. S. S., 7, pt. 2: 151 & 154 (1933) in nota: Hopea diversifola Scheff., l.c., Burck in Ann. Jard. Bot. Buitenz. 6: 239 (1887), et Brandis in l.c. 64, non Miq..,. l.c. 491; Foxw. Mal. For. Rec. 10: 123 (1932) in obs.: Hopea odorata Auct. (1876-1927), partim, quoad syn. H. sangal: Doona odorata Burck in l.c. 233 quoad syn. H. sangal: Doona micrantha Burck in l.c. 234: Doone Gardens Bulletin, S.S. ae <9 javanica Burck in l.c. 235, t. 29, fig. 7: Hopea micrantha Benth. et. Hook. f. ex Heim, Rech. Diptéroc. 64 (1892) in obs.: Hopea Hasskarliana Heim, l.c.: Hopea javanica Heim l.c.; Brandis in l.c. 65: Hopea Curtisu King in lc. 124 et in Ann. Bot. Gard. Calcutta 5: 155, pl. 1878 (1896) ; Brandis in l.c. 65; Burkill, Journ. Str. Br. R. As. Soc. 81: 66 (1920) fig. 113-120; Ridl., Fl. Mal. Penins. 1: 236, fig. 22a (p. 212) (1922); Foxw., l.c. 130: Hopea globosa Brandis in l.c. 61; Ridl. in Journ. Str. Br. R. As. Soe. 54: 26 (1910) et Fl. Mal. Penins. 1: 236 (1922); Foxw., lc. 121: Hopea Lowu Dyer ex Brandis in l.c. 63; Ridl. in Journ. Str. Br. R. As. Soc. 73: 1438 (1916) et Fl. Mal. Penins. 1: 237 (1922) ; Foxw., l.c. 123: Doona micrantha var. macrosepala Berl., MSS. in Herb. Bog. (1900): Hopea fagifolia var. fol. latioribus Berl. MSS. in Herb. Bog. (1900) ; Hopea fagifolia var. javanica Berl., MSS. in Herb. Bog. (1900) : Hopea macrosepala Berl., MSS. in Herb. Bog. (1900): Hopea multiflora Foxw., l.c. 119, partim, non Brandis: Hopea albescens Foxw., l.c. 122, non Ridl. Plate V. . The most complete description of this species is given by Foxworthy under Hopea Curtisi but as that was based upon a limited number of specimens it seems desirable to recast the description to cover all the collections here considered to be H. sangal. Branchlets dark purple when dry, lenticellate, pale fugacious tomentose towards the ends. Leaves usually ovate lanceolate but very variable from lanceolate to broadly ovate, apex acuminate, base slightly unequal rounded to cuneate, usually about 10 cm. X 4.5 cm. but varying between wide limits, grey-green or tawny coloured and shining above. usually darker beneath and glabrous or sparsely stellate puberulous upon the nerves; midrib slightly depressed above, tomentose to glabrous, strongly elevate and _ usually puberulous beneath; secondary nerves 10-14 pairs, strongly elevate on the lower surface, commonly with barbate domatia in the axes of the lower pairs; tertiary nerves rather obscure, joining the secondaries in numerous more or less parallel lines; petioles usually puberulous and rugose, about 1 cm.; stipules linear, about 3 mm. long, falling very early. Panicles axillary and terminal, pendulous, racemose, when axillary shorter than the leaves, grey or tawny tomentose; ultimate branches about .5 cm. apart, up to 2.2 cm. long, each bearing 6—15 closely set secund flowers. Flowers oval in bud, 3 mm. long including the pedicel; pedicel about 1 mm., tomentose. Sepals imbricate; two ovate or oblong, thick, tomentose outside and on the upper half Vol. VIII. (1934). 20 inside; three smaller, more or less rotundate and irregular, tomentose outside only.* Petals contorted in bud, oblong, fimbriate at the apex, tomentose outside on the portion exposed in bud, white. Stamens 10, two appressed to each petal; anthers oblong, the posterior cells almost as large as the anterior; filaments a little longer than the anthers with broad flattened overlapping bases; appendage to connective filiform, about the same length as the anther. Ovary broadly cylindrical truncate, puberulous on the upper portion which may be considered to be the stylopodium, glabrous on the lower; style less than half as long as the ovary, glabrous; stigma minute. Frwit: stalk 1 mm. long, tomentose; calyx lobes united to form a flat open tomentose receptacle; 2 large calyx lobes spathulate, usually about 7 cm. X 1.7 cm., 8-10 nerved, puberulous towards the base; 3 small lobes oblong obtuse or orbicular, striate, puberulous, about half as high as the mature nut; nut globose, apiculate, about 1 cm. diam., sparsely tomentose. (When immature the nut is conical and almost enveloped by the small calyx lobes). Embryo: cotyledons erect, unequal, longitudinally striated ; dorsal cot. slightly embracing the placental cot. and alone reaching the apex of the ovary; placental cot. smaller ; radicle erect, exposed on one side between the lobes of the dorsal cotyledon. Hopea sangal was first collected by Korthals in about 1840 from Borneo. He gave an excellent description of the field characters of the tree sangal, but, being based on limited sterile material, it was hardly adequate from a botanical point of view and led to misinterpretation and confusion by Hance and others with Roxburgh’s Hopea odorata (1832) from India, to which species H. sangal was reduced by Hance in 1876. This reduction appears to have been accepted by all authors, Kurz excepted, for over half a century, and on the strength of it H. odorata has been erroneously recorded from Borneo, until in 1927 van Slooten indicated that separation was necessary. During that period or earlier Korthals’ plant had been recollected on numerous occasions, referred to various species, and described under a variety of names. In 1856 Hasskarl founded his new genus Petalandra and described his P. micrantha from collections from east Java. When Bentham and Hooker [Gen. Plant. I (1862-67) 193] united Petalandra Hassk. with Hopea Roxb., confusion arose between this species and Hopea micrantha Hook. f., (1862) a quite distinct plant. Reference is made to the * On some collections the receptacles and calyx parts show a tendency to enlarge even before. the buds have opened. This is the condition of the flower figured in Plate V, 6a, 8a and 9a. Gardens Bulletin, S.S. = 21 problem by Hance in Journ. Bot. 5: 308 (1876), by King in Journ. As. Soc. Bengal 62, pt. 2: 126 (1893), and by Heim in Rech. Diptéroc. 64 (1892) who proposed the epithet Hasskarliana for Hasskarl’s plant, that of Hooker having prior claim to the trivial micrantha. In 1860 Miquel redescribed H. sangal from a sterile collection of Teysmann from Bangka as “Hopea ? fagifolia’”’ and in 1870 Scheffer referred another collection of Teysmann (“prope Djebees’’) to Miquel’s species without having seen the type. I have not seen the Djebees collection and I am in doubt as to whether it was correctly placed by Scheffer. Scheffer also referred collections of Teysmann from Sumatra to Hopea diversifolia Miq. In this he was wrong. H. diversifolia Mig. is founded on two collections which I have examined and find to represent two distinct species neither of which is H. sangal. In 1877, when Burck transferred Petalandra Hassk. and several species of Hopea to Doona, he created the combinations Doona odorata (under which he retained H. sangal) and Doona micrantha, and reduced H. fagifolia Miq. to the latter species. At the same time he described H. sangal yet again from Javanese collections, under the name of Doona javanica, a name that was altered by Heim in 1892 to Hopea javanica with the transfer of Burck’s Malayan species of Doona to Hopea. In 1899 Berlage, and in 1900 Koorders and Valeton, suggested the union of H. javanica with H. fagifolia Mig. and Hopea (Petalandra) micrantha. The earliest recorded collections of H. sangal from the Malay Peninsula were made by Kunstler in Perak (1885) and Curtis in Penang (1888). Upon these collections King founded his Hopea Curtisu in 1893. Another Peninsula collection from Perak was made by Wray jnr. and described by Brandis in 1895 as Hopea globosa. At the same time he described two collections, one by Lowe in Borneo and the other by Teysmann in Sumatra, as Hopea Lowi using a name given in manuscript by Dyer. In the Flora of the Malay Peninsula (1922) Ridley referred some additional collections to H. Curtisii and H. globosa and recorded H. Lowti from the Peninsula referring several collections to that species. He mentions, however, that the last two are suspiciously alike. Foxworthy, in Mal. For. Rec. 10 (1932), referred most of our broad-leaved collections of H. sangal to H. Curtisii; to H. globosa he referred one collection (C. F. 8023), and to H. Lowi one collection (Ridley 6585—formerly referred by Ridley to H. globosa). As Hopea multiflora Foxworthy Vol. VIII. (1934). 22 described a species closely related to H. odorata Roxb. that I am renaming Hopea semicuneata, but C. F. 5906 upon which he based the fruit description is H. sangal, as also are a number of collections cited under this heading with doubt. The remainder of our collections were cited by Foxworthy under Hopea albescens. H. albescens Ridl. is, however, a very distinct plant which I have reduced to Shorea hopeifolia (Heim) Sym. and discussed in Gdns. Bull. 5. S. 7: 150°(1933). Two reasons mainly account for the fact that H. sangal has been redescribed on so many occasions. The first is the inadequacy of the material upon which early descriptions were based and the second the variation in leaf within the species. The collections upon which King based his dH. Curtisu and to which Beerlage gave the manuscript names Doona micrantha var. macrosepala, Hopea macrosepala, and Hopea fagifolia var. fol. latioribus are typical of the broad- leaved form represented by a number of our collections, most of which were placed by Foxworthy under the first name. Berlage’s manuscript species H. fagifolia var. javanica and our collections placed by Foxworthy under H. albescens are representative of the ovate-lanceolate-leaved form. Miquel’s H. fagifolia and Brandis’s H. globosa were described from collections of immature leaves probably taken from saplings (the fruits of Wray 816 are not attached to the leaves). All intermediate stages are represented in the collections I have examined and I am unable to recognise any clearly defined divisions within the species although I do not doubt that definable local strains do exist. Some remarks seem necessary concerning the affinities of H. sangal. Petalandra micrantha Hassk., which is here shown to be a synonym of H. sangal Korth., was the species upon which Hasskarl founded the genus Petalandra. Bentham and Hooker and later authors have maintained Petalandra as a section of Hopea referring to it those species that differed from the section Euhopea in having, inter alia, 10 as opposed to 15 stamens. This is a convenient division but I am of opinion that the maintenance of Petalandra even as a section tends to obscure natural relationships. I should like to have the support of more detailed study of the anatomy and embryo development before making a final statement, but it seems tome unfortu- nate that H. sangal Korth. and H. acuminata Merr. should be separated from such species as H. odorata Roxb., H. Helferi Brandis (=H. dealbata Hance), H. nutans Ridl. and H. semicuneata Sym. because of a single character—the number of stamens. I have found on several occasions Gardens Bulletin, S.S. 23 that in dipterccarps that normally have 15 stamens, one row may abort, or in those that normally have 10, 15 may be present. It would appear therefore that this character should be treated with reserve when considering natural relationships within this family. Collections examined— SIAM: C. F. 3761, 3835, 3863 & 3874. MALAY PENINSULA: KEDAH: C. F. 12220, 17614, 17858, 27061, 27464 & 27472. PENANG: Curtis 1562 (Auth. spec. of Hopea Curtisii King) ; S’pore 3459, 3473 (H. B. 933.21.5), 3504 (H. B. 983.21.4), 3655, 3715 & 3717; C. F. 28005 (Hort. Penang Tree No. 18) & 28010 (Hort. Penang Tree No. 15). PERAK: King 8161 (Auth. spec. of Hopea Curtisii King); Wray 816 (Auth. spec. of Hopea globosa Brandis) & sine no. (Larut); Harun sine no. (1906) ; C. F. 0118, 8023, 11022, 11590, 13233, 30816, 30913, 31403 & 32261. KELANTAN: S’pore 24839, C. F. 32733 & 32834. TRENGGANU: C. F. 26961. PAHANG: C. F. 1262, 1281, 2754, 4650, 4855, 5160, 5469, 5480 (H. B. 933.21.2 & 7), 6697, 8139, 8887, 10564, 10659, 15147, 15500, 15629, 15714, 15765, 17286, 18653, 18667, 26182, 26501 & 27995. SELANGOR: Ridley, sine no. (Rantau Panjang); Mat, Sine no. (1908); C. F. 8372, 14586, 24906 & 32871. NEGRI SEMBILAN: Clerk sine no. (1912); C. F. 619, 659, 1886, 1991, 4206 & 20106. JOHORE: Curtis 3613; C. F. 5906, 5984 & 12058. SINGAPORE: Ridley 6585. SUMATRA: Teysmann 12039 (H. B. 933. 21.61), 12040 (H. B. 933.21. 59) & 12042 (H. B. 933.21.60) [Auth. specs. of H. diversifolia Scheff], & sine no. in Herb. Kew. [ Auth. spec. of H. Lowit Dyer] ; Grashoff 217 (H. iv deoet.10). (Lo, th. Bb. SSS289)) a,-973..(H. B. 9338.21.52); Remoorin sine no. (H. B. 9383.21.51); sd, Db. 16595..4 H. B.933.21.68) > . B. 933.21.8. BANGKA: Teysmann 3236 (H. B. 9383.21.16) [Auth. spec. of H. fagifolia Miq.]; 1520 (H. B. 933.21.15). JAVA: Teysmann 12043 (H. B. 9383.21.37) [Auth. spec. of P. micrantha Hassk.]; Burck sine no. (H. B. 933.19.2; H. A. R. T. 72511) [Auth. spec. of D. javanica Burck] ; Koorders 13876 (H. B. 9338.21.48), 15566 (H. B. 933.21.40), 16478 (H. B. 9383.21.47), 100046 (H. B. 9338.21.50), 202568 (H. B. 9833.21.46), 203266 (H. B. 933.21.41), 230188 (H. B. 933.21.39) & 2479268 (H. B, Vol. VIII. (1934). 24 933.21.49) ; Uhl 5951 (H. B. 9383.21.14 & 44), & 595la (H. B. 9388.21.13 & 45); Van Steenis 2419 (H. B. 9338.21.48) ; Kalshoven 33 (H. B. 933. 21.36) ; Cult. in Hort. Tiekeumeuk—H. B. 933.21.24 (8-B-6) ; 933.21.56 (4-c-4) & 57 (4-D-4) [H. fagifolia var. fol. latioribus Beerl.|; 983.19.1 (5-D-5) [H. macrosepala Beerl. & D. micrantha var. macrosepala Beerl.| ; 933.21.54 (6-A-6) & 55 (6-D-4) [H. fagifolia var. javanica Beerl.]; 933.21.22 & 23; Cult. in Hort. Bog.—H. B. 9383.21.18 & 27 (VI-c-222) ; 9338.21.19 (VI-c-222a) ; 933.21.20 & 28 (VII-B-22a) ; 933.21.25 (VII-B-65); 9383.21.21 & 26 (VII-2-26a) ; 9383.21.29, 30, 31, 82, 33, 34 apeo. BORNEO: Korthals sine no. (H. B. 933.19.8; H. A. R. T. 72513) [ Auth. spec. of H. sangal Korth.] ; Lowe sine no. in Herb. Kew [ Auth. spec. of H. Lowi Dyer] ; Teysmann 8020 (H. B. 9338.21.12) ; Hallier 1028 (H. B. 9383.21.42) ; Dumont sine no. (H. B. 9338.21.11); H. B. 1904 (933.21.17) ; For. Dept. B. N. B. No. 1234. LOCALITY UNCERTAIN: H. B. 9383.21.33 & 58. Hopea sangal is quite common in the Peninsula, parti- cularly in the north and in Pahang. It is usually found at low elevations near streams and has a dark, closely fissured bark upon which tears of white opaque damar are common. The most commonly applied vernacular names are damar mata kuching, chengal pasir and mesiput (Pahang). Collections of this tree from the Malay Peninsula and Islands have formed the types of a number of species which are shown to be synonymous and are united under the oldest name, Hopea sangal. [lopea semicuneata Symington, sp. nov.: Hopea ? diversi- folia Miq., Fl. Ind. Bat. Suppl. 491 (1860) partim: Hopea sp., V. Sl. ex Merr., Univ. Calif. Pub. Bot. 15: 202 (1929) ; sama rupa chengal, Foxw., Mal. For. Rec. 3: 71 (1927): Hopea multiflora Foxw. Mal. For. Rec. 10: 119 (1932) partim, non Brandis: Hopea ? plagata Sym. in Gdns. Bull. S. S. 7 pt. 2: 154 in obs., non Vidal. Plate VI. Hopea odorata Roxb. affinis, petiolis brevioribus, ovario glabro, stylopodio truncato, stylo breviore differt; H. sangal Korth. similis, nerviis secondariis foliorum paucioribus, ovario glabro, staminibus 15 pro 10 distinguitur. Branchlets terete, almost black when dry, puberulous towards the ends. Leaves* usually oblong lanceolate or ovate lanceolate, apex acuminate, base usually markedly «The very marked similarity between the leaves of this species and those of H. plagata Vidal led me to suggest the possible identity of those species (Symington, l.c.). Gardens Bulletin, S.S. 25 unequal, acute or cuneate on one side, obtuse or sub-acute on the other, about 11 cm. X 4.5 em. but variable, glabrous and usually drying grey-green above, glabrous or sparsely puberulous upon the veins and with a brown tinge beneath ; midrib elevate on both surfaces; secondary nerves 8-10 pairs, inconspicuous above, elevate and frequently with barbate domatia in the axes of the lower pairs beneath; tertiary nerves hardly visible on either surface, joining the secondaries in numerous fine more or less parallel lines; petioles rugose, about 1 cm., puberulous, black; stipules caducous (not seen). Panicles axillary and terminal, pendulous, racemose, about half as long as the leaves, grey or tawny tomentose; ultimate branches about 4 mm. apart, up to 1.6 cm. long, each bearing 3—9 closely set secund flowers. Flowers elliptical or ovate in bud, 3.5 mm. long including the pedicel; pedicel 1 mm., tomentose. Sepals imbricate, sub-equal in height; two oval or ovate, thick, tomentose outside and on the upper portion inside; three slightly broader, irregularly oval or rotundate, thinner, tomentose outside only. Petals contorted in bud, oblong, fimbriate at the apex, tomentose outside on the portion exposed in bud, paie yellow (M. Haniff, S’p. 3729). Stamens 15, in pairs opposite the sepals, solitary opposite the petals, of 3 heights; anthers oval, the posterior cells smaller than the anterior; filaments 2-3 times as long as the anthers, filamentous. above, suddenly widening to the broad flattened basal portion; appendage to connective filiform, about twice the length of the anther. Ovary broadly cylindrical, slightly constricted before the upper third which may be considered as a stylopodium, flattened at the top where there is a papillose ring, glabrous; style short, cylindrical, glabrous; stigma minute. Frwt (immature—C. F. 24644); stalk i mm., puberulous; calyx lobes shortly united to form a flat puberulous disc-like receptacle; 2 large calyx lobes oblong spathulate to broadly elliptical, apex rounded, much constricted and puberulous at the base, about 3.5 cm. long and 1—1.8 cm. broad, with 10-12 longitudinal nerves; 3 small lobes deltoid, obtuse, 2.5 mm. X 2.0 mm., puberulous; nut globose, 5 mm. diam., puberulous. I have given as allies to this species two species, H. odorata and H. sangal, conventionally considered to belong to different sections of the genus. My reasons for this will be evident from my notes on the affinities of H. sangal. H. semicuneata is a most interesting species because it supports the argument in favour of the union of the sections Fuhopea and Petalandra by combining characters typical of each. It has the truncate stylopodium of Petalandra and the 15 stamens of Huhopea. Vol. VIIT. (1934). 26 Collections examined— MALAY PENINSULA: PENANG: S’p. 3729. ; PERAK: C. F. 24644 (Type of young fruit) 25151, 25419, 28589 & 28595. DINDINGS: C, F. 27838. PAHANG: C. F. 2729, 4526 (Type of flower), 5440, 6696, 79438, 7947, 24821, 26001, 26002, 260038, 26038, 29554 & 29555. NEGRI SEMBILAN: C. F. 4403 & 26605. MALACCA: C. F. 18267 & 18268. BORNEO: B. N. B.: Elmer 21382. Hopea semicuneata is the botanical name here given to the tree known in Kinta district, Perak, as sama rupa chengal. From parts of Pahang the names chengal, chengal oatu, penak and penak batu have been recorded. The species was described as H. multiflora in Mal. For. Rec. 10. where, (p. 120) additional information concerning the field characters may be found. Usually this tree is found upon ridges up to about 1,200’ but I have also found it on the peculiar association on sand, at sea level, at Pasir Panjang in the Dindings. The only Bornean collection known records the tree “‘in forests along tidal streams.” Balanocarpus Bedd. The genus Balanocarpus was established by Beddome in 1873 upon two species (B. erosa and B. utilis) from the Tinevelly mountains in southern Madras, that differed from Hopea in having a woody nut enclosed in a 5-lobed woody cup. I have not examined any specimen of B. erosa but, if the specimens of B. utilis kindly lent to me by Mr. Parkinson, may be taken as representative of Beddome’s Balanocarpus, it may reasonably be doubted whether the creation of a new genus was justified. In my opinion the importance of the nature of the fruiting calyx has been greatly overstressed by systematic workers on the Diptero- carpaceae. Thus it is that the genus Balanocarpus, originally created for two wing—less species related to Hopea odorata Roxb., has become the repository for dipterocarps of very diverse groups that have failed to conform to conventional generic definition by possessing, when in fruit, an accrescent calyx cup none of the lobes of which have developed into wings. In 1899, Trimen added to the genus B. zeylanicus which is possibly most closely related to Hopea ferrea Lanessan. In 1893 King added several others from the Malay Peninsula. B. Curtisii King is clearly related to Hopea myrtifolia Mig. and H. bracteata Burck [=B. bracteatus Gardens Bulletin, S.S. 27 {Burck) Merr.*j. B. penangianus King [=B. multiflorus (Burck) Sym.] is, as I have stated elsewhere (Gdns. Bull. S. S. 7, pt. 2: 129 & 152), very evidently related to a group of Shorea including S. Faguetiana Heim and S. hopei- folia (Heim) Sym. This relationship was observed by Heim (Recherch. Diptéroc.: 50) when he founded his genus Richetia upon R. coriacea Heim (=Balanocarpus coriaceus Auct.) and three other species now all reduced to B. multiflorus (Burck) Sym. Balanocarpus anomalus King was evidently placed in this genus at a venture in the absence of fruiting material and has since been reduced by Fox- worthy to Hopea ferrea Lanessan. Balunocarpus maximus King is quite distinct from any of the other species hitherto referred to that genus. Its most close ally appears to be Balanocarpus longiflorus (Brandis) Foxw., both of which species appear to be more closely allied to some species in the Brachyptera section of Shorea than any others. Balano- carpus Heimu King is an anomalous species unlike any other to which I have had the opportunity of giving detailed study. It certainly does not appear to be at ali close to any other species of Balanocarpus and, although its fruits are some- what akin to those of Hopea (Pierrea) pachycarpa (Heim) Sym., it differs in too many respects from that species to be grouped along with it as Heim would have it. Balanocarpus Wrayi King was based upon a collection of immature fruit of B. Heimii and is no longer upheld as a species. King’s last species, Balanocarpus Hemsleyanus was recognised by its author to have affinity with certain species of Shorea but was placed in Balanocarpus solely on account of its fruit. Foxworthy is undoubtedly correct in placing this species in the Brachyptera section of Shorea. In 1918 Foxworthy added two Philippine species to the genus Balanocarpus. B. brachypterus is clearly a close relative of B. Curtisii King which, as I have indicated above, should naturally be included in a group of Hopea containing H. myrtifolia Miq. and H. bracteata Burck. B. cagayanensis Foxw. was unlike any Balanocarpus then described, but again, the character of wingless fruit was allowed to overshadow the fact that the flowers are almost identical with those of Hopea philippinensis Dyer and the many other characters that these species have in common. JHopea philippinensis Dyer, usually referred to the Euhopea section of Hopea, I consider to belong to a group apart which I shall call the Pierrea group because Hopea (Pierrea) pachycarpa * B. Curtisii King has been united with H. bracteata Burck by Merrill [Journ. Str. Br. R. As. Soc. special no.: 407 (1921) ], a union which I venture to question. “Vol. VIII. (1934). 28 (Heim) Sym. is one of its members (vide my notes on A. pachycarpa, p. 30). Ridley made further additions to the genus. In 1920 he added Balanocarpus ovalifolius, described from one flowering and one sterile sheet from Penang. The first is clearly Hopea intermedia King (?=H. Beccariana Burck) and the second is Shorea Maxwelliana King. In 1922 Ridley added B. pubescens which so closely resembles Hopea pachycarpa (Heim) Sym. that careful study is required to separate them. This clearly should be included in the Pierrea group. The most recent additions to Balanocarpus have been made by Foxworthy who, in 1932, described as B. pahang- ensis a plant clearly related to B. multiflorus (Burck) Sym. which, as I have stated, would be most happily placed along with S. Faguetiana Heim in Shorea. At the same time Foxworthy described as B. ovalifolius a plant, distinct from Ridley’s B. ovalifolius, but almost indistinguishable from B. brachypterus Foxw. I have not yet had the opportunity of examining authentic material of Balanocarpus sphaerocarpus Heim, B. bancanus Beerl., nor B. sibogae Berl., all very imperfectly known species, but sufficient has been said to indicate that the disruption of the genus Balanocarpus is imminent and that it will be possible to distribute its members among various groups of Hopea and Shorea. I am inclined to think that the work of Heim in his : “Recherches” sometimes shows a keener appreciation of the : natural grouping of the species usually relegated to the genus Balanocarpus than does that of his critics. It is, however, undoubtedly true that Heim’s excessive subdivision was premature and, as Burkill [Journ. R. As. Soc. Mal. Br. 3: 6 (1925) ] has remarked, his work “needed, and still needs much further prosecution in order to establish it”. Trimen, King, Brandis, and Foxworthy, who have contri- buted largely to the motley collection of species now included in Balanocarpus, have all expressed dissatisfaction with the genus. Handicapped by the inadequacy of their material as they so often were, they were probably wise to leave the disruption of the genus until more complete material was available. Of recent years, not only has herbarium material of the Dipterocarpaceae accumulated rapidly, but many of the species formerly known only from an incomplete herbarium sheet, are now known intimately as jungle trees. The suggestions I have made above are prompted mainly by the examination of external morphological characters of old and recent material but have not a little support from anatomical Gardens Bulletin, S.s." : 29 data and from observations in the field. When this has been supplemented and the careful work on the embryo and seedling development carried out by Burkill has been correlated with these observations, then the suggested change may be executed, but meanwhile it seems desirable to maintain the polymorphous genus Balanocarpus provided it is recognised as merely a temporary arrangement of convenience. Balanocarpus longiflorus (Brandis) Foxw., MS., comb. nov.: ‘Hopea longiflora Brandis in Journ. Linn. Soc. nae 63 (1895): Balanocarpus grandifolius Ridl. MSS. Herb. Kew. Plate VII. As Brandis’s description is rather brief and was based on flowering material only I have prepared the following with the additional help of Igon 00342 and Richards 2441. Branchlets pale or reddish-brown, coarsely lenticellate, puberulous when young. Leaves narrow oblong—lanceolate or long lanceolate tapering gradually from near the base to the acuminate apex, base rounded or sub-cordate, usually about 24 cm. X 4.5 cm. but sometimes proportionately broader, margins revolute, glabrous on both surfaces, duli olive green above, shining yellow-brown beneath, very coriaceous; midrib depressed above; secondary nerves 12-15 prs., very conspicuous on the lower surface; reticulations fine, inconspicuous above, distinct beneath; petioles sulcate, rugulose, puberulous, black, about 1.3 cm.; stipules caducous, not seen. Panicles axillary, glabrous, dark-coloured, 1 or 2 main branches irregularly racemose, ultimate branchlets with 1-5 unilateral flowers. Flowers elongate pointed in bud, over 1 cm. long including the pedicel (1.7 cm. in Hav. 2120) ; pedicel 2 mm. long, glabrous or sparsely puberulous. Sepals sub-equal (3 inner slightly broader than the 2 outer) deltoid obtuse or broadly ovate acute, glabrous or sparsely puberulous. Petals much contorted in bud, linear, slightly fimbriate along the inner margin, puberulous along the outer portion exposed in bud, “brownish purple’ (Richards). Stamens 15, pairs alternating with single stamens, of 3 sizes ; anthers oval; filaments linear, flattened, broader at the base, several times longer than the anthers; appendage to connective filiform, 1-2 times as long as the anther, erect in bud, later becoming deflexed with the anthers. Ovary globose, glabrous, surmounted by an elongate-conic stylo- podium which tapers gradually into the glabrous style; stigma minute, obscurely 3-lobed. Fruit: sub-sessile: receptacle woody, glabrous; calyx lobes equal, embracing the lower portion of the nut but quite free from it, deltoid, obtuse, 8 mm. high, papyraceous above, thickened below, Vol. VIII. (1934). 30 striated, glabrous; nut long elliptic, sharp pointed at apex, 6 cm. X 1.7 cm., glabrous. The earliest collections of this species examined by me are Haviland 2120 and 2254 from Kuching. Upon the former, which bore fully expanded flowers, Brandis based his Hopea longiflora in 1895. The latter has young flower buds and has been given the MS. name of Balanocarpus grandifolius by Mr. Ridley. In the absence of fruit and with the limited knowledge of the systematic groups within the Dipterocarpaceae then possible, Brandis placed the species in the Huhopea section of Hopea. In 1926 a fruiting collection (00342) was made in Setapok F. R., Sarawak, by Dr. Foxworthy’s collector. A sheet of this was examined by Foxworthy who identified it with Hopea longiflora Brandis and suggested the transference to the genus Balanocarpus. The genus Balanocarpus of King, Brandis and others is not a natural genus [vide my notes under Balanocarpus (p. 26)], and I am therefore reluctant to add yet another species to so heterogeneous a group, but until the classifica- tion of the species in all genera concerned is revised upon a more natural basis efforts must be concentrated upon grouping together those species that are clearly related to one another. Of the species examined by me Balanocarpus maximus King appears to be the closest relative of Hopea longiflorus which is therefore transferred to the genus Balanocarpus. Whether these species should remain generically distinct from Shorea macrantha Brandis, S. Hemsleyona King and some other species in the Brachyptera section of Shorea is a question upon which I am not yet prepared to give an opinion. Collections examined— SARAWAK: Haviland 2120 (Auth. spec. of H. longiflora Brandis) & 2254 (auth. spec. of B. grandifolius Ridl. MS.) ; Foxworthy’s No. (Igon) 00342. P. W. Richards 2441; Gunong Balapau, Ulu Tinjar, c. 800-900 m., primary forest; 2nd Nov. 1932; flower. Field note—‘Medang tiong. Tree 26.7 m. high. Calyx greenish yellow; corolla lobes brownish purple’. Vv Hopea pachycarpa (Heim) Symington, comb. nov.: Pierrea pachycarpa Heim in Bull. Soe. Linn, Paris 2: 958 (1891) et Recherch. Diptéroc. 78 (1892) Pl. 7; Brandis in Journ. Linn. Soc. 31: 113 (1895) in nota; Merr. in Journ. Str. Br. R. As. Soc. special no. 408 (1921) ; Gilg. in Nat. Pflanzenfam. 2nd edit. 21: 263 (1925) in nota: Pierreocarpus pachycarpa Ridl., MSS. in Herb. Kew. Plate VIII. Gardens Bulletin, S.S. j dl Branchlets pale tawny tomentose with decurrent elevate lines. Leaves elliptic-oblong or oblong—lanceo- late, gradually tapering to the long acuminate apex, base slightly unequal, usually rounded on sub-cordate on one side, acute on the other, sub-revolute, about 15-20 cm. K 4-7 cm., glabrous except for the midrib and secondary nerves, drying a milk-chocolate colour on both surfaces or slightly glauces- cent beneath; midrib strongly elevate and tomentose above, slightly elevate and puberulous beneath; secondary nerves 10-14 prs. at an angle of 45° to the midrib, slightly depressed and puberulous above, strongly elevate and fugaceous tomentose beneath; tertiary nerves joining the secondaries in numerous more or less parallel lines, hardly distinguishable on either surface; petioles about 0.7 cm., thick, rugulose, tawny tomentose; stipules caducous (not seen). Panicles of 2-3 racemes fascicled in the leaf axes; racemes 3-4 cm. long, black, glabrous, lax; ultimate branchlets about 6 mm. apart, subtended by minute deltoid caducous stipular bracts, each bearing 3-4 sub-secund flowers about 5 mm. apart subtended by caducous bracteoles. Flowers oval in bud, 4.5 mm. long including the pedicel; pedicel 1 mm. long, ‘glabrous, minutely papillose. Sepals sub-equal in size; 2 ovate, rounded or apiculate at apex, thick, glabrous, 3 oval or rotundate, obtuse mucronate at apex, thinner, minutely ciliate along the upper margin, otherwise glabrous. Petals slightly contorted in bud, oblong, erose at the top and along one margin, tomentose along one half outside. Stamens 15, pairs alternating with single stamens; anthers oval, the posterior cells slightly smaller than the anterior; filaments 2-3 times as long as the anthers, broad and flattened below, gradually narrowing to the filamentous upper portion; appendage to connective filiform, about the same length as the filament. Ovary with stylopodium shaped like an hour glass, glabrous; stylopo- dium puncticulate above; style short glabrous; stigma minute, obscurely lobed. Fruit (mature): stalk hardly protruding from the concavity of the enlarged sepal bases, thick, 2 mm. long, glabrous; receptacle and accrescent calyx lobes thick, woody, glabrous, black, united together and to the nut by a resinous cement, forming a 5-lobed cupule, about 2 cm. high X 1.7 cm. broad; nut ovate-conical, mucronate, glabrous, shining, covered ‘with a pale yellow resinous coating : pericarp woody, thickened in the apical portion and splitting readily into 8 equal portions to one of which the placenta is attached. Embryo: cotyledons collateral, the dorsal exceeding the placental and alone reaching ‘the apex of the ovary, markedly lobed; radicle medial, erect, directed into the apex of the ovary. Vol. VIII. (1934). 32 Pierrea pachycarpa was described by Heim from a single specimen with immature unfertile fruit (Becc. 3314) and made the type of a new genus Pierrea. The inadequacy of the material for the foundation of a new genus is, of course, obvious, and no published information has since been added to our knowledge of the species except that Brandis remarked that it “seems to stand nearest to Balanocarpus and Shorea’. The first and only known flowering collection of this species is Haviland 2251*. This was identified with Heim’s plant by Mr. Ridley who proposed (MSS. in Herb. Kew) to rename the genus Puierreocarpus, Pierrea Heim being invalidated by the earlier homonym Pierrea Hance (1877), now reduced to a section of Homalium Jacq. A sheet of Hav. 2251 from the Kuching Museum, in a better state of preservation than the Kew sheet, has enabled me to give a description of the inflorescence and flowers, while Igon (Foxworthy’s collector) No. 00207 has provided excellent material with mature fertile fruits. The assignation of this plant to the genus Hopea requires some explanation. Pierrea pachycarpa Heim belongs to a group of dipterocarps including the following :—Hopea philippinensis Dyer (1878), Hopea mindanensis Foxw. (1911), Balanocarpus cagayanensis Foxw. (1918), Balano- carpus pubescens Ridl. (1922), probably Hopea papuana Diels (1922)+, and Hopea laxa Sym. (1934). These form an essentially homogeneous group with tomentose branchlets, slightly inzeequilateral leaves, pronounced secondary nerves joined by numerous obscure tertiaries, flowers in solitary or fascicled glabrous racemose panicles, sepals nearly glabrous, petals erose at the apex, 15 stamens with broad filament bases and long filiform appendages, glabrous ovaries shaped some- what like an hour glass, and minute stigmas. The fruits are apparently more diverse, those of the species referred to Hopea having two calyx lobes developed into wings, those relegated to Balanocarpus (as Pierrea pachycarpa itself) having no such wings. The nut of all species is, however, essentially the same. The lower portion is enclosed in the thickened sepal bases which are closely appressed to it. The pericarp is hard and woody, very thick in the apical portion, and splits readily along three lines of weakness into three equal portions, to one of which the placenta is attached. The cotyledons are erect and collateral, the dorsal is slightly * The Kew sheet bears young fruits and flowers that are detached from the leaves; the sheet in the Kuching museum has no fruits but the flowers are attached. | Hopea papuana Diels is known to me only from a Serer of the type kindly communicated to me by Dr. van Slooten Gardens Bulletin, S.S. 33 larger than the placental, and the radicle is medially placed and directed into the apex of the ovary. The first of the above species to be described was H. philippinensis Dyer. Certainly this plant has sufficient in common with AH. odorata Roxb. and its closest allies to justify its inclusion in the genus Hopea but I would not include it in the section Huhopea as Heim, Brandis and others have done. I prefer to consider it along with the species listed above as belonging to a distinct ‘“‘Pierrea” group (vide my notes on Balanocarpus, p. 26). To this croup Hopea pachycarpa (Heim.) Sym. belongs. Collections examined— SARAWAK: Becc. 3314 (Auth. spec. of Pierrea pachycarpua Heim.); Haviland 2251; Foxworthy’s No. (lgon) 00207. Iiopea laxa Symington, sp. nov. Plate IX. Affinis H. pachycarpa (Heim) Symington sed follis majoribus multinerviis, paniculis longioribus laxioribus, pedicellis longioribus differt. Branchlets pale tawny tomentose with decurrent lines elevate for 1-2 cm. Leaves oblong or elliptical—oblong, gradually tapering to the acute apex, base slightly unequal, rounded on one side and sub-acute on the other, slightly revolute, about 27 cm. X 8 cm., glabrous, pale yellow-brown or with a green tinge on both surfaces and sometimes glaucescent beneath (v.s.) ; midrib elevate on both surfaces ; secondary nerves 13-17 prs. at an angle of 45° to the midrib, slightly depressed above, with traces of barbate domatia in their axes beneath; tertiary nerves fine and parallel, numerous, distinguishable only on the under surface; petioles 1.2 cm., thick rugulose, tawny tomentose; stipules linear sub-faleate, 5 mm. long, puberulous, persistent. Panicles of 2-3 racemes fascicled in the leaf axes; racemes vp to 18 cm. long, black, glabrous, lax; ultimate branchlets 2 cm. apart, subtended by sub-falcate tomentose stipular bracts of which the lower sometimes persist; each bearing 4—5 flowers 1 cm. apart, arranged spirally or tending to be secund and subtended by caducous bracteoles. Flowers ovate in bud, 8 mm. long including the pedicel; pedicel 2 mm. Jong, glabrous, sparsely papillose. Sepals sub-equal in size and shape, ovate-deltoid, obtuse, the inner three sparsely _ ciliate along the apical margin, otherwise glabrous. Petals slightly contorted in bud, oblong, erose at the top and along one margin, tomentose along one half outside. Stamens 15, pairs alternating with single stamens; anthers oval, the posterior cells slightly smaller than the anterior; filaments 1-2 times as long as the anthers, broad and flattened below, Vol. VIII. (1934). 34 narrowing abruptly to the filamentous upper portion; appendage to connective filiform, as long as or longer than the filament. Ovary with stylopodium shaped somewhat like an hour glass, glabrous; stylopodium puncticulate above; style short. glabrous; stigma minute, obscurely 3-lobed. Fruit unknown. | Hopea laxa belongs to a group of Hopea which I have called the ‘‘Pierrea” group (vide this paper pp. 27 & 32). Hopea pachycarpa (Heim) Sym. seems to be its closest relative but it has much in common with all the members of the Pierrea group. Being known from flowering material only, it is not possible to predict with absolute certainty the form of the fruits of this species, because some of its nearest relatives in the Pierrea group have fruits of the two—winged Hopea type while others have fruits of the wingless Balanocarpus type. The very slight differentiation of its sepals in flower, however, indicates that the fruits will probably be of the latter type. The species of this group producing fruits of the Hopea type, that I have examined, may be detected by a marked dissimilarity in the sepals even in bud. Collections examined— SARAWAK: P. W. Richards No. 2361 (Type of H. laxa Sym.); Dulit, under 300 m.; 29th Oct. 1932; flower. Field Note—Medang kuning. Crest of ridge, primary forest. Tree, 24 m. high, 5 cm. (?) diam. Traces of buttresses”’. Hlopea sp. P. W. Richards No. 1836; Ulu Koyan, ec. 900 m., white sand (“heath”) forest; Sept. 15th, 1932; young fruit. Field note—“Luis bukit. Fruit deep crimson. Tree, 80 cm. diam. No buttresses. Bark reddish, smooth, without fissures. c. 20-25 m. high. Timber used.” This collection, which bears very young fruit, is clearly of a species of Hopea belonging to the section Dryobalanoi- des. I am of opinion that it represents an undescribed species but owing to the limited material and the fact that diagnostic characters within this section are so difficult to define, it seems prudent to await further collections before attempting to describe a new species. In leaf Richards 1836 is very close to collections of Hopea Pierrei Hance from Cambodia and I should be sorely tempted to refer it to that species were it not that the fruit appears to have a slightly longer pedicel and that to find this species in Borneo would be surprising. Brandis and others do actually record H. Pierrei from Borneo, but this is on the strength of the misidentification by Brandis of Gardens Bulletin, S.S. 4 = a * 35 Beccari 2504 and 3050 which are, I consider, H. Dyeri Heim. With this exception no dipterocarp has as yet been recorded from both Indo-China and Borneo. “Vatica dulitensis Symington, sp. nov. Plate X. V. pallida Dyer et V. albiramis V. Sl. similis sed foliis lanceolatis acuminatis, laciniis calycis fructiferi omnino reflexis quam nuce multo longioribus distinguenda. Branchlets glabrous, grey-brown, becoming dark ferruginous puberulous towards the ends. Leaves lanceolate acuminate, base rounded, usually about 8.cm. < 2 cm. but may be up to 10 em. 3.5 cm., glabrous, when dry shining light red-brown or greenish yellow above, paler and dull beneath ; midrib slightly elevate above, strongly elevate and pale yeliow beneath; secondary nerves about 14 pairs, at an angle of 45° to the midrib, yellow beneath, with frequent short intermediary nerves and reticulations conspicuous on both surfaces; petioles 0.8 cm., puberulous, black; stipules minute, caducous. Panicles terminal and axillary (rarely extra—axillary), branching irregularly from the base, up to 5 cm. long, minutely ferruginous tomentose; ultimate branchlets up to 1.5 cm. long, each bearing 2-5 spirally arranged flowers 1.0-2.0 mm. apart. Flowers oblong in bud, 8 mm. long including the pedicel; pedicel 2 mm., ferruginous stellate tomentose. Calyx cup less than 1 mm. high; lobes apparently valvate, equal, deitoid, obtuse, 0.6 mm. long, tomentose outside, puberulous inside. Petals hardly contorted in bud, oblong, rounded at apex, tomentose outside except at base, puberulous on the upper half inside, “creamy white” (Richards, 1343). Stamens 15, in pairs opposite the sepals, solitary opposite the petals; filaments short, very broad and flattened at the base; anthers almost as short, posterior celis much smaller than the anterior ; appendage to connective conical, blunt. Ovary sub-globose, stellate tomentose; style about as long as the ovary, glabrous; stigma, a smooth conical crown consisting of 3 conical lobes. Fruit (nearly mature?): stalk 6 mm. long, puberulous; receptacle thin, disc-like, not enclosing the nut; accrescent calyx lobes reflexed and curved, sub-equal, elliptical or elliptic-oblong, obtuse, about 2.2 cm. X 0.8 cm., fugacious ferruginous tomentose, with 5 inconspicuous longitudinal nerves, “greenish-yellow, suffused with brown’ (Richards) ; nut globuiar, apiculate, 0.5 cm. diam., ferruginous tomentose, obscurely marked with 3 lines—probably the lines of cleavage on germination of the fruit. Vatica dulitensis belongs to the sub-genus Jsauzxis (Arn.) Brandis. It seems to me to have more in common with V. pallida and V. albiramis than other species on Vol. VIIT. (1984). 36. account of the curved and reflexed fruiting calyx lobes, but its fruits are quite distinct from any hitherto described. Collections examined— SARAWAK: P. W. Richards No. 1343; Dulit, under 300 m.; Aug. 19th, 1932; flower. Field note—‘Resak tiong or resak bukit. Primary forest on summit of steep- sided ridge. Tree, 87 ft. high, diam. 15”. Almost unbuttressed. Bark smooth with a few narrow longi- tudinal splits, 34” thick. Wood yellowish. Corolla creamy white. Peduncles reddish brown. Fruit R. 2543. Fls. with sickly sweet scent. Timber useful. Resin when kept for some months used for torches”’. Native Collector No. 2543; Dulit, under 300 m.; Nov. 10th, 1932; fruit. Field note—“Resak tiong or resak bukit. Primary forest. Tree 11.7 m. high, 24 em. diam. Wings of fruit greenish—yellow, suffused with brown, reflexed. Fruit of same species as an earlier number (1343) from neignbouring tree”. Cotylelobium flavum Pierre, Fl. For. Cochinch., t. 258A (1891); V. Sl. in Bull. Jard. Bot. Buitenz. III, 12, pt. 1: 44 (1932); Cotylelobiwm asperum V. Sl. in Lc. IIT, 10: 401, fig. 3 (1930). P. W. Richards No. 2715; Gunong Santubong (1st Division), c. 300 m., low rain forest on rocky mountain side; Dec. 1932; young fruit. Field note—Fallen branch from tree, probably of moderate size. Fruit pale green, wings bright crimson. Leaves coriaceous, yellow brown beneath.” Distribution—Borneo. C. flavum has been reported from the Malay Peninsula by Brandis, Ridley, van Slooten, and others as a result of the misidentification of Ridley 4630 and other collections of C. malayanum Y.S\. Van Slooten corrected the error in January 1932 after study of the type material concerned, but later Foxworthy (Mal. For. Rec. 10: 246) again recorded the species from the Peninsula on the evidence of four collections from Kuala Kemaman. I am of opinion that these collections also are C. malayanum and that there is no evidence of the occurrence of C. flavwm Pierre outside Borneo. Corrigenda , Notes on Malayan Dipterocarpaceae I [Gdns. Bull. 8. 8. Vol. VII, pt. 2 (1933)]. p. 133—C. F. 27801 cited under Shorea platycarpa Heim is S. palembanica Miq. and should be inserted after C. F. 5696 on p. 142. p. 150—Under Shorea hopeifolia (Heim) Sym. add the typonym Hopea hopeifolia VY. Sl. in Bull. Jard. Bot. Buitenz. III, X, 3: 396 (1929). Gardens Bulletin, S.S 4 37 INDEX Names of groups especially dealt with in this paper are shown in heavy type and synonyms are printed in italics. Anisoptera Korth. pa ae Oa SBS Anisoptera brunnea Foxw pS 17 Anisoptera cochinchinensis Lanessan te T2. 12,.08 Anisoptera costata Korth. eG: Lie 83, MAES, te Anisoptera costata Ridl. “fh 9, I4 Anisoptera Curtisii Dyer 3, 15 Anisoptera Curtisiit var latifolia King Sof 16, 17 Anisoptera glabra Kurz ee mL A ee, 2 Anisoptera glabra Pierre ig a. Es Anisoptera glabra Ridl. ah 6, 7, 8 Anisoptera grandifiora Brandis id i, Le Anisoptera levis Ridl. Bee) FOYT Anisoptera marginata Korth. 3, 13, 18 Anisoptera marginatoides Heim i. rk A Anisoptera megistocarpa V. S]. 3, 10, 14, 17 Anisoptera oblonga Dyer 10, 13, 14, 17 Anisoptera robusta Pierre we eo oop &: Anisoptera scaphula (Roxb.) Pierre 2.3, Me Gs 12 Anisoptera thurifera (Blco.) BI. 6G, i t7, 18 Anisoptera thurifera Foxw. Se 6,°7 Anisoptera thurifera Ridl. 4, 6 Balanocarpus Bedd. | 26, 30, 32 Balanocarpus anomalus King : 27 Balanocarpus bancanus Boerl. 28 Balanocarpus brachypterus Foxw. 27, 28 Balanocarpus bracteatus (Burck) Merr. 2 Balanocarpus cagayanensis Foxw. =p Pe Balanocarpus coriaceus Brandis 27 Balanocarpus Curtisii King 26) 27 Balanocarpus erosa Bedd. 2 Balanocarpus grandifolius Ridl. 29, 30 Balanocarpus Heimii King 27 Balanocarpus Hemsleyanus King 27 Balanocarpus longiflorus (Brandis) Foxw. 27, 29 Balanocarpus maximus King .. B77 80 Balanocarpus multiflorus (Burck) Sym. O20 Balanocarpus ovalifolius Foxw. 2 Balanocarpus ovalifolius Ridl. . 28 Balanocarpus pahangensis Foxw. 28 Balanocarpus penangianus King 7 Balanocarpus pubescens Ridl. LGy. 52 Balanocarpus sibogae Boerl. 28 Balanocarpus sphaerocarpus Heim 28 Balanocarpus utilis Bedd. 26 Balanocarpus Wrayi King 27 Balanocarpus zeylanicus Trim. . 26 Brachyptera Heim (section) 2730 Cotylelobium asperum V. S\. ag Cotylelobium flavum Pierre iat 36 Cotylelobium malayanum V. Sl. 36 Doona Thw. 21 Doona javanica Burck IQ, 2! Doona micrantha Burck sige Doona micrantha var. macrosepala Boerl. TO, 22 Doona odorata Burck ee 18, 21 Dryobalanoides Miq. (section) 34 Euhopea Heim (section) > 25, 2%, 430, 33 Glabrae Heim (section) a ye v1 Homalium Jacq. 4a Be 2 Hopea Roxb. Ie i .22,. 20, 277 20, 325-35 Hopea sp. a3 a 34 Hopea acuminata Mert. Vol. VIIT. (1934). Prin A Pe Pl. If PETE C PL IVA BP IE 3 ‘gl ites gage a Pi Lt 38 meen | Hopea albescens Foxw. +3 te 10,22 Hopea albescens Ridl. ae ies 2 Hopea Beccariana Burck pi he 28 Hopea bracteata Burck ae ae 2G. 2a: Hopea Curtisii King ... oy |) ag) See Hopea dealbata Hance a “ oo Hopea diversifolia Mig. Se a 2h aa Hobea diversifolia Schett. a et 18, 27 Hopea Dyeri Heim _... “te Ph 35 Hopea fagifolia Mia. . 18 24; 28 Hofea fagifolia var. fol. latioribus Boerl.. 1, “24 Hopea fagtfolia var. javanica Boerl. ue 10;° 22 Hopea ferrea Lanessan iG Pee 20, 27 Hopea globosa Brandis Eve Rs i yt age Hopea Hasskarliana Heim A ts 19 Hopea Helferi Brandis ws A a5 Hopea hopeifolia V. Sl. tas apt 36 Hopea intermedia King oti. 2 28 Hopea javanica Heim ot as 1g Hopea laxa Sym. i iy 32, 33. PLAX Hopea longiflora Brandis waa ar 29, 30 Hopea Lowi Dyer ... re. ak 19, 28 Hopea macrosépala Boerl. ae,’ ra 1Q,' 2g Hlopea micrantha B. & H. gi uh IG Hopea micrantha Hook. f. tk = 20 Hopea mindanensis Foxw. ee bs 2 Hopea multifiora Foxw. ee ip in LQ of Sik aa Hopea myrtifolia Mig. AS is Be oe. 26, 27 Hopea nutans Ridl. ... fart ot 22 Hopea odorata Auct. oy ce 18, 20 Hopea odorata Roxb. ... ie 20; 22, 25. 2051 am Hopea pachycarpa (Heim) Sym. ... 27, 28, 30, 34 Pl. VIT} Heopea papuana Diels asi Phe 32 Hopea philippinensis Dver rs Be Eee Hopea Pierrei Hance © 4p dae 2 Hopea plagata Sym, % ce farang 5 Hopea plagata Vidal i ee 24 Hopea sangal Korth. ... a *S 18,25. -Plw“V Hopea scaphula Roxb. ty. af rey Hopea semicuneata Sym. ee ee 22, 248° Rhee Isauxis (Arn.) Brandis (section) ae 35 Petalandra Hassk. ... sale MO, Bie Lzune Petalandra micrantha Hassk. ... io kdige eeleeee Pierrea Hance Aes ms my aa Pierrea Heim i pene 32 Pierrea (Heim) Sym. (section) i 27, 20,.. 33,94 Pierrea pachycarpa Heim 2 4 90. 82 Pierreocarpus pachycarpa Ridl. ... co Wa 9S Richetia Heim y; bs 2 Richetia coriacea Heim ay at 27 Scaphula Parker gt lols 2 Scaphula glabra Parker ey i eh Shorea Roxb. a re ee ee Shorea Faguetiana Heim ass Bh 275: sae Shorea Hemsleyana King ie Bs 30 Shorea hopeifolia (Heim) Sym.
;
/ >»,
_- a 1)
eee eee
ai So eT
a
=r
=
4
Hopea semicuneata, Sym.
Plate VII.
Gard. Bull. S.S., Vol. VIII.
Balanocarpus longiflorus (Brandis) Foxw.
Plate VIII.
Gard. Bull. S.S., Vol. VIII.
Hopea pachycarpa (Heim) Sym.
- Gard. Bull. S.S., Vol. VIII.
Hi i / tn
reat
Hin : :
7 ee
4) TATA
! Mh Ty hin Mas
Shady f rN
EY), iain iW}
Uh i, 1 A
Jom.
/
<& iis
My Lei 7
Ss SS >
, a
~xS
+
aS >
——
5
Li
sy
Hopea laxa, Sym.
Plate X.
Vatica dulitensis, Sym.
~
fe
x av Gar ba NAW LD yee 7,
S Y. y y