BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VOLUME XIV OAKES AMES MEMORIAL VOLUME BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS 1949-1951 PRINTED AND PUBLISHED AT THE BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS From a painting by Blanche Ames, 1946 OAKES AMES 1874-1950 YEAR has now elapsed since ProfessorOakes Ames, who served the Botanical Museum as its principal administrative officer from 1928 to 1945, first as Curator, later as Supervisor, and finally as Director, passed away on April 28, 1950. In the intervening months the mem- bers of the staff of the Botanical Museum have been engaged in completing this volume of the Botanical Museum Leaflets which, soon after his death, it was de- cided to dedicate to his memory. The choice of the Leaflets as the medium for a mem- orial volume was a quite natural one since Professor Ames himself founded the Leaflets and since they rep- resent not only his diverse botanical interests, but also his insistence upon high quality in paper, printing, com- position and illustration. Professor Ames deplored the trend toward cheaper publications. ‘‘A scholarly scientific paper,” he said, ‘‘is a jewel worthy of a proper setting. ”’ The entire staff of the Botanical Museum has shared in the production of this volume, and the papers which it contains represent quite accurately the extraordinary diversity of Professor Ames’ botanical interests. Of the fourteen papers, ten are concerned with taxonomy and three of these with the taxonomy of the orchids, Professor Ames’ lifelong special interest. His second professional love, economic botany, is represented by eight papers of which five are also taxonomic and three are in ethno- botany, a field which the Botanical Museum, at Professor Ames’ instigation, has fostered in recent years. Paleo- [ vii ] botany, a traditional interest of the Museum and one which Professor Ames was instrumental in reviving, is represented by one significant paper. Professor Ames’ sustained interest in the course in Economic Botany, which he taught for many years, finds representation in “The Story of Pan Chewing in India’ which was sub- mitted as a term paper by a student enrolled in the course. Only one facet of Professor Ames’ botanical career fails to find expression in this volume; his broad philo- sophical outlook upon the Orchidaceae, a plant family to whose study the major part of his life was devoted. Happily this interest had previously been treated in ‘*Orchids in Retrospect,’” a collection of his essays on the Orchidaceae which was published in 1948 by his colleagues to commemorate the fiftieth anniversary of his service to Harvard University. This book also con- tains a biographical sketch. With deep respect for the memory of Oakes Ames, with constant admiration for the standards of excellence for which he stood, and with a firm purpose of perpet- uating the sound traditions which he established, the members of the staff of the Botanical Museum dedicate this memorial volume. Cambridge, Massachusetts April 28, 1951 [ viii ] TABLE OF CONTENTS NuMBER 1 (June 27, 1949) Degradation of Plant Remains in Organic Sedi- ments By Etso S. BarRGHOORN . NuMBER 2 (October 26, 1949) Plantae Colombianae XII By Ricuarp Evans SCHULTES NuMBER 8 (December 7, 1949) Nomenclatural Notes and New Concepts of Tropi- cal American Orchids By CHARLES SCHWEINFURTH . Noumse_er 4 (February 6, 1950) Studies in the Genus Hevea III By Ricuarp Evans ScHULTES The Identity of Ucuqui By JoAo Murca PirEs aND RicHarp Evans SCHULTES The Correct Name of the Yaupon By Ricuarp Evans SCHULTES A New Variety of [lex vomitoria from Southern Mexico By A. J. SHarp [ ix ] 21 49 87 97 107 Number 5 (March 20, 1950) Plantae Austro-Americanae VI By Ricuarp Evans SCHULTES Numser 6 (November 6, 1950) Notes on Peruvian Cinchonas I By W. H. Hopce Number 7 (December 11, 1950) Prehistoric Plant Remains from Bat Cave By C. E. Smita, Jr. NuMBER 8 (January 15, 1951) The Story of Pan Chewing in India By M. Gowpa NuMBER 9 (March 138, 1951) New Orchids from Africa By V.S. SUuMMERHAYES . A Revision of the Genus Angraecopsis By V. S. SUMMERHAYES NuMBER 10 (April 18, 1951) Whence Came Maize to Asia? By Paut C. MANGELSDORF AND Doucutas L. OLIVER [x ] 109 137 157 181 215 240 263 INDEX OF ILLUSTRATIONS PLATE pamece Cacecnt.c 2 « fa « 5% > OSG MEV Ambelania zschokkeiformis Marcg. . . . . . XXX Brassavola ovaliformis C. Schweinf. . . . . . XVII Cecropia porvenirensis Cuatr. ........ VII Cerro del Castillo, Rio Apaporis, Colombia. . . . X Cinchona glandulifera R. & P. . . . . . XX XVIII Cinchona Pahudiana Howard. . . . . . XXXIX Cinchona parabolica Pav. . ......... XL Combretum karijonorum FP. EH. Schultes . . XXXVI Cranichis ciliilabia C. Schweinf. . . . . . . . . XII Degraded wood ......... LILIUO,IV Discussion of the black-drink, in Catesby’s **Natural History of Carolina, Florida and the Bahama Islands” ......... XXXVI Distribution of Areca Catechu Z. in eahivaiion GD) cs aces sc SS ee A ew De EVE Distribution of the betel-chewing habit (Map) XLVI Earliest publication of the name Stphonia brasiliensis, in Adr. de Jussieu’s ‘‘De Euphorbiacearum generibus” . . . .. . XVIII [ xi ] Epidendrum flexuosissimum C. Schweinf. . . XIV Epidendrum pajitense C. Schweinf. . . . . . . XV Epidendrum strictiforme C. Schweinf. . . . . XVI First Kuropean account of the black-drink, in de Vaca’s ‘‘Relacién y commentarios’””» . XXIV POSSI WOOK, 4.6% oe be ee eR Oe Herrania Camargoana R. FE. Schultes X XIX, XXXII Herrania kanukuensis R. H. Schultes . . . XNXXIII Herrania kofanorum R. E. Schultes XX VIII,X XXIV Herrania Mariae (Mart.) Decaisne ex Goudot var. putumayonis R. Hf. Schultes . . . . . XXX Herrania nycterodendron R. EH. Schultes . . XXXV Herrania pulcherrima Goudot var. pacifica R. E. Schultes ........... . XXVIII Hevea brasiliensis (Willd. ex Adr. de Juss.) DTU MRATER 6c i i we we ea we Ee Ke Ilex vomitoria (Solander in) Mit. . 2. . 2. XXV Lindakeria nitida Aullip & Schultes. . . 2... . IX hbue glomerata Killip & Schultes. . 2... . XI Normal wood. .......... 2.2... 1V, V Octomeria pygmaea C. Schweinf. . . . . . . XIII Parascheelia anchistropetala Dugand. . . . XXVII { xii ] Piper Betle LL. ........2.4..+.,. MLVII Plains of San Augustin, New Mexico, near Pee | CS a ke ee eg eee Plant remains from Bat Cave. . . . XLII-XLIV Pouteria Ucuqui Pires & Schultes . . . XX-XXIII Rapata modesta Maguire ........ =XXXI Rhytidanthera mellifera R. H. Schultes. . . VIII, X Roots isolated from post-glacial peat. . . . .. VI Sandia Valley and Ramospata Ridge (Dept. of Puno) Peru ........... ~XXXVII Solanum quitoense Lam. ...........3X Types of beeda used in pan-chewing . . . . . XLVI Vochysia lomatophylla Standl. . ...... =. VII ZeaMaysh. ............ .XLVIII [ xii | INDEX TO GENERA AND SPECIES ACACIA Catechu Willd., 195,209 ACIOTIS aequatorialis Cogn., 40 polystachys (Bonpl.) Tr., 40 purpurascens (Aubl.) Tr., 40 adike, 185,206 AERANGIS, 240 AERANTHUS trifurcus Reichb.f., 252,253 AEROBION parviflorum Spreng., 255 agase, 210 AGERIA Cassena (Michx.) Raf., 98 atoho-ayo-o, 40 ajuo-wo, 240 allu, 210 AMARANTHACEAR, 172 AMARANTHUS blitoides S. Wats., 172,179 hybridus L., 172,179 Powellii, S. Wats., 172,175, 179 sp:,-172,179 amarilla, 154 AMBELANIA cuneata Muell.-Arg., 135 zschokkeiformis Markg., 135 amber, 197 amli, 209 AMOMUM subulatum Roxb., 206 ANACARDIUM occidentale L., 212 ANETHUM graveolens L., 208 ANGRAECOPSIS, 240,241, 242 amaniensis Summerh., 246, 260,261 Boutoni (Reichb.f.) Perr. de la Bathie, 261 breviloba Summerh., 244,246, 261 sect. Cardiochilus, 243,258 sect. Coenadenium, 244,259 comorensis Summerh., 252 dolabriformis (Rolfe) Schltr. , 245,252,253,255 elliptica Summerh., 245,250 sect. Eu-angraecopsis, 243, 246 Jalcata (Thunb. ) Schltr., 240, 241 gracillima (Rolfe) Summerh., 244,248, 249,250 holochila Summerh., 246,260 ischnopus (Schltr. ) Schltr., 24.0, 245,247,252 macrophylla Summerh., 245, 247,249 [ xv ] occidentalis Schltr., 254 parviflora (Thou.) Schltr., 240,246,255, 256 Pobeguinii (Finet) Perr. de la Bathie, 246,257 pusilla Summerh., 244,246, 258 tenerrima Krénzl., 240,248, 244,247,248 tenuicalear Summerh., 260 Thouarsii Perr. de la Bathie, 252,253 tridens (Lindl. ) Schltr., 245, 252,254 trifurea (Reichb f.) Schltr., 245,250,252,257 ANGRAECUM amaniense Krinzl., 247 Boutoni Reichb.f., 261 dolabriforme Engl., 254 ischnopus Schltr., 247 occidentale Rolfe, 254 parviflorum 'Thouars., 255 tenerrimum Schltr., 247 tridens Lindl., 254 APLOPAPPUS gracilis (Nutt.) 4.Gray, 174, 179 APOCYNACEAE, 43,135 AQUILARIA Agallocha Roxb., 207 ARACEAE, 22 arali, 210 ARECA Catechu L., 185,186,196, 206 var. silvatica Becc., 187 arrow-poisons, 30,41,42,43,120 ARTEMISIA spp., 174,179 ARTOCARPUS altilis (Park.) Fosb., 212 heterophyllus Lam., 212 ASCLEPIADACEAE, 173 ASCLEPIAS sp., 173,179 Assamese maize, 265 asvattha, 210 aswat, 210 ATRIPLEX canescens (Pursh) Nutt., 158, 171,179 BACTRIS balanophora Spruce, 110 bale, 211 battalu adike, 193 beeda, 190,200,201,202 be-ee-cha, 111 BETA vulgaris L., 209 Betel chewing, 181,190 Betel leaf, 181,182,191,195, 204,206 Betel nuts, 181,182,184,185, 191, 192,208,206 Betel palm, 185,186,187,190, 205 Betel pepper, 188,206 Betel vine, 188,189 betre, 188 boia, 136 BomBacackak, 30 { xvi | BoraGinaceak, 158 BOSWELLIA serrata Rovb., 211 BOUTELOUA curtipendula (Michr.) Torr., 159 gracilis (HBK.) Lag. ex Steud., 168,178 hirsuta Lag., 158,159,168, 175,179 BRASSAVOLA nodosa (L.) Lindl., 61 ovaliformis C.Schweinf., 60 Perrinii Lindl., 61 BROTOBROMA aspera Karst. & Triana., 130 BULBOPHYLLUM encephalodes Summerh., 228 lupulinum Lind/,, 230,232 oreonastes Feichb f., 228 porphyrostachys Summerh., 230 unifoliatum DeWildem., 228 buntata, 285 cacao de jacaré, 124 cacao de monte, 129,132 cacao de macaco, 124 cacao gacaré, 123 cacao silvestre, 128,129,130 CacTAcEAk, 173 CALAMAGROSTIS inexpansa 4.Gray, 167,178 capavira, 116 CARUM Bulbocastanum Koch, 208 Carvi L., 208 CASINE yapon Bartr., 98 CASSINE ? amulosa Raf., 98 caroliniana Lam, 98 corymbosa Mill., 105 Paragua Mill, 98,100,101, 103 Peragua L. (1762), 98 Peragua L. (1771), 101 ? ramulosa Raf., 98 vera Floridanorum.... Pluk., 99,100 vomitoria Swanton, 99 yaupon Gatschet, 99 cascarilla crespilla, 153,155 cascarilla crespilla chica, 150 cascarilla crespilla con hajas rugosas, 155 cascarilla del pajonal, 147 cascarilla delgada, 147 cascarilla negra, 147 cascarilla negrilla, 147 CASTILLEJA minor 4.Gray, 158 catechu, 185,195,209 CECROPIA discolor Cuatr., 25 latiloba Miqg., 118 marginalis Cuatr., 25,26 mocoana Cuair., 24 porvenirensis Cuatr., 25 telealbida Cuatr., 118 telenivea Cuatr., 118 tolimensis Cuatr., 118 chakkota, 212 [ xvii ] CHAMAEANGIS, 247 Hariotiana (Kriinzl.) Schltr., 261 Pobeguinit Schltr., 257 CHAMAEDOREA integrifolia (T'rail) Dammer, 110 lanceolata (Ruiz & Pav.) Kunth, 110 champaca, 212 chandan, 210 cha-te-ré, 131 cheape, 212 Cuenopopiaceak, 160,171,179 CHENOPODIUM sp., 171,179 chimbillo, 27 CHLORANTHACEAE, 117 chongaru, 193,196 chua, 197 churu-adike, 194 CINCHONA Calisaya Wedd., 139,141 carabayensis Wedd., 139,140, 141,142,147, 149 Delessertiana Standl., 150,151 glandulifera Ruiz & Pav., 143, 144,145,150 Josephiana Wedd., 141,145 micrantha Ruiz & Pav., 143 Mutisii Lamb. var. B, 151 var. crispa Wedd., 151 officinalis, 147,154 Pahudiana Howard, 141,142, 147,148,149 parabolica Pav. ex Howard, 150,151,152,153,154 pubescens sensu lat., 188,145, 149,154 rugosa Pay. ex Howard, 151, 153 var. crispa Wedd., 151 succirubra Pav. ex Klotzsch, 138 CINNAMOMUM Camphora (L.) Nees & Eberm., 207 Cassia (Nees) Nees ex Bl., 207 zeylanicum Breyn, 207 CITRUS aurantifolia (Chr.) Swgl., 211 grandis (L.) Osbeck., 212 Limon (L.) Burm f., 212 medica L., 212 sinensis (L.) Osbeck, 212 CLIBADIUM asperum (Aubl.) DC., 46 spp., 47 COCOS nucifera L., 207,211 colorada, 154 ComMBRETACEAE, 39,134 COMBRETUM karijonorum R. E.Schultes, 134 laxum Jacq., 39,40 rotundifolium L. C. Rich., 135 ComposITak, 46,158,160,161, 174,179 coqui, 90 CORIANDRUM sativum L., 208 [ xvii | COUSSAPOA intermedia Mart. ex Miq., 119 magnifolia Trécul, 119 CRANICHIS ciliilabia C.Schweinf., 49 crespilla, 153 CROCUS sativus L., 207 croker, 285 CUCUMIS Melo L., 209 CUCURBITA foetidissima HBK., 174,179 Pepo L., 161,162,164,177, 178 CucurBITACEAR, 174 CUMINUM Cyminum L., 208 CUNURIA crassipes Muell.-Arg., 54 curare, 41,42,43,120 CURCUMA longa L., 212 cutch, 198,195 CYPERACEAR, 22,169 CYRTOCHILUM aemulum Kranzl., 73 falcipetalum Krinzl., 68 hastiferum Krianzl., 70 Incarum Krinzl., 70 monachicum, 69 mystacinum Lindl., 66 Pavonii Krinzl., 68 superbiens Krinzl., 73 undulatum HBK., 73 ventilabrum Kranzl., 73 Weberbauerianum Krinzl., 64 CYTOPHAGA, 12 dalchini, 207 dalimbre, 212 darchini, 207 dhanya, 208 DIAPHANANTHE, 247 dodda-jirige, 208 DRYOBALANOPS aromatica Gaertn f., 207 ECCLINUSA Balata Ducke, 91 sanguinolenta Pierre, 91 echenique, 141 ee-ru-che, 42,43,44 ee-sb-pe-ke, 123 EGGELINGIA, 235 clavata Summerh,, 238 ligulifolia Summerh., 236,238, 239 elakki, 206 ELETTARIA Cardamomum (L.) Maton., 206 ELYMUS salina M. EF. Jones, 167,178 ? EMETILA ramulosa Raf., 99 EPIDENDRUM exasperatum Feichb,f., 58 flexuosissimum C.Schweinf., ao pajitense C.Schweinf., 56 sarcodes Lindl., 60 Schumannianum Schltr., 58 [ xix ] strictiforme C.Schweinf., 58 verrucosum Sw, var. myrianthum (Lindl.) Ames & Correll, 58 EPIDORCHIS parviflora O. Ktze, 255 tridens O.Ktze., 254 ERYTHRINA indica Lam., 188,211 spp., 189 EUPATORIUM sp., 174,179 EUPHORBIA penicillata (Millsp.) R.E. Schultes, 29 Tirucalli L., 211 EupuHorBIAcEAk, 29 Facaceak, 170 FALLUGIA paradoxa (D. Don) Endl., 158 See-see-pa-chu, 43 FESTUCA Kingii (8S. Wats.) Cassidy, 167,178 FICUS religiosa L., 193,210 FIMBRISTYLIS miliacea (L.) Vahl, 22 Jish-poison, 47 FLACOURTIACEAE, 36 FOENICULUM vulgare Mill,, 208 FORSTIERA sp., 158 gandham, 210 [ xx ] ganna, 209 GEONOMA interrupta (Ruiz & Pav.) Mart., 111 laxiflora Mart., 111 pycnostachys Mart., 110 geru, 212 GODOYA, 34 godumbi, 212 GONGYLOLEPIS maroana Badillo, 47 GOSSYPIUM arboreum L,, 286 barbadense L., 287 Darwinii Watt, 287 hirsutum L,, 287 Raimondii Ulbr., 286 tomentosum Nutt., 287 GRAMINEAE, 167 granisillo, 117 granisillo del grande, 118 granisillo pequeno, 117 guaco, 47 guamo machete, 27 guava machete, 27 gundi, 197,203 gurvaca, 206 HABENARIA § Diphyllae, 217 egregia Summerh., 217 falciloba Summerh., 215 Keayi Summerh., 217 Lecardii Krénzl., 219 § Multipartitae, 215 pilosa Schltr., 217,219 halasu, 212 harisina, 212 he-de-kd-pe, 120 HEDEOMA Drummondii Benth., 158 HEDYOSMUM toxicum Cuatr., 117 translucidum Cuatr., 117 HEISTERIA cyanocarpa Poepp. & Endl., 119 eurycarpa Stand/l., 120 HELIANTHUS annuus L., 164,175,178 HELICONIA lingulata Ruiz & Pav., 116 psittacorum L,f., 116 Schumanniana Loesn., 116 stricta Huber, 117 HELIOTROPIUM curassavicum L,, 158 HERRANIA albiflora Goud., 131 aspera (Karst. & Triana) Karst., 130 balaénsis Preuss, 128 Camargoana R. EF. Schultes, 120 Dugandii, 128 kanukuensis R. FE. Schultes, 126 kofanorum R.E.Schultes, 126 128 lemniscata (Schomb.) R.E. Schultes, 125,126 laciniifolia Goud. ex Triana & Planch. apud Garcia- Barriga 125 Mariae (Mart.) Deen. ex Goud., 125,129 var, putumayonis R, F. Schultes, 129 nitida (Poepp.) R.E.Schultes, 130 var. aspera (Karst. & Triana) R. E.Schultes,130 forma sphenophylla R. E. Schultes, 131 var. sphenophylla R.E. Schultes, 131 nycterodendron, 129 pulcherrima Goud., 131,132 var. pacifica R. EF. Schultes, 131 purpurea ( Pitt.) R.E. Schultes, 132,133 hess-pda-chu, 41 HEVEA brasiliensis (HBK.) Muell.- Arg., 85 brasiliensis (Willd. ex A. Juss.) Muell.-Arg., 79,80,85 var. acreana Ule, 86 var. angustifolia Ule, 86 mut. Granthami Bartlett,86 var. janeirensis (Muell.- Arg.) Pax, 86 var. latifolia Ule, 86 forma Randiana (Huber) Ducke, 86 var. Randiana (Huber. ) Pax, 86 forma subconcolor Ducke, 86 var. subconcolor Ducke, 86 var. stylosa Huber, 86 forma typica Ducke, 86 Granthami Bartlett, 86 Janeirensis Muell.-Arg., 85 Kunthiana Ule, 80 Randiana Huber, 86 Siebert Warb., 80,81,85 [ xxl ] HIEROPHYLLUS Cassine (Walt.) Raf., 98 HIMATANTHUS articulata (Vahl) Woodson, 43 bracteata (4.DC.) Woodson, 44 hogesoppu, 209 hunise, 209 ILEX caroliniana (Lam.) Loes., 97,99,105 Cassena Michx., 98 Cassine L., 103 Cassine Walt., 98,103 Cassine B L., 98,100 floridana Lam., 98 ligustrina Jacq., 98 Peragua (L.) Trel., 99 religiosa Bartr., 98 vomitoria [Soland. in] Ait., 97,98, 100, 104,105,107 var. chiapensis A. J.Sharp, 107 var. Yawkeyit Tarbox, 99 INGA macrophylla H. & B. ex Willd., 27 setifera DC., 27 jaiphal, 207 jambul, 209 jatri, 207 jayanti, 210 Jharada, 196,203 Jjiluga, 210 Jira, 208 Jirige, 208 JUGLANDACEAE, 170 JUGLANS major (Torr.) Heller, 170, 175,178,179 JUNIPERUS pachyphloea Torr., 165,175, 178 spp., 159,166,178 kabbu, 209 kachu, 185,209 kadapan, 197 kaddipudi, 197 ké-he-pa, 89,90,91 kala zerah, 208 karyirige, 208 karpura, 207 kath, 193,195,196,209 kau-ré, 27 kela, 211 kempugandha chekke, 210 kesar, 207 kesari, 207 khaina, 196 ko-kee-ot-chu, 128,129 kolu kalli, 211 konda mavu, 211 kothamburi, 208 kumkumkesari, 207 ku-na-pe, 47 kun-kun-ni-ai, 30 ku-see-ye-hé-pa, 42 LaBiATAE, 158,173 LAPPULA [ xxii | Redowskii (Hornem.) Greene, 158 lavang, 206 lavanga-churu, 194 LEANDRA divaricata (Naud.) Cogn., 40 le-che-mah, 135 Leeuminosak, 27,158,173 LEMNA valdiviana Philippi, 23 LeMNACEAR, 23 LEPIDOCARYUM tenue Mart., 111 Litiaceak, 169 lime, 197 LINDAKERIA maynensis, 37 nitida Killip & R.E.Schultes, 36 LIPARIS Deistelii Summerh., 226,228 odontochilos Summerh., 226 LISTROSTACHYS parviflora S.Moore, 255 tridens Rehb.f., 254 trifurca Finet, 252 LogGanraceak, 41 laja, 154 lulo, 44,45 LYCIUM sp., 158,173,179 MAHUREA tomentosa Ducke, 133 maiz indio, 273 MALAXIS Chevalieri Summerh., 226 Hirschbergii Summerh., 225 katangensis Summerh., 221 S Katochilos, 221,223,224 Maclaudii (Finet) Summerh., 225,226 pygmaea Summerh., 221,223 MAMMILLARIA sp., 159 MANDEVILLA nerioides Woodson, 44 MANGIFERA indica L., 211 MANIHOT esculenta Crantz, 213 MANILKARA spp., 92 maraganesu, 213 ma-sha-kve, 45 masala, 198,202 mavu, 211 MAYNA glomerata Killip & R.E. Schultes, 38 longifolia Poepp. & Endl., 39 MELASTOMACEAE, 40 MENISPERMACEAR, 120 MENTZELIA pumila (Nutt.) T. & G., 158 MESOSPINIDIUM Wallisii Reichb.f., 67 MICHELIA Champaca L., 212 misreys, 208 mita zerah, 208 [ xxiii | MONARDA sp., 178,179 Moracekagk, 24,118 MORINGA oleifera Lam., 210 MUHLENBERGIA rigens (Benth.) Hitche., 168, 178 muhuri, 208 mun-tai-ru-chee, 42 murphy, 285 MUSA paradisiaca L., 211 subsp. sapientum (L.) O. Ktze., 211 Musaceak, 116 mu-se-nda, 130 MYRISTICA fragrans Houtt., 206 MYSTACIDIUM, 242 dolabriforme Rolfe, 253 gracillimum Rolfe, 248 pedunculatum Rolfe, 255,256 Thouarsii Finet, 252 tridens Rolfe, 254 trifurcum Durand & Schinz, 252 naranjillo, 45 nariyal, 207 NEOFINETIA, 241 nerale, 209 NICOTIANA rustica L., 209 Tabacum, L., 209 NIGELLA sativa L., 208 nimbu, 211 nugge, 210 Ocunackak, 82 OCTOMERIA minuta Cogn., 54 pygmaea C. Schweinf, 53 ODONTOGLOSSUM angustatum Lindl. 1837, 62, 638 angustatum Lindl. 1846, 62 aureo-purpureum Reichbf., 63,64 bellum Schltr., 62 brevifolium Lindl., 64,65 var. Weberbauerianum (Kransl.) C. Schweinf. , 64 compactum Reichb.f., 63 flavescens Rolfe, 65 Koehleri Schltr., 63,64 Loesenerianum Schltr., 62,63 longifolium Lindl., 67 mystacinum ( Lindl.) Lindl., 66 obscurum C.Schweinf., 67 retusum Lindl,, 65 rigidum Lindl., 66 tetraplasium Reichb.f., 62 Wallisii Lind. & Reichb.f., 67 Weberbauerianum Schltr., 64 Wyattianum G. Wils., 67 OECEOCLADES parviflora Lindl., 255 OLacackAk, 119 ONCIDIUM aemulum Reichb.f. & Warsc., 1%,19 bryolophotum Reichb.f., 69,70 chrysopyramis Reichb.f. & Warsc., 72 [ xxiv ] Davisii Reichb.f., 68,69 falcipetalum Lindl., 67,68 hastiferum Reichb.f. & Warsce. 70 heteranthum Poepp. & Endl., 69,70 incarum (Krdnzl.) C. Schweinf. 70 inferlobum Hort., 72 inops Cogn. & Rolfe, 69 ionodon Reichb.f., 68 macranthum Lindl., 71 var. hastiferum (Reichbf. & Warsc.) C.Schweinf., 70 megalous Schltr., 69 monachicum Reichb.f., 69 obryzatum Reichb f. & Warsce. 71 Pavoni Reichb.f. ex Lindl., 68 pyramidale Lindl., 71,72 superbiens Reichb f., 72,73 undulatum (HBK.) Lindl., 738,74 undulatum Salisb., 73 undulatum Sims., 73 undulatum Warner & Williams 12 ventilabrum Reichb, fi & Warsc., 73,74 Weberbauerianum Krinzl., 64, 65 06-le-da, 89,90,91 OPUNTIA spp., 159,173,175,179 ORNITHOCEPHALUS gladiatus Hook., 77 var. peruvianus C. Schweinf. nr if 6-s6-pee-ki, 123 OXYTROPIS sp., 173,179 Patmark, 110 pa-ma, 27 pan, 181,182,196,197,198,199, 203,205,206 pan-chewing, 181,183,184,198, 202 pan dena, 181 pan trinity, 181,198 pangra, 211 pannerale, 212 PARASCHEELIA anchistropetala Dugand, 112 PERICOME caudata A.Gray, 174,179 PERSEA americana Mill., 91 PETALOCENTRUM, 75 angustifolium Schltr,, 76 bicornutum Schltr., 76 pusillum Schltr., 76 PHASEOLUS vulgaris L., 163,164,177,178 PHYLLANTHUS fluitans Benth. ex Muell.-Arg., 29 PIMPINELLA Anisum L., 207 Pinaceak, 165 pinang, 206 pinang asin, 194 pinang blah, 194 [ xxv ] pinang kossi, 194 pinang salai, 194 PINUS, 12 edulis Engelm., 159,165,175, 178 ponderosa Dougl. ex P. Laws., 159,165,178 sp., 165,178 PIPER Betle L., 188,206 PISTIA Stratiotes L., 22 platanillo, 116,117 platanote, 44 PLEUROTHALLIS Broadwayi Ames, 51 gongioglossa Schltr., 52 guadalupensis Cogn., 51 leucantha Schltr., 52 nana A. & S., 51 Sanchoi Ames, 52 Williamsti Ames, 51 POA Fendleriana (Steud. ) Vasey, 167,178 secunda Presl, 167,178 POPULUS tremuloides Michx., 170,175, 179 POUTERIA Ucuqui Pires & Schultes, 87, 90,91 pratie, 285 PROTIUM sp., 211 PSIDIUM Guajava L., 212 PTELEA sp., 178,179 PTEROCARPUS santalinus L., 193,210 puch-pee-d, 89,91 pudi-adike, 199 puh-pia, 89,90 PUNICA Granatum L., 212 QUARARIBEA bracteolosa (Ducke) Cuatr.,32 putumayensis Cuatr., 30 Schultesii Cuatr., 31 QUERCUS, 160 Gambellii Nutt., 170,179 grisea Liebm., 170,175,179 sp., 170,171,179 RaApATEACEAR, 112 RAPATEA longipes Spruce ex Kérn., 112 115 modesta Maguire, 112,115 Spruceana Kérn, 116 RHAMNACEAE, 160,173 RHAMNUS sp., 173,179 RHAPHIDORHYNCHUS Pobeguinii Finet, 257 RHUS trilobata Nutt. ex T. & G., 158 RHYTIDANTHERA, 32 magnifica (Gleason) Dwyer, 32,34,35 mellifera R. E.Schultes, 33,34 35 [ xxvii | splendida (Planch.) van Tieghm., 34 suleata van Tieghm., 34 RIBES ? aureum, 172,179 sp., 158,172,179 RONDELETIA rupicola R.Schum., 46 var. chiribiquetana FR. E. Schultes, 46 Rosacear, 160,172,179 RuBIACEAE, 46 ruktochandan, 210 RUTA graveolens L., 213 Rutaceak, 160,173 SACCHARUM oficinarum L., 209,210 SACCOLABIUM parviflorum Cordemoy, 255 occidentale Krinzl., 254 salei, 211 SALICACEAR, 170 SALSOLA pestifer 4. Nels., 158 sampig, 212 saora, 210 SAPINDACEAR, 30 satapushpa, 208 saunf, 208 SAUVAGESIA erecta L., 36 SAXIFRAGACEAR, 172 SCIADOTENIA toxifera Krukoff & Smith, 120 SCIRPUS Olneyi A.Gray, 169,175,177, 179 validus Vahl, 169,175,178, 179 sp., 169,179 see-a-se-sep, 42 SENECIO longilobus Benth., 158 SEQUOIA, 12 SERJANIA dasyclados Radl., 30 var. sibundoya R. E. Schultes 30 SESAMUM indicum L., 210 SESBANIA aculeata Pers., 210 aegyptica Pers., 210 grandiflora Pers., 210 spp., 189 shah-zerah, 208 Shunti, 212 SIGMATOSTALIX, 74 bicornuta Rolfe, 76,77 graminea Poepp. & Endl., 75 hymenantha Sedltr., 75 macrobulbon Krénezl., 75 peruviana Rolfe, 76,77 pusilla Schltr., 76 SIPHONIA brasiliensis HBK., 79,80,82, 83,85 brasiliensis Willd., 82,83,85 Jjaneirensis (Muell.-Arg. ) Cook, 86 Kunthiana Baill., 79,85 Ridleyana Cook, 81,86 [ xxvii ] SITANION Hystrix (Nutt.) J.G.Smith, 167,178 SMILAX calophylla Wall, 203,208 SOLANACEAR, 44,173 SOLANUM quitoense Lam., 44,45 sonp, 208 SORGHASTRUM nutans (L.) Nash, 168,179 SORIDIUM Spruceanum Miers, 110 SPECKLINIA graminea Poepp. & Endl., 75 SPHYRARHYNCHUS, 233 SPIRODELA polyrrhiza (L.) Schleid., 23 SPOROBOLUS airoides (Torr.) Torr., 158, 168,175,178 Wrightii Munro ex Scribn., 168,178 STERCULIACEAR, 120 STRYCHNOS amazonica Krukoff, 41 Erichsonii Rich. Schomb., 41 guianensis (Aub/.) Mart., 42 javariensis Krukoff, 42 Jobertiana Baill., 42 Peckii B.L. Robinson, 43 subcordata Spruce ex Benth., 43 SUAEDA suffrutescens S. Wats., 171, 179 su-he-sé-pa, 43 sunnadu-dabbi, 198 supari, 206 su-sé-pe, 43 SWARTZIA bracteosa Mart. ex Benth., 28 SYZYGIUM aromaticum (L.) Merr. & Perry, 206 Cumini (L.) Skeels, 193,209 Jambos (L.) Alston, 212 TAMARINDUS indica L., 208 tambul, 189,206 tambula kodi, 181,189 tata, 285 tengu, 207 TERNSTROEMIACEAR, 138 THEOBROMA Cacao L., 133 purpureum Pitt., 133 til, 210 TITHYMALUS penicillatus Millsp., 2 tobacco, 196,197,203 tombacu, 196,197,202,209 toronja, 45 TRICERATORHYNCHUS, 232 viridiflorus Summerh. , 233,234 TRIDACTYLE anthomaniaca (Reichb,f. ) Summerh., 236 Scottellii (Rendle) Schltr. , 236 TRIPSACUM, 279 XXVI111 [ ] TRISETUM Wolfii Vasey, 167,178 TRIURIDACEAE, 110 TROCHODENDRON, 12 tsau-hess, 111 TYPHA latifolia L., 166, 169,175,177, 178 TYPHACEAE, 166 tzu-se-hé-pa, 42 ucuqui, 87,89,90,91,92 ucuquirana, 91 UNCARIA Gambir (Hunt.) Roxb., 196, 209 unde-adike, 193 upakunchica, 206 u-su-se-é-pa, 43 VANILLA africana Lindl., 221 imperialis Krdnzl., 219,221 polylepis Summerh., 219 ramosa Rolfe, 221 VELLOZIA phantasmagoria R. Fk. Schultes, 23 VELLOZIACEAE, 23 VERBENA ciliata Benth., 158 VERBESINA sp., 174,179 VIBURNUM laevigatum, 103 sp., 101 VOCHYSIA lomatophylla Standl,, 28 obscura Warm., 28 vismiifolia Spruce ex Warm., 29 VoOcHYSIACEAR, 28 yaupon, 97 YUCCA baccata Torr., 169,179 spp., 161,169,179 yuct, 91 ZEA Mays L., 178 ZINGIBER officinale Rosc., 212 sira, 208 [ xxix | ERRATA page 25, line 31 for ramanentibus, read remanentibus page 115, line 33 for amaz6nica, read amaz6nico page 134, line 30 replace the comma with a period page 150, lines 7, 10 and 15 for Matthews, read Mathews page 195, line 31 for Areca, read Acacia page 242, line 23 for Kraenzlin, read Kraenzlin page 249, lines 5 and 9 for E., read A. page 252, line 32 for H. Perrier, read Perrier de la Bathie page 253, line 32 for Excell, read Exell page 281, line 7 for Coons, read Coon Plate XXXI for 4, read 5; for 5, read 6; for 6, read 7; add 4 to figure at the left of 3 Issued April 28, 1951 oe at BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 14, No. 1 CamsripGr, Massacnusetts, JuNE 27, 1949 DEGRADATION OF PLANT REMAINS IN ORGANIC SEDIMENTS BY Eso S. BARGHOORN One of the least understood but at the same time fun- damental processes which operates in nature is the al- teration of plant residues into economically important substances such as peat, lignite and coal. In its broadest terms this degradation of plant substance must be re- garded as a phase of the major organic cycle of carbon. It is obvious, however, that in ‘‘geologically permanent”’ large-scale accumulations of organic complexes such as the fossil fuels, we are concerned essentially with major deviations from, rather than participation in, the carbon cycle. In a certain sense, therefore, the study of the aceumu- lation of plant residues and their subsequent alteration comprise an area of botanical investigation which lies between microbiology on the one hand and the geologi- cal and chemical aspects of sedimentation on the other. Microbiological studies of degradation, however, are more often centered on specific organisms or on the meta- bolic products of their activity, rather than on an analy- sis of the effects produced on various naturally occurring substrata. In addition, emphasis in microbiological stud- les is quite logically placed more on the degradative as- pects than on the accumulative aspects of the organic cycle. [1] The ultimate fate of organic residues in nature is de- termined in large part by diverse factors of the environ- ment: physical, chemical and biological. The vast bulk of plant and animal remains is quite rapidly reduced, through complex energy-releasing processes, to simpler organic and inorganic compounds; and_ conversely, through complex energy-consuming processes these are re-incorporated into the tissues of living organisms. Con- spicuously active and effective in the biochemical decom- position changes are the fungi and bacteria. The rate of degradation due to microbiological activity is ordinarily most rapid in a warm, moist, highly aerobic environment. Favorable conditions are probably fulfilled to the greatest degree on or just below the surface of the soil. The ratio of accumulation versus degradation in the soil environment provides variable amounts of the so- called humus or organic component of the soil. If the environment is excessively wet, and more particularly if the availability of oxygen becomes deficient, the rate of degradation of plant tissues is greatly reduced. Such conditions reach their extreme in stagnant, poorly drained and relatively shallow basins such as swamps and bogs. Under such circumstances the accumulative phase exceeds the degradative phase and there results a gradual accretion of modified plant residues. In discuss- ing the degradation and preservation of plant tissues it is therefore essential that reasonable distinction be drawn between environments in which aerobic microbiological changes may proceed at a rapid rate, as in soil, and sub- aqueous environments of submergence and oxygen de- ficiency. This distinction seems to be of fundamental importance since the character of the degradative changes is greatly influenced, if not primarily determined, by the degree of these contrasting conditions of the biological and physical environment. It might almost be said that [2] at one extreme of the spectrum of organic decay there are the various processes which operate in the soil and lead to the formation of soil organic matter, while at the other extreme there are the conditions of anaerobic sub- mergence. In all cases microbiological processes induce physical and chemical changes in the plant tissues in- corporated in the substratum, but the degree of change and the duration of the change vary profoundly. Some understanding of the degradation of plant tissue is essential to a satisfactory interpretation of many types of plant fossils, since the variously modified remains of plant parts constitute the bulk of our knowledge of the paleontological record of plant life. For the most part these source materials are fragments or occasionally en- tire plants which fortuitously entered basins of deposition in which degradative processes were retarded and event- ually inhibited. Under certain unusual and poorly known conditions the infiltration of mineral salts in solution and their subsequent precipitation resulted in an unusually perfect preservation of the original structure of the plant tissues. This is well known in the case of many fossil woods and other plant parts which are silicified or calci- fied. The sequence of events preceding mineralization, however, appears to have been basically similar to that which prevails in any accumulation of plant fragments, as in the formation of peat and other carbonaceous sedi- ments. There seems to be no evidence that special envi- ronmental conditions have been responsible for the ‘‘bio- logical fixation’’ of mineralized plant fragments, i.e., their unusual retention of organic structure. Rather, the unusual conditions have been those leading to precipita- tion of the mineral matrix itself, as in the calcareous nodules, or ‘‘coal balls,’’ of certain coal seams, and the formation of silicified nodules. In anticipation of the evidence to be shown later in [3 ] this paper, it may be stated here by way of a generali- zation that: (1) The anaerobic degradation of the plant cell wall in sediments involves fundamentally similar struc- tural changes in all plant tissues; and (2) These degrada- tive changes are directly related to basic physical, chemical and structural features of the cell wall. The significant features in the general pattern of struc- tural change in the anaerobic degradation of wood were described in a previous study of numerous woods recov- ered from an archeological site in Boston, Massachusetts (Bailey and Barghoorn, 1942). Since the publication of this earlier study, a far wider range of material from various geological horizons has been examined anatomi- cally and, in part, microchemically. A more detailed de- scription of these studies is in the course of publication (Barghoorn 1949),' but it seems desirable to review here the salient features of these investigations and their re- lation to various paleobotanical and anatomical problems as well as their possible bearing on dating archeological remains. In wood of varying age, entombed in mineral or or- ganic sediments, and permanently submerged, there oc- curs a gradual reduction of the amount of cellulose con- tained in the original unmodified cell walls. Depending on the availability of oxygen during degradative changes, and its effect on microbiological activity, a variable de- gree of ‘‘humification’’ results. Stumps of trees, sub- merged by sudden inundation, commonly occur embed- ded in peat deposits, yet may show little evidence of the ‘‘humification’’ so characteristic of the peat. For exam- ple, in the fresh water peats and ‘‘buried forests”’’ of the ‘As a section of the second monograph on the archeology, geology, stratigraphy and paleobotany of the Boylston Street Fishweir (Papers of the Robert S. Peabody Foundation for Archeology). [ 4] New England coast, now under tidal influence, numerous stumps and roots may be recovered which retain to a great extent the color of the original wood, yet whose tissues are altered to a soft, almost cheese-like consist- ency. Similarly, the permanently submerged basal parts of wooden piles driven into sediments in both marine and fresh water environments show a peripheral zone of de- graded wood which may be relatively unchanged in color or gross appearance. Archeological remains of wood pre- served in the anaerobic environment of estuarine sedi- ments have been shown to undergo almost no ‘‘humifi- cation’’ or pronounced change from their original color (Bailey and Barghoorn, 1942). On the other hand, plant fragments from autochthonous peats or other organic accumulations in which aerobic degradation has played an important part almost invariably exhibit pronounced changes in color and the other physical and chemical modifications which feature the concept of ‘‘humifica- tion’’ (Waksman, 1938). However, all degraded plant tissues, whether ‘‘humi- fied” or visibly unaltered, are characterized by a signifi- cant reduction of their original cellulose, the extent of loss of cellulose being fundamentally an index of the de- gree of chemical and physical degradation. In view of these facts, which are well supported by numerous ana- tomical and chemical investigations (Mitchell and Ritter, 1934; Jurasky, 1988; Cartwright and Findlay, 19438; Waksman, 1988; Jahn and Harlow, 1942; etc.), it is of interest to examine in detail the structural changes which occur during degradation of the cell wall. In order to orient the significance of these anatomical changes it is desirable to note briefly certain fundamental aspects of the structure of plant cell walls. Among higher plants the presence of a cell wall is the most conspicuous visible feature in the organization of tissues, organs and, [5] in general, of the entire plant body. The cell wall is com- posed initially, in its development from pre-existing cells, of a primary wall, This primary wall is a cellulosic layer (or series of layers) which ordinarily increases in surface area and may vary in thickness during the growth and enlargement of the cell. After growth and enlargement have ceased, a secondary wall may be formed, always in- ternal to the primary wall. In many plant tissues, partic- ularly in reproductive organs, secondary walls are not formed, and, in such cases the primary wall may become thick and conspicuous, comprising the bulk of the cell wall. In the cell walls of wood and woody plant tissues, however, thick secondary walls are ordinarily developed, the major topographic and structural feature of such tis- sues being their greatly thickened secondary walls. The secondary walls of cells in woody tissues consist, except in certain cases, of three morphologically distinct layers or lamellae (Bailey and Kerr, 1985; Bailey, 1988; 1940). The innermost and outermost of these lamellae comprise relatively thin aggregations of cellulosic wall substance in which the crystalline cellulose aggregates tend to be oriented more or less transversely, or in helices of low pitch. The central and ordinarily thickest layer of the secondary wall, on the other hand, is composed most commonly of crystalline cellulose which tends to be oriented more nearly vertically, or in helices of steep pitch (Bailey, 1940). All three cellulosic lamellae may be in- filtrated to varying degrees with encrusting substances such as lignin, terpenes, resins or other ‘‘protective”’ chemical complexes which retard biological and chemical degradation. In the anaerobic degradation of wood it has uniformly been found that a consistent sequence of degradative changes occurs. This sequence shows a significant and fundamental relation to the lamellar structure of the cell [ 6 ] wall. The incidence of degradation of the lamellae is in the following order: (1) Central layer of the secondary wall (2) Innermost layer of the secondary wall (8) Outermost layer of the secondary wall (4) Primary wall’ Progressive stages in the deterioration of the cell wall are shown in Plates I-IV. Three outstanding features may be noted in these thin sections of degraded wood: (1) areduction of the major portion of the secondary wall to a granular, virtually amorphous residue; (2) the reten- tion of a structurally intact cell wall ‘‘layer’’ or ‘‘mem- brane’’ which corresponds in position to the primary wall; and (8) the presence of isolated or contiguous groups of cells whose secondary walls are relatively un- altered. That the conspicuous granular remnants of degraded wall layers are essentially lignin residues may be demon- strated by their total extraction during delignification (compare Plate I, Figs. 1, 2 and 8 with Plate III, Figs. 1, 2 and 38). The visibly intact, persistent layers, or oc- casionally entire cell walls, are truly cellulosic, as may be shown by their brilliant birefringence in the microscope when viewed between crossed Nichols. Birefringence in the thinner cellulosic residual layers is greatly accentu- ated by delignification. Because of their extremely tenuous character, certain structural residues, such as those in Plate I, Figs. 1, 3 and 4 and Plate III, Figs. 1, 2 and 8, might be assumed to consist solely of primary walls. The composition of these tenuous ‘‘membranes,’’ however, can be shown to com- ‘The term primary wall refers here specifically to the wall layers of individual contiguous cells exclusive of the true intercellular sub- stance which is often erroneously, and confusingly, included as a part of the ““primary wall’’ (Kerr and Bailey, 1934). [7] prise five layers viz., the outermost secondary wall layer and the primary wall of the two contiguous cells plus the true intercellular substance between the adjacent primary walls. The five-layered structure of these tenuous ‘‘mem- branes”’ is revealed by their relation to the pit border of bordered pit-pairs. This relationship is most easily ob- served in the late wood tracheids of coniferous woods, as shown in Plate III, Figs. 1, 2 and 3 (also compare Plate ILI, Fig. 8 with Fig. 5). The pit border is solely a struc- tural feature of the secondary wall. It is apparent, there- fore, that the tenuous innermost layer of the pit border, shown in Plate III, Figs. 2 and 8, is the resistant outer- most or first formed layer of the secondary wall. Hence, it may be deduced that the persistent fraction of the cell wall in these tissues consists in large part of the remaining outermost layers of the secondary wall of two adjacent cells, including of course the retained and still more ten- uous primary walls and true intercellular substance. That the persisting or visibly intact cell wall is actually a very minor part of the original unmodified secondary wall is shown in thin sections in which cells possessing intact secondary walls are intermingled with cells exhibiting extensively degraded secondary walls (Plate I, Figs. 1 and 2 and Plate III, Fig. 1). The reason or reasons for the retention of the entire secondary wall in certain cells which are in direct contact with others in which these wall layers are extensively degraded remain obscure. It should be emphasized in connection with its greater resistance to degradation that the outermost or first formed layer(s) of the secondary wall possesses quite dif- ferent staining reactions, optical properties and physical behavior from those of the broad central layer. The first formed layer is more heavily lignified than the central layer and shows far less tendency to swell in strong min- eral acids, such as 72 per cent sulfuric acid or 40 per cent [ 8] hydrochloric acid. It is probable, therefore, that signi- ficant chemical as well as physical differences exist be- tween the first formed and the subsequently formed lamellae of the secondary wall. In these respects the outermost layer of the secondary wall more closely re- sembles the primary wall than it does the later formed lamellae of the secondary wall. The existence of a pronounced difference in the rate of degradation of the various cellulosic layers of a single cell wall presents a seeming paradox in the interpretation of cellulose decomposition. wo questions immediately arise: (1) what are the factors, biological or chemical, which cause the decomposition; and (2) what substance or substances impart such differential resistance to de- gradation within the cell wall itself? The fact that dif- ferential degradation of wall layers is not an isolated phenomenon, but instead a fundamental feature in the anaerobic decomposition of plant tissues emphasizes the significance of these questions and their bearing on the problem of cellulose decomposition in general. In a previous study by the author these questions have been discussed in some detail, particularly with reference to environmental conditions and the possible causes of degradation (Barghoorn, 1949). It has been proposed, though not proven, that the degradation of wood deeply submerged in marine sediments is probably due in large part to the hydrolytic breakdown of cellu- lose rather than directly to microbiological attack.’ It is probable, although again not proven, that the relatively high concentration of hydrogen sulfide (and associated hydro-sulfuric acid) in many carbonaceous marine sedi- ments is responsible in part for the gradual chemical 'This conclusion does not question the apparently well established evidence confirming the existence of anaerobic bacteria at considera- ble depths in peat deposits (Thiessen and Strickler, 1934; Waksman, 1930). Lo] hydrolysis of cellulose. Whether similar or comparable environmental factors are operative in terrestrial carbon- aceous sediments remains to be determined. The general uniformity of degradation in wood after prolonged submergence in sediments of diverse compo- sition attests to the operation of a widely diffused set of factors which result in the gradual hydrolysis of the less resistant cellulosic fractions. In the case of the archeolo- gical remains previously noted it was found that wooden stakes, driven through successive strata of marine silt, peat and glacial blue clay, were uniformly degraded throughout, the physical condition of the stakes being identical in their entire length regardless of the surround- ing media and the duration of submergence (Bailey and Barghoorn, 1942). Hence, the hydrolytic degradation of the cellulose appears to have progressed at an approxi- mately similar rate regardless of the matrix. Similarly, entire stumps, logs or branches which have been sub- merged for periods exceeding thousands of years may be- come uniformly degraded and softened throughout, ex- hibiting no significant differences between their periphe- ral and interior parts. Occasionally the innermost por- tions of larger stumps may retain a core of intact or incompletely degraded wood. When freshly removed from its matrix the wood often shows little or no com- pression failure, nor even any significant change from its original volume; upon drying, however, it contracts ex- cessively. Anatomical and chemical study shows that such degraded wood consists primarily of the lignin res- idue of the original wood substance, its cellulose content being reduced to a small fraction of the original, fre- quently on the order of three to five per cent. The per- sisting fraction of the cellulose, significantly, however, is found restricted primarily to those more resistant layers of the cell wall previously described. All of this evidence [ 10 ] indicates that gradual hydrolysis of the cellulosic matrix of the cell walls has proceeded at a rather uniform rate, the degradative changes first affecting the peripheral zone and slowly extending inward. Ultimate complete degradation of the cellulosic frame- work of the wood cell wall may be delayed for enor- mously longer periods than have elapsed since the be- ginning of post-glacial time (Mitchell and Ritter, 1934; Jurasky, 1988; Barghoorn and Bailey, 1938). Recent extensive studies of lignites of Tertiary age show that in various genera of hardwood trees the lignitized wood retains from two and one half to six per cent of its origi- nal carbohydrate fraction, the major constituent of this fraction being degraded cellulose. In the same deposit lignitized fruits and seeds may show retention of as much as 45 to 50 per cent of their original carbohydrate frac- tion, predominantly cellulose.’ One aspect of the general problem which is of consid- erable theoretical, if not practical, interest is that dealing with the causes of selective retention of cellulose in cer- tain lamellae of the cell wall. Two possible explanations for the selective degradation of cellulose are immediately apparent: (1) the presence in the cellulosic matrix of substances which effectively retard the hydrolysis (either microbiological or chemical) of cellulose; and (2) actual chemical differences in the cellulosic framework of the more resistant lamellae of the cell wall. The first of these possible explanations finds much sup- port from various observations and lines of evidence. For example, it has been established quantitatively by sev- eral investigators that the degree of lignification corre- lates with resistance to decay in wood and other woody tissues (Waksman and Cordon, 1986; Olson, Peterson ' Unpublished data from analyses of the Brandon lignite of Brandon, Vermont. [ 11 ] and Sherard, 1937; Virtanen, Koistinen and Kiuru, 1988). Additional evidence confirming the ‘‘protective’’ effect of lignin has been obtained by the author in a series of experiments utilizing a species of the cellulolytic bac- terium Cytophaga. In a series of liquid cultures each containing mineral salts, filter paper and thin sections of various woods (Pinus, Sequoia and the hardwood T'rocho- dendron), it was found that the filter paper was com- pletely broken down within a period of a week, whereas the various wood sections were unaffected after six weeks incubation. In addition to lignin, other normal constituents of the cell wall may serve to inhibit biological degradation of cellulose, viz., resins, terpenes and tanniniferous sub- stances. Differential resistance of cell wall layers might logically be ascribed, therefore, solely to the effect of these encrusting ‘‘protective’’ substances of which lignin is undoubtedly the most significant in anaerobic degra- dation of woody tissues. In further support of this ex- planation of the selective breakdown of the cell wall, it should be emphasized that the most resistant layer of the secondary wall, viz., the outermost or first formed layer, is often far more heavily lignified than the broad central layer (Bailey and Kerr, 1985). Moreover, the primary wall, the most resistant of all the lamellae of the cell wall, is intensely lignified (Ritter, 1925; Kerr and Bailey, 19384). The effect of degrees of lignification in different lamel- lae of the cell wall often produces striking morphological effects on the mode of attack of wood by various fungi. Fungi belonging to diverse genera of the Pyrenomycetes and Fungi Imperfecti commonly degrade wood by dis- solving the central and often the innermost layers of the secondary wall, but leaving the outermost layer and the primary wall visibly unaltered (Bailey and Vestal, 1937; [ 12 ] Barghoorn, 1944). Early stages of this restricted break- down of the cellulosic matrix are illustrated in Plate IV, Figs. 1 and 2, showing the invasion of wood by fungal hyphae. Similar in nature are the effects of certain brown rots on the degradation of wood. These are shown, per- haps more diagrammatically in Plate IV, Figs. 8 to 6. In some cases the entire secondary wall (with the excep- tion of the tenuous outer layer) is completely removed, leaving a structural residue consisting of five layered ‘‘membranes’’ similar in composition to those previously described from wood degraded under anaerobic condi- tions (compare Plate III, Figs. 1 and 8 with Plate IV, Figs. 3, 4, 5 and 6). All of this evidence indicates that the deterioration of wood is primarily a process of removal of cellulose, the rate of loss being, in many cases, significantly correlated with the degree of lignification or the protection of cel- lulose by other substances. A fundamental exception to this relatively simple ex- planation, however, may be determined by an anatomical study of plant remains extracted from peats. In defloc- culated, and subsequently delignified (‘‘dehumified’’), samples of fibrous peats innumerable delicate plant frag- ments are often released. In grass and sedge peats these fragments are most commonly cellulosic epidermal resi- dues of roots and rhizomes. Microscopic examination often reveals preservation of even the most delicate epi- dermal structures such as root hairs, root cap cells and epidermal papillae (Plate VI, Figs. 1 to 6). Although these tenuous residues of roots have retained minute ana- tomical features in certain parts, they are devoid of their thick walled fibrous or conductive tissues (Plate VI, Figs. 1 and 2). However, in undelignified preparations of comparable roots there may be found degraded rem- nants of the originally thick walled conducting tissues [ 13 ] and fibers, whose secondary walls have been reduced to granular residues. The secondary wall residues and other degraded or ‘‘humified’’ decomposition products are readily extracted (with little or no deleterious action on the remaining cellulose) by means of various techniques utilizing sodium chlorite as the delignifying agent (Barghoorn, 1948). Studies of fibrous peats of varying post-glacial age consistently show a cellulosic residue, often representing avery small fraction of the original sample. The amount of cellulose recovered is much influenced by the tech- nique employed in lignin extraction; direct repeated chlorination or treatment with acidified hypochlorite so- lutions may result in oxidation and partial (or complete) solution of the persisting degraded cellulose residues. Anatomically, however, structural cellulosic residues pre- served in peats consist for the most part of thick, previously unlignified primary cell walls. Tissues possessing thick, lignified secondary cell walls may undergo extensive de- gradation in environments in which even delicate cellu- losic cell walls are incompletely broken down. In recog- nition of these facts it seems a paradox of note that the most delicate tissues of various plant organs may be far more resistant to anaerobic degradation than are cells or tissues possessing thick, frequently heavily lignified sec- ondary cell walls. In view of these observations, which are consistently supported by anatomical studies as well as by chemical analyses of degraded plant remains, it seems quite diffi- cult to interpret selective degradation of different por- tions of the plant cell wall except in terms of chemical differences in the successively formed lamellae of the cell wall. In other words, the ewceptional resistance of the primary cell wall to degradation may be due to chemical rather than physical ‘‘protective’’ factors. Such an inter- [ 14 ] pretation is strongly supported by the fact that primary wall cellulose often persists after degradation of the cel- lulose of heavily lignified layers of the secondary wall. For this and other reasons it seems difficult to interpret selective degradation of cellulose in different parts of the plant cell wall except in terms of chemical differences and resistance to hydrolysis in successively formed lamellae. The persistence of the primary and occasionally the outermost layers of the secondary cell wall is strikingly shown in many silicified woods. Not uncommonly, sili- cified wood, regardless of geologic age, contains unusual amounts of organic residues, occasionally present to such an extent that the demineralized wood may be embed- ded, sectioned and stained much as living tissue (Arnold, 1981; 1941). In such material of diverse groups of plants which may range in age from Devonian to Ter- tiary, there is no cellulose remaining; it seems likely that silicification took place at a time when degradation of the cell wall had not yet passed the stage of the partial retention of a cellulosic structural residue. In other words, the cellulosic framework of the tissues was re- tained for a sufficient length of time to allow preservation of structure before silicification began. After or during the silicification process the remaining cellulose was lost from the tissue, leaving a modified but coherent lignin residue. ‘This modified and silicified residue simulates the original cellulosic residue of anaerobically degraded wood (compare Plate I, Figs. 1 and 8 and Plate II, Figs. 2 and 8 with Plate V, Figs. 1, 2 and 5). Whether the greater resistance to degradation of the cellulosic lamellae of the primary wall and the outermost secondary wall of plant cells is due to intrinsic chemical factors or primarily to physical protection by extraneous non-cellulosic substances cannot be determined without further codrdinated histological and chemical study. [15 ] However, it is evident from anatomical and microchem- ical investigation that resistant fractions of cellulose oc- cur in many organic deposits of various geologic ages and that these resistant fractions are directly related to pre- determined structural features of the plant cell wall; hence they have their basis in the biochemistry of growth and differentiation of the cell wall. Since the ultimate loss of cellulose in plant remains may, under certain conditions, be delayed to varying degrees for immense periods of geologic time, it is appar- ent that simple generalizations can scarcely be made re- garding the complex processes of degradation in organic sediments as a whole. Chemical analyses of peat and humus have supplied data showing gross trends in the chemical alteration of plant remains in deposits (W aks- man, 1938). However, such empirical analyses do not correlate microchemical and anatomical aspects of degra- dation, and in general they interpret natural decompo- sition of plant remains in terms of plant substance rather than in terms of the complex organization of cells and tissues. It is significant, however, that the less favorable the conditions are for microbiological processes, the greater is the retention of the original cellulose in plant accumulations. With increasing geologic age, total loss of cellulose occurs through ‘‘coalification’’ and various changes associated with the transformation of plant resi- dues into higher rank coals. The problems of the earlier stages of cellulose degradation in anaerobic environments and of the processes of ‘‘humification’’ in general are in need of further co6rdinated anatomical, microbiologi- cal and chemical studies. Such investigations will be es- sential before a clear understanding of the major biologi- cal and chemical changes occurring between plant source materials and their geologic accumulation in the form of fossil fuels will be possible. [ 16 ] me SUMMARY . Alteration of plant residues under anaerobic condi- ditions in organic sediments is a fundamental but inadequately known phase of the accumulation of carbon in nature. Details of anatomical changes during decomposition of plant tissues are of importance in the morpholo- gical interpretation of many structurally preserved plant fossils. . Anaerobic degradation of the plant cell wall involves certain basic structural changes which appear to be similar in all plant tissues; these changes are directly related to basic physical and chemical features in the organization of the cell wall. Degradation of the cell wall, either aerobically or anaerobically, is characterized primarily by loss of cellulose; the removal of cellulose most frequently follows a uniform sequence determined by the lamel- lar organization of the wall. The incidence of degradation of cell wall lamellae in lignified tissues occurs in the following sequence: 1. Central layer of the secondary wall 2. Innermost layer of the secondary wall Outermost layer of the secondary wall . Primary wall eo The greater resistance of certain lamellae of the cell wall may be logically assigned to protection of the cellulosic framework by lignin and other ‘‘encrust- ing’’ substances. The persistence of unlignified pri- mary walls in fibrous peats, however, cannot be ex- plained by such mechanical protection, and demands an alternate explanation. [17 ] 7. It is proposed, although it cannot be demonstrated within the scope of this study, that the greater re- sistance to hydrolysis of the primary wall and the outer secondary wall layeris due to chemical differ- ences in the organization of the cellulosic matrix, rather than solely to a physical relation with lignin and other protective complexes. 8. The ultimate loss of cellulose in organic deposits may be delayed for periods measured in terms of geologic time, as shown by its presence in lignites of Tertiary and older age. The rate of loss, however, is greatly influenced by the initial biological condi- tions of deposition as well as by subsequent environ- mental changes accompanying ‘‘coalification. ”’ 9. Mineralization of plant remains apparently often oc- curs at a stage when cellulosic structural residues are still retained. Infiltration and precipitation of min- erals, such as silica and calcium carbonate, occasion- ally cause a ‘‘fixation’’ of structure which, in min- eral form, retains a large measure of biological detail. Such petrifactions may or may not retain a rather high percentage of the original, though chemically modified, organic matter during subsequent geologic change. 9 10. Many aspects of the problem of ‘‘humification’”’ and ‘*eoalification’’ of plant residues are known only in empirical terms. An adequate understanding of these complex processes will necessitate extensive coordinated microbiological, anatomical and chemi- cal investigation. Institute for Research in General Plant Morphology Harvard University [ 18 | LITERATURE CITED Arnold, C.A. 1931. On Callixylon Newberryi (Dawson) Elkins et Wieland. Contrib. Museum of Paleontology, Univ. of Mich. 3: 207-232. —— 1941. The petrifaction of wood. The Mineralogist. 9: 323-324; SOO=SI0O% Bailey, I.W. 1938. Cell wall structure of higher plants. Indust. Eng. Chem. 30: 40-47. —— 1940. The walls of plant cells. Amer. Assoc. Adv. Sci. Publ. no. 14; 31-48. and E. S. Barghoorn. 1942. Identification and physical condi- tion of the stakes and wattles from the fishweir. (In Johnson et al. : The Boylston Street Fishweir.) Papers of the Robert S. Peabody Foundation for Archeology. 2: 82-89. and T. Kerr. 1935. The visible structure of the secondary wall and its significance in physical and chemical investigations of trach- eary cells and fibers. Journ. Arn. Arboretum 16: 273-300. and Mary Vestal. 1937. Significance of certain wood-destroying fungi in the study of the enzymatic hydrolysis of cellulose. Journ. Arn. Arboretum 18: 196-205. Barghoorn, E.S. 1948. Sodium chlorite as an aid in paleobotanical and anatomical study of plant tissues. Science. 107: 480-481. —— 1949, Paleobotanical studies of the fishweir and associated de- posits. Papers of the Robert S. Peabody Foundation for Archeology. (in press) and I. W. Bailey. 1938. Occurrence of Cedrus in the auriferous gravels of California. Amer. Journ, Bot. 25: 641-647. and David Linder. 1944. Marine fungi: Their taxonomy and biology. Farlowia. 1: 395-467. Cartwright, K. St. G. and W. P. K. Findlay. 1943. Timber decay. Biol. Reviews. 18: 145-158. Freudenberg, K. 1932. The relation of cellulose to lignin in wood. Journ. Chemical Education. 9: 1171-1180. Jahn, E.C. and W.M. Harlow. 1942. Chemistry of ancient beech stakes from the fish weir. (In Johnson et. al.: The Boylston Street Fishweir) Papers of the Robert S. Peabody Foundation for Arche- ology. 2: 90-95. [ 19 ] Jurasky, K.A. 1938. Wandlungen des Holzes auf dem Weg zur Kohle. Geologischen Rundschau. 29: 441-461, Kerr, T. and I. W. Bailey. 1934. The cambium and its derivative tissues. No. X. Structure, optical properties and chemical compo- sition of the so-called middle lamella. Journ. Arn. Arboretum. 15: 327-349, Mitchell, R. L. and G. J. Ritter. 1934. Composition of three fossil woods mined from the Miocene auriferous gravels of California. Journ. Amer. Chem. Soc. 56: 1603-1605. Olson, F. R., W. H. Peterson and E. C. Sherard. 1937. Effect of lig- nin on fermentation of cellulosic materials. Ind. Eng. Chem. 29: 1026-1029, Ritter, G.J. 1925. Distribution of lignin in wood, Ind. Eng. Chem. 17: 1194-1208. Thiessen, R. and H.S. Strickler. 1934, The distribution of micro- organisms in four peat deposits. Coop. Bull. no. 61. Mining and Metallurgical Investigations U.S. Bureau of Mines, Pittsburgh, Pennsylvania. Vertanen, Koistinen and Kiuru. 1938. Suomen Kemistilehti. 11: 30 (not seen, abstract in Papier Fabrik 38: 22, 1940). Waksman, Selman A. Humus. 1938. pp. 93-129; pp. 261-288. Williams and Wilkins, Baltimore. ——— 1930, Chemical composition of peat and the réle of microorgan- isms in its formation. Amer, Journ. Science, 19: 32-54. and T. C. Cordon. 1938. On the nature of lignin. Journ. Amer. Chem. Soc. 58: 969-972. ILLUSTRATIONS EXPLANATION OF THE ILLUSTRATIONS Piare I. Fia. 1. Transverse section of the wood of a sub- merged white pine stump, age about 2000 years, showing differential degradation of the cell wall. Note the presence of some intact cell walls, others incipiently degraded and many cells in which the secondary wall is largely reduced to a lignin residue occupying the center of the lumen, 70. Fic. 2. Transverse section of the same material showing incipient degradation of the secondary wall of an isolated tracheid (center) surrounded by cells in which the secondary wall has been largely degraded. 170. Fic. 38. Transverse section of the same, but more highly magnified. Note the discrete mass of the ‘‘coagulated’’ rem- nants of the major part of the secondary wall in the center of each cell. 870. Fic. 4. Tangential longitudinal section of the same material. The dark, heavily stained amorphous material represents the degraded remains of the central and inner layers of the sec- ondary walls. 100, Fic. 5. Transverse section of a spruce pile, submerged in marine silt for 100 years. This section from the outer rings of the pile shows incipient degradation of the central and inner layers of the secondary wall. 500. Fic. 6. Transverse section from a different part of the white pine stump shown in Fig. 1. Note the complete breakdown of the central and inner layers of the secondary wall and the retention of the primary wall and the outermost layers of the secondary wall. 500. PLATE [ EXPLANATION OF THE ILLUSTRATIONS Pirate Il. Sections of wood of submerged white pine stump. Fic. 1. Transverse section. Note the incipient degradation of the inner lamellae of the secondary wall of the tracheids, which shows compression wood, or “‘rotholz’” structure. The heavily stained layers between the degraded inner lamellae and the less heavily stained outer secondary wall are non- cellulosic layers of the secondary wall. »* 500. Kia. 2. Transverse section. The visibly intact layers of the cell wall consist of the intercellular substance, primary wall and the outer secondary wall. The lignin residues of the central and innermost layers of the secondary wall form a discrete cylinder in the lumen of each cell. In this section, as also in Fig. 8, the outermost secondary wall is unusually thick and appears to consist of more than one layer. Unde- lignified, stained with Ruthenium red. “800, Kia. 3. Same section shown in Fig. 2, except more highly magnified, Undelignified, stained with Ruthenium red, 1400, Kia. 4. Transverse section showing varying extent of de- gradation of the secondary wall lamellae. ~ 800. Fic. 5. Transverse section of less degraded wood from inner parts of the stump. Note the varying degrees of degradation of the secondary wall and the retention of a structural frame- work of modified cell walls. Unstained preparation. 220. kia. 6. Same specimen as shown in Fig. 5, but from more extensively degraded wood. Unstained preparation, 530, PLATE EXPLANATION OF THE ILLUSTRATIONS Prare IIL, Fic. 1. Transverse section of wood of white pine stump. The section shows the cellulosic structural residue of degraded wood after extraction of lignin and lignin resi- dues by sodium chlorite treatment. Four tracheids possess visibly intact secondary walls. All other cells are repre- sented by walls consisting of intercellular substance, primary walls and outermost secondary walls. Stained with Rutheni- um red. 500, Fig, 2. Same specimen as Fig. 1. The cellulosic structural residue consists of intercellular substance, primary walls and outermost secondary walls. Evidence for the presence of a secondary wall residue is provided by the persisting first formed layers of the pit borders. Delignified section stained with Ruthenium red. * 870. Fic. 3. Same specimen as Fig. 1. Evidence that the cellu- losic residue consists in large part of the outermost layer of secondary wall is shown by the conspicuous remnants of the borders of bordered pit-pairs. Note the retention of the pit torus. Delignified, stained with Ruthenium red. » 870. Fie. 4. Tangential longitudinal section of a degraded late wood tracheid of white pine showing the granular lignin residue of the secondary wall, Note the remnants of the bordered pit-pairs. Such cell wall residues, when delignified and viewed in transverse section, show the organization rep- resented in Figs. 1, 2 and 8. * 500. Fic. 5. Bordered pit-pairs of a normal spruce tracheid. Com- pare the lamellae with those shown in Figs. 2 and 3.) 870. Fic. 6, Transverse section of “‘rotholz’’ of degraded white pine wood after treatment with 72 per cent sulphuric acid. The more heavily lignified primary walls and outer secondary walls yield a coherent structural residue comparable to that Ys shown in Plate II, Figs. 2 and 8.» 800. kia. 7. Same specimen as shown in Fig. 6, but from differ- ent portion of the stump. Section treated with 72 per cent sulphuric acid. Note the lignin residue of the pit borders and compare with the cellulosic residue of the pit borders shown in Fig. 8.0 500. ae Be PLATE EXPLANATION OF THE [ILLUSTRATIONS Prare IV. Fie. 1. Transverse section of xylem of Laurelia aromatica Juss., showing enzymatically produced cavities in the secondary wall. Note the partial localization of action within the central layers. (After Bailey, I. W. and Mary Vestal: Journ, Arnold Arboretum 28, Plate 209: 19387.) ~990, kia, 2. Transverse section of hard pine wood which was ex- posed to the sea and to the action of marine fungi. Enzy- matic attack on the cell wall is concentrated within the cen- tral layers. Compare with anaerobic degradation as shown in Plates I, Il and III. 500. Fic. 8. ‘Tangential section of xylem of U/mus crassifolia Nutt. showing normal wood structure. * 100. Kia, 4. Tangential section of xylem of Ulmus americana L. degraded by an unidentified brown rot fungus. The structu- ral residue closely resembles delignified residues of anaero- bically degraded wood as shown by comparison with Plate III, Figs. 1, 2 and 8. 100. Fra, 5, ‘Transverse section of xylem of U/mus americana LL. showing normal wood, with scattered gelatinous fibers. LOO. Fic. 6. Transverse section of xylem of Ulmus americana L. degraded by unidentified brown rot. Same specimen as shown in Fig. 4. The structural residue is lgnified cellulose as shown by lignin extraction, ~ 100, PLatrE IV Gy os EXPLANATION OF THE ILLUSTRATIONS Prare V. Fie. 1. Transverse section of an unidentified fos- sil wood of Cretaceous age. The wood is an intensely silici- fied organic residue of the original woody tissue. The cell wall layers retained closely resemble the cellulosic residues of anaerobically degraded wood as shown by comparison with Plates I and II. Ground section. » 182. Kia. 2, Same specimen as shown in Fig. 1, but more highly i y @6y. . ry : . magnified. Note the “lignin’’ residue of the secondary wall. 732. Fig. 38. Robinia Pseudo-Acacia LL. Portion of vessel wall showing structure of the vestured pits. The vestured portion of the pit consists of primary wall. ~< 1532, hic. 4. Robinia sp. Silicified wood from the Tertiary of Mon- tana. Portion of vessel wall showing retention of structure of the vestured pits. Compare with normal wood of Robinia Pseudo-Acacia L. shown in Fig. 8. Ground section. 860. Fie. 5. Same specimen as shown in Figs. | and 2, but with varying degrees of degradation of the cell wall previous to silification. Ground section. 732. Kia. 6. Tangential longitudinal section of same specimen as shown in Fig. 4. The beaded structure of the more resistant primary wall is preserved as an organic residue, consisting of modified lignin and embedded in silica. Ground section. 225, Fig. 7. Same as Fig. 6, except more highly magnified. Ground section, » 860, \ PLATE EXPLANATION OF THE ILLUSTRATIONS Prare VI. Fic. 1. Portion of root of Juncus sp. isolated from postglacial peat. Age approximately 4000 to 5000 years. The heavily lignified fibrous and conductive tissue in the center of the root has been largely degraded, whereas the delicate unlignified epidermal structures are preserved. 132. ia, 2. Root epidermis of Scirpus sp. from same peat. Note preservation of epidermal cells. 126. Fig, 3. Root epidermis of Carex sp. from same peat. * 126. Fic. 4. Root tip of unidentified monocotyledonous root from same peat. Note preservation of minute structural details of root cap. ‘Treated with acidified sodium chlorite. ~ 126, Fig. 5. Cells of a root apex meristem from the same peat showing preservation of primary cell walls and the residue of cell contents. Thin section of peat, unstained, 500. Fic. 6. Root epidermis of Carex sp. showing preservation of unlignified primary walls of epidermal cells. 126. PLATE VI [4 L pres bate ae : ue Re eae ae Be OS Bess Camprinar, Massacnuserts, Ocroper 26, 1949 BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY PLANTAE COLOMBIANAE XII DE PLANTIS PRINCIPALITER AMAZONIAE COLOMBIANAE INVESTIGATIONES BY Ricuarp Evans Scuuutres! THE Amazonia of Colombia, vast and, for the most part, still unknown, has a surprisingly varied flora. Bo- tanical explorations of this northwesternmost part of the great Amazon Valley have been pitifully few in relation to the size of the area and the wealth of its flora. Colom- bia’s Amazon region is divided into the Comisarias of the Putumayo, Caqueté, Amazonas and Vaupés. It com- prises about 404,000 square kilometers or one-third the area of the entire country. It is at once obvious how vital to our understanding of the flora of the rest of Co- lombia, as well as of the whole Amazon basin, is a clear conception of the phytogeographical relationships of the vegetation of this immense area. In the course of identifying recent collections from the Colombian Amazonia, I have found that a number of these represent hitherto apparently undescribed concepts or concepts previously unknown either from Colombia or from its Amazonian regions. It seems advisable to publish, from time to time, collected notes of this type ' Botanist, Bureau of Plant Industry, Soils, and Agricultural Engi- neering, Agricultural Research Administration, United States Depart- ment of Agriculture; Research Fellow, Botanical Museum, Harvard University. [ 21 | Voi. 14, No. as a contribution towards our understanding of the flora of Colombia. I wish to thank the following botanists for their collab- oration: Dr. José Cuatrecasas of the Chicago Museum, who has studied the representatives of Cecropia and Quararibea; Dr. Robert E. Woodson of the Missouri Botanical Garden, who identified the apocynaceous spe- cies; Padre Lorenzo Uribe of Medellin and Bogota, Colombia, who determined the specimens of Inga; Dr. B. A. Krukoff and Mr. Joseph Monachino who have attended to the collections of Strychnos; and Dr. F. A. Stafleu of Proefstation voor de Java Suikerindustrie, Pasoeroean, Java, who identified the representatives of Vochysia. 1 have also to thank Dr. Frederick J. Hermann of the United States Department of Agriculture for placing at my disposal the notes and identified plant specimens which he collected, some for the first time in Colombian territory, in the Leticia area in 1944. CYPERACEAE Fimbristylis miliacea (L.) Vahl Enum. 2 (1805) 287. Collections of Mimbristylis miliacea have been made in British and Dutch Guiana, Venezuela and Ecuador. Hermann 11318 represents apparently the first collection to be reported from Amazonian Colombia. Cotomsra: Intendencia del Amazonas, Leticia, April, 1944, Fred- erick J. Hermann 11318. ARACEAE Pistia Stratiotes Linnaeus Sp. Pl. (1758) 963. This very widespread tropical American plant is known from almost all parts of Middle and South America. Al- though collections have been made from nearly every other district of Colombia, Hermann 11301 would seem [ 22 to be the first reported from the Amazon watershed of the Republic. CotomsrA: Intendencia del Amazonas, Rio Loretoyacu, floating on the river near its mouth, April, 1944, Frederick J. Hermann 11301. LEMNACEAE Lemna valdiviana Philippi in Linnaea 33 (1864) 239. The collections cited below seem to be the first of this minute floating plant from the Amazon watershed of Colombia and adjacent Peru. Cotompia: Intendencia del Amazonas, Rio Hamacayacu, April, 1944, Frederick J. Hermann 11312. Peru: Departamento de Loreto, Rio Amazonas, Isla Cacao, April, 1944, Frederick J. Hermann 11313. Spirodela polyrrhiza (L.) Schleiden in Linnaea 18 (1839) 392. This interesting little aquatic plant has apparently never before been collected in the Colombian Amazon watershed. It has been known from Boyacé and Santa Marta in Colombia and from Guayas Province in Ecua- dor and Loreto in Peru. Cotompia: Intendencia del Amazonas, Leticia, April, 1944, Fred- erick J. Hermann 113165. VELLOZIACEAE Vellozia phantasmagoria PR. I. Schultes in Bot. Mus. Leafl. Harvard Univ. 12 (1946) 130, pl. 19, 20. A specimen of Cuatrecasas 7700 has recently come to my attention during studies in the United States National Herbarium. It would appear to represent Vellozia phan- tasmagoria, a dominant shrub on Mount Chiribiquete which is located slightly southeast of San José del Guaviare and which has a number of floral elements in common with the granitic hill at San José. The sterile [ 23] condition of the specimen, however, precludes definite identification. Nevertheless, it represents the northwest- ernmost collection of the Velloziaceae from the Amazon Valley and from South America. Cotompra: Intendencia del Meta, Rio Guaviare, San José del Gua- viare, terrenos graniticos, sabana, alt. 270 m. **Ramificada, 1-2 m. alt.’’ December 11, 1939, J. Cuatrecasas 7700. MoRACEAE Cecropia mocoana Cuatrecasas sp. nov. Arbor parva. Ramuli scabri, pilis conicis rigidis in- clinatis et pilis longioribus hamatis praediti. Folia ampla, membranacea, peltata, longe petiolata. Petiolus robus- tus, 50 em. longus, striatus, copiose hirsutus cum pilis longis saepe hamatis. Lamina palmata, profunde (8/4— 4/5) 8-lobata; lobis majoribus obovato-oblongis, minori- bus obovato-subrotundatis, omnibus abrupte acuminatis, basi contractis, margine integra vel leviter irregulariterque sinuata, lobo maximo ab insertione petioli usque ad api- cem 54cm. longo et 30 cm. lato, lobo minimo 24 em. longo, 15 cm. lato, parte concrescenti 10-19 cm; supra viridis, strigosa, pilis conicis brevibus et pilis longis acu- tissimis inclinatis munita, nervis principalibus et secun- dariis notata; subtus albida, nervis primariis crasso- eminentibus valde hirsutis, secundariis angulo obtuso 1-3.5 cm. distantibus prominentibus hirsutulis, tertiis roseis transversalibus et reticulatis demum hirsutulis, in- ter hos lamina dense albo-arachnoideo-tomentosa. In- florescentiae masculae geminatae. Pedunculus rigidus, mediocris, 6.5-7 cm. longus, strigosus, pilis hamatis acu- tis satis armatus. Spatha ovoideo-oblonga, acutiuscula, circa 13 cm. longa, extus villoso-hirsuta, intus puberula, Amenta masculina circa 28 fasciculata, graciliter pedi- cellata, praefloratione 7-10 cm. longa, postea usque ad 17 em. longa, 3 mm. lata. Perigonium tubulosum, 2 [ 24 ] mm. longum, glabrum vel apice setis parcissimis minutis donatum. Pedicelli subcapillares, 12-20 mm. longi, dense scabro-hirsutuli, pilis hamatis instructi. Dr. Cuatrecasas writes: ‘*‘Cecropia mocoana is closely related to C. marginalis Cuatr. and C. discolor Cuatr. From both, it differs in having leaves which are a little thinner and more abundantly pubescent on the upper surface and in having more conspicuous scales (small conic hairs) which cause the leaves to be rough. The leaves on the under surface are more like those of C. marginalis because of the nerves and venation which are more or less hirtose and not covered with arachnoideous tomentum; in C. discolor the arachnoideous tomentum is brightly white and covers the network. The spathes of Cecropia mocoana are densely villous-hirsute as in C. discolor (in this species they are glabrous or only puberu- lent inside, contrary to the statement in the original de- scription), whereas in C. marginalis they are covered with a very short and obsolete tomentum. Only further abundant collections can indicate whether Cecropia mo- coana represents a male specimen of one of these earlier species, ”’ Cotoms1a: Comisaria del Putumayo, Mocoa, alrededores al norte, 750-850 m. ‘‘Small tree.’? December 38-7, 1942, Richard Evans Schultes & C. Earle Smith 2003 (Tyre in Chicago Museum). Cecropia porvenirensis Cuatrecasas sp. nov. Arbor magna. Ramuli seabri, pilis conicis et longiori- bus hamatis satis instructi. Folia membranacea vel sub- coriacea, peltata. Petiolus 32 em. longus, juvenilis albo- arachnoideus, deinde hirtus, pilis longis tenuibus rigidis patentibus inferne hamatis tantum ramanentibus. La- mina palmata profunde (2/38) 7-vel 6-lobata, lobis obo- vatis vel obovato-oblongis, basi constrictis, apice abrupte acuminatis, margine integra vel leviter irregulariterque [ 25 ] sinuata, lobo maximo ab insertione petioli usque ad api- cem 85 cm. longo, 15 em. lato, lobo minimo 14 cm. longo, 8 cm. lato, parte centrali concrescenti 7-13 cm. ; supra viridis, scabra, pilis conicis minutis satis atque tuberculis acutis et pilis sparsis longis praedita, nervis primariis secundariisque valde conspicuis; subtus alba, nervis principalibus in foliis juvenilibus tomento arach- noideo deciduo deinde sparse hirtis, nervis secundariis prominentibus, adscendentibus, 1-2 cm. distantibus, plusminusve arachnoideis et sparsissime hirtulis vel gla- brescentibus, reticulo venularum obsoleto et lamina re- liqua tomento arachnoideo albo dense tecta. Stipula in statu juvenili 8 cm. longa, acuta, villoso-hirta, margine utrinque glabra, intus villoso-sericea. Inflorescentiae femineae geminatae. Pedunculus 6-8 cm. longus (fructi- fer usque ad 17 em. longus) scabrido-hirtulus, tuberculis minutis atque pilis rigidis fragilibus patentibus rectis et hamatis munitus. Spatha cylindrica, apice longe angus- tata, acuta, 16 cm. longa, extus sparse puberula et arach- noidea, leviter scabrida, intus pubescenti-arachnoidea. Amenta feminea vulgo 5 vel 6 (4-7), leviter pedicellata, 11 cm. longa, 4 mm. lata, fructifera in sicco usque ad 23 cm. longa, 7-8 mm. crassa. Perigonium minute tu- berculatum. Achaena elliptica, 2 mm. longa, 1 mm. lata, laevia. Pedicelli crassi, 3-4 mm. longi, scabri, pilis conicis brevibus et pilis longioribus hamatis instructi. Concerning the relationship of this new species, Cua- trecasas writes: ‘‘Cecropia porvenirensis is closely related to C. marginalis Cuatr., but it differs by having the upper surface of the leaves very scabridous, the under surface more densely white-arachnoideous with the nerves much less hispid, and the lower part of the petiole with hooked hairs; the amenta are pedicellate, longer and usually five or six, while in C. marginals they appear in groups of four and are sessile. The stipule of Cecropia porveniren- [ 26 | sis is villose inside, whereas in C. marginalis it is gla- brous.”’ Cotomsia : Comisaria del Putumayo, Puerto Porvenir cerca de Puerto Ospina, alt. 250 m. “‘Large tree. Yarumo.’’ May 20, 1942, Richard Evans Schultes 3697 (Tver in Chicago Museum). LEGUMINOSAE Inga macrophylla Humboldt & Bonpland ex Willdenow Sp. Pl. 4 (1806) 622. Inga macrophylla, which ranges from Venezuela to Peru and Bolivia, was first reported from Colombia by Uribe (in Caldasia 8, no. 14 (1945) 347), on the basis of material from the Intendencia del Meta. The second and third Colombian collections, cited below, greatly extend the known range of this species in Colombia. In the Colombian Amazonia, this species, cultivated for the edible pulp around the beans, is called guava machete or guamo machete. In the Tikuna Indian language, it is known as pa-md. Cotompr1a: Comisaria del Amazonas, Rio Karaparand, between E] Encanto and the mouth, May 22-28, 1942, Richard Evans Schultes 3857.—Comisaria del Amazonas, Leticia. ‘‘Bushy shrub. Flowers white. Fruit long. Guamo machete. Tikuna: pa-md’’. September 20, 1945, Richard Evans Schultes 6542. Inga setifera de Candolle Prodr. 2 (1825) 432, 615. According to Uribe (in a letter to the writer), Inga setifera has not hitherto been reported from Colombia. It is known to be common in the Departamento de Loreto in Peru. In Leticia, the plant is called chimbillo by the Spanish-speaking population, kau-ré by the Tikuna Indians. CotomBia: Comisaria del Amazonas, Leticia. “‘Flowers yellow. Pulp of fruit edible. Chimbillo. Tikuna: hau-ré.’’ September 20, 1945, Richard Evans Schultes 6543. [ 27 | Swartzia bracteosa Martius ew Bentham in Mar- tius Fl. Bras. 15, pt. 2 (1870) 20. A species of the western half of the Amazonia, Swart- zia bracteosa is now known, through the collection cited below, well within Colombian territory. It is possible that the type was from Colombia, for it was collected by Martius ‘‘in sylva Yapurensibus. ”’ Cotompia: Intendencia del Amazonas, Rio Igaraparand, los alrede- dores de La Chorrera, alt. ca. 180 m. ‘Small tree. Fruit red.”’ June 4-10, 1942, Richard Evans Schultes 3939. VOCHYSIACEAE Vochysia lomatophylla Standley in Field Mus. Publ. Bot. 22 (1940) 150. Vochysia lomatophylla was described from material collected on the Rio Mazin in Loreto, Peru; it has also been known from the Rio Solimdées in Brazil (Stafleu in Rec. Trav. Bot. Néerl. 41, bk. 2 (1948) 505). The col- lections cited below are the first recorded from Colombia and greatly extend the known distribution of the species to the headwaters of the Rio Negro basin. Cotompta: Comisaria del Vaupés, Upper Vaupés River, alt. about 300 m. (?) January, 1944, Gabriel Gutiérrez & Richard Evans Schultes 504.—Comisaria del Vaupés, Upper Vaupés River, near Miraflores, alt. about 300 m. (?) ‘‘Enormous tree on floodland along banks.’’ January, 1944, Gabriel Gutiérrez & Richard Evans Schultes 701. Vochysia obscura Warming in Martius FI. Bras. 18, pt. 2 (1875) 73, t. 13. Widely distributed in the Amazon basin of Brazil and Peru and in southern Venezuela, this species of savannas and caatinga-forests has not hitherto been reported from Colombia. The collection cited below extends the known range very far to the northwest. Cotompia: Comisaria del Vaupés, Cerro Chiribiquete, January, 1944, Gabriel Gutiérrez & Richard Evans Schultes 688, [ 28 ] Vochysia vismiifolia Spruce ex Warming in Mar- tius Fl. Bras. 18, pt. 2 (1875) 99. This species of savannas or open caatinga-like forests is widely distributed throughout the Amazon Valley of Brazil and in southern Venezuela, but has hitherto ap- parently not been recorded from Colombia. Cotomsra: Comisaria del Vaupés, Cerro Chiribiquete, January, 1944, Gabriel Gutiérrez & Richard Evans Schultes 638, EUPHORBIACEAE Euphorbia penicillata (Millspaugh) R.E. Schultes comb. nov. Tithymalus pencillatus Millspaugh in Field Mus. Publ. Bot. 4 (1925) 95. Although the binomial Huphorbia penicillata has been used in several enumerations (Herrera: ‘‘Catdlogo alfa- bético de los nombres vulgares y cientificos de plantas que existen en el Perti’’ (1989) 192; Herrera: ‘‘Sinop- sis de la flora del Cuzco’’ 1 (1941) 291), it seems that the combination has never properly been made. The specific epithet was originally published as pencillatus, an obvious typographical error which was corrected in the index of the volume. Phyllanthus fluitans Bentham ex Mueller-Argo- viensis in Linnaea 82 (1868) 36. Described from Brazil, Phyllanthus fluitans has ap- parently not hitherto been reported from the flora of Colombia. CotomsiA: Intendencia del Amazonas, near mouth of Rio Loreto- yacu, April, 1944, Frederick J. Hermann 11298. Peru: Departamento de Loreto, Rio Amazonas, Isla Cacao, April, 1944, Frederick J. Hermann 113144. [ 29 | SAPINDACEAE Serjania dasyclados Radlkofer var. sibundoya R. EE. Schultes var. nov. Frutex scandens, a Serjania dasyclados principaliter foliolis subtus densius tomentoso-pilosis, supra dense et sordide adpresso-pilosis et tactu mollibus differt. This concept would seem to represent either a south- ern variant of Sexjania dasyclados or an endemic variety possibly confined to the mountain-encircled Valley of Sibundoy where a large number of endemic plants have been found. It is at once distinguished from Serjania dasyclados by having dense and appressed hairs of a dirty brown color on the upper surface of the leaflets. The more northern collections of the Colombian Serjania dasyclados (the type and a cotype from Cundinamarca at 700 and 1000 m. respectively; a collection from Cal- das at 1800-2100 m.; one from Santander at 2000-2600 m.; one from Norte de Santander at 2000-2500 m.) have no indumentum on the upper surface of the leaflets. CotomsiA: Comisaria del Putumayo, Valley of Sibundoy, Sibundoy, alt. about 2225-2300 m. ‘*Vine. Flowers white; fruits ripen red. Hairs on both leaf surfaces.’’? May 29, 1946, Richard Evans Schultes & Mardoqueo Villarreal 7652 (Tyre in Herb. Gray). BoMBACACEAE Quararibea putumayensis Cuatrecasas in Lloydia 11 (1948) 186. This collection is topotypical. The species is known only from the Comisaria del Putumayo and adjacent regions of Peru. The Kofan Indians employ the roots of this bush in preparing arrow-poison. CotomsiA: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. **Flowers white. Fruit edible. Bush. Kofan: hun-kun-ni-ai.’’ March 23-25, 1942, Richard Evans Schultes 3420, [ 30 ] Quararibea Schultesii Cuatrecasas sp. nov. Frutex. Ramuli badii, glabri, sublaeves, novelli pal- lide virides, pilis stellato-pubescentibus vestiti. Folia grandia, membranacea, simplicia, alterna, petiolata, pal- lide viridia. Petiolus robustus, 1.5—2 em. longus, teres, dense stellato-tomentosus. Lamina obovata, plusminusve oblonga, basim versus attenuata, basi cuneata, apice subito contracta et cuspidata margine integro; tri-quin- tuplicinervis, nervis secundariis quattuor, utroque latere duo tertio superiores dispositis aseendentibus prope mar- gine arcuato-anastomosatis reliquis nervis transversis et reticulum formantibus; supra prospectu glabra sed pilis sparsissimis simplicibus munita, nervis principalibus con- spiculs, reticulo obsoleto; subtus pilis sparsis simplicibus et stellatis praecipue ad nervos praedita, nervis primariis secundariisque valde eminentibus, reticulo bene emi- nenti; 18-30 cm. longa, 10-15 cm. lata. Stipulae lan- ceolatae, 9 mm. longae, 2.5 mm. latae, dense stellato- tomentosae, mox deciduae. Flores solitarii, oppositifolii, pedunculati. Pedunculi mediocres, flexuosi, circa 18 mm. longi, striolati, dense adpresse stellato-tomentosi, apice tres bracteas, late lanceolatas, dense stellato-tomentulo- sas, circa 1 cm. longas, 2—4.5 mm. latas, basi calycis ad- natas ferentes. Calyx tubuloso-conicus; tubus 13 mm. longus, basim versus minute striatus, apice breviter 5- lobulatus, dorso alis quinque membranaceis, apicem ver- sus dilatatis, extremis rotundatis, calycis tubum superan- tibus instructus, extus tubo alisque stellato-tomentulosis intus adpresse albido-sericeis. Petala 8 em. longa, supra glabra, subtus villoso-lanata cum pilis stellatis radiis ten- uibus longis intricatis, verosimiliter albida; ungui late lineari, glabro, 16 mm. longo; limbo apice exserto, spathulato, 14 mm. longo, 5-6 mm. lato, apice rotun- dato, reflexo vel patenti. Staminorum columna quam calyx plus duplo longior, glabra, cirea 3 cm. longa, ex- [ 31 ] trema in lacinias quinque antheriferas lineares, 4—5 mm. longas producta; laciniis sex antherarum loculis, oblon- gis, 1-1.2 mm. longis, valde contiguis munitis, et loculis alteris (1-2) parte superiore tubo concrescentibus. Stylus longe villosus, antherarum lobulis parum brevior vel ae- quilongus. Stigma capitatum, 1.8 mm. crassum. In connection with the foregoing description, Cuatre- casas states: ‘‘Quararibea Schultesiu belongs to the small group of species which have wings on the calyx. The most similar to it is Quararibea bracteolosa (Ducke) Cuatr. but Q. Schultesiu differs in having longer petioles and pedicels, by its smaller flowers, by the shorter mem- branaceous calyx of which the wings are more expanded and even more membranaceous, by the herbaceous- membranaceous lanceolate bracts, and by the indumen- tum of calyx and bracts which is shorter and thinner. Quararibea bracteolosa, on the contrary, has bracts which are ovate and coriaceous or subcoriaceous, a calyx tube which is much thicker and dense. I have seen an isotype of Quararibea bracteolosa which is preserved at the Chicago Museum of Natural History. Probably Ducke 1781 and 1782 (fruiting specimens), distributed as Quar- aribea bracteolosa, do not belong to that species. ”’ Cotomsi1A: Comisaria del Amazonas, Trapecio amazénico, Loreto- yacu River, alt. about 100 m. ““Bush.’’ October 20-30, 1945, Richard Evans Schultes 66694 (Tyrer in Chicago Museum). OcHNACEAE Rhytidanthera magnifica (Gleason) Dwyer in Lloydia 9 (1946) 53, t. 2, a-g. In his monographic treatment of Rhytidanthera and related genera, Dwyer (loc. cit.) reported but one collec- tion of R. magnifica, the type, Lawrance 599, from the El Umb6o region, Departamento de Boyaca. Cuatrecasas 13370 extends the known range of Rhyti- [ 32 | danthera magnifica into the Departamento Norte de San- tander and very near to the Venezuelan boundary. A|- though the type was described as a tree up to 65 feet tall, the species may also be, as reported by Cuatrecasas, a small tree (‘‘arbolito’’). Dwyer states that Rhytidan- thera magnifica ‘‘is readily distinguished from the two other species. .. . by its narrower leaflets which taper to- ward the base....’’, but the leaflets of Cuatrecasas 13370, while very narrow, are broadly rotund at the base. CoLtomsia: Departamento Norte de Santander, regién del Sarare, hoya del rio Marque entre Junin y Cordoba, 920-1240 m. alt. “‘Ar- bolito. Pétalos blancos. Estambres amarillos.’’? November 22, 1941, J. Cuatrecasas 13370, Rhytidanthera mellifera R. HL. Schultes sp. nov. Arbor parva, usque ad quindecim pedes alta, ramosis- sima. ‘Truncus tres ad quattuor pollices in diametro, rufo- brunneo cum cortice ruguloso et minute lenticellato. Folia imparipinnata, aliquid pendula; petioli 16.5-24.5 em. longi; foliola duodecim vel plerumque tredecim (cum petiolulis 1-2 mm. longis), lateralia alterna, 2.2-2.5 em. distantia, anguste lanceolata, apice acuta, basi ro- tundata, 8.5-14 (plerumque 10.5) cm. longa, 2.2-3.2 (plerumque 2.5) cm. lata; foliolum terminale liberum, cuneato-lanceolatum, apice obtusum, 10-12 cm. longum, usque ad 4 cm. latum; alia marginem versus valde re- trocurvata, margine grossiuscule serrata, dentibus apice incurvis et plerumque 6 mm. distantibus, supra nitida venis non elevatis, subtus pallidiora, venis prominenter el- evatis, secundariis plusminusve viginti; stipulis caducis. Inflorescentiae terminales, parvae, quam folia multo breviores, 20 cm. longae, rhachide 8 mm. in diametro, pauciflorae (floribus usque ad viginti); rhachidis rami usque ad 8 cm. longi; pedicelli ad rhachidem articulati, robustiores et lignosi, usque ad 10 mm. longi, apice sub- clavati, basi 1 mm. et apice 2 mm. in diametro. Flores [ 33 ] solitaril, aromatico-fragrantes, mucilaginosi, alabastris 16 mm. longis, 7-8 mm. latis. Sepala quinque, imbricata, concava; duo exteriora subcoriacea sed marginem versus membranacea, rotundata, margine irregulariter lacerata, apice profunde retusa, supra minute pulverulento-strio- lata (non strigillosa), 7 mm. longa, 7 mm. lata, basi cum plusminusve quindecim glandulis minutis digitaliformi- bus, usque ad 1 mm. longis, 0.25 mm. in diametro; in- teriora majora, membranacea vel apicem versus aliquid papyracea, oblonga, apice profundissime fissa, margine integra (vel sepalo extimo minute lacerato), 18 mm. longa, 7 mm. lata. Petala quinque, alba, membranacea, leviter inaequalia, oblongo-spathulata, apice saepe obliqua (ala- bastrum aliquid subcuculliforme), margine integra, 17 mm. longa, 7 mm. lata. Stamina sexaginta quattuor, brunneo-flava, subaequalia, usque ad 12 mm. longa. An- therae 10 mm. longae, 0.8 mm. latae, longitudinaliter rugulosae, filamentis 2 mm. longis, basi 0.6 mm. latis. Ovarium crassum, nigrum, inverse claviforme, apicem versus in stylum indeterminatum, apparenter erectum, luteum, paulatim angustans, 183 mm. longum, 2.5 mm. in diametro, quinque cum stigmatibus sessilibus. Capsu- lae immaturae solum visae, usque ad 2 cm. longae, 2.5— 3 cm. latae, praecipue apicem versus anguste subfalcatae. Rhytidanthera, closely allied to Godoya, is the only compound-leaved genus of the Ochnaceae. Hitherto, there have been but three species known: Rhytidanthera magnifica, R. splendida and R. suleata. The genus is en- demic to Colombia. Rhytidanthera mellifera represents the first collection of the genus from the Amazon Valley, the other species being Andean in their distribution. This is an excellent example of the curious type of plant distribution and relationship of the plants on the are which forms the so- called Venezuelan-Guianan land-mass. Of this distribu- [ B34 | tion, Dr. Lyman B. Smith (in Contrib. Gray Herb. no. 161 (1946) 81) has said: ‘‘....movement on the are was from the Andean region eastward. Finally most of the species [i.e. of certain bromeliads and cyperaceous plants he was discussing ] are narrow endemics, indicating that the movement was relatively ancient. ’”’ Rhytidanthera mellifera is an endemic of the cretaceous, quartzite hills of the upper Apaporis basin. We might well expect Rhytidanthera to occur in collections from other sandstone mountains in eastern Colombia and southern Venezuela, but even now we may venture to say that it is not acommon element of the flora of these repositories of endemics. On the Cerro Castillo, Rhyti- danthera mellifera is extremely rare. Special search for more abundant flowering material failed to furnish addi- tional individuals, so the species is at present known only from one bush. This bush, growing on the edge of a rock- ledge covered with a scrubby vegetation, was visible to us far below as we were climbing the face of the moun- tain. Like the other species of the genus, it makes an unusual display with the large, white caesalpinaceous-like flowers. Rhytidanthera was not seen on the nearby moun- tains, Chiribiquete and Campana. Rhytidanthera mellifera, so named because of the sweet and mucilaginous exudation of the flower which attracts swarms of wild bees to the plant, is immediately set apart from the other known species by having 64 stamens, a much larger number than has hitherto been found in the genus. The closest ally of Rhytidanthera mellifera would ap- pear to be BR. magnifica (Gleason) Dwyer, from which it differs in having twelve or thirteen alternate (instead of nine or ten opposite) leaflets ; a very broadly cuneate- lanceolate, obtuse (instead of a narrowly lanceolate- elliptic, acute) terminal leaflet; much smaller lateral [ 35 ] leaflets; minutely lacerate (instead of entire) sepals; smaller outer and larger inner sepals; and generally smaller petals which, in the bud, are subcucullate at the apex. There are also minor floral and vegetative differ- ences which more abundant material may show to be of importance. Cotomsi1a: Comisaria del Caquetaé, Apaporis River, Cerro Castillo (below confluence of Ajaju and Macaya), quartzite base, alt. of col- lections 350-1000 feet above the forest floor, 1250-1900 feet above sea-level. ““Flowers white, centres yellow-brown, aromatic fragrant. Treelet 15 feet. On exposed rock. Much branched. Basal diameter 3- 4 inches. Leaflets very coriaceous, marginally inrolled.’’ July 27, 1943, Richard Evans Schultes 5657 (Tyrer in U.S. Nat. Herb.). Sauvagesia erecta Linnaeus Sp. Pl. (1753) 203. Sauvagesia erecta, the most widespread species of the genus, is rather variable and presents several general geo- graphical variants. It has been collected in Colombia in the Departamentos de Norte de Santander, Huila, To- lima and Cauca (Dwyer in Bull. Torr. Bot. Club 72 (1945) 580). The collection cited below is apparently the first from the entire Amazon basin of Colombia. There are numerous collections from the adjacent Amazonian regions of Brazil and Peru. Cotomsia : Comisaria del Amazonas, Rio Igaraparanaé, near La Chor- rera, alt. about 180 meters. ‘‘Small shrub under 8 inches tall. Flow- ers pinkish.’’ June 4-10, 1942, Richard Evans Schultes 3955. FLACOURTIACEAE Lindakeria nitida Willip & R. I. Schultes sp. nov. Arbuscula usque ad viginti quinque pedes alta, ramo- sissima, ramis longis laxisque. Folia ovata, basi cuneato- obtusa, apice attenuato-acuminata, integerrima, 10.5—14 cm. longa, 4.5—5.5 em. lata, supra nitidissima, vivo ap- parenter atroviridia cum venis non conspicuis, subtus aliquid pallidiora cum venis prominentibus, utrinque glaberrima, nervis secundariis plusminusve quinque. Pet- [ 36 ] ioli 2.5-8.5 em. longi, teretiusculi, ad laminam incrassati. Racemi axillares, apparenter erecti et stricti; rhachides usque ad 10 cm. longae, folia subaequantes, compresso- angulatae. Pedicelli distichi, 6-7 mm. longi, solitarii vel saepe duo vel tres. Flores 10 mm. in diametro vel mi- nores, alabastro globoso. Sepala tria, triangularia, apice subacuta, margine valde involuta, 4.5 mm. longa, basi 4 mm. lata, extus apparenter minute aspero-pustulata et basi pilosa. Petala sex, alba, membranacea, lanceolata, apice acuta, 7 mm. longa, 1.8 mm. lata, extus rarenter dense albido-leprosa. Stamina plusminusve quadraginta, 4—5 mm. longa; filamenta libera, laxa, aliquid compla- nato-filiformia, 2 mm. longa; antherae lineares, 2.5-8 mm. longae. Ovarium ovoideum, echinato-tuberculatum, 2 mm. in diametro. Stylus ovario circiter duplo longior, 3.5-3.8 mm. longus, complanato-filiformis, glaber, stig- mate minute trilobo. Fructus adhue ignotus. Lindakeria nitida is at once distinguished from all other species of the genus by its very long and lax inflo- rescence which gives the plant a completely unique as- pect. This species, growing in the Magdalena watershed of Colombia, is perhaps most closely allied to the rather widespread Amazonian Lindakeria maynensis. In addi- tion to the length of its inflorescence and its highly lus- trous leaves, Lindakeria nitida differs from L.maynensis in the shape and size of the sepals and slightly in the structure of the stamens. The collection Haught 1917 has, in general, much larger leaves and flowers than the type. The leaves of this cotype are lanceolate-ovate, up to 26 cm. long and 8 cm. wide; the petioles measure up to 7 em. in length, and the rachis is 16 ecm. long. Additional collections will be needed before we can determine whether the type represents the normal condition of the species, but it would appear from an examination of the available ma- [ 37 | terial that Haught 1917 was taken from an individual in an open, favorable site or else from a new flush. Cotompia: Departamento E] Magdalena, 17 km. north of El Banco on road to Chimichagua. “‘Large (5 m. high) shrub with many long, lax branches. Inflorescence very conspicuous. Corolla white, anthers yellow. Inswampy forest, along road.’*? August 3, 1938, Oscar Haught 2221 (Tyre in U.S. Nat. Herb. No. 1706900).—Departamento de Santander, El Centro, vicinity of Barranca-Bermeja, Magdalena Val- ley, between Sogamoso and Carare Rivers, alt. 100-500 meters. ‘‘Large shrub (4 m.) very conspicuous white flowers and buds.’’? August 26, 1936, Oscar Haught 1917. Mayna glomerata Killip & R. EH. Schultes sp. nov. Arbor parva, usque ad duodecim pedes alta. Rami ramulique verrucoso-striati, griselli; ramuli juveniles cinereo-virides, squamiferi. Folia alterna, simplicia, lan- ceolato-elliptica, usque ad 22 cm. longa (sine petiolo), 7-7.5 cm. lata, integra, apice longe acuminata, basi cu- neata, chartacea, nervis secundariis valde arcuatis duo- decim vel tredecim, omnino glabra, supra atroviridia et apparenter nitida cum nervis leviter prominentibus sed non elevatis, infra pallide viridia, non nitida, nervis valde elevatis et fulvis. Petioli robusti, usque ad 8 cm. longi, 1.5-1.8 mm. in diametro, glabri. Stipulae non visae. Dioecia. Flores in fasciculis caulinis, glomeratis, inflores- centiis usque ad plusminusve decemfloris. Alabastra glo- bosa vel elongato-globosa, plusminusve 2 mm. in diam- etro, minute et subdense pilosa. Bracteae minutae. Pe- dunculus usque ad 8 mm. longus, dense albo-pilosus. Flores staminiferi albi, conspicui, usque ad 2 em. in di- ametro. Sepala tria, leviter inaequalia, valde concava, ovata, integra, apice subacuta, extus pubescentia, intus glabra, quinque cum venis brunneis longitudinaliter stri- ata, sepalum majus 3.8-4 mm. longum, 2.7-2.9 mm. latum. Petala quinque ad septem, aequalia, subspathu- lata, margine irregularia, apice late rotundata et valde in- voluta, alba, membranacea, usque ad 6.5 mm. longa, apice [ 38 | circiter 2 mm. lata, basi 0.4 mm. lata, omnino glabra. Stamina circiter vigintiquinque, 2.6 mm. longa, filamen- tis liberis, erectis, plerumque 0.8 mm. longis, et antheris leviter arcuatis, 1.8 mm. longis, 0.5 mm. latis, flavis, apice poricidis. Fructus adhuc ignotus. Mayna glomerata is easily distinguished from the other species of this genus by its unusual inflorescences which are in small, closely packed, glomerate fascicles. There are other notable differences, as well, such as the curious folding of the upper third of the petals, the shape of the petals, and the number of the stamens. Cotompra: Departamento de Santander, vicinity of Barranca Ber- meja, Magdalena Valley, between Sogamoso and Carare Rivers, Vis- cania Creek, 32 km. south of El Centro, at about 100 meters alt. **Small tree, about 4 m. high. Flowers white with yellow centres, showy, closely covering twigs.’’ November 14, 1936, Oscar Haught 2070 (Typr in U.S. Nat. Herb. No. 1742387). Mayna longifolia Poeppig & Endlicher Nov. Gen. ac Sp. 3 (1845) 64, t. 271. Mayna longifolia is extremely abundant in the forests of the southern half of the trapecio in areas above flood- level. The collection Llewelyn Williams 2711 from La Victoria (now ‘‘Granja Caldas’’) near Leticia, cited as from Peru in Macbride’s ‘‘ Flora of Peru’’ (Field Mus. Publ. Bot. 18, pt. 4, (1941) 18), but actually from Co- lombian territory, is referable to Mayna longifolia. CoLoms1a: Comisaria del Amazonas, Trapecio amazénico, Loreto- yacu River, alt. about 100 m. ““Treelet 25 feet.’’ October 1945, Richard Evans Schultes 6700.—Comisaria del Amazonas, same locality and date. ‘“Tall treelet. Leaf larger than normal.’’ Richard Evans Schultes 6701.—Comisaria del Amazonas, same locality and date. “Leaves very narrow. Fruit on trunk up to top.’’ Richard Evans Schultes 6705. COMBRETACEAE Combretum laxum Jacquin Enum. Pl. Carib. 19 (1760) 115. [ 39 | Combretum laxum appears to be an aggregate species in which several clearly distinct ‘‘tendencies’’ can be seen. As now understood in this broad sense, the species occurs all over tropical America. Several collections from the Departamento de Loreto, Peru, represent the same variant as Schultes 3924 from adjacent Colombia. When further material is studied, this variant can probably be shown to be specifically distinct from true Combretum laxum. In the Igaraparana, the Witoto Indian name of Com- bretum lawum is qjuo-wo. In another Witoto dialect from Loreto, Peru, the native name has been recorded as aioho-ayo-o (in Field Mus. Publ. Bot. 18, pt. 4 (1941) 225). This attractive vine often covers riverside vegetation and, especially at night, gives off an extraordinarily strong and sweet perfume. Cotompra: Comisaria del Amazonas, Rio Igaraparana, alrededores de La Chorrera. “‘Vine. Flowers yellow, fragrant.’’ June 4-10, 1942, Richard Evans Schultes 3924. MELASTOMACEAE Aciotis aequatorialis Cogniaux in Martius FI. Bras. 14, pt. 8 (1885) 464. Known from the Pacific coast of Colombia, from An- tioquia and from the Intendencia del Meta, Aciotis ae- quatorialis has hitherto apparently not been reported as an element of the flora of Amazonian Colombia. A ciotes polystachys (Bonpl.) Tr. and A. purpurascens (Aubl.) Tr. have been collected in Colombia only a few kilo- meters from the locality of the specimen of 4. aequa- torialis cited below. CoLomspia: Intendencia del Amazonas, Rio Hamacayacu, April, 1944, Frederick J. Hermann 11303. Leandra divaricata (Naud.) Cogniaux in Martius Fl. Bras. 14, pt. 4 (1886) 196. [ 40 | Not infrequent in British Guiana, Venezuela, Ama- zonian Peru and elsewhere in Colombia, Leandra divari- cata has been collected hitherto on the Amazonian water- shed of Colombia but twice: at Puerto Porvenir, on the upper reaches of the Rio Putumayo (Cuatrecasas 10666), and at Mocoa (Cuatrecasas 11316). Hermann 11308, from the southeasternmost point of the Colombian Am- azonia, would seem to indicate a wide distribution for this species in eastern Colombia. CoLompia: Intendencia del Amazonas, Rio Hamacayacu, April, 1944, Frederick J. Hermann 11308. LOGANIACEAE Strychnos amazonica Krukoff in Brittonia 4 (1942) 284. This collection, determined with some reservation as representing Strychnos amazonica, is the first of the spe- cies from Colombia. Krukoff and Monachino (in Caldasia 4 (1946) 45) enumerated Strychnos amazonica as a species to be expected in the Colombian Amazonia. It is one of the plants the root of which is used by the Kofin Indians in preparing arrow-poison. CoLtomsia: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. 6 . . r r ‘ . , ‘Vine. Root used in Kofan Indian arrow-poison. Kofan name: hess- pd-chu.”’’ April 25, 1942, Richard Evans Schultes 3689. Strychnos Erichsonii Richard Schomburgk Faun. Fl. Br. Guiana (1848) 1082, nomen: ex Progel in Mar- tius Fl. Bras. 6, pt. 1 (1868) 274. First reported from Colombia on the basis of a collec- tion from the Rio Sucumbios (in Bot. Mus. Leafl. Har- vard Univ. 13 (1949) 290), Strychnos Erichsonii is appar- ently one of the most abundant species of this genus in the upper Putumayo basin where it is sought by the Kofan Indians for use in preparing cuvrare. [ 41 ] Cotomsia: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. “*Vine. Bark of stem rasped and used for arrow-poison. Kofin name: ee-ru-che.’’ April 25, 1942, Richard Evans Schultes 3682.—Same lo- cality and date. ‘*Vine. Root used in arrow-poison, Kofaén name: see-a-se-sep.’’ Richard Evans Schultes 3682.—Same locality and date. ““Vine. Pieces of stem used in preparing Kofan arrow-poison. Kofan name: mun-tai-ri-chee.’’ Richard Evans Schultes 3685. Strychnos guianensis (4ub/.) Martius Syst. Mat. Med. Bras. (1848) 121. Known from Colombia previously from one collection made in the Rio Vaupés, Strychnos guianensis is now registered from the southwesternmost point of the Am- azon basin of Colombia where it serves the Kofin Indians as one of the lesser ingredients in arrow-poisons. Cotomsi1A: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. ee e . . 4 , 99 Vine. Root used in preparing curare. Kofin name: ku-see-ye-hé-pa. April 25, 1942, Richard Evans Schultes 3688. Strychnos javariensis Avrukoff in Brittonia 4 (1942) 279. The following collections are apparently the first to be recorded from Colombia. The species is one of the lesser ingredients of Kofan curare. Cotompra: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. ‘*Vine. Root used in arrow-poison. Kofan name: tzu-se-hé-pa.’’ April 25, 1942, Richard Evans Schultes 8690.—Same locality and date. **Vine. Root used in Kofan arrow-poison.’’ Richard Evans Schultes 3691. Strychnos Jobertiana Bail/on in Adansonia 12 (1879) 367. Although previously reported from two localities in the Colombian Amazonia, Strychnos Jobertiana is now known to be very abundant in the upper Putumayo basin and to be one of the preferred ingredients of curare amongst the Kofan Indians. (42) Cotomspia: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. **Vine. Stem employed as ingredient in arrow-poison. Kofaén name: u-su-se-é-pa.’” April 25, 1942, Richard Evans Schultes 3692.—Same locality and date. ““Vine. Root used to make arrow-poison. Kofan name: fee-see-pd-chu.’’ Richard Evans Schultes 3695. Strychnos Peckii B. L. Robinson in Proc. Am. Acad. 49 (1918) 504. This is the second collection of Strychnos Peckii from Colombian Amazonia. It is one of the preferred species amongst the Kofan Indians for preparing arrow-poison. Cotoms1a: Comisaria de] Putumayo, Rio Putumayo, Puerto Ospina. “Vine. Root used in arrow-poison. Kofan name: su-he-sé-pa.’’ April 25, 1942, Richard Evans Schultes 3684. Strychnos subcordata Spruce ex Bentham in Journ. Linn. Soc. 1 (1856) 106. This species has never been reported from Colombia. It is considered one of the strongest of the curare ingre- dients by the Kofan Indians. Cotoms1A: Comisaria del Putumayo, Rio Putumayo, Puerto Ospina. “*Vine. Root used in arrow-poison formula. Kofaén name: su-sé-pe.”’ April 25, 1942, Richard Evans Schultes 3687. APOCYNACEAE Himatanthus articulata (Vah/) Woodson in Ann. Missouri Bot. Gard. 25 (19837) 196. The collection cited below is apparently the second from Colombia, the first being Cuatrecasas 9001 from Florencia, capital of the Comisaria del Caquetaé. The species is rather widespread in the Guianas, Amazonian Brazil, the Orinoco drainage-area of Venezuela and in Middle America. . , , oe oe | 6é CotompBia: Comisaria del Vaupés, Cerro Chiribiquete. Flowers white.’’? January, 1944, Richard Evans Schultes 5735, [43] Himatanthus bracteata (4.DC.) Woodson in Ann. Missouri Bot. Gard. 25 (1937) 200. Known from British Guiana and several stations in the Amazon of Brazil (where it is apparently most abun- dant in the Rio Negro basin) as well as from Pernam- buco and Bahia, Himatanthus bracteata has hitherto not been reported from Colombia. Schultes 5814, tentatively determined as representing this species, has, according to Woodson, leaves which ‘‘are larger than usual.’’ In the Vaupés, the plant is known as platanote, due to the superficial resemblance of the fruit to the banana. Cotomsia: Comisaria del Vaupés, Cerro Chiribiquete. ‘‘Flowers white.’’ January, 1944, Richard Evans Schultes 5738.—Comisaria del Vaupés, Rio Vaupés, Cafio Pact (below Cerro Circasia). ‘‘Small tree up to 15 feet tall. Latex abundant, white. Platanote.’’ March 6, 1944, Richard Evans Schultes 5814. Mandevilla nerioides Woodson in Ann. Missouri Bot. Gard. 28 (1941) 274. Mandevilla nerioides is known only from the Inten- dencia del Meta and the Comisaria del Vaupés in Colom- bia. The type collection was made on the Guaviare River at San José del Guaviare by Cuatrecasas. The following collections establish the occurrence of this endemic spe- cies on the summit of the remnant quartzite hills of the upper Apaporis River which have, in common with the hills at San José del Guaviare, a number of. striking endemic elements. CotomB1a: Comisaria del Vaupés, Cerro Chiribiquete. ““Vine. Flow- ers pink. Latex white, sticky.’’ May 15-16, 1943, Richard Evans Schultes 5442.—Same locality. “*Vine. Flowers pinkish white. Latex white.’’ January, 1944, Richard Evans Schultes 5740. SOLANACEAE Solanum quitoense Lamarck Llustr. 2 (1798) 16. Although the /u/o, the fruit of this species, is known [ 44 ] in many parts of southern Colombia and has been the subject of several recent articles (Chalons in Agric. Amer. 4 (1944) 110-112; McCann, ibid. 7 (1947) 146- 149; Hodge in Rev. Fac. Nac. Agron. 7 (1947) 147- 154; Hodge in Journ. N.Y. Bot. Gard. 48 (1947) 155- 159), there are apparently no Colombian collections of Solanum quitoense in the larger herbaria of Colombia and the United States. The lulo is used as a fruit and is cultivated in much of Andean Colombia (Hodge, loc. cit.), especially in the southern part of the country—in El Valle, Cauca and Narifo (Pérez-Arbelaez, ‘‘Plantas medicinales y venen- osas de Colombia’’ (1987) 246; ‘‘Plantas ttiles de Colom- bia’? (1947) 451), where it is known also as naranjillo and toronja. Lacking specimens or a critical taxonomic study of the lulo from the numerous areas of Colombia, we cannot be certain that only one species is involved. Solanum quitoense, as represented in our herbaria, is rather common in the Andean highlands of Ecuador and Peru. The collections Schultes & Villarreal 7616 and Schultes 3271 from the southernmost Andes of Colombia not far from the boundary of Ecuador are taxonomi- cally indistinguishable from the material from Ecuador and Peru. It is obvious that an extensive taxonomic study of the lulo in Colombia is needed. Certainly the possibilities of this delicious fruit would indicate the ad- visability of such an investigation. Co.tomsra: Comisaria del Putumayo, Valley of Sibundoy, Sibundoy, alt. about 2225-2300 m. “‘Large shrub. Stem and under surface of leaf purplish, hairy. Flowers white, anthers yellow. Fruit orange colored and size of oranges. Naranjillo. Kamsa Indian name: ma-sha- kve.’’ May 29, 1946, Richard Evans Schultes & Mardoqueo Villarreal 7616.—Same locality, February 16, 1942, Richard Evans Schultes 3271.—Departamento de Antioquia, Rio Negro, ‘“La Granja.’’ Cul- tivated, November 2, 1946, W. H. Hodge 67172. [ 45 ] RUBIACEAE Rondeletia rupicola R. Schumann var. chiribi- quetana Rk. EH. Schultes var. nov. Frutex a Rondeletia rupicola stipulis foliaceis multo minoribus; foliis elliptico-lanceolatis (non longe lanceo- latis), minoribus, 4-5 cm. longis, 1-1.3 em. latis (non 5-8 cm. longis, 1.5 cm. latis), acutis (non acutissimis, mucronatis), supra prominentius bullatis, prope nervos majores densius sericeo-strigillosis, subtus prope nervos atque in lamina ipsa maxime densissime sericeo-strigillo- sis; corollae laciniis majoribus, extus densius strigillosis ; antheris majoribus; fornicibus papillosis majoribus den- sissime hirsutis differt. Capsula globosa, coriacea, 2mm. in diametro, in fructu calycis laciniis usque ad 1.5 mm. longis densissime strigillosis coronata, fusco-brunnea; semina numerosissima, minuta, angulata, A study of the original description of Rondeletia rupi- cola and an examination of a duplicate type (Spruce 3392) preserved at the Gray Herbarium and a photograph of the Spruce specimen at Berlin indicate that there are a number of minor variations which serve to separate the new variety described above. The capsule of Rondeletia rupicola is unknown, but the type of var. chiribiquetana has both flowers and fruit. The color of the flowers of Rondeletia rupicola is not designated; that of var. chiri- biquetana is pink. Cotomsia: Comisaria del Vaupés, Macaya-Ajaju River confluence, Mount Chiribiquete, quartzite base, summit 800-1200 feet above forest floor, 1700-2100 feet above sea level. ‘‘Flowers rose-pink. Leaves dark green above, pale below. Hair silvery. Shrub.’’ May 15-16, 1943, Richard Evans Schultes 5448 (Tyrr in Herb. Gray). COMPOSITAE Clibadium asperum (4 1b/.) de Candolle Prodr. 5 (1836) 506. [ 46 ] This species of Clibadiwm is abundantly cultivated in the vicinity of La Pedrera and in the Rio Miritiparana for use as a fish-poison. It is known locally as guaco by the Spanish-speaking inhabitants and as /v-na-pe by the Mirafia Indians. Several other species of Clibadium are widely cultivated in the Amazon regions as fish-poisons. CoLtompia: Comisaria del Caqueté, Rio Caquetdé, vicinity of La Pedrera, April, 1944, Richard Evans Schultes 5879. Gongylolepis maroana Badillo in Bot. Soc. Venez. Ciéne. Nat. 8 (19438) 287. Recently, two stations for this species were reported for Colombia, both in the Comisaria del Vaupés (in Bot. Mus. Leafl. Harvard Univ. 13 (1949) 810). The collec- tion cited below is the third from Colombia and extends the known range of the plant to the west on the same interrupted sandstone hills on which previously reported collections were made. Cotomsp1a: Comisaria del Vaupés, Ajaju River, Cerro de La Cam- pana, quartzite base, summit about 800-1200 feet above forest floor, 1700-2100 feet above sea-level. ‘‘Composite shrub. Common on top of mountain,’’ June 1-6, 1943, Richard Evans Schultes 5568. [ 47 ] ILLUSTRATIONS EXPLANATION OF THE ILLUSTRATION Pirate VII. (Upper figure). Habit photograph of CrcROPIA PORVENIRENSIS Cuatr., Puerto Porvenir, Rio Putumayo, Colombia. (Lower figure). Flowering branch of Vocuysta Lo- MATOPHYLLA Stand/, the tree from which Gutiérrez & Schultes 701 was collected, Rio Vaupés, Colombia. Photographs by R. E. Scuutters Pruare VII EXPLANATION OF THE ILLUSTRATION Puare VILE. Ruyrrpanrurra Meviirera PR. FE, NSchultes. 1, leaf one half natural size. 2, inflores- cence, one half natural size. 3, flower, natural size. 4, stamen, about four times natural size. 5, anther tip showing apical pores, about twenty times nat- ural size. Drawn by KE. W. Suivi PLaTE VIII mellifera RFE. Schuttes EXPLANATION OF THE ILLUSTRATION Pirate IX. Linpakerta nitipa Killip & Schultes. 1, flowering branchlet (Haught 2227), one half natu- ral size. 2, budding branchlet (Haught 1719), one half natural size. 3, flower, about two and one half times natural size. 4, stamen, about five times nat- ural size. Drawn by EK. W. Suiru PuaTE IX LINDAKERIA nitida Fal lip & Schultes Q of, ) Qu. Qy TY 7 A EXPLANATION OF THE ILLUSTRATION Piare X. (Upper figure). Soranum qurromnse Lam. Photograph of the plant from which Schultes & Vil- larreal 7616 was collected. (Lower figure). An escarpment of Cerro del Cas- tillo, Rio Apaporis, Colombia. RuyTIpANTHERA MELLIFERA Occurs on these open, exposed quarzitic escarpments. Photographs by R. F.. Scuu.tres PLATE X EXPLANATION OF THE [TLLUSTRATION Prare XI. Mayna aromerata Killip & Schultes. 1, Howering branchlet, one half natural size. 2, flow- er, about three times natural size. Drawn by Kb. W. Suivi PLaTE XI y MAYNA glomerata v Lillip & Schultes ; 7’, ee ee ee ae, eee A + ee eee eS ee ee eee pee eee BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 14, No. 3 CamsripGe, Massacuusetts, DecemMBer 7, 1949 NOMENCLATURAL NOTES AND NEW CONCEPTS OF TROPICAL AMERICAN ORCHIDS BY CHARLES SCHWEINFURTH Durinec the course of recent studies, I have found it necessary to make numerous changes in the status of published concepts, to amplify certain descriptions and to describe several orchids which appear to be new. This addition to the number of recorded species and varieties, together with many similar ones, shows that tropical America from the northern boundary of Mexico to the southern borders of Peru and Brazil is a veritable store- house of plants new to science. Only the collector and the taxonomist are necessary to bring these to light. The arrangement of the species follows the order pro- posed by Dr. Rudolf Schlechter in ‘‘Notizblatt des Bo- tanischen Gartens u. Museums Berlin-Dahlem 9 (1926) 563-591. Cranichis ciliilabia C. Schweinfurth sp. nov. Herba terrestris, gracilis. Folia plura, plerumque ba- salia et petiolata; lamina ovata, acuta vel breviter acu- minata, parva. Caulis basi glaber, supra dense glanduloso- pubescens, vaginis pluribus remotis ornatus. Inflorescen- tia laxe pluriflora cum floribus parvis membranaceis. Sepala elliptica vel ovalia, acuta. Petala oblique elliptica, pro genere lata, dense ciliata. Labellum simplex, cuneato- [ 49 ] obovatum, antice late rotundatum, concavum, dense longe ciliatum. Columna perbrevis. Plant terrestrial, slender, about 21.5 cm. tall. Roots tuberous, fasciculate, slender, lanuginose. Leaves several, the four basal blades rosulate, slender-petioled; lamina ovate, acute or short-acuminate, broadly cuneate to sub- truncate at the base, commonly 3.2-3.5 cm. long, up to 2.2 cm. wide; petiole channelled, slightly dilated above and below, about 2 cm. or less long. A fourth similar but sessile leaf rises near the base of the stem. Stem slender, glabrous in the lower portion, densely glandular- pubescent above, bearing four distant sheaths of which the lower ones have a close long-sheathing tubular base and the lowermost is dilated into a small foliaceous blade. Raceme loosely 17-flowered, about 4.5 cm. long, with the apical portion (bearing immature flowers) nodding. Flowers small, but medium-sized for the genus, mem- branaceous. Sepals very sparingly glandular-pubescent on the outer surface. Dorsal sepal elliptic, acute, con- cave, 3-nerved with the lateral nerves branching below, about 5.9 mm. long and 8.2 mm. wide when expanded. Lateral sepals oval, acute, concave, 5-nerved, about 5.6 mm. long and 8.2 mm. wide. Petals obliquely elliptic, acute or subacute, 4-nerved, densely ciliate with the cilia longer on the anterior margin, about 5 mm. long and 2.7 mm. wide. Lip simple, cuneate-obovate, broadly rounded in front, minutely acute at the apex, densely long-ciliate, about 4 mm. long and 3 mm. wide near the front in natural position, rather deeply concave especially above, 3-nerved, the lateral shorter nerves recurved in the middle of the lamina and having two or three short recurved branches. Column very short, with the stalk of the anther arising near the base. The long-ciliate lip seems to be unique in the genus and the ciliate petals are remarkably broad. [ 50 | Mexico: Chiapas, at 4000-6000 feet altitude, “‘moist bank of trail above Liquidambar,’’ flowers white, November 9, 1945, A. J. Sharp 45977 (Tyrer in Herb. Ames No. 65446). Pleurothallis Broadwayi 4 mes Orch. 2 (1908) 267. Pleurothallis guadalupensis Cogniaux in Urban Symb. Antill. 6 (1909) 432. Pleurothallis Williamsti Ames Orch. 7 (1922) 120. Pleurothallisnana A. & 8. in Sched. Orch. 8 (1925) 29. The above synonymy, except for Pleurothallis nana, has already been published (C.Schweinfurth in Bot. Mus. Leafl. Harvard Univ. 8 (1940) 41). At the time when Pleurothallis nana was erected, it was thought to be distinct from the concepts referable to P. Broadwayt. It seemed to have longer and more branched stems than the other species of this group. It also appeared to be separable by reason of its longer pedi- cels, ‘‘snow white’’ rather than yellow or greenish yellow flowers and especially because of its narrower and more acuminate sepals. A recently acquired Mexican collection (Nagel & Juan G. 6447), although it appears to be inseparable vegetatively from the Costa Rican P. nana, has long pedicels and sepals intermediate in form between those of typical P. Broadwayi and P. nana. It is further noted as having pale yellow-green flowers, as in P. Broadway. A Costa Rican collection (Austin Smith H541) also has pedicels of a length intermediate between the two con- cepts and flowers noted as pale lemon yellow as in P. Broadway; but it has sepals of quite the form of P. nana. Therefore, in view of the great variability naturally to be expected of a wide ranging plant, it seems advisable to regard P. Broadwayi as representing one polymorphic species. It occurs in Mexico (Oaxaca), Honduras, Costa Rica (type of P. nana), Panama (type of P. Williamsii), [ 51 ] Cuba, Martinique, Guadeloupe (type of P. gwadalupen- sis) and Venezuela (Island of Margarita). Pleurothallis leucantha Schlechter in Fedde Re- pert. 10 (1912) 353. Pleurothallis Sanchot Ames in Sched. Orch. 4 (May 1923) 26. Pleurothallis gonioglossa Schlechter in Fedde Repert. Beih. 19 (November 1928) 189. Judging from the type description of the Guatemalan Pleurothallis leucantha (amplified by an analytical draw- ing made under the supervision of Dr. Schlechter), that species differs from the Costa Rican P. Sanchoi (repre- sented by the type and several subsequent collections), by having generally longer and more acuminate leaves, smaller flowers, somewhat dissimilar straighter petals and an obtuse (not apiculate) lip. In the Ames Herbarium there is a Guatemalan collec- tion (Harry Johnson 609) taken to represent P. leucan- tha which came from the same region of Coban as the type and at about the same altitude. This collection, which may be called atopotype of P./eucantha, has some- what larger flowers than specified as typical, slightly curved petals and an apiculate lip which very nearly coincides with that of P. Sanchot. Moreover, one Costa Rican collection (Standley & Valerio 45983) is entirely typical of P. Sanchoi, except that the petals are nearly straight, as in typical P. leucantha. The concept described as Pleurothallis gonioglossa, which is represented by an isotype and analytical draw- ings in the Ames Herbarium, appears to be an exact vegetative counterpart of P. Sanchoi, although it has slightly larger flowers than the latter species. Except for inconsequential floral differences, this species appears to be inseparable from P. leucantha. [ 52] It seems logical, therefore, to regard Pleurothallis leu- cantha as a rather widespread and variable species in Central America. Octomeria pygmaea C. Schweinfurth sp. nov. Herba minutissima, epiphytica, corticicola. Caules perbreves, uniarticulati, vaginis duabus membranaceis tubulatis evanidis omnino obtecti. Folium anguste ellip- ticum, acutum, sessile, rigide nervosum, crasse margina- tum. Inflorescentiae saepissime duae vel tres,abbreviatae, uniflorae. Flos membranaceus, perparvus sed pro planta magnus. Sepala persimilia, triangulari-lanceolata, acuta vel acuminata, trinervia, omnino libera. Petala lanceo- lata vel elliptico-lanceolata, acuta vel acuminata, uni- nervia. Labellum in circuitu rhombico-ovatum, medio profunde trilobatum ; lobi laterales faleato-dolabriformes, porrecti; lobus medius obovato-oblongus, late obtusus, lobos laterales valde excedens. Columna in pedem sub- aequalem extensa. Plant minute, caespitose, epiphytic in the cracks of the bark. Roots fibrous, relatively stout, glabrous. Stems numerous, clustered, abbreviated, about 4.8 mm. or less long, 1-jointed near the base, entirely concealed by two tubular scarious evanescent sheaths. Leaf solitary, ses- sile, narrowly elliptic, up to about 8 mm. long and 2.1 mm. wide, acute, conduplicate at the base, coarsely mar- ginate, rigid-nervose, with the mid-nerve enlarged and prominent beneath. Inflorescences abbreviated, 1-flow- ered, apparently two or three to a stem. Flower very small, but large in relation to the plant, membranaceous. Sepals free, 3-nerved, concave. Dorsal sepal triangular- lanceolate, acute or acuminate, about 1.75 mm. long and 0.75 mm. wide. Lateral sepals very similar, but a little longer and wider, oblique. Petals lanceolate to elliptic- lanceolate, acute or acuminate, 1-nerved, about 1.5 mm. [ 58 ] long and 0.5 mm. wide. Lip rhombic-ovate in outline, deeply 8-lobed near the middle, cuneate at the base, about 1.1 mm. long and 0.87 mm. wide across the lateral lobes, with a pair of fleshy keels through the lower por- tion; lateral lobes faleate-dolabriform, subequal to the mid-lobe, obtuse, porrect, erect in natural position; mid- lobe obovate-oblong, broadly obtuse, much exceeding the lateral lobes. Column about half as long as the petals, produced into a subequally long foot. Pollinia eight, pyriform. This species is probably the most minute member of the genus Octomeria yet described and ranks with the most insignificant orchids known to science. It appears to be allied to O. minuta Cogn., but differs in having acute sepals (when expanded), the lateral ones being free, as well as in having a widely dissimilar lip. I take the liberty of quoting from the notes of the collector of this remarkable little orchid, as follows: ‘*A fter making camp one afternoon I put on the climb- ing irons in order to collect a specimen of a beautifully flowering Cunuria crassipes, a relative of Hevea rubber. When half way up the ninety foot tree, I stopped to lean back in the belt and rest. My glance was caught by what seemed to be a tiny orchid growing, together with the ever-present mosses and lichens, in the deep furrows of the shaggy dark brown bark of the tree. Carefully I plucked it out. Examination with a lens proved that it was indeed a very beautiful little orchid less than one half an inch high with yellowish flowers, of a genus un- known to me. Calling to Francisco to stop his cooking, I put the little plant in my note-book and let it fall to him. All other thoughts dismissed, we spent the rest of the afternoon searching the trunks of all the trees in the vicinity, but only a few additional specimens of the or- chid were found. [ 54 | ‘It is a coincidence that here occurred examples of the two extremes of the jungle plants which interest me most—the most diminutive of epiphytic orchids and the most gigantic of trees growing together, my interest in the latter leading to the discovery of the former.’ Brazi.: Upper Rio Negro drainage-area, Rio Dimiti, at the base of Mt. Dimiti, epiphytic in cracks of the bark of Cunuria crassipes, flowers yellow, May 12-19, 1948, Richard Evans Schultes & Francisco Lopez 10003 (Tyrer in Herb. Ames No. 65206), Epidendrum flexuosissimum C. Schweinfurth sp. nov. Herba parva, epiphytica, crassior. Rhizoma repens. Caules breves, robusti, approximati, vaginis tubulatis vel vaginarum fibris omnino tecti, apice bifoliati. Folia subopposita, late patentia, ovalia vel orbicularia, sessilia. Inflorescentia supra paniculata, laxe multiflora, cum pe- dunculo rachideque valde fractiflexa. Flores parvi, cum segmentis late patentibus. Sepalum dorsale elliptico- lanceolatum vel oblongo-lanceolatum, acuminatum. Se- pala lateralia paulo majora, obliquissime elliptico-lan- ceolata, complicato-acuminata. Petala linearia, leviter sigmoidea, dense ciliata. Labellum columnae valde ad- natum; lamina carnosa, medio trilobata; lobi laterales erecto-incurvi, aliformes; lobus medius_ triangularis, porrectus, Plant small, rather stout, about 8-17 em. high. Rhi- zome creeping (usually fragmentary in our specimens). Roots numerous, fibrous, glabrous, slender, commonly unbranched. Stems short, stout, suberect to lightly flex- uous, approximate and spreading from the rhizome, 1- or 2-jointed, entirely concealed by two or three tubular imbricating sheaths or the fibres of sheaths, about 1-3 em. long, 2-leaved at the apex. Leaves subopposite, widely spreading, oval to orbicular, sessile, apparently always rounded at the apex and sometimes with a small [ 55 mucro, coriaceous, about 1.5-8.6 cm. long, up to 2.5 cm. wide. Inflorescence divaricately paniculate, much surpassing the leaves, with the peduncle and rachis mark- edly fractiflex; peduncle 3-4.5 cm. long, complanate and narrowly bialate with two to four prominent spread- ing bracts which are conduplicate, broadly winged on the back and 1.2 em. or less long. Bracts subtending the branches of the panicle similar but smaller. Panicle loosely several- to many-flowered, with spreading or re- curved few-flowered branches. Flowers small, pale green- ish yellow, with spreading segments. Dorsal sepal elliptic- lanceolate or oblong-lanceolate, acuminate or short- acuminate, 3-nerved, about 8 mm. long and 2.2—2.9 mm. wide, dorsally keeled toward the apex. Lateral sepals very obliquely elliptic-lanceolate, complicate-acuminate above, concave, 3-nerved, prominently keeled toward the apex, 8.1-8.4 mm. long measured along the posterior margin, about 38 mm. wide. Petals linear, lightly sig- moid, acuminate, 1-nerved, with the margins densely ciliate, about 7 mm. long and 1.1 mm. wide. Lip adnate to the column nearly to the apex; lamina fleshy, concave below, 3-lobed near the middle, cordate at the base, acute or acuminate at the tip, 4-5 mm. long in the center; lateral lobes erect-incurved, semiovate or aliform, with or without a free apex; mid-lobe about equally large with the lateral lobes, porrect, triangular or ovate-triangular. Column stout and dilated upward when seen from the side, subentire at the apex, about 4mm. long at the back. This little species does not appear to have any near allies. Panama: Coclé, Cerro Pajita, hills north of El Valle, at 1100 meters altitude, October 27, 1946, Paul H. Allen 3780 (Tyrer in Herb. Ames No. 65445). Epidendrum pajitense C. Schweinfurth sp. nov. Herba elata. Caulis suberectus, gracilior, vaginis arctis [ 56 | tubulatis pustulatis omnino velatus. Folia disticha, late patentia, superne solum praesentia, oblongo-lanceolata vel anguste elliptico-lanceolata. Inflorescentiae laterales et terminales, laxe pluriflorae. Florum segmenta valde patentia. Sepala elliptico-oblonga, lateralia obliqua. Pe- tala cuneato-spathulata, subacuta. Labellum columnae valde adnatum; lamina trifida, cum lobis lateralibus ob- lique obovato-oblongis et lobo medio flabellato-cuneato, profunde bilobato. Columna apice quadrialata. Plant rather tall, “‘saprophytic.’” Rhizome apparently abbreviated. Roots fibrous, numerous, rather stout, gla- brous. Stem suberect, rather slender, with one strict flower-bearing branch above (the main stem broken off), entirely concealed by close tubular pustulose sheaths which are leaf-bearing above, about 49 cm. high. Leaves several to numerous, distichous, spreading, oblong-lan- ceolate or narrowly elliptic-lanceolate, obtuse to sub- acute, sessile at the shortly cuneate base, 8.5-6.2 em. long, about 1.3 cm. or less wide, coriaceous. Inflores- cences lateral and terminal; the lateral on the upper part of the stem, racemose, short, loosely few-flowered; the terminal one loosely paniculate with ascending few- flowered branches. Floral bracts minute, concave, ovate or oblong-ovate. Flowers medium-sized, rose-pink, with widely spreading more or less convex segments. Dorsal sepal oblong or elliptic-oblong, broadly obtuse to sub- acute, about 7-nerved, rather fleshy, about 9.3 mm. long and 3.9 mm. wide. Lateral sepals similar, obliquely elliptic-oblong, obtuse to subacute with a blunt dorsal mucro, about 10.5 mm. long, measured from the base of the dorsal margin to the tip, and 4 mm. wide, about 7- nerved. Petals cuneate-spatulate, abruptly rounded above with a subacute apex, 3-nerved with the lateral nerves branching, slightly oblique, about 10 mm. long and 3.2 mm. wide above. Lip adnate to the column up [ 57 ] to the tip; lamina 3-parted, much exceeding the sepals, submembranaceous; lateral lobes widely spreading, ob- liquely obovate-oblong, with a more or less truncate and obscurely lobulate outer margin, separated by a pair of small complanate calli, about 7.5 mm. long through the middle lengthwise and 3.9 mm. wide above; mid-lobe porrect, narrowly flabellate-cuneate, conspicuously bi- lobed with a minute blunt apicule between the oblong- rounded lobules and with a small callus at the base be- tween the lateral calli, about 8.2 mm. long to the tip of an apical lobule and 7.2 mm. wide across the apical lob- ules. Column short, very stout and dilated upward when viewed from the side, about 6.5 mm. long at the back including the erect 4-lobed wing, with the middle lobes oblong-subquadrate and irregularly lobulate on the trun- cate apex. This species has three rather close allies. Hpidendrum exasperatum Reichb.f. differs in having smooth cauline sheaths, verrucose outer surfaces of the sepals and flowers of another color. 17. Schumannianum Schltr. has broader leaves, broader petals, relatively smaller lateral lobes of the lip and spotted flowers. 2. verrucosum Sw. var. myrianthum (Lindl.) Ames & Correll shows much nar- rower and relatively longer leaves and a lip with dissim- ilar details. Panama: Coclé, Cerro Pajita, hills north of El Valle, at 1100 me- ters altitude, October 27, 1946, ‘“saprophytic,’’ in dense shade, Paul H. Allen 3784 (Tyrer in Herb. Ames No. 65444). Epidendrum strictiforme C. Schweinfurth sp. nov. Herba mediocris, epiphytica, recta. Rhizoma abest. Caulis in siccitate crassus, foliorum vaginis ancipitibus omnino vestitus. Folia disticha, quinque ut videtur, ovata vel oblongo-ovata, apice rotundata, basi lata amplexi- caulia. Inflorescentia erecta; pedunculus spathis condu- plicatis, erectis, imbricatis, duabus vel tribus omnino [ 58 | obtectus; racemus dense multiflorus. Flores parvi. Se- palum dorsale oblanceolatum, acutum. Sepala lateralia oblique oblongo-oblanceolata, acuta vel breviter acumi- nata. Petala oblanceolato-linearia, leviter incurva, acuta vel breviter acuminata. Labellum columnae valde ad- natum; lamina trilobata, carnosa; lobi laterales rotun- dato-dolabriformes, extus irregulariter crenulati; lobus medius ovatus, acutus. Columna generis, recta. Plant medium-sized for the genus, stout, epiphytic, up to 87 cm. tall. Roots and rhizome not present. Stem apparently stout, entirely concealed by strongly compla- nate tubular leaf-sheaths. Leaves five, loosely disti- chous, ovate or oblong-ovate, rounded and minutely bi- lobed at the apex, amplexicaul at the broad base, about 5-7.5 em. long, 2.8-3.4 cm. wide, spreading. Inflores- cence erect, racemose above; peduncle apparently about 11 em. long, entirely concealed by two or three strictly erect, conduplicate, imbricating spathes which are rounded to acute above ; raceme densely many-flowered ; floral bracts lax, linear-lanceolate, the lower ones about equaling the slender pedicellate ovary. Flowers small, pale brownish green flushed with purple, spreading. Dorsal sepal oblanceolate or elliptic-oblanceolate, acute, 5-nerved, about 9-10 mm. long and 3-3.4 mm. wide above. Lateral sepals obliquely oblong-oblanceolate, lightly sigmoid, acute or short-acuminate, longitudinally concave, 5-nerved, about 10.3-11 mm. long on the pos- terior margin and 38-3.3 mm. wide just above the middle. Petals oblanceolate-linear, lightly incurved, acute or short-acuminate, l-nerved, about as long as the dorsal sepal and 1 mm. wide above. Lip adnate to the column up to its apex; lamina sharply 3-lobed, fleshy, cordate at the base, about 5-5.8 mm. long in the center and 6.4-7.7 mm. wide across the lateral lobes; lateral lobes rounded-dolabriform, with irregularly crenate outer mar- [ 59 | gins; mid-lobe ovate, acute, protuberant, very fleshy, about 3-3.5 mm. long and 2.5—2.9 mm. wide at the base ; dise with a pair of small fleshy conical calli at the base. Column straight, dilated above in front, retuse in the center at the summit, about 6.5-7 mm. long on the dor- sal surface, produced on each side into an obliquely semiorbicular-ovate lobe. This species apparently lacks any close allies. It may be related to Mpidendrum sarcodes Lindl., but differs in having the peduncle entirely concealed by the spathes, and in having acute sepals, ete. Peru: Hudnuco, Carpish, at 2800 meters altitude, on a tree in cloud forest, leaf not thick but very hard and stiff, tip of column white, September 1946, F. Woytkowski 37014 (Tyre in Herb. Ames No. 65451). Brassavola ovaliformis C. Schweinfurth sp. nov. Herba epiphytica, pro genere mediocris, deserticola. Rhizoma repens, nodulosum. Caules approximati, plus- minusve elongati, vaginis tubulatis imbricatis scariosis omnino obtecti. Folia subteretia, gracilia, elongata, in siccitate arcuata. Flores duo, grandes, in pedunculi apice approximati. Sepala petalaque late patentia, valde sim- ilia, lanceolato-linearia, sensim angustata, apice acuta. Labellum multo brevius, inferne leviter involutum, ex- pansum ovale-ovatum, acutum, ecarinatum, margine in- tegro. Columna perbrevis, apice trialata, cum alis later- alibus faleato-lanceolatis et recurvis. Plant epiphytic. Roots fibrous, covered with a stout glabrous velamen. Rhizome creeping, nodulose, consist- ing of the approximate swollen bases of the stems. Stems crowded, up to 12.7 cm. or more long (the longest one incomplete), several-jointed, entirely enveloped by long tubular scarious imbricating sheaths, gradually dilated upward, 1-leaved at the apex. Leaves very slender, sub- terete, channelled, arcuate in the dried specimen, about [ 60 ] 27.2-88.2 cm. or more long (the longest blade incom- plete), apparently about 3 mm. in diameter. Peduncle exceeding 2.5 cm. in length (incomplete), with two flow- ers at the apex. Floral bracts very small, lanceolate- ovate, acuminate, deeply concave, amplexicaul, much shorter than the pedicellate ovary which is about 4 cm. long. Flowers large with widely spreading segments. Dorsal sepal lanceolate-linear, long-narrowed to an acute tip, about 5.8-5.8 em. long and 5.5 mm. wide near the base. Lateral sepals closely similar, about 5.9 ecm. long and 5 mm. wide below. Petals similar to the sepals, linear-lanceolate, about 5—-5.8 em. long and 5.4 mm. wide below. Lip much shorter than the other segments, oval-ovate or lanceolate-oval in outline when expanded, about 4.7-4.9 em. long and 2.9 em. wide across the mid- dle, acute, ecarinate, with the lower half gradually in- rolled in natural position, the margins being entire. Column minute, with an erect 8-lobed wing at the apex, about 6.5 mm. high at the back; lateral wings falcate- lanceolate and recurved ; middle wing low and irregular, abruptly extended in the center into a ligulate, apically truncate and denticulate projection; anther oblong- ellipsoid, 2-celled, with four narrowly oblong-ellipsoid pollinia in each cell. This species appears to have no near allies. In the whiteness of the flowers and in the apex of the column it resembles the widespread Brassavola nodosa (.) Lindl., but it lacks the elongated narrow serrated basal portion of the lip of that species. B. Perriniit Lindl., from Brazil, differs in having smaller green flowers, a somewhat carinate lip and dissimilar column-wings. Peru: Amazonas, La Peca, northeast of Jaen, at 1000 meters al- titude, on dwarf trees in dry “‘desert forest full of cacti,’’ flower uni- formly white, the lip having a median greenish yellow line above and ee, e eas beneath, interior parts ... pale green save the whitish column and the two formations immediately below, which are whitish with a brown [ 61 ] hue, leaves dark green, tapering, stiff, round like thick wire,’’? No- vember 18, 1947, F. Woytkowski 37016 (Tyre in Herb. Univ. Calif.). Odontoglossum angustatum Lindley in Bot. Reg. 23 (1887) sub t. 1992; Fol. Orch. Odontoglossum (1852) 17, no. 48, non O. angustatum Lindley Orch. Linden. (1846) 17, no. 90; Bateman Monog. Odontog. (1874) t. 26. Odontoglossum tetraplasium Reichenbach filius in Gard. Chron. n.s. 8 (1875) 558. Odontoglossum bellum Schlechter in Fedde Repert. Beih. 9 (1921) 108; ex Mansfeld in Fedde Repert. Beih. 57 (1929) t. 127, nr. 499. Odontoglossum Loesenerianum Schlechter in Fedde Repert. Beih. 9 (1921) 110; ex Mansfeld in Fedde Repert. Beih. 57 (1929) t. 128, nr. 502. A sketch of the panicle with a drawing of a solitary flower (natural size) and a single enlarged lip of Odonto- glossum tetraplasium from the Reichenbach Herbarium shows that this concept cannot reasonably be separated from O. angustatum Lindl. (1837), as represented by a photograph of the type specimen from the Lindley Her- barium bearing a pen drawing of the callus on the lip. The vegetative parts were lacking in the type specimens of both species, but these were supplied for O. angusta- tum by a subsequent collection mounted with the Lind- ley type. Judging from the description and floral analysis, Odon- toglossum bellum is also a form of this variable species. O. bellum appears to be quite as large throughout as typical O. angustatum, but the sepals are not so elongate- acuminate as in many forms of the latter. The lip which is described and shown as obtuse and apiculate, is not very dissimilar to the acute lip of Vargas 2888 and Vargas 3664 which have been determined as O. angustatum. [ 62 | The floral analysis of Odontoglossum Loesenerianum indicates that this concept is surely referable to O. an- gustatum as first described by Lindley, the only apparent discrepancies being that the bracts are noted as being equal to or a little exceeding the ovary (they are about half as long in O. angustatum) and the flowers are de- scribed as somewhat smaller. This difference in size is probably due to an immature condition. In the several collections here referred to O. angusta- tum, there appears to be a wide variation both in vege- tative and floral size and in the form of the lip. The most striking variations from the usual form are seen in Var- gas 2888, where the obviously immature flowers are smaller than those of O. Loesenerianum and have a lip which is more or less lightly pandurate; and in Weber- bauer 7797, where the carinate calli at the base of the lip are broader, blunter and thicker than in the usual form. All of the Peruvian collections included in this species, however, have a general similarity of lip calli, but there is apparently lacking from all of them one extra pair of tubercles present in the type. It seems to me, therefore, that Odontoglossum angus- tatum is a conspicuous example of the polymorphism that makes the tropical orchids so difficult. Odontoglossum aureo-purpureum (as auropur- pureum) Reichenbach filius in Linnaea 22 (1849) 848; Lindley Fol. Orch. Odontoglossum (1852) 15, no. 44. Odontoglossum compactum Reichenbach filius in Gard. Chron. n.s. 3 (1875) 492. Odontoglossum Koehleri Schlechter in Fedde Repert. Beih. 9 (1921) 109; ex Mansfeld in Fedde Repert. Beih. 57 (1929) t. 128, nr. 501. The concept Odontoglossum compactum is based in part on Peruvian specimens collected by W. Lobb which [ 63 ] were formerly referred by Lindley (Fol. Orch. Odonto- glossum p. 15) to O. aureo-purpureum. It was separated from the earlier species by its more compact densely flowered panicles of larger flowers. However, two recent Peruvian collections (Vargas 2879 and Metcalf 30742) have the loose panicles seen in the typical Venezuelan O. aureo-purpureum, but they have larger flowers than those of the type which they appear to match in floral details. Odontoglossum Koehleriis described as having a rather densely flowered panicle, like O. compactum; but its floral segments, except for being slightly broader, coin- cide well with those of typical O. aureo-purpureum. It seems to me, therefore, that O. awreo-purpureum is a widely variable species which includes the above concepts. Odontoglossum brevifolium Lind/. var. Weber- bauerianum (Kriinzl.) C. Schweinfurth comb. nov. Oncidium Weberbauerianum Krinzlin in Engler Bot. Jahrb. 37 (1906) 389. Cyrtochilum Weberbauerianum Krinzlin in Notizbl. Bot. Gart. Berlin 7 (1917) 95, nomen nudum, in clavi; in Engler Pflanzenreich LV. 50, pt. 2 (Heft 80) (1922) 60, fig. 5H, a-c. Odontoglossum Weberbauerianum Schlechter in Fedde Repert. Beih. 27 (1924) 109. In Fedde Repert. Beih. 9 (1921) 170, Dr. Schlechter cites Oncidium Weberbauerianum as a synonym of the well-known Odontoglossum brevifolium Lindl. I have not seen any figure of Oncidium Weberbaueri- anum, but, after examining an iso-type collection of Odontoglossum brevifolium, there appear to be several marked differences between that species and the descrip- tion of Cyrtochilum Weberbauerianum (supplemented by [ 64 | a floral analysis). It seems to me, therefore, that the above varietal designation is preferable, for the reasons indicated in the following table of comparisons. Odontoglossum brevifolium Pseubobulbs ovoid to cylindric- ovoid, up to 5 em. long. Leaves oval to oblong-elliptic, up to 15 em. long and 7.6 cm. wide. Sepals suborbicular to round- obovate, little longer than broad. Lip deeply bilobed at the apex. Oncidium Weberbauerianum Pseudobulbs “‘linear,’’ up to 10 em. long. Leaves oblong, up to 20 em. long and 4.5 em. wide. Sepals oblong, much longer than broad. Lip obscurely lobulate in front. Odontoglossum flavescens fPo/fe in Orch. Rev. 12 (1904) 92. This species was too inadequately described to furnish a definite concept that can be visualized or keyed. The description, which gives no hint of vegetative parts, merely states that the flowers are clear yellow, that the segments are elliptical-oblong and subconnivent, while the blade of the lip is suberect, rounded below, with an oblong obtuse apex and has a very large 2-lobed callus appressed to the base of the column. It is noted as being allied to O. retuswm Lindl., and was introduced by Messrs. F. Sander & Co., presumably from Ecuador and Peru. A photograph of the type specimen in the Herbarium at Kew shows fragments of an elliptical-oblong pseudo- bulb which is about 8.5 em. long and appears to be bifo- liate or trifoliate. The single leaf shown is oblong-linear, short-acuminate, slightly narrowed to a sessile base, over 51cm. long and about 2.4cm. wide, with the mid-nerve very prominent beneath. The inflorescence shows a very loosely branched panicle with two horizontally spreading branches. ‘The flowers are distant, secund, on long ped- icels (up to 2 cm. long) much exceeding the minute bracts, and the segments appear to be about 1 cm. long. [ 65 ] Even with this additional information, however, it seems advisable to consider this concept among the ob- scure species. Odontoglossum mystacinum (Lindl.) Lindley Fol. Orch. Odontoglossum (1852) 6, no. 14. Cyrtochilum mystacinum Lindley in Bot. Reg. 24 (1838) Mise. 30, no. 38; in Bot. Reg. 25 (1839) t. 62. Odontoglossum rigidum Lindley in Benth. Pl. Hartw. (1844) 152; Fol. Orch. Odontoglossum (1852) 7, no. Ve Judging by an excellent photograph of the type of Odontoglossum rigidum from the Lindley Herbarium, this concept cannot reasonably be separated from the species depicted as Cyrtochilum mystacinum in the Botan- ical Register 25, t. 62. A slight difference is that the pseudobulb of the latter concept is represented as uni- foliate, whereas that of O. rigidum is bifoliate with very unequal leaves. Recent Peruvian collections which are referable to this concept have sometimes unifoliate and sometimes bifoliate pseudobulbs. Although O. rigidum has a more branched and fractiflex panicle than that of Cyrtochilum mystacinum, the plate was obviously drawn from an immature specimen. A minor difference is that the wings of the column in the latter concept are rather broad and multifid, whereas they are drawn as rather narrow and serrate in O. rigidum. Both species, however, have similar pseudobulbs and yellow flowers with almost identical segments. The se- pals are lanceolate, the petals ovate-lanceolate and the lip pandurate-obovate. Moreover, the column wings are not easily seen in the specimens examined and appear to be variable. This species appears to be limited to Peru, Ecuador, and perhaps Bolivia. [ 66 | Odontoglossum obscurum C. Schweinfurth nom. nov. Mesospinidium Wallis Reichenbach filius in Linnaea 41 (1876) 106, non Odontoglossum Wallisii Linden & Reichb. f. (1870). As illustrated by a series of floral analyses from the Reichenbach Herbarium, Mesospinidium Wallisii has the basal part of the lip parallel to the column with the an- terior half abruptly reflexed. It seems, therefore, to be clearly referable to the genus Odontoglossum. Since the specific epithet, Wadllisii, has already been used under Odontoglossum, the new name obscurum is proposed, in allusion to the fact that the pseudobulb and leaves were unknown when the species was described. Except for the lip, which is described and illustrated as retuse at the apex and emarginate in the middle [of each side], this concept might reasonably be referred to Odontoglossum longifolium Lindl. Odontoglossum Wyattianum G. Wilson in Orch. Rev. 36 (1928) 47. In the diagnosis of this species there is no description of the vegetative parts of the plant nor of the inflores- cence. With regard to the flowers, only the size and color are noted; there is no mention of the exact shape of the parts. Finally, the precise origin of the species is doubt- ful, for the description merely states that ‘‘the plant had been obtained from the Rev. Paul Wyatt, Bedford, to whom it had been forwarded by a friend in Peru, doubt- less its native country. ”’ It seems advisable, at the present time, to relegate this concept to the class of obscure species. Oncidium falcipetalum Lindley Orch. Linden. (1846) 14, no. 76; Fol. Orch. Oncidium (1855) 5, no. 7: [ 67 ] Cogniaux Dict. Icon. des Orch. Oncidium (1899) t. 21. Oncidium Pavoni Reichenbach filius ex Lindley Fol. Orch. Oncidium (1855) 5, sub no. 7, nomen nudum; Kriinzlin in Engler Pflanzenr. LV. 50, pt. 2 (Heft 80) (1922) 38. Oneidium ionodon Reichenbach filius in Linnaea 41 (1876) 23. Oncidium Davisn Reichenbach filius in Linnaea 41 (1876) 24. Cyrtochilum faleipetalum Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 87, fig. 1 G, a-e. Cyrtochilum Pavoni Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 388. Oneidium Pavoni, although lacking any original diag- nosis, has been considered by the several writers men- tioning the name as a form of O. falcipetalum. Judging from the description (limited to the flower), the Peruvian Oncidium ionodon differs from O. falcipet- alum only in having an emarginate dorsal sepal, clawed petals and porrect violet lateral lobes of the lip. How- ever, the plate of O. falcipetalum illustrated in Cogniaux’ work (I.c.) shows a crisped dorsal sepal, which might well be taken for emarginate, and petals that appear to be more or less clawed. Also, in specimens reasonably re- ferred to O. falcipetalum, the petals (when examined closely) are very shortly clawed. In general, too, the color of the flower attributed to O. tonodon coincides with that shown in the plate indicated. Onecidium Davis, of which there is in the Ames Her- barium a tracing of the floral analysis from the Reichen- bach Herbarium, is almost an exact counterpart of O. ionodon, except that the dorsal sepal is noted as acute, the petals as very shortly clawed, and the lateral lobes of the lip deflexed. It is a strange coincidence that both O. ionodon and [ 68 | O. Davisii are cited as synonyms of one species, Cyrto- chilum monachicum, by Krinzlinin Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 47. However, the latter con- cept, based upon Oncidium monachicum Reichb.f., is de- scribed as having the dorsal sepal reniform, and thus is quite dissimilar to that of O. falcipetalum or its forms. Apparently Oncidium falcipetalum is a variable species as regards floral size and the contour of the petals which varies from flat to conduplicate. Originating in Venezuela (Merida), this species is also recorded from Colombia and Peru. Oncidium heteranthum Poeppig & Endlicher Nov. Gen. ac Sp. 1 (1886) 34, t. 60; Cogniaux in Martius FI. Bras. 8, pt. 6 (1905) 878, t. 87; Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 175. Oneidium bryolophotum Reichenbach filius in Gard. Chron. (1871) 788; Krinzlin in Engler Pflanzenr. LV. 50, pt. 2 (Heft 80) (1922) 181. Oncidium inops Cogniaux & Rolfe in Journ. des Orch. 3 (1893) 846; 4 (1893) 74. Oneidium megalous Schlechter in Fedde Repert. 9 (1911) 30. The above species is extremely variable. This fact has led to the proposal of several later concepts, none of which, however, seems valid. Oneidium bryolophotum, a native of Costa Rica and Panama, and represented in the Ames Herbarium by Reichenbach’s detailed analyses, by authentic material examined by Kriinzlin, and by many Central American collections, cannot logically be separated from the earlier O. heteranthum trom Colombia, Peru and Bolivia. In general, the Central American plants referred to O. bryolophotum have a mid-lobe of the lip which is rel- atively larger than that of O. heteranthum, but one Costa [ 69 | Rican collection (Standley 33741) has a lip which is quite similar to that of the South American O. heteranthum. One Peruvian collection of O. heteranthum (Vargas 2531) varies from the usual form in lacking the separate calli on the lateral (or basal) lobes of the lip and in hav- ing the apex of the column-wings narrowed to an acu- minate point rather than having the typical more or less broad apex. Another Peruvian collection (Schunke s.n., Herb. Field Mus. No. 571665) has a row of detached calli extending onto the lateral lobes on each side of the middle callus. Other differences noted in the various collections ap- pear in the vegetative size, in the oblong-ovoid to cylin- dric pseudobulbs, and in the leaves which vary from linear-oblong to elliptic. The other concepts cited above have already been re- duced to the synonymy of O. bryolophotum. Oncidium incarum (Ariinzl.) C. Schweinfurth comb, nov. Cyrtochilum Incarum Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 58. Since it seems unwise to regard the concept Cyrtochi- lum as distinct from the large and variable genus Oncid- ium, the above transfer is necessary. Oncidium macranthum Lindley var. hastiferum (Reichb.f. & Warse.) C. Schweinfurth comb. nov. Oncidium hastiferum Reichenbach filius & Warscewicz in Bonpl. 2 (1854) 102; Reichenbach filius in Walp. Ann. 6 (1863) 708. Cyrtochilum hastiferum Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 34. Judging by the descriptions of Oncidium hastiferum and by a drawing in the Reichenbach Herbarium, made from cultivated material, it appears preferable to con- [ 70 ] sider this concept as a variant of the widespread O. macranthum. It seems to differ from O.macranthum only in having one large median keel flanked by a smaller lamella on each side (instead of three equal lamellae) and in having narrower ligulate wings on the column. Oncidium obryzatum Reichenbach filius & Warsce- wicz in Bonpl. 2 (1854) 108; Regel in Gartenfl. 27 (1878) t. 925; Kriinzlinin Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 239, fig. 20 B, a-d. The Peruvian collection cited below shows several dis- crepancies from typical examples of O. obryzatum in the Ames Herbarium. The pseudobulb is bifoliate, not unifoliate as described. The leaves are longer than exemplified, being up to 52.5 em. long. The branches of the peduncle appear to be rather more distant than otherwise shown, being about 5-6.5 cm. (instead of 2-4 cm.) apart. The sepals and petals, although closely similar to those of the typical form, are somewhat abruptly acute, rather than obtuse or rounded or even retuse. The narrow middle part of the lip is somewhat broader than usual and the anterior portion or middle lobe is narrower relative to the basal portion. The column wings are merely acute, not acu- minate, above. In view of the general agreement of this specimen with the typical form, however, this collection is incor- porated with O. obryzatum. Peru: Cajamarca, Cutervo, about Socota, on Socota River, at 2800 meters altitude, epiphyte, December 11, 1938, H. E. Stork & O. B. Horton 101852. Oncidium pyramidale Lindley in Ann. & Mag. Nat. Hist. 15 (1845) 384; Fol. Orch. Oncidium (1855) 29, no. 98; Veitch Man. Orch. Pl., pt. 8 (1892) Oncid- ium 73; Kranzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 195, fig. 16 J, a-c. [71 ] Oncidium chrysopyramis Reichenbach filius & Warsce- wicz in Bonpl. 2 (1854) 108; Lindley Fol. Orch. On- cidium (1855) 29, no. 97; Veitch Man. Orch. PIl., pt. 8 (1892) Oncidium 25; Krinzlin in Engler Pflanzenr. LV. 50, pt. 2 (Heft 80) (1922) 196, fig. 16 F, a-d. Judging from drawings of the habit and floral analyses of Oncidium chrysopyramis from the Reichenbach Her- barium, supplemented by the descriptions, it does not seem reasonable to separate this concept from the earlier O. pyramidale. The latter species appears to differ from O. chrysopyramis in having a longer panicle with the lower branches often compound and many-flowered, not simple and 8- to 5-flowered—a character which is varia- ble and therefore unimportant as a basis for specific sep- aration. The basal portion of the lip is described as being broader than the anterior portion in both species, con- trary to remarks made by Reichenbach following his description of O. chrysopyramis. Again, the wings of the column, a character which was considered of great consequence in separating the concepts, appear to be highly variable in Reichenbach’s own figures. Indeed, the column-wings shown in the pen drawing on the type collection of O. pyramidale appear to be a close approximation to some of those de- picted by Reichenbach for O. chrysopyramis. Oncidium superbiens Reichenbach filius in Lin- naea 22 (1849) 843; Hooker filius in Bot. Mag. 98 (1872) t. 5980; Warner & Williams Orch. Alb. 6 (1887) t. 276. Oncidium aemulum Reichenbach filius & Woarscewicz in Bonpl. 2 (1854) 102. Oncidium inferlobum hort. ex Gard. Chron. (1872) 904, in synon. Oncidium undulatum Warner & Williams Orch. Alb. [ 72 ] 8 (1889) t. 368, non O. undulatum (HBK.) Lindl. (1842). Cyrtochilum aemulum Wrinzlin in Notizbl. Bot. Gart. Berlin 7 (1917) 98, nomen nudum, in clavi; in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 47, fig. 3 C, a—e. Cyrtochilum superbiens Krinzlin in Notizbl. Bot. Gart. Berlin 7 (1917) 93, nomen nudum, in clavi; in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 49, fig. 3 E, oe After a careful comparison between Oncidium super- hiens (represented by a photograph of the type and the type description) and O. aemulum (represented by a photograph of Lindley’s concept, supplemented by the type description), I am unable to separate the two species. All the specimens of O. superbiens noted appear to have the petals yellowish or whitish, more or less barred below with brown or purplish brown, whereas there is no record of such coloring in O. aemulum. The figures of these concepts in the Pflanzenreich (under Cyrtochilum) confirm my conclusions that they are conspecific. Oncidium ventilabrum Reichenbach filius & War- scewicz in Bonpl. 2 (1854) 101; Lindley Fol. Orch. On- cidium (1855) 6, no. 12. Cyrtochilum undulatum Humboldt, Bonpland & Kunth Nov. Gen. et Sp. Pl. 1 (1816) 849, t. 84; Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 57. Oncidium undulatum Lindley Sert. Orch. (1842) sub t. 48, no. 1; Reichenbach filius & Woarscewiez in Bonpl. 2 (1854) 108; Lindley Fol. Orch. Oncidium (1855) 7, no. 15, nee Oncidium undulatum Sims (1804) nec O. undulatum Salisb. (1812). Cyrtochilum ventilabrum Kriinzlin in Notizbl. Bot. [ 73 ] Gart. Berlin 7 (1917) 92, nomen nudum, in clavi; in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 39, fig. 2C, a-c. In his original description of Oncidium ventilabrum Reichenbach admitted that this species was close to O. undulatum Lindl., but he adduced several supposedly weighty reasons for separating them. Among other fea- tures he said, ‘‘Zuniichst hat unsre Art zweimal so grosse Bliithen, die seitlichen Sepala breiter, das obere linger. Der mit dem Fuss der Siitule verwachsene Lippennagel is viel linger... .”° In the Ames Herbarium there are several records of Oncidium ventilabrum (a drawing of a flowering branch and numerous floral analyses) from the Reichenbach Herbarium, and these show a rather striking agreement both in size and form of sepals and petals with those shown in the plate of the type of Cyrtochilum undulatum (l.c.). Moreover, the lip is depicted as sessile in both concepts. The form of the lip and the calli at the base of the lamina are supposed to be different in the two species, but in the drawings of O. ventilabrum there is a rather wide range in these features. Furthermore, in the remarks about O. undulatum (in Bonpl. 2, p. 108) Reich- enbach described a lip-callus which is rather similar to that shown in O. ventilabrum. It appears to be the wise procedure to eliminate the necessity of straining for supposed differences between these two concepts and to consider them conspecific. However, since there is an earlier and different On- cidium undulatum, it is necessary to reject this specific epithet and to adopt the next earlier name, Oncidiwm ventilabrum. Sigmatostalix Reichenbach filius in Bot. Zeit. 10 (1852) 769; Schlechter in Fedde Repert. 15 (1918) 139; [ 74 ] Krinzlin in Engler Pflanzenr. LV. 50, pt. 2 (Heft 80) (1922) 301. Petalocentrum Schlechter in Fedde Repert. 15 (1918) 144; Kranzlin in Engler Pflanzenr. [V. 50, pt. 2 (Heft 80) (1922) 312. The concept Petalocentrum was separated from the variable genus Sigmatostaliv by reason of three alleged differences (Cf. Schlechter in Fedde Repert. 15 (1918) 145), viz. the sessile lip, the spur-like outgrowth at the base of the petals and the ascending rostellum. The characterization of the lip in the type description of Sigmatostalix (1.c), to be sure, specifies ‘‘labellum un- guiculatum.’’ However, the lip in the typical species, S. graminea (as Specklinia graminea Poepp. & Endl. in Nov. Gen. ac Sp. 1 (1836) 51, t. 89 B), appears to be cuneate below to a sessile base; and, even if the analysis is discarded as being inadequate (on the basis of later representations), that organ seems to be at most only very shortly unguiculate. At any rate, some species which were described as be- longing to Sigmatostalivx, such as the Central American S. hymenantha Schitr., have the basal part of the lip more or less cuneate to a sessile base (apparently very broadly cuneate). And other species, such as S. macro- bulbon Kriinzl., have an absolutely sessile lip. In the face of such wide variation, therefore, it seems advisable to discard this character, at least as to its generic value. As regards the spur-like outgrowth on the petals, this character appears to be quite absent in some species with a strikingly sessile lip which were referred to Sigmato- stalix, such as iS. macrobulbon Krinzl. Nor does this character appear in S. hymenantha Schltr. with the broadly cuneate base to the lip. It would seem, there- fore, that this outgrowth on the petals, seen in one poly- morphic species of this alliance, is scarcely of generic weight when taken alone. Finally, the factor of an ascending rostellum appears to be a matter of degree, even if not too recondite and obscure for ordinary recognition in the dried specimen. It appears to me that the wise course is to relegate Petalocentrum to the synonymy of the variable genus Sigmatostala. The concept, Petalocentrum angustifolum, the only one of the genus not previously described as representing Sigmatostalix, was considered identical with Sigmatosta- liv pusilla Schltr. (cf. Kriinzlin in Engler Pflanzenr. LV. 50, pt. 2 (Heft 80) (1922) 312), and the generic identity of the concepts Petalocentrum and Sigmatostalia was once suspected by Kriinzlin (ef. lc. 318). Sigmatostalix peruviana /folfe in Kew Bull. (1910) 371. Sigmatostalix pusilla Schlechter in Fedde Repert. 10 (1912) 392. Sigmatostalix bicornuta Rolfe in Kew Bull. (1913) 842. Petalocentrum pusillum Schlechter in Fedde Repert. 15 (1918) 145; Krinzlin in Engler Pflanzenr. IV. 50, pt. 2 (Heft 80) (1922) 312; ex Mansfeld in Fedde Repert. Beih. 58 (1980) t. 59, nr. 285. Petalocentrum angustifolum Schlechter in Fedde Re- pert. 15 (1918) 145; ex Mansfeld in Fedde Repert. Beih. 58 (1930) t. 59, nr. 234. Petalocentrum bicornutum Schlechter in Fedde Re- pert. Beih. 9 (1921) 179. All of the concepts cited above appear to be closely similar vegetatively, although Sigmatostalia pusilla is described as having markedly shorter pseudobulbs and leaves than the others. It is significant that the sepals and petals of all of the concepts are of almost exactly the [ 76 | same form and size, the only discrepancy being in the degree of acuteness of the apex. In 8. peruviana there is no mention of a horn at the base of the petals (which occurs in all of the others), but this species seems surely identical with S§. bicornuta, even the color of the flowers being nearly identical. The lip seems to vary somewhat in size in the various species, but it appears to be of very similar form throughout. A rather wide diversity is de- scribed in the callus on the lip in the various concepts, but it is noteworthy that this feature appears to be often very variable in a single species of Sigmatostalia. Since, without actual specimens, it would be extremely difficult to recognize differences between these species, the wise course suggests their union. Ornithocephalus gladiatus Hook. var. peruvi- anus C. Schweinfurth var. nov. Planta scapis brevioribus folia non valde excedentibus, sepalorum mucronibus valde prominentibus, sepalo dor- sali quam sepalis lateralibus majore, ac praesertim labelli callo margines basales solum paulo superanti a specie differt. Plant small, with the general appearance of Ornitho- cephalus gladiatus Hook., but differing in having scapes shorter than or subequaling (not markedly exceeding) the leaves, in having a very conspicuous mucro on the sepals, in having the dorsal sepal somewhat larger than the lateral sepals, and particularly in having a less prom- inent and spreading basal callus on the lip which but slightly exceeds the margins. Peru: Junin, La Merced, at about 610 meters altitude, in mon- tania, on tree trunk, flowers white with green markings, August 10- 24, 1923, J. Francis Macbride 5480 (Tyre in Herb. Field Mus. No. 536520). < ILLUSTRATIONS EXPLANATION OF THE ILLUSTRATION PLare XII. Cranicurs crrmLasia C. Schweinfurth. 1, plant, natural size. 2, flower expanded, from the front, four times natural size. 8, half of the lip, cut longitudinally, from the side, seven times natural size. Drawn by E. W. Smiru PLATE XII CRANICHIS ~~ (720, litlab ~ C2 Ca chweinf. EXPLANATION OF THE ILLUSTRATION Pirate XIII. Ocromerta pyemara C, Schweinfurth. 1, plant, eight times natural size. 2, dorsal sepal, sixteen times natural size. 3, lateral sepal, sixteen times natural size. 4, petal, sixteen times natural size. 5, lip expanded, from above, thirty-two times natural size. Drawn by EK. W. Suiri PuatE XIII OCTOMERIA PYG MACH vf (2 C. Schwe EXPLANATION OF THE ILLUSTRATION Pirate XIV. EerpenpruM FLEXUOsIssIMUM C. Schwein- Jurth, 1, plant, three fourths natural size. 2, flower from the side, natural position, two and one half times natural size. 3, lip and column from the side, natural position, three and one half times natural size. 4, lamina of lip expanded and apex of col- umn, from the front, five times natural size. 5, dor- sal sepal, three and one half times natural size. Drawn by FE. W. Surru PLaTE XIV EPIDENDRUM flexuosr SS Lmnunme Cu chweinf- EXPLANATION OF THE ILLUSTRATION PLate XV. Eprpenprum pasiTensE C.Schweinfurth. 1, plant, one half natural size. 2, flower partially expanded, from the front, one and one half times natural size. 3, lip and column from the side, nat- ural position, one and one half times natural size. 4, lamina of lip expanded, from the front, twice natural size. 5, lateral sepal, twice natural size. Drawn by E. W. Surru PLaTE XV me ' A \= : M, | - EPIDENDRUM Pes # tCMNSE pai CS chweinfr EXPLANATION OF THE ILLUSTRATION Pirate XVI. Eprpenprum strictirorMe C,Schwein- Jurth. 1, plant, seven sixteenths natural size. 2, flower expanded, from the front, two and one half times natural size. 3, column and lip from the side, natural position, two and one half times natural size. 4, lateral sepal, two and one half times nat- ural size. Drawn by E. W. Suitru PLATE XVI EPIDENDRUM strictiforme C8 chweinf- EXPLANATION OF THE ILLUSTRATION Prare XVII. Brassavota ovatirormis C.Schwein- Jurth. 1, plant, one half natural size. 2, column from the side, twice natural size. 3, column from above, twice natural size. 4, lip expanded, from the front, one half natural size. Drawn by KE. W. Smiru PLatTE XVII BRASSAVOLA ——— ovaliformis ~ fg C. Schwe BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CampripGr, Massacuusetts, Fesruary 6, 1950 Voi. 14, No. 4 STUDIES IN THE GENUS HEVEA III BY RicHarp Evans ScHuutres! ON THE USE OF THE NAME HEVEA BRASILIENSIS For some time, it has been believed by certain author- ities that Hevea brasiliensis, the name long used to de- note the well known cultivated rubber tree, is untenable if the International Rules of Botanical Nomenclature be strictly interpreted. Several botanists have written on the complicated problem which underlies this presumed untenability, treating it from differing points of view and with diverg- ing conclusions. This divergency of opinion has led to uncertainty amongst taxonomists not only concerning the actual status of the name Hevea brasiliensis, but also as to the authorities to whom it should be attributed. In 1858, Baillon (Etude Euphorb. (1858) 326) pointed out that the name Stphonia brasiliensis HBK. had been applied apparently to two distinct plants: one from the Orinoco and one from the lower Amazon. He proposed to reserve Stphonia brasiliensis for the latter—which is our cultivated species—and published a new name (:S?- phonia Kunthiana Baill.) for the former. Later, in 1900, Warburg (Kautschukpflanzen (1900) 'Botanist, Bureau of Plant Industry, Soils, and Agricultural Engi- neering, Agricultural Research Administration, United States Depart- ment of Agriculture; Research Fellow, Botanical Museum, Harvard University. [ 79 ] 26) discussed this same problem and resolved, in ac- cordance with what appeared to him to be priority, to keep the name Hevea brasiliensis for the Orinoco species and to give the cultivated plant an entirely new name (Hevea Sicberi Warb.). Shortly thereafter, Huber (in Bull. Soc. Bot. France 49, ser. 2 (1902) 43) thought it a better policy to con- serve Hevea brasiliensis for the cultivated species, and he actually made the proposal that it be so conserved, In 1905, Ule studied the circumstances and also con- cluded (in Engler Bot. Jahrb. 85 (1905) 664) that the name Hevea brasiliensis should be kept for the Brazilian species and Hevea Kunthiana used for the concept rep- resented by the Venezuelan collections. In reviewing the problem recently, Burkill (Dict. Econ. Prod. Malay Penins. 1 (1985) 1159), stated that: ‘It would lead to much confusion were botanists at this date to displace the... . name [ Hevea brasiliensis), on the ground that Siphonia brasiliensis Kunth is not the plant which everyone now calls Hevea brasiliensis; yet, if the rules of nomenclature are followed strictly, that, it seems, should happen. ”’ In 1986, Chevalier (in Rev. Bot. Appl. Agric. Trop. 16 (1936) 620) published a most complete review of the historical aspects of the problem. Basing his opinions ona study of the literature, combined with an examina- tion of authentic collections preserved in Paris, he came to the conclusion that the valid name for the cultivated species of Hevea is H. brasiliensis. In his own words, his conclusion is: ‘‘Celle [the collection] du Para [as opposed to the Orinoco material] doit garder le nom de Hevea brasiliensis (Willd.) Muell.-Arg. (excl. syn. ALK.) Although Chevalier intimated that Willdenow’s Sv phonia brasiliensis had been validly published to denote [ 80 | what we today call Hevea brasiliensis, he did not enter into a discussion of the reason why it was so according to the Rules and why the name cannot, under any cir- cumstances, be applied to the Orinoco material. In spite of the fact that Chevalier’s paper constitutes the most important contribution of the century towards a clarifica- tion of this question, it has apparently not received the attention it deserves. Baldwin (in Journ. Hered. 38 (1947) 54; ibid. 40 (1949) 47) accepted Chevalier’s con- clusions, but other investigators who have recently pub- lished on Hevea (Schultes in Bot. Mus. Leafl. Harvard Univ. 12 (1945) 7; Seibert in Ann. Missouri Bot. Gard. 34 (1947) 305), by using ‘‘(HBK.) Muell.-Arg.’’ as authorities for the binomial Hevea brasiliensis, have in- dicated acceptance of the long-established belief that Kunth’s publication of Stphonia brasiliensis was the earliest. Cook, in 1941, went much farther than all who had previously discussed this problem. He _ proposed (in Journ. Wash. Acad. Sci. 31 (1941) 46) to substitute the new name Siphonia Ridleyana Cook for our cultivated rubber tree. He rejected the generic epithet Hevea on the basis of faulty reasoning and an erroneous under- standing of the meaning of the term homonym. Even were a new name needed, Cook’s substitute specific epi- thet would be superfluous in view of Warburg’s Hevea Siebert of 1900. Cook did not mention Warburg’s work in his rather extensive discussion, nor did he indicate by citation or by context that he was familiar with Cheva- lier’s convincing article. To help end the continued uncertainty in regard to the name of such an important economic plant, and to reiterate Chevalier’s conclusion and connect the reasons for it with the corresponding authorizing Article of the International Rules of Botanical Nomenclature, I shall [ 81 ] restate the pertinent historical facts and present as com- plete a synonymy as possible. In 1800, Humboldt collected material of Hevea along the Rio Tuamini near Javita and also along the Rio Orinoco at San Fernando de Atabapo, both localities in the upper Orinoco basin of southern Venezuela. ‘These collections were described by Kunth in 1825 under the name Siphonia brasiliensis. This has been taken almost unanimously as the first valid publication of the binomial. In the first decade of the nineteenth century, Count Hoffmannsegg received, presumably from the traveller- collector F. G. Sieber, who worked in Para from 1801 to 1807, specimens of Hevea which he turned over to Willdenow for study. There is strong reason to believe that this material came from the lowermost course of the Rio Amazonas. Willdenow annotated the material with the name ‘‘Siphonia brasiliensis Willd.’’ and deposited it in his herbarium which was preserved at Berlin-Dahlem. These specimens represent the concept which we have come to know as Hevea brasiliensis. When Kunth published the name Siphonia brasiliensis, he described the two Venezuelan collections and cited them as the only basis for the description. He did not cite the Brazilian material of Sieber, but he did include in synonymy “‘Stphonia brasiliensis Willd. herb.”’ with the following footnote: ‘‘In specimine brasiliensi a Willdenoicum cel. Beauvois communicato (inque Museo Lessertiano asservato) foliola multo minora, subtus pal- lide viridia (nec albida).*” Also in synonymy, he included ‘*Siphoniae species brasiliensis Adr. de Juss. Kuphorb. p. 40°° and ‘‘S. foliolis oblongis, acuminatis. Willd. mss.’’ Willdenow had the habit of making such abbre- viated descriptions on herbarium sheets or on envelopes containing specimens, and it is entirely probable that Kunth, who visited the Willdenow herbarium in Berlin, [ 82 | had copied this himself, for it seems not to have been published elsewhere. It is quite apparent from a thorough study of the Kunth publication of Siphonia brasiliensis that, although he pointed out in the footnote reproduced above that the Brazilian material differed in several characters from the two Venezuelan collections, Kunth himself considered the three collections to represent the same concept. The description, it is also apparent, was based upon the Vene- zuelan material which, as we now know, definitely does not represent the common cultivated plant but rather a rare species which has, as yet, acquired no commercial importance. A critical examination of Kunth’s treatment discloses an inconspicuous point which seems to have been over- looked and which alters our interpretation of the prob- lem. In their ‘‘Nova genera et species plantarum,” Humboldt, Bonpland and Kunth were accustomed to indicate names which they were publishing as their own for the first time with a small dagger. In the preface (loc. cit. 1 (1816) vi), they state: ‘‘Species et genera nova signo fF indicantur.’’? Siphonia brasiliensis is not marked with a dagger. This, coupled with their citation in synonymy of Willdenow’s Siphonia brasiliensis (which had been written on an herbarium sheet) would seem to indicate that Kunth was publishing an unpublished Willdenow name. ‘This puzzling situation is completely clarified if, remembering the lack of the dagger in Kunth’s publication, we refer to an article by Adr. de Jussieu. It then becomes apparent that Jussieu validly published Willdenow’s Stphonia brasiliensis in 1824, one year before the appearance of Kunth’s description. In his “De Euphorbiacearum generibus. ...’’ (1824) tab. 12, fig. 88b, 1-6, Jussieu published a plate consisting of diag- nostic drawings of the staminate calyx, the stamens with [ 83 ] EXPLANATION OF THE [ILLUSTRATION Piare XVIII. The earliest publication of the name Siphonia brasiliensis in Adr. de Jussieu’s “De Euphorbiacearum generibus. .. .”’ (1824) t. 12, fig. 38b, 1-6. we AD. dS. dal? 36. ALEURITES ambinusr. 4d 7. ANDA GJOMCEM, 38.A. SIPHONIA elastia. B.S. brastonets. 3 é L 6 ‘ a Ve pore seagp' ALVId IIIAX EXPLANATION OF THE ILLUSTRATION Prare XIX. A specimen of the type collection of Hevea brasiliensis preserved in the Paris Herbarium. Piate SX 4 Siphoud, bone i ttehy Wes ~ HEED. MUS. PARI r i/ ppherua , oe , ow a ete at Aweta od Aras lous j pv lleom,') | 255 [oe - ce Dede Miho snentiat, 190) - BS rn oe ; the anthers and suprastaminal column, the _ pistillate flower, the pistil and the ovary of ‘‘Siphonia brasiliensis W. (in herbariis).’” In accordance with Article 44 of the Rules, this constitutes valid publication, for the name of a species is validly published if it be accompanied ‘‘by a plate or figure with analyses showing essential charac- ters; but this applies only to plates or figures published before January 1, 1908.”’ When Mueller transferred the specific epithet from Siphonia to Hevea in 1865, he was probably unaware of the discrepancy between the Orinoco and the Amazon collections—and this in spite of Baillon’s insistence on that point in 1858. In making the new combination, Mueller based it on ‘‘Stphonia brasiliensis Kunth in Humb. et Bonpl. Nov. Gen. et Spec. 7, p. 171°” and cited “*S. Kunthiana Baill.’ as a synonym. It is signi- ficant, however, that the Brazilian material (‘‘In Brasilia paraénsi, Hoffmansegg in hb. Willd. fol. 17986. p. 1°’) was cited before the Venezuelan collections. It is because the author of this combination, apparently unaware of Jussieu’s paper, attributed the earliest publication to Kunth that the author citation of Hevea brasiliensis has been erroneously written as ‘‘(HBK.) Mueller-Argovi- ensis’’ by almost all taxonomists. In summary, we may say that the proper and valid name of the cultivated rubber tree is Hevea brasiliensis (Willd. ex Adr. de Juss.) Mueller-Argoviensis in Linnaea 84 (1865) 204. Its synonymy is as follows: Siphonia brasiliensis Willdenow ex Adr. de Jussieu Euphorb. Gen. (1824) t. 12, pl. 88b., fig. 1-6. Mon Siphonia brasiliensis HBK. Nov. Gen. et Sp. 7 (1825) wee Hevea janeirensis Mueller-Argoviensis in Martius F]. Bras. 11, pt. 2 (1874) 706. Hevea Sieberi Warburg Kautschukpf. (1900) 82, fig. [ 85 |] Hevea Randiana Huber in Bol. Mus. Goeldi 4 (1906) 656. Hevea brasihensis var. stylosa Huber (loe. cit. 4 (1906) 640. Hevea brasihensis var. janeirensis (Muell.-Arg.) Pax in Pflanzenr. 4 (1910) 121. Hevea brasiliensis var. Randiana (Huber) Pax loe. cit. 4 (1910) 123. Hevea brasiliensis mut. Granthami Bartlett in Bot. Gaz. 84 (1927) 200. Hevea Granthami Bartlett loc. cit. 84 (1927) 200, nomen alt. anteriori. Hevea brasiliensis forma typica Ducke in Arch. Jard. Bot. Rio Janeiro 6 (1988) 55. Hevea brasiliensis var. subconcolor Ducke loc. cit. 6 (1988) 55. Hevea brasiliensis torma subconcolor Ducke in Arch. Inst. Biol. Veg. Rio Janeiro 2 (1985) 224. Hevea brasiliensis forma Randiana (Huber) Ducke loc. cit. 2 (1985) 224. Siphonia Ridleyana Cook in Journ. Wash. Acad. Sci. 31 (1941) 46. Siphonia janeirensis (Muell.-Arg.) Cook loc. cit. 31 (1941) 61. The varietal names which follow, included by Seibert (loc. cit. 805-806) as synonyms of Hevea brasiliensis, are here omitted pending additional studies which may pos- sibly indicate that they are distinct concepts. Hevea brasiliensis var. angustifolia Ule in Tropen- pflanz. Beiheft. 6 (1905) 8. Hevea brasiliensis var. latifolia Ule loc. cit. 6 (1905) 8. Hevea brasiliensis var. acreana Ule in Engler Bot. Jahrb. 50 (1914) 14. For his friendly counsel in this as in many other prob- lems which have arisen in my botanical research, I am deeply grateful to the late Mr. Charles A. Weatherby of the Gray Herbarium of Harvard University. [ 86 ] THE IDENTITY OF UCUQUI BY Joao Murea Pires' and Ricuarp Evans ScHULTES?’ ONE of the results of recent field work in the upper Rio Negro basin of Brazil has been the identification of a useful plant of that area—the ueuqui. The fruit of this tree has an edible and delicious mesocarp and is an impor- tant part of the diet of the native peoples of the region. Investigation has shown that the uweuqui is an unde- scribed species of the sapotaceous genus Pouteria. It is altogether fitting that, in publishing a description of this food plant, we employ as a specific epithet the common name which refers exclusively to this species over the greater part of its range. Pouteria Ucuqui is immediately set apart from all other species of the genus by the excessively developed disk which surrounds the ovary. Pouteria Ucuqui Pires & Schultes sp. nov. Arbor enormis, usque ad centum viginti pedes alta, radicibus tabularibus, trunco columnari usque ad _ tres pedes in diametro, cortice crasso, molli, extus atrobadio ‘Chief, Section of Biology, Instituto Agronémico do Norte, Belém do Para, Brazil. Botanist, Bureau of Plant Industry, Soils, and Agricultural Engi- neering, Agricultural Research Administration, U.S. Department of Agriculture; Research Fellow, Botanical Museum, Harvard Univer- sity; Collaborator, Instituto Agronémico do Norte. [ 87 ] et intus sanguineo cum latice albo aquosoque. Ramuli juniores, inflorescentiae, petioli et foliorum nervi indu- mento ferrugineo-pulverulento vel ferrugineo-furfuraceo obtecti. Folia alterna, bene coriacea, elliptica, basi et apice acuta vel obtusiuscula, plerumque cum acumine 7-10 mm. longo, margine integra, 11-20 ecm. longa, 5.5-10 em. lata, supra nitidula, subtus opaca (statu bene juvenili ferruginea), minutissime ferrugineo-pilosula; costa leviter striata, supra prominens, subtus valde elevata et robusta; nervi secundarii utrinque octo ad quindecim, supra leviter insculpti, subtus valdissime prominentes ; venulae superficie utraque paulo impressae. Petiolus elongatus, siccitate striatus et leviter rugosus, saepissime 3.5-5.5 em. longus, 2-83 mm. in diametro. Inflorescen- tiae axillares, valdissime congestae, fasciculatae. Flores mordaciter fragrantes, flavo-virides, gemmationis statu longe persistentes sed flores ipsi fugaces. Sepala quin- que, imbricata, subrotundata, plerumque 2 mm. longa, utrinque minutissime adpresso-pilosa, intus demum gla- brescentia. Petala quinque, per + vel $ longitudinis par- tem connata, imbricata, oblonga, apice acutiuscula vel obtusa, margine integra, plerumque 3 mm. longa, 1 mm. lata, extus pilosiuscula sed demum glabrescentia, intus subglabra, tubo vulgo 0.5-1 mm. longo. Stamina quin- que, petalis opposita, filamentis vix 1 mm. longis; an- therae apice acutae, basi rotundatae, connectivo mediano crasso, loculis duobus, sublinearibus, rimis oppositis, de- hiscentibus. Staminodia petalis alterna, apice acuta, 1 mm. longa vel minora, aliquando inconspicua. Ovarium biloculare, minutissimum, pilosum, disco valde hispido, circiter 0.5 mm. alto circumdatum; stylus circiter 1.2 mm. longus, pilosus vel glabrescens, cum stigmatibus inconspicuis. Fructus maximus, 9-13 cm. longus, 5-7 cm. in diametro, monospermus; semen magnum, pler- umque 7-10 cm. longum, 3.5 cm. latum, 3 em. in di- [ 88 | ametro, cum testa crustacea, nitida, plusminusve 2 mm. crassa, area umbilicali magna. Arbor a regionis typicae brasiliensis incolis weuqué (in lingua nheenagtti), puch-pee-d (in lingua tukanorum tri- bus); a regionis colombianae incolis 06-le-da (in lingua kuripakanorum tribus), /a-he-pa (in lingua mirahorum tribus) appellatur. CoLLECTIONS EXAMINED: Brazit: Estado do Amazonas, Upper Rio Negro, Igarapé Uaba, opposite mouth of Rio Xié. ‘‘Immense tree, 110 feet tall, with wide- spreading crown. Large buttress roots up to 6 feet. Tree columnar, 24 to 3 feet in diameter. Bark dark brown, shaggy externally, red internally, soft, }—? inch thick. Latex sparse, white. Wood of medium hardness, compact, white. Flowers in extremely dense clusters, fall- ing easily. Sepals greenish; petals white; anthers brown, apparently odourless. Common name: ucuqut.’’ January 5, 1948, Richard Evans Schultes & Francisco Lopez 9553 (Tyre in Herb. Gray; Dupticatr ryprs in Herb. Inst. Agron. Norte (Belém do Parad); U.S. Nat. Herb. ; Kew; Econ. Herb. Oakes Ames).—Rio Issana basin, between Rio Aiari and Rio Caiari (Uaupés), near Serra Tunui. “‘Arvore 25 m. 40 em. Frutos estimados como alimento indigena. Ucuqut.’? October 13, 1945, Ricardo de Lemos Frées 21388.—Rio Negro, near Uaupés (Sao Gabriel), Serra de Uanari ‘‘Ucuqui. Arvore muito grande. Latex branco, pegajoso, pagina inferior da folha revestida por leve indumento escamoso ferrugineo, fruto amarelado.’’ October 26, 1947, Joao Murca Pires 773.—Same locality. ** Ucuqué. Arvore muito grande. Latex branco, pegajoso, pagina inferior da folha revestida por leve indumento escamoso ferrugineo; fruto amarelo.’’? October 30, 1947, Joao Mur¢a Pires 799.—Same locality. *‘ Ucuqut. Arvore muito grande no alto da serra. Floras quasi brancas (botoes).’? November 17, 1947, Joao Mur¢a Pires 1165.—Rio Negro, near confluence with Rio Uaupés. ‘‘ Ucuqut. Arvore muito grande. Madeira dura. Latex branco, pegajoso. Pagina inferior da folha revestida por leve indumento escamoso, ferrugineo, amarelado.’’ November 5, 1947, Joao Murga Pires 833,— Rio Uaupés, Serra Uapici. ‘‘Ueuqué. Arvore grande.’’ November 17, 1947, Joao Murca Pires 1150.—Rio Negro, near Uaupés (Sao Gabriel). ** Ucuqui. Arvore platinhas novas, colhidas sob a arvore.’’ October 30, 1947, Joao Murga Pires 792.—Rio Uaupés, Ipanoré **Puh-pid. Arvore grande.’’ November 15, 1947, Joao Murca Pires 1057.—Rio Uaupés, Taracua. [ 89 ] **Ucuqui, (lingua geral), Puh-pid (Tukano), November 8, 1947, Joao Murca Pires 905.—Rio Uaupés, Serra Uapici. ** Ucuqui. Arvore grande.’’ November 17, 1947, Joao Murca Pires 1149.—Rio Negro, Sao Gabriel. “‘Coqui. Enormous tree. Buds brownish. Fruit edible.”’ November 20-25 1947, Richard Evans Schultes & Francisco Lépez 9184. —Rio Negro, Sao Gabriel. “‘Ucuquéi. Enormous tree.’’ January 14, 1948, Richard Evans Schultes & Francisco Lopez 9618.—Middle Rio Negro basin, Rio Curicuriari. “‘Ucuqui.’’? Richard Evans Schultes & Francisco Lopez 9710.—Rio Negro, Ipanoré, Caatinga forest beyond town. ‘‘Coqut. Enormous tree. Fruit edible.”” November14—15, 1947, Richard Evans Schultes & Joao Murca Pires 9096. Cotomsta : Comisaria del Vaupés, Rio Negro, opposite Piedra del Co- cuy. ‘Enormous buttressed tree, 110 feet tall, diameter 3 feet. Flowers greenish yellow. Bark shaggy, brown. Fruit edible.’’ December 28, 1947, Richard Evans Schultes & Francisco Lépez 9484.—Rio Guainia, near Sejal. ““Tree 100 feet tall, buttressed. Kuripaka name: 06-/e-da.”’ June 1948, Richard Evans Schultes & Francisco Lépez 10058a.—Com- isaria del Amazonas, Rio Caqueté, near La Pedrera. ‘* Enormous tree. Fruit edible. Mirafia name: kd-he-pa.’’ July 15, 1948, Richard Evans Schultes & Francisco Lopez 10215, VenezurLa: Territorio del Amazonas, Rio Negro, base of Piedra del Cocuy. ““Enormous tree 120 feet tall, 24 feet in diameter, slightly buttressed. Bark brown, shaggy, soft, scarlet within. Latex sparse, white. Buds greenish yellow.’’ December 24, 1947, Richard Evans Schultes & Francisco Lopez 9458. Pouteria Ucuqui is very common in the northwestern- most part of the Brazilian State of Amazonas along the upper Rio Negro, from 'Tapurucuara (Santa Isabel) up- stream, and along its affuents: the Uaupés, Issana, Tikié, Curicuriari, Dimiti, and probably many others. It is known to occur in Colombia in the Rios Guainia (constituting the source of the Rio Negro) and on the Rio Caqueta. It has also been found on the Japura in Brazilian territory and has been reported from the Rio Solimoes (Le Cointe, P. ‘‘Arvores e plantas uteis’’ (1934) 457). According to reliable reports, it occurs in certain places far into Venezuelan territory along the lower course of the Rio Guainia and on the Casiquiare itself. [ 90 | In this entire area, the plant—an enormous, heavily crowned tree which fruits profusely once a year—is found in abundance in the virgin forest on high, well drained soil. [t is also often encountered in a state of apparent cultivation, since, in clearing for house sites, the tree has been spared. When completely ripe, the thick, fleshy mesocarp of the fruit, which resembles that of the avo- cado (Persea americana Mill.) to a striking degree, is very palatable. When green, however, it is full of latex and is extremely sticky. Pouteria Ucuqui is known in the entire Brazilian part of its range by the name weugu?, a word originating from the Lingua Geral or Nheengatti language which is spoken widely in the State of Amazonas, and especially in the Rio Negro. The Tukano Indians of the Rio Uaupés re- fer to the tree as puch-pee-d (the ch being soft as in the German ?ch). In Colombia, the native name for Pouwteria Ucuqui amongst the Kuripaka Indians of the Rio Guainia is 00-le-da; and amongst the Miranias of the Rio Caqueta (La Pedrera), hd-he-pa. The tree is called yuew by the Spanish-speaking population of the Venezuelan town of San Carlos on the uppermost Rio Negro. The term weuqu?, so far as we have been able to ascer- tain, refers exclusively to Pouterta Ucuqui and should not be confused with weuquirana (i.e., ‘false weuqut’’ ), a name widely applied in the same area to the sapotaceous Ecclinusa sanguinolenta Pierre (EH. Balata Ducke)—also ‘alled abiwrana—a tree which is actively exploited as the source of a type of balata. The two trees are completely distinct from all points of view. Pouteria Ucuqut is a very tall and robust tree, usually with comparatively large buttresses at the base; the low- est branches are at a great height from the ground. These conditions make the study and collection of herbarium material rather difficult, especially since the flowers, [ 91 ] which are borne on the branchlets, are very minute and ‘annot be seen from the ground, even with binoculars. Dr. Adolpho Ducke, who has visited the upper Rio Negro basin several times, was unable to collect flowering material of weuqui, although he devoted special attention to this tree which he believed represented an undescribed species. Without flowering material, however, there was some question as to even its generic affinities. When we began our work in the upper Rio Negro in late 1947, Dr. Ducke counseled us to try to find flower- ing material which would settle definitively the identity of ueugui. With this in mind, we studied a number of individuals over a wide area. It was almost always neces- sary to use the balatero’s climbing-irons (employed dur- ing the extraction of balata from species of Manilhara), since the girth of the tree usually prevented the use of the ‘‘peconha”’ (a band of pounded bark which the In- dians place on the feet to aid them in climbing). The expertness of our assistant, the late Francisco L6pez, in the use of the climbing-irons greatly increased the num- ber of trees which we could study. During October and November, this work was carried on principally by Murea Pires. At first, sterile specimens only were obtainable. Then, gradually, we began to se- cure material which was in bud. No flowers, however, were found, and we later learned that the species develops the flower with extreme slowness, persisting in bud some- times for three or four months. Many experiments were tried to force the buds to open: sun, artificial heat, soak- ing in hot and warm water; but all these efforts failed. It is interesting in this connexion to note how little the natives know about the life-history of this species, even though it is one of their common and useful plants. Repeated questioning on our part brought forth the most divergent and amusing remarks concerning the flowering [ 92 ] of weuqut. Many, when asked during what month the tree flowers, replied that they had never seen it flower and that, therefore, it did not flower. Others insisted that the tree blooms during the night and immediately drops all the flowers. This second ‘‘explanation”’ is, in a way, ingenious, because, as we later witnessed, the forest floor under an weuqui tree, where there had even the day previously been no indication of blossoms, would sud- denly be covered with literally hundreds of thousands of flowers or parts of flowers. Our studies showed that, in spite of the very long period of aestivation, the flowers are almost ephemerous in Pouteria Ucuqut. During the last few months of 1947, Murea Pires revisited trees in bud several times in an attempt to col- lect fertile specimens. When he returned to Belém, Schultes and Lopez continued the search. In late De- cember, it was noted that the buds on a number of trees previously examined were swelling rapidly, in spite of the fact that for at least a month previous there had been no appreciable alteration in their size. On January 5, we found our first flowering tree. It was very early in the day, about six o’clock in the morn- ing, when our attention was called to the tree by the patter of falling flowers. Upon climbing the tree, we noted a very strong aromatic and pungent odor. It is significant to note that the buds never had a fragrance. There were innumerable large bees and several other kinds of smaller insects visiting the flowers even at this early hour. Later, we encountered other trees in flower, but the pronounced odor was not noted. It is possible that the strong, aromatic fragrance is given off only for a short period during the very limited blossoming time. As accurately as we could calculate, open flowers per- sist for no longer than three days on an individual tree and usually are much shorter lived. [ 93 | EXPLANATION OF THE ILLUSTRATION PLrate XX. Pourerta Ucugut Pires & Schultes. A, flowering branch, about one half natural size. B, fruit. C, seed. D, flower. E, dissection and view of interior of corolla. F, dissection of flower show- ing disk around ovary. G, dissection of flower showing pilosity around ovary. H, anther. Each drawing has a millimeter scale to indicate magni- fication. Drawn by Perera Fitno Qf Instituto Agronomico do Norte PRATE XX POUTERIA EXPLANATION OF THE ILLUSTRATION Prare XXII.) Pourerta Ucugut Pires & Schultes. Basal portion of trunk of type tree, showing the buttress roots. Photograph by R. EB. Scuvurres 4 | ’ ‘ PLATE Martyn, having corrected Miller’s ortho- graphic error, united two names for the Old World con- cept correctly under the older name Cassine Peragua. But he then proceeded to misapply this name to the New World concept which we now know as Ilex vomi- toria which is ‘‘native of Virginia and Carolina.’’ In con- sideration of these circumstances, then, Miller's specific epithet Paragua cannot be available for transfer to Zea. The confused use of the name I/ex Cassine, which has likewise hampered our understanding of J/ea vomitoria, is attributable to the unfortunate fact that this binomial has been published by two authors for different concepts. The earlier lev Cassine Linnaeus (Sp. Pl. (1753) 125) was an unfortunately chosen name, for while the bi- nomial itself refers without any doubt to the dahoon of the southeastern United States, the specific epithet was taken from one of the common names of the yaupon holly and applied to the other holly by Linnaeus in er- ror. Ilex Cassine L., therefore, is the accepted technical name for the dahoon holly. A few years later, Walter (FI. Carol. (1788) 241) pub- lished Ilew Cassine to refer to the concept we now call I. vomitoria, as ascertained by reference to a photograph [ 103 ] of the type specimen (Schubert [photographer]: ‘* The Herbarium of Thomas Walter at the British Museum (Natural History), South Kensington, London, Eng- land,’ (1946-7) I/eaw Cassine [a bound volume of photo- graphs in the Gray Herbarium]). Although Walter’s specific epithet is correctly applied to the plant which is widely known by that common name, and although it antedates Ilex vomitoria by one year, the name falls because it is a homonym. This dual use of the same binomial has, as Loesener (in Bot. Centralbl. 47 (1891) 161-168) pointed out, led to contusion, although even arapid examination of the case would have made clear all points of uncertainty. E. M. Hale’s very excellent historical treatment of the black- drink plant (‘‘Z/ew Cassine, the aboriginal North Ameri- can tea’’: Bull. No. 14, U.S.D.A. Div. Bot. (1891)), for example, suffers from his use of the name Ilea Cassine without even one reference to indicate whether the bi- nomial of Linnaeus or that of Walter is being accepted ; Hartwich’s monumental work on narcotic plants (‘‘Die Menschlichen Genussmittel’’ (1911) 468) uses [lea Cas- sine L.: and Wehmer (‘‘Die Pflanzenstoffe’’ (1911) 456 ; [ed. 2] 2 (1981) 718) accepts I. Cassine Walt. The first chemical report (Smith, H. M.: in Am. Journ. Pharm. 44 (1872) 216) of the discovery of caffeine in a North American Jlex is of doubtful value partly because of an uncertainty as to which Ilew Cassine was being examined. In 1888, still another report of a chemical analysis (Ven- able, F. R.: in Chem. News 52 (1885) 172) employed the name J/ex Cassine without author citation, thus leav- ing room for uncertainty as to the exact species which was analysed. An even more interesting result of the confusion of the L/ew Cassine of Linnaeus with that of Walter is found in a standard pharmacological work (Tschirch, A.: ‘‘Handbuch der Pharmacognosie”’ 38, [ 104 J Abt. 1 (1928) 482), in which its writer ‘‘affirms and denies the existence of caffeine in I/ex Cassine all in the same paragraph’’ (Maxwell, M. M.: ‘‘Tlea Cassine L. ... 7’ ms. report (no date) in Libr. Econ. Bot. Harvard Univ. ). Loesener (in Bot. Centralbl. 47 (1891) 168) rejects Ilex vomitoria in favor of Ilex caroliniana (Lam.) Loes., a combination which he made on the basis of Lamarck’s Cassine caroliniana. This procedure is not permissible. Lamarck’s binomial Cassine caroliniana (1788) is not available, since it is a substitute for the validly published Cassine corymbosa Miller (Dict., [ed. 8] (1768) Cassine No. 1; [cones (1760) 88, t. 1). It might be pointed out in passing that Miller’s Cassine corymbosa does not reter to the concept Ilex vomitoria. The other synonyms of J/ew vomitoria have not, for the most part, entered into the chemical, pharmacologi- ‘al and anthropological literature. They are included in the foregoing enumeration for the sake of completeness. Several of the Rafinesque names are referred to Ilex vomitoria with some reserve. They are not based on specimens which might be examined, and the descrip- tions are brief and vague in the extreme. It is fortunate that they do not need to be considered nomenclatorially in determining the correct name of the yaupon, but, be- ing later, can be relegated immediately to synonymy. [ 105 | .o A NEW VARIETY OF ILEX VOMITORIA FROM SOUTHERN MEXICO’ BY A. J. SHARP Ilex vomitoria [ Soland. in] Aiton var. chiapensis A. J. Sharp var. nov. Haec variatas ab Ice vomitoria principaliter foliis pu- bescentibus et ramulorum villis longioribus differt. From the species, this new variety differs primarily in having loosely pubescent leaves and longer (about twice as long) hairs on the branchlets. There is a tendency for the leaves to be a little more ovate and obtuse at the apices, although many leaves match in shape those from North American material of the species. The material of Ilex vomitoria in the Gray Herbarium (all from south- eastern United States) has been examined, and the leaves are glabrous, except for occasional puberulence on the mid-rib at the base. This report adds another species to the growing list of plants which occur both in central or southern Mexico and in southeastern United States. In Mexico, their ranges are disjunct from their areas in the United States, and the largest number of such species seems to be found in the highlands toward the Gulf of Mexico.