BOTANICAL MUSEUM LEAFLETS

HARVARD UNIVERSITY

PRINTED AND PUBLISHED AT THE

BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS

BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY

VOLUME XVIII

BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS 1957-1959

TABLE OF CONTENTS

NuMBER 1 (June 7, 1957)

The Identity of the Malpighiaceous Narcotics of South America By Ricuarp Evans SCHULTES. ....... 1

NuMBER 2 (June 26, 1957) The Effects of Certain Genes on the Outer Pistil- late Glume of Maize By WALTON C. GALINAP . Joe 4 «cu a « « BT Number 8 (January 27, 1958) Orchidaceae Neotropicales IV

By Ricnarp Evans ScCHULTES . afi Studies in American Orchids III

By Leste A. GARAY .... +. «s+. « » 108 A New Lepanthopsis from Venezuela

By CuHarRLeEs SCHWEINFURTH ........ . 109

NumBeEr 4 (April 4, 1958)

Plantae Austro-Americanae X

By Ricuarp Evans SCHULTES. ...... .118

NuMBER 5 (November 20, 1958)

How Absurd Can a Nomenclatural Proposal Be? By Lesuige A. Garay AND RicHarp Evans SOUDINGS . <4). Goel. se ew ee

[v ]

Studies in American Orchids IV

By Lesuizg A. Garay ........ . . . 186 Two Additions to the Orchid Flora of Venezuela By CHarLes SCHWEINFURTH .. . . . . . . 219

A Little-Known Cultivated Plant from Northern South America By Ricuarp Evans Scuuures. . . . . . . . 299

Number 6 (January 16, 1959)

New Considerations in an Old Genus: Datura By Arrnur S. Barchay ....... . . . 245

The Present Distribution of Narcotics and Stimu- lants Amongst the Indian Tribes of Colombia By Nesror Uscatrecur M. ..... . . . . 278

Plantae Colombianae X V By Ricuarp Evans ScHuutrEs. ..... . . 805

On the Validity of the Generic Name Cochleanthes Raf. By Ricnarp Evans Scuutres AND LesuiE A. GARAY ............2.2. 2.2. .821

NuMBER 7 (February 20, 1959)

The Origin of Corn. I. Pod Corn, the Ancestral Form By Paun C. MaNGELsporr AND Rosertr G. REEVES. 2. 2...) B29

NuMBER 8 (March 10, 1959)

The Origin of Corn. II. Teosinte, a Hybrid of Corn and Tripsacum By Roperr G. Reeves anp Patt C. MANGELSDORF .

os It ~

[ vi ]

Numser 9 (April 8, 1959) The Origin of Corn. III. Modern Races, the Prod-

uct of Teosinte Introgression By Paut C. MANGELSDORF AND RosBert G. Wve faeces « & & «ahs (See BS BS

Numser 10 (April 17, 1959)

The Origin of Corn. IV. Place and Time of Origin By Paut C. ManGELSDORF AND Rospert G. REEVES. ...... ee ee ee ee ee ALB

The Origin of Corn. V. A Critique of Current Theories By Ropert G. REEVES anpD Paut C. MANGELSDORF ........ +... + « 428

[ vii ]

INDEX OF ILLUSTRATIONS

PLATE Anthurium apaporanum PR. 17. Schultes. 2 2 . XIX

Anthurium polyschistum RB. 2. Schultes & Idrobo LX

Aristolochia georgica R. 7. Schultes . . . 2. XXVI Aristolochia xerophytica R. 7. Schultes . . XXVII

Banisteriopsis Caapi (Spruce ex Griseb.) Morton II, [V, V, V1]

Banisteriopsis inebrians Morton . . . . . . IDL, VI Banisteriopsis Rusbyana (Ndz.) Morton . . . . VII Bifrenaria minuta Garay... . . . OX Calliandra vaupesiana Cowan . . . . . . . XXIX

Caularthron bicornutum (/Zo0ok.) Rafinesque X11,X 111

Caularthron bilamellatum (Pehb.f.) R. Ee. Schultes XIV, XV Conomorpha lithophyta R. 2. Schultes XXXI, XXXII, X XXIII

Cyclanthus indivisus R. H. Schultes . . LVI, LIX Datura inoxia Miller 2...) CI Datura kymatocarpa Barclay . . . . . . LIL, LIV Datura meteloides DC. ew Dunal . . . . . . LL, Ll

[ vill |

Datura reburra Barclay... .. . . . LILI, LIV

Datura vuleanicola Barclay... ... . LV, LVI Duckeella pauciflora Garay. . . . . . . . XXXIV Epidendrum superpositum Garay . . . . XXXVIII Hexadesmia Dunstervillei Garay . . . . XXXIX Inga involucrata Cowan . ....... XXVIII Inga longifolia Cowan . . . ...... XXVIII Jacqueshuberia amplifolia Cowan . . . . . XXIX Lepanthopsis vinacea C. Schweinf... . . . . XVII Lorostemon colombianum Maquire ..... XXX Makuna witch-doctor ........... =VIII Native tribes of Colombia who use stimulants

and narcotics(Map) ........... LVII Notylia Norae Garay ........... XLII Octomeria deltoglossa Garay . . . . . . XXXVII Oncidium vagans C. Schweinf. . . . . . . . XLV Philodendron colombianum R. #7. Schultes . . . XX Platystele ornata Garay .........XXXVI Polyotidium Heubneri (Mansf.) Garay . . . XVI Rhytidanthera regalis R. H. Schultes. . . . . LXII Rodriguezia Teuscheri Garay... ..... XULI

[ ix ]

Solanum Topiro HBK. Adv, AVI, RLVIIL, XIX

Stelis nexipous Garay... .. ..... XXXV

Stenospermatium verecundum R. L2. Schultes XXI, XXII

Telipogon Teuscheri Garay... .. . . XLII Tribes reported to use malpighiaceous narcotics (Map) I Urospatha antisylleptica R. WH. Schultes. . . . LXI

Urospatha somnolenta R. 7. Schultes XXIII, XXIV, XXV

Vargasiella venezuelana C. Schweinf. . . . . XLIV Zamia madida R. H. Schultes . . 2 2...) XVII Zea Mays L. Outer pistillate glumes . . IX, X, XI

[x ]

INDEX TO GENERA AND SPECIES

AERANTHUS Lehmanni Rehb.f., 209,212

ALTERNANTHERA Lehmanii Hieron., 37,42,289

ambil, 289,291,293 ambird, 280,281,283 ambire, 280 AMMOPHILA arundinacea, 69

Andropogoneae, 57,67 ANTHOSIPHON Schltr., 214

ANTHURIUM

apaporanum R, E.Schultes, 115,118,120

eminens Schott, 311

fontoides R. E.Schultes, 121

gracile (Rudge) Engl., 121

Harrisii (Grah.) Engl., 120

nanum FR. E.Schultes, 120

napaeum Engl., 120

polyschistum R.E.Schultes & Idrobo, 310,311,312

ApocyNacEAk, 15,39,42,113, 168,317

ARACEAE, 115,310,316

ARETHUSA

sinensis Rolfe, 184 japonica A.Gray, 185

ARISTOLOCHIA

cornuta Mast., 138

georgica R.E.Schultes, 125, 138,139,140

iquitensis O.C.Schmidt, 139

rumicifolia Mart. & Zucc.,142

sect. Exstipulatae, 138

sect. Pseudostipulosae, 139

sp., 18,42

xerophytica R. E. Schultes, 139,142,144

ARISTOLOCHIACEAR, 18,125 AVENA, 72

ayahuasca, 1,4—7,10,11,14-25, 30, 34,36,41,42, 290,292,293

ayawasca, 15,19,20

BACTRIS Sp., 291

BANISTERIA, 15,22,30,31 Caapi Spruce ex Griseb., 415 Jerruginea, 21 inebrians, 31 lutea Ruiz ex Griseb., 25

BANISTERIOPSIS, 7,16,20, 21,23,24,30,31,35,37,38,40, 41,42, 294 Caapi (Spruce ex Griseb.) Mor- ton, 4,5,7,8,10,11,14-17, 19-26,28, 30-32, 34,37-41 44,48,278

inebrians Morton, 12,21,22, 25,30, 35-37,41,45,46,278 289,290

longialata Ruis ex Ndz., 22, 30

metallicolor (4. Juss.) O°’ Don- nell & Lourteig, 24,25

quitensis (Ndz.) Morton, 19- 22,25,36,37,41,289

Rusbyanae (Ndz.) Morton, 22, 30,35,38,41,45,46,278, 289,290

spp., 230,292,296

barley mutants, 65 bejuco de oro, 36 be-td-ka, 242

BIFRENARIA, 208 grandis (Kriinzl. ) Garay, 208 minuta Garay, 206,207 picta (Schlir.) Schweinf., 208 sect. Lindleyella, 208

BLETIA, 182,163 florida R. Br., 185 Jormosana Hayata, 183 gebina Lindl., 185 hyacinthina (Sm.) R.Br., 185 kotoensis Hayata, 183 morrisonicola Hayata, 184

BLETILLA, 182,183

chinensis Schltr., 184

florida Rehb.f., 183,185

JSormosana (Hayata) Sehltr., 183

gebinae (Lindl.) Rehb.f., 183, 185

japonica (A.Gray) Schltr.,185

kotoensis (Hayata) Schltr., 183

morrisonicola (Hayata) Schltr. 184

ochracea Schltr., 184

scopulorum (W.W.Sm. ) Schltr., 184

sinensis (Rolfe) Schltr., 184 striata (Thunb.) Rehb.f., 185 sselschuanica Schltr., 185 yunnanensis Schltr., 185

var, Limprichtii Schltr. ,185

bo-po, 242 BOLLEA Wendlandiana Hort., 323 borrachero, 25,286,287 borrachera, 37,279,285 BROMUS, 72 Burseraceak, 156 caapi, 1,4-6,11,14—-16, 18,19, 21-25,30,39-42,293,295,298 caapi-pinima, 15,40,41 cagna, 16 CALANTHE gebinae (Lindl.) Lodd., 185

CALLIANDRA, 113,142,143

vaupesiana Cowan, 142,154

CAMARIDIUM, 208 equitans Schltr., 208

CAMPYLOCENTRUM Lehmannii (Rehb.f.) Schltr. , 212

Capanemieae, 104 capi, 6

CAPSICUM frutescens L., 230

CASSIA, 113,142 gigantifolia (Britt. & Killip) Cowan, 146

CAULARTHRON, 77 81,86,102

[ xii |

bicornutum (Hook.) Raf., 82, 84, 86-90

bilamellatum (Rchb.f) R.E. Schultes, 86,88,89,90,92, 94,96-98, 100

umbellatum Raft, 80

CECROPIA peltata L., 292 spp., 292

CENCHRUS setigerus Vahl., 380

CENTRONIA laurifolia D.Don, 161 reticulata T7'riana, 160,161 vaupesana Wurdack, 160,161

chaco, 16 chagro panga, 30,35,36,46 cha-pe-nas, 291

CHAMAEDOREA sp., 291

CHAMAEFISTULA gigantifolia, Britt. & Killip, 146 chicha, 37 chimo, 280

CHONDRORHYNCHA, 321

aromatica (Rchb.f.) P.H. Allen, 323

calloglossa (Schltr.) P.H. Allen, 327

discolor (Lindl.) P.H. Allen, 324

Lipsicombiae Rolfe, 326

marginata (Rehb.f) P.H. Allen, 326

CLADONIA, 122

CLIBADIUM asperum (Auwbl.) DC., 230

coca, 229,278,280-288,29 1-299

COCHLEANTHES, 321,322 amazonica (Rchb.f. & Warsc.) R.E.Schultes & Garay, 322 aromatica (Rchb.f.) R.E. Schultes § Garay, 323 bidentata (Ichb,f. ex Hemsl.) R.E.Schultes §& Garay, 323 candida (Lindl.) R. E. Schultes & Garay, 323 digitata (Lem.) R. E.Schultes & Garay, 324 discolor (Lindl.) R. E. Schultes & Garay, 324 flabelliformis (Sw.) R.E. Schultes & Garay, 324 Jragrans Raf., 322,324 heteroclita (Poepp. & Endl.) R.E.Schultes & Garay, 325 ionoleuca (Ichb.f.) R.E. Schultes § Garay, 325 Klugii (C. Schweinf.) R.E. Schultes & Garay, 325 Lueddemanniana (Ichb.f.) R.E.Schultes & Garay, 326 marginata (Rehbf.) RE. Schultes § Garay, 326 trinitatis (4mes) R.E.Schultes & Garay, 326 Wailesiana ( Lindl.) R.E. Schultes § Garay, 326

cocona, 231,232,233,242

corn archaeological, 343-347 ,377-— 379,396-400,422—424 corn grass, 337,338, 352,430- 431

{ xii |

fossil pollen, 378,419—420, 423 modern races, 389-406 origin of, 329-440 current theories, 428-437 place, 413-423 time, 423-424 papyrescent, 339,340,352, 429-430 pod, 329-353 prehistoric, 343-347 reconstructing ancestor of, 349-351

CREPIS, 3865 sect. Ixeridopsis, 365 sect. Pyrimachos, 365

Cryptocentreae Garay, 213

CRYPTOCENTRUM, 212, Jamesonii Benth., 212 Lehmannii (Rehbf.) Garay,

209 pseudobulbosum Schweinf, 212

CUCURBITA Pepo, 282

CycaDACcKAR, 114 CYCLANTHACKAR, 305

CYCLANTHUS bipartitus Poir, 310 indivisus R. EF. Schultes, 305, 306,308,310

CYMBIDIUM flabelliforme (Sw.) Sw., 324 Alabellyfolium Sw. ex Griseb., 325 hyacinthinum Sm., 184 striatum (Thunb.) Sw., 184

DATURA, 245,285,286,289

arborea L., 22,279

candida ( Pers.) Safford, 279, 287

discolor Bernh., 259,260

dolichocarpa (Lagerh. ) Safford, 279,287

inoxia Miller, 245,247,252,

254,255

kymatocarpa Barclay, 256, 262,264

meteloides DC, ex Dunal, 245- 248 250,254,255

reburra Barclay, 258-260, 262,264

sanguinea Ruiz & Pav., 270, 271,279,287

sect. Stramonium, 260

sect. Dutra, 260

sp., 6,7,10,42

suaveolens H. & B, ex Willd., 25,279,285

vuleanicola Barclay, 260,266, 268,270,271

Wrighttt Regel, 246,247,254

dd-pee-ka-he, 168 de-twa, 242

DIACRIUM Benth., 77,79,80 amazonicum Schltr., 87,90 bicornutum (Hook.) Benth., 78,87,93,96

bidentatum (Lindl.) Hemsl., 102

bigibberosum (Rchb.f.) Hemsl. 92

bilamellatum (Rehb.f.) Hemsl. 93 var. Reichenbachianum

Schltr., 93 bivalvatulum Schltr., 93,100

[ xiv ]

indivisum (Bradf. ex Griseb. ) Broadway, 78,93,96,97

Ulmekei Krinzl., 102

venezuelanum Schltr., 93,100

djerabe, 291

DUCKEELLA Adolfii Campos Porto & Brade, 187 pauciflora Garay, 186-188 ELYONURUS tripsacoides, 67,72 EKPIDENDRUM, 77-81 bicornutum Hook., 73-80,86, 96,97 bidentatum (Lindl.) Hemsl., 102 bigibberosum Rehb.f., 92,96 bilamellatum Rchb.f. ,92,93,96 Boothianum Lindl., 102 flabelliforme Sw., 324 indivisum Bradf. ex Griseb., 92,96 obliquum Schiltr., 203 sect. Diacrium Lindl., 77,80 soratae Rchb. fi, 203 Steinbachii Ames, 203 stenopetalum Hook., 80

striatum (Thunb.) Thunb., 184

superpositum Garay, 203,204 tuberosum Lour., 184

ERIOCAULACEAE, 124

ERIOCAULON, 113 vaupesense Moldenke, 124,125 ERIANTHUS spp., 57 ERYTHRODES

cylindrostachys Garay, 190, 191

quadrata Garay, 192 xystophylla (Rchb f.) Ames, 194 xystophylloides Garay, 193 zeuxinoides (Schiltr.) Ames, 193 ERYTHROXYLON Coca Lam., 229,278 novogranatense (Morris) Hieron., 278 EUCHLAENA, 57,379-381 EULOPHIA cochleata Knight ex Hooker, 324

cochlearis (Lindl.) Steudel, 324

Jloripondio 25 GOSSYPIUM, 368 grano turco, 419 guanguala, 282 guanguara, 282 guanto, 6 guay-ee-ga-m0-yoo-ke-ree, 39 guayaroche, 16 GUTTIFERAE, 113,158 GYAS

humilis Salisb., 184

HACKELOCHLOA spp., 57 HAEMADICTYON, 15

amasonicum Bentham, 14,15, 40,42

hayawasca, 19

hayo, 280,281

[ xv ]

HERRANIA nitida (Poepp.) R.E.Schultes, 230 HEXADESMIA Dunstervillei Garay, 203, 205,206 Jiminezii Schltr., 206

hi(d)-yati(d)yahe, 11

HOEHNEELLA Gehrtiana (Hoehne) Ruschi, 327 HUNTLEYA candida Hort. ex Lindl., 323 imbricata Hort. ex Rehb.f., 325 imbricata Pinel ex Lem., 324 marginata Hort. ex Rehb.f., $26 Huntleyinae, 321

HYBOCHILUS, 103 Huebneri Mansf., 1038-105, 108 tahi, 10 INGA, 113,142 inflata Ducke, 147 involucrata Cowan, 146,148 longifolia Cowan, 147,148 strigillosa, 150 IXERIS, 365 alpicola, 365 JACQUESHUBERIA, 113,142 amplifoliola Cowan, 150,154 quinquangulata, 151 JIMENSIA, 182,183 formosana (Hayata) Garay & R.E.Schultes, 183 kotoensis (Hayata) Garay & R.E.Schultes, 183

morrisonicola (Hayata) Garay & R.E.Schultes, 183

nervosa Rafi, 185

ochracea (Schltr.) Garay & R.E.Schultes, 184

scopulorum (W,W.Sm.) Garay & R.E.Schultes, 184

sinensis (Rolfe) Garay & R.E.Schultes, 184

striata (Thunb.) Garay & R.E.Schultes, 184

szetschuanica (Schiltr.) Garay & R.E.Schultes, 185

yunnanensis (Schltr.) Garay & R.E.Schultes, 185

kaapi, 11,18

kahée, 38

kahi, 24

kali, 11

katyja, 24

LACAENA grandis Krinzl., 208

LEGUMINOSAR, 142

LEPANTHOPSIS vinacea C.Schweinf., 109,110

LIMODORUM striatum Thunb., 184

LIPARIS elegantula Krédnsl., 202 neuroglossa Rcehb,f., 202 serpens Garay, 201,202

LOROSTEMON, 158 bombaciflorum Ducke, 158, 159 colombianum Maguire, 159, 162

Lorostemonoideae, 158

[ xvi ]

lulo, 233-235 ma ma, 282

macuira, 279,280

MACAIREA albiflora Cogn., 165 arirambae Hub., 164,165 seabra Cogn., 165 Schultesii Wurdack, 164,165 theresiae Cogn., 165

MACROLOBIUM, 113,142 microcalyx Ducke var. minimum Cowan, 151

maicoma, 7,10 maikoa, 10

maize outer pistillate glumes, 57-76 normal glumes, 58-64 papyrescent glumes, 64-66 tunicate glumes, 66-68 vestigial glumes, 68-72

(See corn) maizillo, 395

MALAXIS tipuloides (Lindl.) O. Ktze., 201 var. exigua Garay, 201

MALOUETIA

Tamaquarina, 39,42 MALPIGHIACEAR, 2,19,22,32,41

MANDEVILLA annulariifolia Woodson, 168, 169,170 Benthamii, 169 cuneifolia Woodson, 169 nerioides, 178

seabra, 176

stephanotidifolia Woodson, 176

Steyermarkii, 177

thevetioides Woodson, 177

manilla, 279

MANIHOT esculenta Crants., 229 utilissima, 283 MANISURIS, 72 spp., 67 MARSUPIARIA equitans Hoehne, 208 MASCAGNIA psilophylla (Juss.) Griseb. var. antifebrilis (Ruiz. & Pav.) Ndz., 19,20,41

MAXILLARIA, 208 equitans (Schiltr.) Garay, 208 heteroclita Poepp. & Endl.,325 Sophronitis (Rchb f.) Garay, 208 vandiformis Schweinf., 208

Maxillarieae, 212,213

Maydeae, 57,66,72,434

me-ne-ka-heé-ma, 38

MEeELASTOMACEAR, 113,160

METHYSTICODENDRON Amesianum R. E. Schulltes,

279,287

mets-kwat borrachera, 279

mihi, 11

Moronobeae,

muémueli, 24

MyrsINAcKAE, 166

narcotics, 1—-56,273-304

nalema, 7,10,11,23,24,41

[ xvii ]

nepe, 7,10,24,41 nepi, 7 NICOTIANA Tabacum L., 39,230,279 no, 280 noat, 280 NOTYLIA lilacina Arédnsl., 214 Norae Garay, 214,215 OcuNacrar, 317 oco-yagé, 30,35 ONCIDIUM

vagans C.Schweinf., 223,224,

227 Warmingii Rchb f., 227 OCTOMERIA

deltoglossa Garay, 199,200 rhodoglossa Schiltr., 199

ogfa, 36 ORNITHIDIUM Sophronitis Rehb.f., 208 Ornithocephalae, 213 pa-ree-ka, 294 pajuil, 39 Papperitzieae, 104 PARAPROTIUM, 113 amazonicum Cuatr., 156,157 vestitum Cuatr., 157 paricad, 278,294-296 PAULLINIA Yoco R. EF. Schultes, 279 PEGANUM Harmala, 22 pejt, 25 PENNISETUM ciliare (L.) Link, 380

PHILODENDRON colombianum R. E. Schultes, 121,126 PHREATIA, 212 PHYLLANTHUS spp., 220 PHYSOSIPHON, 195 PHYSOTHALLIS, 195 ptache, 279,280 pinde, 7,10,23,24,41 PIPTADENIA peregrina Benth., 278,295, 299,300 PLATYSTELE ornata Garay, 195,198 PLEIONE scopulorum W.W.Sm., 184

Pleurothallidinae, 195

POLYOTIDIUM Garay, 105 Huebneri (Mansf.) Garay, 105, 106 PRESTONIA amazonica ( Benth.) Macbride, 14,15,17,18,20,23-25,40, 42 Lindleyana Woodson, 179 vaupesana Woodson, 179 PROTIUM heptaphyllum, 297 pujana, 24 QUEKETTIA, 104 reé-ma, 38

RHYTIDANTHERA regalis I. F.Schultes, 317,318

roca, +

[ xviii ]

RODRIGUEZIA corydaloides Krinzl., 209 Teuscheri Garay, 209,210

sameruja, 6 Saundersieae, 213

SCELOCHILUS corydaloides (Kranzl.) Garay, 208

SEPALOSIPHON, 213

SIPHANTHERA capitata Gleason, 166 Hostmannii, 165,166 paraensis Huber, 166 pratensis Mg f., 166 ramosissima Cogn. ex Hoehne,

166

subtilis Pohl, 166 Tatei Gleason, 166 tenera Pohl, 166 vaupesana Wurdack, 165,166

snuff, 293-296,298-300

SOBRALIA candida (Poepp. & Endl.) Rehb f., 190 lancea Garay, 187,190 SOLANUM, 230 hyporhodium 4. Br. & Bouché, 232355 quitoense Lam., 233 Topiro HBK., 229-234,236, 238, 240-242, 244 SORGHUM, 65,66,74 papyrascens Stlapf., 65 sp.; 2383 STELIS, 195 nexipous Garay, 194,196 STENOSPERMATIUM

verecundum R. FE. Schultes, 121,122, 128,130 Ulei, 122

sugii, 283

TABERNAEMONTANA coriacea Link ex R. & S., 180 rimulosa Woodson, 179,317

TACHIGALIA, 113,156 multijuga Benth., 153 polyphylla Poepp. & Endl.,

153 Schultesiana Dwyer, 152,153

TELIPOGON semipictus Rchb.f., 218 Teuscheri Garay,214,216,218 Wallisii Rchb,f., 218

teosinte, 357-383,431-433 introgression, 389-406

TEPHROSIA

toxicaria Pers., 230

TETRAPTERYS methystica FR. F.Schultes, 40, 41,278,298 sp., 33,41,42 THEOBROMA subincanum, 295 tobacco, 279-284,286,288,289, 29 1-296,298-300 ponga, 279,285 too-chée-ka-pee, 139 topiro, 233,234 TRIPSACUM, 57,69,72 australe, 370,402,403 dactyloides, 359,360,362, 369-371, 381

maizar, 370

[ xix |

tupiru, 233,235

UROSPATHA

antisylleptica R. EF. Schultes, 311,314-316

decipiens Schott, 123,316

Hostmannii, 316

sagittifolia (Rodach) Schott, 122

somnolenta RP. FE. Schultes, 123,124,132,134,136

Spruceana Scholt, 123

VARGASIELLA peruviana C. Schweinf. , 223 venezuelana C. Schweinf. , 219,

220

VIROLA calophylla Warburg, 278, 294,295 calophylloidea Markgraf, 278, 294,295

WARREA, 321 candida Lindl., 328 digitata Lem., 324 discolor Lindl., 321,324 marginata Rehb.f., 326 quadrata Lindl., 326 Wailesiana Lindl., 326

WARSCEWICZELLA, 321,322

amazonica Rehb.f. & Warsce., 322

aromatica (Rehb.f.) Rehb.f., 323

bidentata (Rchb.f. ex Hemsl.)

calloglossa Schltr., 327 Schltr., 323

cochlearis (Lindl.) Rehb.f., 324

cochleata Barb.-Rodr., 325

digitata Barb.-Rodr., 324

discolor (Lindl.) Rehb.f., 324

Jlabelliformis (Sw.) Cogn. , 325

Gehrtiana Hoehne, 327

Gibeziae (N.E. Brown) Stein, 325

heteroclita (Poepp. & Endl. ) Hoehne, 325

ionoleuca (Rehb.f.) Schltr. 825

Lindenti Hort., 323

Lueddemanniacana Rehb.f. 826

marginata Rchb.f., 326

velata Rehb., 326

Wailesiana (Lindl. ) Rehb.f, ex Morren, 327

Wendlandi (Rehb.f.) Schltr. , 323

,

’

yageuco, 35

yagé, 24,25,30, 34-36

yageé cultivado, 36

yagé del monte, 35

yahé, 11,24

yaeé, 1,6,7,10,11,14, 16-25, 30, 33, 35-38, 41,42,44,46,278, 288-290, 292, 296,298,300

yajen, 14

ya-kee, 278,294

yako-borrachero, 25

yarumo, 292

yato, 278,294

yera, 291

yeras, 291

yoco, 279,288-300

yopo, 278,299

yuca, 229,283

[ xx ]

yurupari, 25

ZAMIA madida R. F.Schultes, 114, 115,116

ZEA, 372,380,381 Mays L., 57 mexicana, 330,357,380,381, 389

Zygopetalinae, 321

ZYGOPETALUM, 321 amazonicum (Rehb.f. & Warsc.) Rehb.f., 323 aromaticum Rehb.f., 323 bidentatum Rchb.f. ex Hemsl.,

discolor (Lindl. ) Rehb.f., 324

flabelliforme (Sw.) Rehb.f., $25

Gibsiae N.E.Brown, 325

ionoleucum Rehb.f., 325

Klugii C.Schweinf., 326

Lindenii (Hort.) Rolfe, 323

Lueddemannianum (Rchb.f. ) Rehb.f., 326

makat H., 321

quadratum Pfitz, 326

rhombilabiumC. Schweinf. ,326

sect. Cochleare Rchb.f., 322

sect. Warscewiczella Rchb f., $22

trinitatis Ames, 326

323 Wailesianum (Lindl.) Rehb.f. candidum (Lindl. ) Rehb.f., 327

323 Wendlandi Rehb.f., 323 cochleare Lindl. , 321,322,324 cochleatum Paxt., 324 ya-to, 294 conchaceum Hoffmannsegg yato, 278

ex Rehb.f., 325

Issued June 3, 1959

Ee8

ERRATA

Page 65, line 138 for Sholz read Scholz Page 105, line 19 for cornoso read carnoso Page 106 add Drawn by Leslie A. Garay Page 151, line 8 for basem read basim Page 123, line 6 for lanceoloto read lanceolato Page 172 reverse upper and lower legends Page 278 after Banisteriopsis inebrians add Banisteriopsis quitensis (Ndz. ) Morton, Colombia, Ecuador, Peru

Page 283, line 7 fo

utilissima read esculenta

_

Page 310, line 1 for florium read florum

Page 316, line 9 for Hostmanti read Hostmannii

Page 317, line 18 fo

specimine read speciminis

=

Page 317, line 26

for essensial read essential Page 323, line 22

for Gary read Garay

=

Page 324, line 1

for (Lehm.) read (Lem.)

—

Page 325, line 8 for Barb. Rod. read Barb.-Rodr.

Page 375, line 14 for terminal read subterminal

[ xxii |

BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY

VoL. 18, No. 1

CampBripGce, MassacnusetTts, JUNE 7, 1957

THE IDENTITY OF THE MALPIGHIACEOUS NARCOTICS OF SOUTH AMERICA

BY Ricuarp Evans ScCHULTES

This is all I have seen and learnt of caapi or aya-huasca. . . . Some travel- ler who may follow my steps with greater resources at his command, will, it is hoped, be able to bring away ma- terials adequate for the complete anal- ysis of this curious plant.

Richard Spruce

I

Tue New World narcotics to which man has attributed the most extraordinarily bizarre powers of altering the state of his body and mind hid out successfully from pry- ing European eyes and were not discovered and identified botanically until about one hundred years ago. They remain, even to-day, the most poorly understood of the hallucinogens of the Americas and the narcotics whose identification is most baffling. Indeed, we may truthfully say that we stand merely on the threshold of our inves- tigations into the botany, ethnology, history, pharma- cology, chemistry and therapeutics of that complex of intoxicants known as ayahuasca, caapt or yaje.

II

The purpose of this paper is to summarize what Is known about the identity of the malpighiaceous narcot-

[1]

ics and to try to make some order from the rather chaotic state which at present obtains. A survey of the literature has been made, and the information from this source is evaluated in the light of field experience in the north- western part of the Amazon Valley which seems to be the centre of use of these narcotics. It is noteworthy that such a standard work as Hartwich (29) fails even to men- tion any intoxicant from the Malpighiaceae.

This present study, however, leaves much to be de- sired. Since references to the malpighiaceous narcotics are widely scattered throughout botanical, anthropologi- cal, geographical and travel literature, I can make no claim to completeness of coverage. By far the greatest number of the references in regard to the sources of these narcotics are of limited or doubtful value; nevertheless it has been thought best to cite all that have come to light. I might say categorically, however, that much careful and sustained field work must be done before anything approaching a full understanding of the problem may be attained.

Since this paper is concerned only with the identifica- tion of source-species, all discussion of the uses, effects, chemical constitution and ethnological importance of the narcotics themselves has been omitted.

The line drawings were made possible by a grant from the American Academy of Arts and Sciences, and pub- lication of the illustrations has been done through a grant from the National Science Foundation. I express my appreciation to Miss Ruth Barton and Mr. Elmer W. Smith for the drawings reproduced herein.

II]

When, in 1851, the botanical explorer Richard Spruce penetrated the upper Rio Negro basin in Amazonian Brazil, a new and fertile epoch opened for natural science

[2]

PLATE I

3° _200 ace aaa geo = pr t re —

1 Guahibo 13 Bora 25 Sukuna | j

2 Zapare uXarihona 26. aiwane - s¥ngalere 15 uber aa ne ‘4 Mazan toSiukano 2Andoke »5 Correguahe wAbaku 2» Bare

6 Jivaro 1 Desano wHaniva 7 Colorado 9GYwanano 31 Jariana_,

stayana rwekwano xfiyuka | *9JIngano wkakuna s3Ftoje _to-Jtona drasana dian 23 fanimuka Pit

aNMira i

_

TRIBES REPORTED TO USE | MALPIGHIACEOUS NARCOTICS

(CAAPI, AYAHUASCA , YAJE)

in the New World. Undoubtedly, one of Spruce’s great- est contributions to science was his discovery and ethno- botanical study of the curious caapt. He found the Tu- kanoan Indians of the Rio Uaupés and its affuents using caapt to induce for prophetic and divinatory purposes a narcosis characterized, amongst other strange effects, by frighteningly realistic colored visual hallucinations and a feeling of extreme and reckless bravery. This con- tribution included a precise identification, through flow- ering material, of a source of the narcotic, an hitherto undescribed species of a malpighiaceous liana. The liana, originally named Banisteria Caapi Spruce ex Griseb., is now correctly called Banisteriopsis Caapi (Spruce ex Griseb.) Morton. Although Spruce’s observations were written down in his wonted preciseness in 1852, they were not published until the posthumous account of his travels appeared in 1908 (90). In 1852, finding caapi cul- tivated in quantity on the Rio Uaupés, he noted: ‘‘There were about a dozen well-grown plants of caapi, twining up to the tree-tops along the margin of the roca [a cul- tivated plot] and several smaller ones. It was fortunately in flower and young fruit, and I saw, not without sur- prise, that it belonged to the order Malpighiaceae and the genus Banisteria, of which I made it out to be an undescribed species and therefore called it B. Caapi.”’ (90)

Two years later, Spruce found caapi in use amongst the Guahibo Indians of the upper Orinoco in Colombia and Venezuela. Here the natives ‘‘not only drink an in- fusion, like those of the Uaupés, but also chew the dried stem, as some people do tobacco.’ Again, in 1857, whilst working in the Peruvian Andes, he encountered the Za- paro Indians using a narcotic known as ayahuasca. He stated that he ‘‘again saw caapi planted”’ in the north- eastern Peruvian Andes and that ‘‘it was the identical species of the Uaupés, but under a different name.’ (90)

[ 4]

The earliest published record of the use of any mal- pighiaceous narcotic of which I am aware dates from 1858, when Villavicencio (95) reported that the drug was em- ployed by the Zaparos, Angateros, Mazanes and other tribes of the upper Rio Napo in Amazonian Ecuador for sorcery, witchcraft, prophecy and divination. Although apparently no specimens were taken and no reference to a botanical determination was made in Villavicencio’s excellent and complete account, which included a report on self-intoxication, the common name ayahuasca was used, and the plant was described as a liana. Later work has definitely shown that ayahuasca in Peru and Ecuador belongs, in great part, at least, to the same genus as the Brazilian caapi. Indeed, as Spruce noted, although ‘‘of the plant itself?’ Villavicencio ‘‘could tell no more than that it was a liana or vine,’’ his ‘‘account of its proper- ties’’ coincided ‘‘wonderfully with what I had previously learnt in Brazil.’’ (90)

In alist of plant names in the Tupi language of Brazil, the explorer von Martius (98,99) discussed caapi, stating that the Indians of the Rio Uaupés prepared from the fruits of the caapi plant an intoxicating drink, which they employed in their dance-ceremonies, and identify- ing it as Banisteriopsis Caapi. Since von Martius had never visited the Rio Uaupés, this report must be ac- cepted as an indirect one. He may have seen specimens of the caapi sent by Spruce to Europe and knew that Spruce had reported them as the source of the narcotic beverage. The statement that the fruits of the liana are used as the source of the intoxicants must be read with extreme reservation, even with doubt; for none of the reports of travellers and explorers who have seen the plant in use mention the fruits as the part of the plant employed.

The few incidental references to caapi or ayahuasca

[5]

found in travel books have almost invariably neglected the fundamental question of the botanical identity of the source of the narcotic.

Orton, who travelled widely in the Andes in the last half of the last century, reported (61) the narcotic use of ayahuasca amongst the Ziparos of the eastern slope of the Andes of Ecuador, but he failed to shed any light on the botanical source of the drug.

Writing on his travels in South America, Crévaux reported (15) that, in southeastern Colombia, the Corre- guahes of the Rio Caqueta prepared an intoxicating bev- erage from the bark of a plant called yayé. He likewise noted, as had Spruce, the use by the Guahibos inhabit- ing the Rio Inirida of Colombia of an intoxicating drink prepared from a root called capi, apparently unaware that this capi and the yajé were probably botanically the same. Many years passed before it was known that yajé and ayahuasca referred to the same intoxicant.

This report by Crévaux injected a new uncertainty into the picture. Was yajé from the same source as aya- huasca and caapi, or was it made from a different plant ? Although, as time went on, the belief that the three were at least generically identical became stronger, the prob- lem was not subjected to careful study until the late 1920s.

In 1866, Simson reported (89) that the Piojes of Ecua- dor, like their neighbors, the Zdparos, ‘‘drink ayahuasca mixed with yajé, sameruja leaves and guanto wood, an indulgence which usually results in a broil between at least the partakers of the beverage.’’ He failed to iden- tify any of the ingredients botanically, but we may safely assume, I think, that gwanto refers to a tree-species of Datura. This report is, however, of great significance in being apparently the earliest to mention the mixture of anumber of elements in preparing the ayahuasea drink

[ 6 |

and the first intimation that yqj7é and ayahuasca might be different plants. In relation to Simson’s report that the Ziparos mix a number of plant materials to make the narcotic drink, we should point out Spruce’s definite statement that, so far as he could discover, the Banis- teriopsis was used alone by these Indians.

Still a further complication arose when, in 1890, Ma- gelli, a missionary who had spent much time in the Pas- taza and Bobonaza areas of Amazonian Ecuador, reported (50) an intoxicating drink which the Jivaro Indians called natema. Concerning the botanical source of natema, Magelli wrote: ‘‘Natema is a red bark which, when boiled 24 hours, yields a decoction which deprives those who drink it of their senses for three days... .’’ Ma- gelli seems to have confused natema with another narcotic —maicoma—which he stated is ‘‘a little less toxic than natema.’’ It would appear from Magelli’s notes that he used the term natema to refer to a species of Datura and maicoma to refer to ayahuasca. As Reinburg (69) has suggested, Magelli undoubtedly erred and confused two plants whose uses were very similar. Later workers have established the fact that natema is actually referable to Banisteriopsis, whereas maicoma is, in reality, the much employed tree- Datura narcotic of the Jivaros.

Tyler (92) reported, in 1894, the use and preparation of ayahuasca by the Zaparos of the Rio Napo area of Ecuador, stating merely that ‘‘the beverage is a decoc- tion of a certain species of liana. . .”’ and giving a de- scription of the effects of the drug.

Writing of the Colorados of Ecuador in 1905, Rivet (73) referred to nepi (nepe) as a febrifuge, a violent emetic and an intoxicant and stated that its source was a “‘liana.”’ Later, in a glossary published in 1907 (74), he identified the Colorado nepe and the Cayapa pinde as Banisteriop- sis Caapi. We might here note that a more recent work

ae

EXPLANATION OF THE ILLUSTRATION

Piate II. Banistertopsis Caapr (Spruce ex Grised. ) Morton. Drawing of a specimen of the type col- lection. 1, flowering branch, about one half nat- ural size. 2, flower, about two and one half times natural size. 3, fruit, somewhat larger than one half natural size.

Drawn by Eimer W. Siri

Prare fi

BANISTERIOPSIS Caapi (Spruce ex Griseb,) Morton

on these Indians by von Hagen (96) likewise refers the cultivated nepe to Banisteriopsis Caapi. It is again doubt- ful that any of these determinations were made on the basis of botanical specimens.

In 1905, Rocha (77) published an account of his trip to the headwaters of the Rios Caquetaé and Putumayo in Colombia and reported that the Inga and Siona In- dians of the region, which to-day is included in the Comisaria del Putumayo, employed as a narcotic a “‘little bush” or ‘‘liana’’ called yajé. He stated that the natives would not allow white men to see the plant and that, for this reason, he did not know it from personal observa- tion. His account of its properties, however, coincided very closely with those described for ayahuasca, and it was widely assumed that the two were identical.

In 1907, Rivet (75) wrote on the Jivaros of the Rios Pastaza and Bobonaza in eastern Ecuador. He indicated that they had a narcotic liana which they called natema and which he referred to Banisteriopsis Caapi. He clearly distinguished between the narcotics natema or yajé on the one hand and maicoma or maikoa (Datura sp.) on the other. Rivet’s identifications were apparently not based on botanical specimens.

Later, Karsten (35) stated his belief that the ayahuasca of the Indians of the Rios Napo, Curaray, Bobonaza and Pastaza; the natema of the Jivaros; the pinde of the Cayapas; the nepe of the Colorados; and the caapi of the Rio Negro are all referable to one species: Banisteriop- sis Caapi. Ina letter to Reinburg (69), Karsten reiterated that he had no doubt that natema, ayahuasea, nepe and pinde are identical and may be referred to Banisteriopsis Caapi, even though he had available no identifications made on the basis of specimens, and that he could not offer a botanical determination of the ‘‘idhi’ (yajé?), an intoxicating liana which the Jivaros never used alone but

[ 10 ]

always mixed with natema. Furthermore, in 1926 (36) and again as recently as 1935 (87), he set forth his belief that ayahuasca and natema were the same and were refer- able to Banisteriopsis Caapi.

Early in the present century, Koch-Griinberg, whose extensive travels (1903-1905) and investigations in the upper Rio Negro basin and in the Apaporis-Caqueta area are classic, stated that ‘‘kaapt’’ was prepared from a mal- pighiaceous shrub (39). I have been unable to find in our herbaria a specimen collected by Koch-Griinberg. Un- doubtedly basing his statement on Spruce’s earlier work in the same area, he referred the intoxicant to Banisteri- opsis Caapi. According to Koch-Griinberg (89), the Tukanos of the Vaupés distinguish two species of caapi; but for the second he ventured no botanical determina- tion. He also reported that the Karihonas (Hiandkoto- Umiua) of the headwaters of the Apaporis knew Banis- teriopsis Caapi, calling it yahe and hi(d)-yati(d)yahe. Since he did not penetrate to the sources of this river, he had to assume the ‘‘identification’’ without specimens. He stated, in a letter to Rivet (69) that, for the yayé of the Hiandkoto and the mihi of the Kubeos, he unfortunately did not have botanical knowledge and could not say whether it was Banisteriopsis Caapi or some other plant, but that he believed them both to be from this malpigh- iaceous species.

During his studies amongst the Yekwanas of the upper Orinoco basin of southern Venezuela, Koch-Griinberg (40) found these people using two ‘‘species”’ of narcotic vines, one cultivated and one wild. Both were called kali. The stronger narcotic of the two was the cultivated kind. In his letter to Rivet (69) he stated his belief that these both represented Banisteriopsis Caapi.

Hardenberg (28) and Whiffen (100), writing on the Colombian region lying between the Rios Caqueta and

Ci

ee

EXPLANATION OF THE ILLUSTRATION

Pirate III. Banisreriorsis ineprtans Morton. Draw- ing of the type specimen. Branch and fruit about one half natural size.

Drawn by Etmer W. Siri

[ 12 ]

PuaTeE III

BANISTERIOPSIS a tnebrians Morton

SSRN . SSGGRAQAHyw SSQoon7yr sx SSS 7M) my S77! //) ) a mm!

Putumayo, inhabited chiefly by Witoto and Bora In- dians, definitely reported the use of the narcotics in this area of the northwest Amazon, but neither was able to offer a botanical determination. Hardenberg stated merely that the liana was called ayahuasca or yajén; Whiffen, that it was known as caapi north of the Caqueta or ayahuasca south of this river. Both of these explorers thus intimated that these sundry names referred to one plant.

In 1917, Safford (83), who had devoted much study to New World narcotic plants, referred both ayahuwasea and caapi to Banisteriopsis Caapi, after an evaluation of the literature on the subject.

A suggestion that ayahuasca and yajé might be differ- ent plants, however, can be found in Reinburg’s study of the tribes inhabiting the region between the Rios Napo and Curaray in Pert. In 1921, he wrote (69) that the nar- cotic drink was an infusion of a few fragments of ayahu- asca, a liana the diameter of a man’s thumb, and leaves of yajé, “‘un petit arbuste, de 1m.50 de haut, 4 feuilles pétiolées (pétiole de 15 mm.) entiéres, ovales, longues de 20 cm., larges de 7 cm., réguliéres et terminées par une pointe de 2 cm.”’” On the basis of specimens collected, he held that ayahuasca, the source of which was always, according to his report, a wild, forest liana, and caapi were conspecific and represented Banisteriopsis Caapi. Still on the basis of specimens, he suggested that the yajé of the Curaray in Peru could, with reservation, be referred to the apocynaceous Haemadictyon (approach- ing, in some respects, Hl. amazonicum Benth.) or a re- lated genus. '

It was apparently Spruce (90) who first suggested that

'The generic concept Haemadictyon has been united with Prestonia, and the proper name of the plant to which Reinburg referred is now Prestonia amazonica (Benth.) Macbride (47).

[ 14 |

Prestonia amazonica might enter into the caapi picture. * He said that the eaapi-pinima (*‘ painted caapi’’) is ‘*an apocynaceous twiner of the genus Haemadictyon, of which I saw only young shoots, without any flowers. The leaves are of ashining green, painted with the strong, blood-red veins. It is possibly the same species . . . dis- tributed by Mr. Bentham under the name of Haemadic- tyon amazonicumn. sp. It may be the caapi-pinima which gives its nauseous taste to the caapi. . . and it is prob- ably poisonous... , but it isnot essential to the narcotic effect of the Banisteria, which (so far as I could make out) is used without any admixture by the Guahibos, Zaparos and other nations out of the Uaupés.”’

In 1922, ayahuasca or ayawasca was reported as a nar- cotie for the first time from Bolivia, from the Rio Beni, by White (101), thus greatly amplifying our knowledge of the range of the use of the intoxicant. White defi- nitely identified it as Banisteriopsis Caapi. Although he collected material for chemical analysis, it is not clear from his account whether or not his identification was based on an herbarium specimen. Much of White’s bo- tanical material is preserved in the Botanical Museum of Harvard University, but a search through the Eco-

? However, Spruce’s original field notebook, preserved at the Royal Botanical Gardens at Kew, shows a discrepancy, in this regard, with the published report. The field notes say: “9712, Banisteria Caapi Mss. From this is prepared an intoxicating drink known to all the natives on the Uaupés by the name of caapi. The lower part of the stem, which is the thickness of the thumb swollen at the joints, is the part used. This is beaten in a mortar with the addition of water and a small quantity of the slender roots of the Apocynac. (apparently a Haemadictyon) called caapi-pinima or painted caapi, from its lvs. being stained and veined with red . . . Query? May not the peculiar effects of the caapi be owing rather to the roots of the Haemadictyon (though in such small quantity) than to the stems of the Banisteria? The In-

dians, however, consider the latter the prime agent, at the same time admitting that the former is an essential ingredient. The two plants

rls

are planted near all mallocas (villages)...

[15 |

nomic Herbarium of Oakes Ames at this institution fails to reveal a specimen of ayahuasca collected by White.

In his report, White stated that the intoxicant was prepared either exclusively from the stems of ayahuasca, an immense liana with greenish white flowers, or else from ayahuasca stems boiled with the leaves of the chaco, a shrub with small, globose, red-yellow fruits. Leaves of plants locally called cagna and guayavoche may also be used with ayahuasca, but White could not ascertain whether these were other names for the plant known as chaco or whether they represented different species. It is unfortunate that so much uncertainty surrounds White’s report, but we may feel rather sure (from the common name ayahuasca, from his account of the intoxication and from recent plant collections in adjacent areas) that a species of Banisteriopsis is likewise the source of the nar- cotic in this Bolivian locality.

In several reports, the botanist Rusby, who had accom- panied White on the Bolivian expedition, detailed the physiological effects of caapi and stated that the drug was derived from Banisteriopsis Caapi (80,81,82). Ina pharmacological report published in 1924, Seil and Putt (88) reported the isolation from Rusby’s material of a ‘‘fine powder”’ with at least three alkaloids (both phenolic and non-phenolic), but they offered no botanical deter- mination for the material which they had studied.

In the same year, 1922, the Belgian botanist-explorer Claes, who had gone to the upper reaches of the Rio Caqueta, investigated the yaqjé of the Correguahe Indians of this area of southern Colombia (10). He learned that the yayé, hitherto usually described in the literature as ‘‘a small bush,’’ was an enormous forest liana. Claes argued—lI think quite correctly—that those who had de- scribed yajé as a small bush had seen young, cultivated individuals and not the vine in its wild state (9).

[ 16 ]

The intoxicating yajé of the Correguahe Indians was made by boiling pieces of the trunk itself together with the crushed bark of the trunk, but later the stem and leaves of a plant which the Indians did not permit him to examine were added to the mixture. This brew pro- duced, as Claes witnessed amongst the natives, very defi- nite narcotic symptoms. It is extremely unfortunate that Claes was unable to procure specimens of the leaves and stems which were used together with the liana, for we now know, from the studies of more recent investigators, that the narcotic yaje drink of this region of Colombia is sometimes of a more complex preparation than that of other parts of the Amazon; that very frequently a non- malpighiaceous ingredient may be employed.

Claes did not offer a botanical determination of yajé (68). He mentioned that, according to De Wildeman, yajé ‘‘might be’? Prestonia amazonica. There is no speci- men of this species in the Rijksplantentuin in Brussels, and Claes himself states (68) that he did not obtain ma- terial for determining yajé, adding: ‘‘It thus is necessary to leave for others the task of collecting materials of the definitive classification of the plant.’” Michiels and Clin- quart (54), publishing their observations made during pharmacological experiments with Claes’ material, sug- gested that the stems with which they were working appeared to belong to Prestonia amazonica. The French pharmacologist, Rouhier, feeling (although apparently with no sound basis) that yajé and ayahuasca were one and the same thing, but that they might, physiologically, act differently, put forth the suggestion that ayahuwasca represents Banisteriopsis Caapi, whereas yqjé could rep- resent the unknown plant which Claes saw the Corre- guahe Indians add to the brew.

A Colombian chemist, Fischer, was apparently the first to isolate a crystalline alkaloid from yajé (22).

[17 ]

Fischer admitted that botanical determination of yajé had not been made, but he stated that, to judge from anatomic and histologic details, it seemed to bea species of Aristolochia. In 1924, shortly after Fischer’s work was done, Rouhier (78) accepted this ‘‘identification,’’ stat- ing that his specimens ‘‘reminded one of a liana, proba- bly a species of the Aristolochiaceae.’’ In a subsequent work, however, Rouhier (79) attributed ayahuasca to Banisteriopsis Caapi but yajé to a ‘‘liana,’’ the identifica- tion of which was still under discussion. He pointed out that in its physiological action and its geographical range, yajyé was comparable to ayahuasca, and he dismissed its reference to Prestonia amazonica as ‘‘doubtful.’’ At about the same time, Barriga-Villalba (6), a Colombian chem- ist, and Albarracin (1), a Colombian pharmacologist, actively investigated yaqjé, but neither shed any further light whatsoever on the botanical identity of the drug. Albarracin described the source of his material as a ‘‘climbing shrub’’ and asserted that the natives did not cultivate yajé because it abounded in the forests.

Reporting in 1926 on the caapi ceremony of the Tu- kano Indians of the Rio Tikié, a Brazilian affluent of the Uaupés, MacCreagh (49) described the intoxicant as ‘‘a thin, almost colorless liquid ... concocted from the leaves of a vine.’ He, apparently, failed to collect her- barium material of the vine. It is worthy of note that the Rio Tikié is the same area where Koch-Griinberg had made his observation on caapi. It is very near the locality where Spruce first met with the drug (90). Com- menting on ‘‘haapi,’’ which he had encountered in his travels in the same general region, the English explorer McGovern (53) failed to venture a botanical determina- tion, merely mentioning that the drug was made from **a root.”

A rather inclusive survey of what had been accom-

[ 18 |

plished up to that date on the botanical identification and chemical constitution of this complex of malpighia- ceous narcotics was published in 1927 by two French pharmacologists, Perrot and Hamet (66,67). They pointed out that even more confusion reigned in the chemical field of investigation than in the botanical, prin- cipally because pharmacologists and chemists consistently disregarded taxonomic accuracy in identifying the sources of plant materials under study. Perrot and Hamet con- cluded that (a) yajé, ayahuasca and caapi refer to one species of plant: i.e., Banisteriopsis Caapi; and that (b) no apocynaceous plant is at all concerned in the problem of the source of the narcotic known under these three names. Notwithstanding the meritorious efforts of Per- rot and Hamet, their review brought little clarification into either the botanical or the chemical picture.

In reply to the article by Perrot and Hamet, the Ger- man botanist Niedenzu (59) published several very in- teresting observations based on herbarium specimens preserved in the Berlin Herbarium. These specimens are, of course, no longer extant, but the notes are of special importance because of the authenticity of the determi- nations at the hand of an outstanding specialist in the Malpighiaceae. The collection Tessmann 5424 from Ya- rina Cocha on the Rio Ucayali in eastern Peru repre- sented, according to Niedenzu, a mixture: leaves of Mascagnia psilophylla (Juss.) Griseb. var. antifebrilis (Ruiz & Pav.) Ndz. and, in a paquet attached to the sheet, samaras of Banisteriopsis quitensis (Ndz.) Morton; on the label was annotated the vernacular name of haya- wasca or ayawasca and the observation ‘‘Stiicke der Liane gekocht, dann kalt genommen.’’ Niedenzu further called attention to three specimens of Banistertopsis quitensis. Eggers 15485, from eastern Ecuador, consisted of leaves, scanty flowers and plentiful fruit and had a note that,

[ 19 ]

from the stems, an intoxicating brew was prepared. Niedenzu stated that the characters of Mggers 15485 did not seem to correspond very closely with Banisteri- opsis quitensis as it was described. Tessmann 4974 and 5525, from the Rios Pastaza and Itaya, respectively, were referred without question to Banisteriopsis quitensis. Both were called ayawasca and were reported to be the source of a narcotic drink employed by witch-doctors. Tessmann identified Tessmann 5325 as Banisteriopsis Caapi, but Niedenzu pointed out that the collection was distinct from this species. In summary, Niedenzu stated that these wild and cultivated plants of Banisteriopsis from eastern Ecuador and Peru and the uses reported for them would seem to indicate that yageine, the alkaloid whose source was still in doubt, might well have come from similar species and that, on the basis of his studies, it would seem that three sources of the narcotic ‘‘aya- wasca’’ ought to be considered: Mascagnia psilophylla var. antifebrilis, Banisteriopsis quitensis and B. Caapi. Another attempt to put some order into the confusion resulting from field observations and fragmentary speci- mens occasionally gathered by ethnologists was made by Gagnepain, who, in 1930, published his botanical opin- ions relative to these narcotics (24). He pointed out (a) that, according to Reinburg, ayahuasca was ‘‘probably”’ Banisteriopsis Caapi, but that yayé could not be referable to this species; (b) that yayé seemed to approach Pres- tonia amazonica; (c) that fragments received as yajé by the chemist Rouhier in 1924 showed the plant to be an ‘““opposite-leaved vine’’; (d) that both Rivet and Rein- burg sent in material which seemed to represent the same malpighiaceous plant. Later, Gagnepain received, through Rouhier, a specimen from the Departamento de KI Valle in Colombia, where the plant was cultivated under the name of yajé.. The specimen, with leaves and

[ 20 ]

inflorescence attached, was determined as Banisteriopsis Caapi. This led Gagnepain to the rather extraordinary assertion that the yajé of Colombia was the same species as the caapi of Brazil, but different from the yayé of Ecua- dor. He could state with assurance only that the Ecua- dorian yajé represented a species of Banisteria ‘‘near ferruginea’”’ and that an exact clarification of the botan- ical sources of yayé was still far from realization.

In 1929 and 1930, the Russian botanist Hammerman (26,27) published a rather complete survey of the prob- lem. His statements were founded on an evaluation of reports in the literature in the light of a study of material gathered in 1925-1926 by G. N. Varonof and S. V. Juzepezuk along the Rio Orteguaza, in the Colombian Comisaria del Caqueta. Pointing out that Zerda Bayén’s assertion that the Indians of the Caquetaé mixed four kinds of leaves to make a brew which they called yq7é, Hammer- man indicated that the material of Voronof and Juzepezuk gave a variety of results when chemically analyzed. He further stated that there seemed to be several species of Banisteriopsis involved, even though most of his material apparently was referable to Banisteriopsis quitensis. Any one of the species could be a source of the narcotic prep- arations of the Indians. He was, withal, careful to em- phasize that only Spruce had seen in flower an actual vine definitely known to be used to make the hallucinating drink. He intimated that the known variation in prep- aration, use and effect of the narcotics called caapi, yajé and ayahuasca might be due to differences in composition.

In 1931, Morton (58) described a new species of Ban- isteriopsis from southern Colombia, naming it B.inebrians Morton because of its use by the Indians of the Com- isaria del Putumayo as an inebriant. Basing his state- ments on the meticulous field observations of the late Guillermo Klug, a German plant collector who worked

[ 21 ]

in eastern Peru and adjacent parts of Colombia, Morton indicated that at least three species are employed in this region: Banisteriopsis Caapi, B.inebrians and B.quiten- sis; and that Banisteria longialata Ruiz ex Ndz. and Banisteriopsis Rusbyana (Ndz.) Morton may enter cer- tain of the narcotic preparations as additional ingredients.

Several reports on chemical studies of caap? published during the late 1920’s and the 1930's considered the bo- tanical identification of the intoxicant under investiga- tion.

Reutter, reporting in 1927 (72) that he had isolated yageine and yagenine from the vegetal parts of yajé or ayahuasca, accepted Rouhier’s ‘‘identification’’ as Pres- tonia amazonica, stating that some tribes add to their yajé drink Banisteriopsis Caapi or ‘‘Datura arborea 1..”’ Lewin, a year later (44), wrote that ayahwasca was ‘‘a member of the Malpighiaceae.”’

In 1929, Keller and Gottauf (38) worked with material of ayahuasca from Bolivia and Peru, referring it to ‘‘a Banisteria,’’ (although they had no leaves nor flowers available) and isolated a harmine-like alkaloid.

Costa and Faria (18,14) held that yajé, ayahuasca and caapt were the same narcotic and were derived from Bani- steriopsis Caapt. Three years later, Chen and Chen (8) summarized briefly the literature references and indicated their belief that caapi, yajé and ayahuasca represented Banisteriopsis Caapi, ‘‘a woody climber that attains a height of 8 to 4 metres and attaches itself to the trunks of large trees’’ or that is ‘‘grown in the North-western regions of South America.’’ Chen and Chen identified the alkaloids telepathine, yajeine and banisterine with harmine from Peganum Harmala, corroborating experi- mentally the earlier suggestions of Wolfe and Rumpf (104). The real importance of Chen and Chen’s paper, however, lies in the chemical study, apparently for the

[ 22]

first time, of vegetal material which seems to have been identified on the basis of botanical specimens. For the ‘‘twigs, leaves and roots of caapi’’ and the ‘‘decoction just as used by the Indians’ which were subjected to chemical analysis had been collected by the botanist Williams near Iquitos, Peru, and were determined as Banisteriopsis Caapi (4,102,103).

For the most part, recent investigators who have de- voted critical attention to the study of the South Amer- ican malpighiaceous narcotics have concurred with Ham- merman, Gagnepain and Klug that several species of Banisteriopsis, if not of other genera, may be involved in some regions (12).

Nevertheless, some contemporary writers tend to sim- plify the problem of identification, even though their ‘‘determinations’’ are seldom, if ever, based on speci- mens. While they are certainly not specialists, they are often quoted uncritically in technical works. Some, as in the case of Reko (70,71), have stated that ayahwasca, pinde, natema, caapi and yajé are all derived from Bani- steriopsis Caapit.

In his dictionary of Amazonian plant names, Le Cointe (41) indicated a belief that ayahwasca and caapt were Banisteriopsis Caapi and that yaqjé corresponded to ‘‘another plant that enters into the composition of the caapi-drink as prepared by some tribes,’’ and he pointed out that some writers identify yajé as Prestonia amazon- ica. Von Hagen (96,97) considers both natema and caapi to be Banisteriopsis Caapi.

In 1936, Pardal (62) referred caapi to Banisteriopsis Caapi and yajé to Prestonia amazonica. The following year, however, he stated (68) that caapi, yqjé and aya- huasca are Banisteriopsis Caapt. Maxwell (52), after an evaluation of the literature concerning the identity of caapi, concluded that the narcotic was a species of Ban-

[ 23 ]

isteriopsis, that ‘‘other vegetable ingredients are some- times included in the making of the beverage’ and that their identity is still in question.

Lewin (42,45,46) identified natema, yagé, yahé, nepe, “*kahv’ and pinde as Banisteriopsis Caapi and enumerated the following Indian tribes as ‘‘addicted’’ to the use of this narcotic species: Guahibo, Tukano, Correguahe, Tama, Ziparo, Vaupé, Yekwana, Baré, Baniva, Manda- wake, Tariana, Siona, Jivaros, Colorados and Cayapas. Although Lewin held the principal ingredient to be Ban- isteriopsis Caapt, he admitted that it was sometimes used together with other plants. Amongst these plants, he named, as one possibility, Prestonia amazonica (43), al- though he strongly doubted that the name yajé referred exclusively to Prestonia amazonica. He further indicated that some Indians probably also add tobacco-water to the beverage.

A most interesting note on ayahuaseca in Peru ap- peared in 1943 (94) and included a ‘‘recipe’’ for making the intoxicating beverage. Unfortunately, all of the plants employed were indicated with Indian names, with no hint as to their identity. The recipe commonly used in the vicinity of Iquitos has, as its principal ingredient, the liana or ‘‘death vine,’’ ayahuasca, which is undoubt- edly a Banisteriopsis. Into the decoction of this vine, the leaves of a species of muémueti are put; the name muém- ueti is said to refer to several plants, only one of which is used at atime. According to this account, the muémueti is responsible for the visual hallucinations. Alsoemployed in the decoction is an evil-tasting tuber called katija. To sweeten up the ayahuasca drink, leaves and seeds of a plant known as pujana are added.

Herrera (31) listed the ayahuasca of Valle de Lares as Banisteriopsis metallicolor (A. Juss.) O’ Donnell & Lour- ‘planta t6xica.””

[ 24 ]

teig and reported it as a

Caller (7), in a consideration of the historical literature on the botanical sources of ayahuasca, concluded that the species used to prepare this narcotic drink is Banisteri- opsis Caapi and that B. quitensis is a synonym of B. Caapi. Other recent writers on Peru (18,97) attribute ayahuasea exclusively to Banisteriopsis Caapi.

Padre Placido, who spent many years as a missionary in the Putumayo of Colombia, did not identify yajé botanically (17), but he stated that the Sionas often add to the narcotic drink of this name the pe or yako- borrachero—the floripondio of Mocoa, or, probably, Da- tura suaveolens H. & B. ex Willd.

Sandeman (84) mentioned yajé casually, attributing it to Prestonia amazonica.

Allen (2), in an excellent description of the yurupari ceremony of the Kubeo Indians of the Rio Vaupés in Colombia, indicated Banisteriopsis Caapi as the source of the narcotic beverage upon which this frenzied ritual is based. Taylor (91), following Allen, has attributed the malpighiaceous narcotics to Banisteriopsis Caapi.

In arecent and authoritative work on plant alkaloids, Henry (30) identified the narcotics known as yajé, caapt and ayahuasca as Banisteriopsis Caapi, B. metallicolor or Banisteria lutea Ruiz ex Griseb.

Manske and Holmes (51) attributed all three to Ban- isteriopsis Caapi. In 1946, Ducke (20) identified the fa- mous narcotic of the natives of the extreme northwest of the Amazon Valley, called caapi in the Brazilian part and yagé in the Colombian part of the region, as Bani- steriopsis Caapi. Hesse (82) attributed ayahuasca and yajé both to Banisteriopsis Caapi, and Moller (55) stated that caapi was exclusively referable to this species.

Recently, Macbride (48), probably following Morton (58), pointed to Banisteriopsis Caapi, B. inebrians and B. quitensis as the principal sources of the alkaloid vari-

[ 25 ]

EXPLANATION OF THE ILLUSTRATION

Pirate IV. Banistertopsis Caari (Spruce ex Griseb. )

Morton. Photograph of the plant from which the

collection Schultes & Cabrera 18156a was made. Photograph by Hrrnanvo Garcia-Barrica

[ 26 ]

Prats TV

EXPLANATION OF THE ILLUSTRATION

Prare V. Cultivation of Banisteriopsis Caapi (Spruce ex Griseb. ) Morton (Schultes & Cabrera 17.209) by the Barasana Indians in the headwaters of the Rio Pira- parana, Comisaria del Vaupés, Colombia.

Photograph by Ricuarn Evans Scuvvres

PLATE V

ously known as telepathine, yageine or banisterine. He stated, however, that it might also be found in ‘ta num- ber of related plants or forms** and quoted Morton in citing Banisteriopsis Rusbyana and Banisteria longialata as admixtures known by the names of oco-yagé and chagro panga.

O’Connell and Lynn (60), working on material of Ban- istertopsis inebrians from the Putumayo of Colombia, found that the stems contain harmine and the leaves ‘‘an alkaloid which was partly identified as harmine.’’ The work of O’Connell and Lynn is noteworthy as represent- ing apparently the second chemical investigation carried out on material of Banisteriopsis identified through her- barium material.

Mors and Zaltzman published a most interesting con- tribution in 1954 (56) in which they indicated their opin- ion, based on chemical examination, that yageine was different from harmine. On the basis of a brief review of the literature, they concluded that eaapi and ayahuasea were referable to Banisteriopsis Caapi but that yagé was not the same narcotic.

Most recently, Fabre (21) published an historical re- view of caapi, ayahuasca and yajé, indicating his belief that only one species

Banisteriopsis. Caapi--is_ the source of these narcotics, though citing Spruce that, amongst certain tribes, another plant may oftentimes be added in the preparation of the narcotic drink.

IV

Although Morton (57) has convincingly shown why the generic epithet Banisteria is not available for the group of malpighiaceous plants which should be called Banisteriopsis, in accordance with the lucid arguments of Robinson (76), there is a tendency to persist in the incorrect use of the name Banisteria. This tendency is

[ 80 ]

not always attributable to ignorance or uncertainty. It is frequently the result of a deliberate flouting of the International Rules of Botanical Nomenclature. Its continuation will not lead to standardization or to clarity but to the further beclouding of the exact identity of the malpighiaceous narcotics as well as to confusion in tax- onomic and floristic works.

Ducke (19,20), in refusing to use the generic epithet Banisteriopsis, stated that his reason was an objection to the multiplication of names in botany. Macbride (48) not only chose to use Banisteria; he made the new combina- tion Banisteria inebrians, attributing it erroneously*® to Morton who has been most outspoken against this use of the name Banisteria. And recently Baldwin, in an article on the chromosomes of Banisteriopsis Caapi (5) chose to use the epithet Banisteria.

Non-botanical writers, faced with this lack of standard- ization in technical papers, often continue to use Bani- steria instead of Banisteriopsis in connection with the narcotic species under discussion (21,380,382, 46,51,62,68, 64,91).

An increasing number of technical writers, neverthe- less, are employing the correct generic epithet (12,28, 33, 34,65,85,93).

V

Although little new information has been published since 1981, a study of material preserved in our herbaria and museums, and recent field observations and collec- tions by several plant explorers have contributed other data which should be made available. It would seem to be especially necessary to do this, since, as the foregoing literature review has shown, little if anything of an ac-

3 ** Ranisteria inebrians Morton, Journ. Wash. Acad. Sci. 21: 485 (1931).””

curate nature may be looked for from the usual type of anthropological field investigation.

In utilizing data found on herbarium labels, caution must be taken. Fora field botanist might jot down ‘‘used as a narcotic’’ on the basis of information supplied by a native and without sufficient checking or without himself having seen the plant thus employed. Nevertheless, such data may be of great value as an orientation in problems as complex as the one at hand.

A further drawback—and a serious one—to the use of sterile specimens for botanical identification is the diffi- culty, in a genus of such vegetatively similar species, of arriving at a satisfactory determination.

Usually, all we have to work with in studying the mal- pighiaceous narcotics is sterile material. The cultivated plants seem rarely or never to flower (probably because of constant cutting back), and the forest lianas blossom sporadically and are seldom found in flower by collectors. In one hundred years, for example, we have only one flowering collection of Banisteriopsis Caapi from the field (that is, excluding plants brought to flower in experi- ment stations or botanical gardens), and that is the type collection made by Spruce. My determinations of most of the older specimens concur closely with the identifica- tions made by Dr. C. V. Morton, who specialized in the Malpighiaceae. 1 have identified the numerous sterile herbarium specimens considered below with reserve, even though I have put in more than ten years of field study on the problem. As the result of prolonged study of liv- ing plants, one acquires some familiarity with certain of the variations which these species may show under dif- ferent natural habitats. But I must further point out that this long period of field study in itself has made me rather cautious about drawing categorical and far-reaching con- clusions from what herbarium material we have at hand

[82 |

at the present time. No one can feel more certain than I that we are just beginning our serious taxonomic studies of the complex of malpighiaceous narcotics.

Perhaps the most revealing observation based on bo- tanical material to have been made since Spruce’s time has not hitherto been published. I was fortunate in dis- covering in the Museum of Economic Botany at the Royal Botanic Gardens at Kew several samples of the ‘“‘yajé plant,’’ comprising twigs and leaves (Econ. Mus. Kew 60-1913) from the Caqueta, Colombia. These were sent in 1918 by Perey EK. Wyndham, Esq., His Majes- ty’s Minister in Bogota. The leaves were determined by Dr. T. A. Sprague who wrote, in a communication at- tached to the specimens, that they represented species of Malpighiaceae (possibly Tetrapterys) and that ‘*the twigs no doubt belonged to one of them.’’ I have care- fully compared the leaves with specimens of the Mal- pighiaceae at Kew and would agree (insofar as it is pos- sible to determine such limited sterile material) with Sprague that they may be referable to Tetrapterys. This is the first time that a malpighiaceous genus other than Banisteriopsis has been mentioned as the source of a narcotic.

With the Wyndham material at Kew, there are sev- eral communications concerning the specimens. It is ap- parent that Wyndham was moved to gather specimens of the yajé plant as the result of interest aroused by a newspaper article (3). This article spoke rather optimisti- cally of yajé as a cure for beri-beri, and it published a long letter from Dr. Rafael Zerda Bayén, a Colombian pharmacist, who put forth extraordinary claims concern- ing the telepathic properties of the vine. No botanical determination was offered. Sprague, in one of the com- munications with the Kew specimens, wrote: ‘*. . . Dr. R. Z. Bay6én says he has deposited specimens of the

[ 38 ]

‘Bejuco de Yagé’ with the Ministry of Public Works and that he doesn’t think it desirable to send the speci- mens abroad! Apparently, the climber has not been identified. ”’

The collection Killip & Smith 27385, from Iquitos, Peru, represents Banisteriopsis Caapi and reports the following information: ‘‘Ayahuasca. Woody vine. Strong narcotic tea brewed from the leaves and stem produces fanciful dreams; also used as a cure for many diseases and as an intoxicating beverage.’’ Killip & Smith 29486 and 29825, from the same region, likewise repre- sent Banisteriopsis Caapi, but in connection with these two collections, the botanists made no notes as to their uses. Ducke 25258 is a flowering collection of Banisteri- opsis Caapi from a plant gathered in Amazonian Peru and cultivated in the Jardim Botdnico in Rio de Janeiro. Wilhams 3741 and 8224, also from the Iquitos area, like- wise are referable to Banisteriopsis Caapi, and are the basis of Williams’ report (108) that ‘‘the lower part of the stem and its leaves are crushed and boiled in water. When sufficiently triturated, the infusion is passed through a sieve to remove fibrous material, and to the residue water is added to render it drinkable. The resulting greenish- brown infusion has a disagreeable, bitter taste and is said to contain narcotic properties. ”’

From this wealth of material, we may judge that Ban- isteriopsis Caapi is rather commonly cultivated in Ama- zonian Peru. Still another collection, Seibert 2173 (also from the Amazonian basin of Peru but from the Rio Tahuamant in the Madre de Dios, far from the Iquitos area) has been determined as Banisteriopsis Caapi, even though its leaves are slightly atypical, due possibly to the fact that the plant was a wild liana in the jungle and not a cultivated shrub in open gardens. The collector reported: ‘‘Ayahuasca. Liana in forest. Infusion taken

[ 84 ]

internally and preparation from bark produces illusions. Used by Indians. Narcotic. ”’

It is from the Amazonian regions of Colombia, how- ever, that the greatest number of collections of the mal- pighiaceous drugs has recently been taken. Klug 1971 and Cuatrecasas 10597, both from the Comisaria del Putumayo, represent apparently Banisteriopsis Rusby- ana, 2 species usually characterized by large leaves. The former collection, from the village of Umbria, is reported to be called ‘‘chagro-panga or oco-yagé’’ ; the latter, from near Puerto Ospina on the Rio Putumayo, is known amongst the natives as yageuco (undoubtedly the same as oco-yagé) and is reputedly cultivated by the Kofan Indians who use the leaves in the preparation of the nar- cotic yajé. In 19538, I collected what appears to be Ban- isteriopsis Rusbyana near Mocoa, capital town of the Putumayo, under the name chagropanga; the leaves of this plant are reputedly used together with the pounded bark of B. inebrians in the preparation of the drink yaye.

In 1931, Klug discovered in Umbria what turned out to be an undescribed species of Banisteriopsis. Described by Morton (58) as Banisteriopsis inebrians, this forest liana, which grows toa length of ninety feet, goes under the epithet of yagé del monte amongst the Ingano In- dians of the Putumayo. Klug collected ample botanical material of the yagé del monte and observed (58): ‘‘One of the most interesting plants found in the region of the upper courses of the Rios Putumayo and Caqueta is the yagé. The Indians make a beverage of either the wild or cultivated yagé, boiling it in a large earthenware vessel an entire day, until there is formed a sort of liquid, like the syrup of sugar cane. They add to the yagé the leaves and the young shoots of the branches of the oco-yagé or chagro-panga (No. 1971), and it is the addition of this plant which produces the ‘bluish aureole’ of their visions. *’

[ 35 |

More recently, Banisteriopsis inebrians has been re- ported as the source of a narcotic in the same region. Cuatrecasas 10598, collected near Puerto Ospina, appar- ently represents this species and is cultivated under the name yagé by the Indians as the ‘‘principal ingredient”’ of the narcotic drink, the unboiled stem (‘‘tallo crudo’’) being employed. Likewise, Cuatrecasas 11061 was taken from a plant cultivated by the Kofiin Indians of the nearby Rio Sucumbios. In 1942, I found the same In- dians using cultivated Banisteriopsis inebrians, called oo- Jd in the Kofan language, as the source of a narcotic drink at Puerto Ospina and on the Sucumbios, and I ex- perimented in both localities with the intoxicant prepared from vines from which the collections Schultes 3452 (from a cultivated plant) and 3474 (from a wild liana) were made. The collection Schultes 3346, likewise apparently referable to Banisteriopsis inebrians, was taken from a vine pointed out by the Ingano Indians of Puerto Limoén on the nearby Rio Caquetaé as the plant from which, without admixture, they prepare their yajé. From Mocoa, capital town of the Putumayo and centre of the Inganos, still other collections of Banisteriopsis inebrians were made (Schultes & Smith 3037; Schultes & Cabrera 19113) with the field annotations that they were called ayahuasca or yajé and bejuco de oro (‘‘golden vine’’), that they were narcotic and that the leaves were used as a strong pur- gative. I have found Banisteriopsis inebrians to be used with and without the admixture of any other species of the genus, but decoctions of this species had marked nar- cotic effects each time I drank them, whether or not any admixtures had been used.

What is probably Banisteriopsis quitensis was reported by Klug (Klug 1934) as yagé cultivado (58) in the Um- bria region. Later, Cuatrecasas (Cuatrecasas 10599) noted that the Kofiins near Puerto Ospina grow this spe-

[ 36 |

cies and use it in the preparation of the intoxicating yajé drink. He stated that the plant material is boiled to pre- pare the beverage. Notes with the collection Garefa- Barriga 4634a, which has been identified as probably referable to this same species, indicate that the Indians who live between Mocoa and Umbria (Inganos) call the vine yajé and prepare an intoxicant from it, with the ad- mixture of two other plants. Garcia-Barriga (25) stated that Banisteriopsis quitensis is ‘‘cultivated near their dwellings in order to have it at hand during the rainy season’? and that ‘‘they make with the stems a drink or beverage which they call yaje.’’ One of the admixtures, according to Garcia-Barriga, is the amaranthaceous A/- ternanthera Lehmanii Hieron., locally known as borra- chera or chicha (both of which terms refer to ‘‘intoxi- cant’? in Spanish); the other admixture could not be determined. Alternanthera Lehmanii is said by Garcia- Barriga to be added also to native beers or chichas as a condiment to increase their intoxicating properties. A sterile collection which seems to represent Banisteriopsis quitensis (Pérez- A rbeldez 639) was stated to be used in the region of Florencia on the Rio Orteguaza in preparing an inebriating drink. Williams (103) has reported Ban- isteriopsis quitensis to be used in the same way as B. Caapt in Amazonian Peru, where it is gathered from both wild and cultivated plants.

The Indians of the Comisaria del Vaupés in Amazon- ian Colombia have conserved many of their aboriginal customs, such as the use of the malpighiaceous narcotics. Unlike the natives of the Putumayo, at the eastern foot- hills of the Andes, who use species of Banisteriopsis in a concentrated decoction made by boiling the plant ma- terial, the Indians of the Vaupés prepare a cold-water infusion of the bark. Banisteriopsis quitensis and B. in- ebrians seems to be lacking in the flora of the Vaupés,

[ 37 |

but B. Caapi and what appears to represent B. Rusbyana are cultivated for use as intoxicants. The collections cited below are all sterile and identification is, therefore, not certain, but, in each case, they were taken from plant material which I know, from personal experimentation, to possess narcotic properties.

The 'Taiwano and Kabuyari Indians of the Rio Kana- nari, an affluent of the middle course of the Apaporis, prepare their kaheé from two kinds of Banisteriopsis, both cultivated. One vine had no leaves at the time of the Baile de la Sabaleta, when yajé is drunk, but the stems obviously belonged to a species of Banisteriopsis. The other is represented by the collection Schultes & Cabrera 13156a and corresponds in all vegetative charac- ters with Banisteriopsis Caapi.

Those Makunas who live along the Rio Popeyaca, also an affluent of the Apaporis, prepare the drink in two ways: either with one species of Banisteriopsis or with two species of this genus, both cultivated. The drink is called haheé by the Makunas. When it is made from one species, Banisteriopsis Caapi (represented by Schultes & Cabrera 15587) is employed: this species is known in Makuna by the name reé-ma. When two species are used in the preparation of the drink, Banisteriopsis Caapi is mixed with a vine called me-ne-ka-heé-ma (‘‘vine of ha-heé™’) and referable probably to B. Rusbyana (Schultes & Cabrera 15588). Of Banisteriopsis Caapi, only the rasped bark is employed, but of B. Rusbyana either the leaves or the bark or both are utilized. With these natives, I took yajé twice; once made with bark from Schultes & Cabrera 15587 and leaves from Schultes & Cabrera 15588, and once with bark of Schultes & Cabrera 15587 alone. Intoxication was induced in both eases, and I was unable to note that one preparation had dif- ferent or stronger effects than the other. Both of these

[ 38 |

preparations are taken by all men of the tribe during dances. When the medicine-man employs the narcotic for purposes of diagnosis or to enable him to work evil on others, he takes a strong preparation of Banisteriopsis Caapi alone. The Makunas of the Popeyaca report that, in difficult cases of diagnosis, the medicine man will add a few crushed leaves of a tree which is abundant along the flood-banks of the river: gway-ee-ga-mo-yoo-he-reé (‘‘tree of the gill of fishes’’). This tree, represented by Schultes & Cabrera 15556, has been determined by Dr. R. W. Woodson as probably Malouetia Tamaquarina. The leaves contain an abundance of sticky, white latex and, as the species belongs to the A pocynaceae, possibly are poisonous in large doses. Indeed, there are persistent reports in the upper Amazon that the bones of the pa- juil are, at the time when Malouectia Tamaquarina sets fruit (which this bird is said to eat), highly poisonous to dogs which may eat them (86). I have never witnessed the use of Malouetia and was not able to corroborate the report with those who understand the properties of me- dicinal plants either amongst the Makunas or neighbor- ing peoples.

The Indians of the headwaters of the Rio Piraparana cultivate caapi in almost every plot around their houses. Schultes & Cabrera 17209 from a Barasana Indian site on the Cano Teemeefa represents Banisteriopsis Caapi. A cold-water infusion of the rasped bark of this plant to which was added water in which dried tobacco leaves (Nicotiana Tabacum 1.) had been steeped had highly narcotic effects.

Banisteriopsis Caapi is apparently the commonest source of the narcotic caapi drink in adjacent parts of Brazil. It was from Ipanoré on the Rio Uaupés near the Colombian boundary that Spruce collected the type specimen of this species. A recent collection, Ducke 153,

[ 89 |

taken from a plant cultivated in Mandos from material gathered along the Rio Curicuriari, an affluent of the upper Rio Negro, represents Banisteriopsis Caapi.

During a year’s stay in the upper Rio Negro and its affuents in Brazil in 1947, I heard, on several occasions, reference to two kinds of caapi. As has been pointed out in detail above, it was from this region that Spruce re- ported a second kind of caapi, known locally as caapi- pinima, which he suggested might be referable to the apocynaceous vine Prestonia amazonica (Hlaemadictyon amazonicum). And it was likewise in this region that Ikoch-Griinberg found that the Tukanos distinguish two species of caapi, for only one of which (Banisteriopsis Caapi) he ventured a determination.

It was my good fortune in 1948 to be able to witness the preparation of and to take a narcotic drink amongst the nomadic Maku Indians of the Ira-Igarapé, an affluent of the Rio Tikié which flows into the Rio Negro below Ipanoré. Specimens (Schultes & Lopez 10184) taken from the flowering vine, from the bark of which a cold-water infusion was made without the admixture of other plants, were found to represent an undescribed species of a mal- pighiaceous genus allied to Banisteriopsis— Tetrapterys methystica R. K. Schultes (87), The beverage prepared from Tetrapterys methystica has a yellowish hue, quite unlike the coffee-brown color characteristic of all prep- arations of Banisteriopsis which I have seen or taken. A small amount of stem material for chemical study which I gathered from this wild vine was lost in the overturning of my canoe, so nothing, unfortunately, can be stated concerning the chemical nature of the plant. ‘wo impor- tant points, however, should be emphasized in connection with this discovery: (1) T'etrapterys methystica may pos- sibly represent the second kind of caapi reported by Spruce and Koch-Griinberg, and it might be that the

[ 40 |

epithet caapi-pinima (‘‘painted caapi’’) alludes not to ‘‘painted leaves but to the unusual yellowish hue of the drink prepared from it; and (2) the tentative identifica- tion as ‘‘possibly Tetrapterys’’ of the sterile material sent to Kew by Wyndham from the Caqueta of Colom- bia would seem to be strengthened by the employment, even at such a distance, of a species of T'etrapterys in the elaboration of a drink with proven narcotic properties.

SUMMARY

1. The narcotic drink known in the western Amazon- ian regions as caapi (Brazil and Colombia), yayé (Colom- bia) and ayahwasca (Ecuador, Peru and Bolivia) is made basically from the same or closely related plants of the Malpighiaceae. It is probable that the Indian names natema, nepe and pinde are synonymous with the three more commonly used names mentioned above.

2. The most widely employed species are members of the genus Banisteriopsis. The species most frequently used in Brazil, easternmost Colombia and much of the Amazon basin of Peru and Bolivia is Banisteriopsis Caapi, but B. Rusbyana seems also to be utilized in easternmost Colombia. In the westernmost fringe of the Amazon basin, along the Andean foothills of Colombia, Keuador and Peru, Banisteriopsis quitensis, B. mebrians and B. Rusbyana seem to be the species most preferred.

3. The closely related genus Tetrapterys is employed in Brazil and possibly in Colombia. The only species of Tetrapterys definitely identified as the source of the nar- cotic caapi is T. methystica.

4. Mascagnia psilophylla var. antifebrilis has been in- dicated as one possible source of ayahuasca, but this re- port is open to very serious doubt.

[ 41 ]

5. The identification of yajé as a species of Aristolo- chia is definitely without foundation.

6. Prestonia amazonica (aemadictyon amazonicum) of the Apocynaceae has frequently been named as the source of yaqjyé and caapi. There is little or no reliable evidence that this vine is ever employed, at least as the prime ingredient, in preparing the narcotic drink.

7. The species of Banisteriopsis and Tetrapterys are known to be employed alone and to have narcotic prop- erties when thus used. Two species of Banisteriopsis may also occasionally be utilized together.

8. Non-malpighiaceous plants are known occasionally, but apparently not frequently, to be added as admixtures together with Banisteriopsis in some areas. Prestonia amazonica has been reported to be thus used in Brazil. The solanaceous Datura and the amaranthaceous A/fer- nanthera Lehmanii have been indicated as an added ingredient in Colombia. The apocynaceous Malouetia Tamaquarina may enter into the preparation of the nar- cotic in the Vaupés of Colombia, but this report could not be verified. Several unidentified plants have been mentioned as admixtures in Bolivia, Colombia and Peru.

[ 42 ]

EXPLANATION OF THE ILLUSTRATION

Pirare VI. (Upper). Makuna Indian witeh-doctor gathering stems of Banisteriopsis Caapi (Spruce ex Griseb.) Morton (Schultes & Cabrera 15587) for preparation of narcotic drink. Rio Popeyaca, Com-

isaria del Amazonas, Colombia.

(Lower). Banisteriopsis inebrians Morton (Schultes & Cabrera 19113) cultivated in an Indian garden near Mocoa, Comisaria del Putumayo, Colombia. The rasped bark of this plant, which is referred to as yaje, is added to the leaves of Banisteriopsis Rusby- ana in the preparation of the narcotie drink.

Photographs by Ricuary Evans Scuunres

[ 44 ]

|

ri ATI

EXPLANATION OF THE ILLUSTRATION

Prare VIL. Banisrertopsts Ruspyana (Ndz. ) Morton (Schultes & Cabrera 19112) growing in secondary forest at Mocoa, Comisaria del Putumayo, Colom- bia. The leaves of this plant, which is called chag- ropanga, are added to the rasped bark of Banister- iopsis inebrians to prepare the yajé drink in the Mocoa area.

Photograph by Ricuarp Evans Senurres

A

PLATE

EXPLANATION OF THE ILLUSTRATION

Prare VILL. Makuna witch-doctor under the in- fluence of caapi prepared in a cold-water infusion of the bark of Banisteriopsis Caapi (Spruce ex Griseb.) Morton (Schultes § Cabrera 15587) with no admixtures. Rio Popeyaca, Comisaria del Ama- zonas, Colombia.

Photograph by Guitiermo Cano O,

(48 |

PLATE VIII

10.

i.

12.

LITERATURE CITED

Albarracin, Leopoldo. Contribucién al estudio de los alkaloides del yagé. [Bogota] (1925).

Allen, Paul H. Indians of southeastern Colombia. Geogr. Rev. 87 (1947) 580.

Anonymous. The yajé plant. Times (London) South American Supplement, Aug. 27, 1912.

Anonymous. Toxicologist completes study of Peru’s ““death vine. ”’ Field Mus. News 10: no. 3 (1939) 5.

Baldwin, J[ohn T.]. Banisteria Caapi Spruce: Its chromosomes. Bull. Torr. Bot. Club 73 (1946) 282.

Barriga-Villalba, A. M. Un nuevo alkaloide. Bol. Soc. Col. Cienc. Nat. (1925) 31-36.

Caller Ibérico, Clorinda. Ayahuasca. Bol. Mus. Hist. Nat. Javier Prado 5 (1941) 313.

Chen, A. L. and K. K. Chen. Harmine, the alkaloid of caapi. Quart. Journ, Pharm. & Pharmacol. 12 (1939) 30.

Claes, Florent. Quelques renseignements sur les coutumes des huitotos et des correguajes de Colombie. Bull. Soc. Améric. Belg. (Dec. 1931) 39.

Claes, Florent. Chez les huitotos et correguajes. Bull. Soc. Roy. Belg. Geogr. 56 (1932) 25-51.

Clinquart, E. Contribution A l’étude de la liane yajé et de son

alkaloide. Journ. Pharm. Belg. 8 (1926) 671.

Cooper, John M. Stimulants and narcotics. in Handbook of South American Indians. [F.d. J. H. Steward] Smithson, Inst. Bur. Am. Ethnol. Bull. 143 (1949) 552, map 14.

. Costa, Oswaldo de Al. and Luis Faria. A planta que faz sonhar.

Rev. Fl. Med. 2: no. 10 (1936).

. Costa, Oswaldo de Al. and Luis Faria. Rev. Assoc. Brasil. Farm.

17 (1936) 265.

. Crévaux, J. Voyages dan 1’ Amérique du Sud. (1883) 362, 536.

Elger, F. Helv. Chim. Act. 11 (1928) 162.

[ 51 |

24.

25.

26.

PRS) I

. de Calella, Placido. Apuntes sobre los indios Sionas del Putu-

mayo. Anthropos 35-386 (1944) 749.

. de Szyszlo, Vitold. La naturaleza en la América ecuatorial. (1955)

250.

Ducke, Adolfo. O cabido Para. Arqu. Serv. Forest. 2(1943) 13.

Ducke, Adolfo. Plantas de cultura pre-colombiana na Amazonia brasileira. Notas sobre as especies ou formas espontiéneas que supostamente lhes teriam dado origem. Bol. Téen. Instit. Agron. Norte, no. 8 (1946) 5.

Fabre, René. Quelques plantes médicinales de 1’Amérique La- tine: leur utilisation thérapeutique. Rev. Gen. Sci. Pures Appl. 62 (1955) 49.

Fischer Cardenas, G. Estudio sobre el principio activo del yagé.

[Unpublished thesis, Univ. Nac., Bogota. } (1923).

23. Fosberg, F. R. Principal economic plants of tropical America.

in Plants and Plant Sciences in Latin America. [Kd. F. Verdoorn] (1945) 28,

Gagnepain, F, Observations sur le yajé. Rev. Bot. Appl. Colon. 10 (1930) 293.

Garcia-Barriga, Hernando. Conferencias de botdénica médica [Mimeographed] (1950) 108, 151-152.

Hammerman, A. F. Bull. Appl. Bot. Gen. Pl. Breed. (1929) 165,

Hammerman, A. F. Le yagé en Amazonie. Rev. Bot. Appl. Agric. Colon. 10 (1930) 600,

Hardenberg, W. EK. The Putumayo, the devil’s paradise. (1913) 86. Hartwich, C. Die menschlichen Genussmittel. (1911).

Henry, Thomas Anderson. The plant alkaloids. Ed. 4 (1949) 488.

Herrera, Fortunato L. Sinopsis de la flora de Cuzco. 1 (1941) 288.

Hesse, Erich, Narcotics and drug addiction. (1946) 112.

Hill, Albert F. Economic botany. Ed. 1 (1937) 294; ed. 2(1952) 283, 511, t. 143.

Hoehne, F. Plantas e substancias vegetais t6xicas e medicinais do Brasil. (1939) 171.

[52 ]

38,

39;

40.

41.

42,

Or to

Karsten, Rafael. Beitrige zur Sittengeschichte der siidamerik- anischen Indianer. II. Berauschende und narkotische Getranke unter den Indianern Siidamerikas. Act. Acad. Aboensis Hum. Abh. 1 (1920) 39.

Karsten, Rafael. The civilization of the South American Indians. (1926) 324 ff., 473. Karsten, Rafael. The head-hunters of western Amazonas. (1935) 343-346, 432-451.

Keller, O. and Franz Gottauf. Uber einigen in Heimatlande arzneilich benutzte bolivianische Drogen. Arch. Pharmaz. 267 (1929) 373.

Koch-Griinberg, Theodor. Zwei Jahre unter den Indianern. 1 (1909) 298.

Koch-Griinberg, Theodor. Vom Roraima zum Orinoco, 1 (1917) 324,

Le Cointe, Paul. A Amazonia brasileira III. Arvores e plantas uteis (1934) 70, 471.

Lewin, Louis. Sur une substance enivrante, la banisterine, extraite de Banisteria Caapi. Comptes Rend. Acad. Sci. 186 (1928) 469.

. Lewin, Louis. Untersuchungen iiber Banisteria Caapi Spr. Arch. f.

exp. Pathol. Pharmakol. 129 (1928) 133.

. Lewin, Louis. Arch. f. exp. Pathol. Pharmakol. 129 (1928) 324.

Lewin, Louis. Gifte und Vergiftungen. Ed. 4 (1929) 687.

Lewin, Louis. Phantastica— narcotic and stimulating drugs. (1981) 140.

. Macbride, Francis J. Flora of Peru. Field Mus, Nat. Hist. Publ.

288, Bot. Ser. 11 (1931) 34.

. Macbride, Francis J. Flora of Peru. Field Mus. Nat. Hist. Publ.

653, Bot. Ser. 13 (1950) 821 ff.

. MacCreagh, Gordon. White waters and black. (1926) 373-384.

Magelli, José M. Carta sobre las misiones dominicanas del Oriente. Carta sexta. La misién de Macas. (1890) 17.

Manske, R. H. and H.L. Holmes. The alkaloids — chemistry and physiology. 2 (1952) 393.

Maxwell, Mary Mitchell. Caapi, its source, use and possibilities. [Unpublished ms. ] (1937).

[ 53 |

Or

Lon) ©

61. 62.

66.

67.

68.

~l

McGovern, William Montgomery. Jungle paths and Inca ruins. (1927) 262.

Michiels, M. and E. Clinquart. Sur des réactions chimiques d’ identification de la yagéine. Bull. Acad. Roy. Méd. Belg., s. 5, 6 (1926) 19.

Moller, Knud O. Rauschgifte und Genussmittel. (1951) 78.

Mors, Walter B. and Perola Zaltzman. Sdbre o alkaléide de Ban- isteria Caapt Spruce e do Cabi paraensis Ducke. Bol. Inst. Quim. Agric. no, 34 (1954) 17.

Morton, C. V. A new Banisteria from Brazil and British Guiana. Proc, Biol. Soc. Wash. 43 (19380) 157.

Morton,C. V. Notes on yagé, a drug plant of southeastern Colom- bia. Journ, Wash. Acad. Sci. 21 (1931) 485 ff.

Niedenzu, F. Uber die Stammpflanzen des Yageins. Pharmaz. Zeit. 73 (1928) 141.

. O'Connell, F. D, and E. V. Lynn. The alkaloids of Banisteriop-

sis inebrians Morton. Journ. Am. Pharm. Assoc. 42 (1953) 753. Orton, James. The Andes and the Amazon. (1871) 171.

Pardal, Ramon. Las drogas estupefacientes e ilusidgenas del indio americano, Rev. Geogr. Amer. 8 (19386) 1.

Pardal, Ramon. Medecina aborigen americana. Humanior, Sece. C, 3 (1937) 2783.

Penna, Meira. Diccionario brasileiro de plantas medicinais. Ed. 8 (1946) 55,

Pérez-Arbelaez, Enrique. Plantas utiles de Colombia. Ed. 1 (1985) 44; ed. 2 (1947) 826, 334; ed. 3 (1956) 460.

Perrot, KE, and R. Hamet. Le yagé, plante sensorielle des indiens de la région amazonienne de 1’ Equateur et de la Colombie. Comp- tes Rend. Acad. Sci. 184 (1927) 1266.

Perrot, E. and R. Hamet. Yagé, ayahuasca, caapi et leur alca- loide: télépathine ou yagéine. Trav. Lab. Mat. Méd. Pharm. Galén. 18: pt. 2 (1927) 1.

Pynaert, L. Florent Claes — botaniste belge, explorateur. Publ.

Jard. d’Agrément, Bull. Mens. Doc. Bot. Hort. d’Hort. Prat. Régions temp. trop. (July 1937) 45.

Reinburg, P. Contribution A ]’étude des boissons toxiques des indiens du Nord-ouest de l’Amazone, l’ayahuasca, le yajé, le

[ 54 |

80.

81.

82.

88.

huanto. Journ. Soc. Améric. Paris, n.s., 13 (1921) 25-54, 197— w16;

Reko, Victor A. Ayahuasca, der Trank der grauenhaften Triume. Heil-Gewurzpfl. 15: pt. 3 (1982-83) 185.

Reko, Victor A. Magische Gifte — Rausch- und Betiubungsmittel der Neuen Welt. (1936) 89.

Reutter [K.]. Du yagé ou aya huesca. Schweiz. Apotheker Zeit. 25 (1927) 289.

Rivet, Paul. Les indiens Colorados. Journ. Soc. Améric. Paris 2 (1905) 201.

Rivet, Paul. Contributions 4 étude des langues Colorado et Cay- apa. Journ. Soc. Améric. Paris 4 (1907) 49.

Rivet, Paul. Les indiens Jivaro. L’Anthrop. 18 (1907) 592. Robinson, C. B. in Small: North American Flora 25 (1910) 131.

Rocha, Joaquin. Memorandum de viaje (regiones amazonicas), El Mercurio (Bogota) (1905) 43.

Rouhier A[lexandre]. Le yagé: plante télépathique. Paris medi- cal, partie para médicale 52 (1924) 341.

Rouhier, A[lexandre]. Les plantes divinatoires. Rev. Métapsych. (1926) 325-331.

Rusby, H. H. South American exploration, Journ. Am. Pharm, Assoc. 12 (1928) 54.

Rusby, H. H. The aboriginal uses of caapi. Journ. Am, Pharm. Assoc. 12 (1923) 1123.

Rusby, H. H. The pharmacodynamics of caapi. Journ. Am. Pharm. Assoc. 13 (1924) 98.

Safford, W. E. Narcotic plants and stimulants of the ancient Amer- icas. Ann. Rept. Smithson. Inst. 1916 (1917) 411-413.

Sandeman, Cristopher. Thyme and bergamot. (1947) 24.

. Schery, Robert W. Plants for man. (1952) 288. . Schultes, Richard Evans. Plantae Austro-Americanae VIII. Bot.

Mus. Leafl. Harvard Univ. 16 (1953) 90.

. Schultes, Richard Evans. Plantae Austro-Americanae IX. Bot.

Mus. Leafl. Harvard Univ. 16 (1954) 202, t. 30, bottom figure.

Seil, Harvey A. and Earl B. Putt. The chemical examinination of caapi. Journ. Am. Pharm. Assoc. 13 (1924) 99.

[ 55 |

100,

101,

102.

103.

104.

Simson, Alfred. Travels in the wilds of Ecuador. (1886) 196.

. Spruce, Richard. Notes of a botanist on the Amazon and Andes.

[Ed. A. R. Wallace] 2 (1908) 414-425.

. Taylor, Norman. Flight from reality. (1949) 175.

. Tyler, Charles D. The river Napo. Geogr. Journ. 3 (1894) 480.

Uribe Uribe, Lorenzo. Botanica. Ed. 3 (1948) 210.

. Villarejo, Avencio. Asi es la selva. (1943) 121,

. Villavicencio, Manuel. Geografia de la Reptblica del Ecuador.

(1858) 371.

. von Hagen, Victor Wolfgang. Scarlet men. Nat. Hist. 40 (1937)

389,

. von Hagen, Victor W[olfgang]|. Realm of the Incas. (1957) 110.

. von Martius, Carl F. P. Nomina plantarum in lingua tupi. Bull.

K. Bayer. Acad. Wissensch. (1858) 1-6.

. von Martius, Carl F.P. Zur Ethnographie Amerikas zumal Brasil-

iens. (1867) 388. Whiffen, Thomas. The northwest Amazon, (1915) 139.

White, O. E. Botanical exploration in Bolivia. Brooklyn Bot. Gard. Rec. 11 (1922) 102.

Williams, Llewellyn. The death vine—ayahuasca. Field Mus.

News 2, no. 8 (1931) 8.

Williams, Llewellyn. The woods of northeastern Peru. Field Mus. Nat. Hist. Publ. 877, Bot. Ser. 15 (19386) 257.

Wolfe, O. and K. Rumpf. Uber die gewinnung von Harmin aus einer siidamerikanischen Liane. Arch, Pharmaz. 266 (1928) 188.

[ 56 ]

Cam

BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY

BRIDGE, MAssaAcHUsETTS, JUNE 26, 1957

THE EFFECTS OF CERTAIN GENES ON THE OUTER PISTILLATE GLUME OF MAIZE BY Watton C. GALINAT

A comparison of the effects which various genes have on the structure of a given plant-organ should sometimes reveal which of the loci concerned were involved in the previous evolution of that organ and which of them might, during domestication of the species, contribute to its further evolution. he extreme susceptibility of the outer pistillate glume’ in the maize tribe (Maydeae) and the related tribe 4 ndropogoncae to evolutionary modifi- cations suggests that the genetic variation affecting de- velopment of this organ in maize (Zea Mays 1) may reflect some of its evolutionary changes. The changes in glume structure were drastic during evolution leading to the formation of the cupulate fruit case (Galinat, 1956). This organ evolved from a long foliaceous bract (d’rian- thus spp.) to a shorter, coriaceous structure marked by various types of sculpturing (Manisuris spp. ; Hackeloch- loa spp.) and, finally, to a highly lignified glume which is specialized in shape, texture and plane of divergence so as to bring about a closure of the narrow opening of a cupulate rachis-segment (Muchlaena and Tripsacum).

‘Further mention of the glumes of maize will refer only to the outer pistillate ones.

Thus, we might expect that a study of the various glume phenotypes of modern maize would be revealing as to the past and possible further evolution of this glume dur- ing domestication.

Three dominant or semi-dominant genes which affect principally the outer pistillate glume were included in this study. They are as follows: (1) Papyrescent (Pn gene on chromosome-7); (2) Tunicate (Zw gene on chromosome-4); (8) Vestigial glume (/¢ gene on chrom- osome-1). These genes were studied in heterozygous con- dition after they had been incorporated, by repeated backcrossing, into an isogenic background of a sweet corn (maize) inbred (Purdue 389). Their effects on the histological structure of this glume were examined in material prepared in the following manner: Normal (typical) and variant specimens were fixed at 18 days after pollination. ‘This material was later embedded in paraffin, according to the usual procedure, and then cut in cross-section at 12 uw. The sections were then stained by the standard safranin-fast-green technique. Finally, a projection apparatus was used in making tracings from comparable slides.

Normal glumes. The glumes and other floral bracts of modern maize are so reduced that the mature grain emerges naked above them. Protection for the grain, which is provided by long floral bracts in most other grasses, is supplied by husks (modified leaf-sheaths) borne below the pistillate inflorescence (ear) on a condensed branch (shank).

The normal glume has papery lateral wings (Plate LX, fig. 3) and the entire structure is homologous to a leaf- sheath with sheath-auricles. The glume-sheath, which is the counterpart of the leaf-sheath, may be coriaceous in texture (in many South American varieties) or, more commonly, it may resemble its counterpart in teosinte

58 |

in being corneous or indurate (in teosinte-contaminated varieties from North and Central America). The degree of this induration has been used as an estimate of teosinte- introgression in archaeological maize, after the reliability of the method has been established (Galinat et a/, 1956).

Internally, the mesophyll of the glume is divided into definite specialized regions. The induration is confined to a region of smaller cells extending from the outer epi- dermis to a line delimited by a row of vascular bundles about midway through a cross section (Plate X, fig. 8). As indicated by strong safranin staining, these smaller indurated cells are characterized by an accumulation of lignin in the secondary walls. The remainder of the glume, extending to the inner epidermis, consists of large parenchyma-cells. During final maturation of the glume, this parenchymatous tissue collapses from desiccation. The resulting shrinkage of the parenchyma toward the bundles causes an outline of the vascular system to be revealed as parallel ridges along the inner epidermis.

The vascular arrangement of the normal glume differs slightly from that which occurs in a vegetative leaf- sheath. The glume has smaller bundles and, since this organ is determinative, they converge at the apex. In the leaf-sheath these principal bundles remain parallel as they continue on into the blade. As in the leaf-sheath, the parallel bundles are of two sizes with the larger ones alternating with one or two smaller ones. ‘The small cross connections which anastomose between the parallel bun- dles are confined to the apical region of the glume as compared to a distribution along the entire length of both the blade and sheath of the leat. The glume wings are non-vascular as are the ligules and sheath-auricles of the leaf-sheath. It seems that these vascular differences between glumes and leaf-sheaths are more in the nature of minor modifications.

[ 59 |

EXPLANATION OF THE ILLUSTRATION

Prate IX. The outer pistillate glume of maize, as effected by various genes. Left and right speci- mens show dorsal and ventral views. 1, papyres- cent (Pn); 2, tunicate (Tu); 3, normal; 4, ves- tigial glume (Vg). Twice natural size.

Photograph by Frank Wuire

[ 60 |

IX

PLATE

EXPLANATION OF THE ILLUSTRATION

Prare X. Projection tracings of comparable cross- sectional views of the outer pistillate glume as this organ is affected by certain genes. In each figure the outer (abaxial ) epidermis is uppermost. 1,papy- rescent (Pn); 2, tunicate (7'w): 3, normal; 4, vestigial glume (J/g). About 80 times natural size.

Drawn by Wavron C. GALinat

Ba

e westecte ge 2240 stlyanseeg > ABET | Weta’ o

As in the leaf, stomata occur along both epidermal surfaces. In the glume, however, they are vestigial and obscure along the outer epidermis. Such reduction may result from the extreme lignification of the underlying tissue. Rather well-developed stomata are found on the inner surface directly opposite the larger vascular bun- dles. The fact that the maize-glume has stomata and yet does not have an opportunity to function in photosyn- thesis, seems to emphasize the leaf-like nature of this organ.

The outer epidermis has a glabrous cuticle and may acquire a brown, red or purple coloration from the action of certain genes. Epidermal pubescence is greatly re- duced, except in the lateral wings.

Papyrescent glumes. This dominant mutant is charac- terized by glumes which upon final maturation become papery and similar in texture to that which occurs in lem- mas, paleas and lateral wings of normal glumes, as well as in the ligules of normal leaves (Plate LX, fig. 1). These papyrescent glumes are usually slightly longer than the mature grain. A condition which is superficially similar to the papyrescent character develops in normal ears which are poorly matured. This simulation results from the protrusion of the lemmas, paleas and papery wings from normal glumes above chaffy or small grains.

The papyrescent character is likewise similar or iden- tical with the ‘‘palee sviluppate’’ of Bonvicini (1932) and apparently with the recessive ‘‘semivestidos’’ of Andrés (1950). Our character for papyrescent glumes may have been incorrectly identified by Andrés as a recessive fac- tor since the effects of this semi-dominant gene are not always readily apparent in the heterozygote. It has also been confused by Weatherwax (1954) with half-tunicate, from which, as will be pointed out later, it is histologi- cally distinct. Furthermore, we find a counterpart of the

[ G4 |

papyrescent condition in Sorghum, where it once led to the describing of a new species (Sorghum papyrascens Stapf), an error later corrected by Rangaswami and Panduranga (1986) who correctly identified it as a single gene mutant characterized by defective glume develop- ment. Papery glumes have also appeared in two X-ray induced mutants (seminudoides and subnudoides) of bar- ley (Scholz, 1956). In these two recessive mutants, the floral bracts fragmentize away from the grain during threshing. Ordinarily the floral bracts of barley remain about the grain following threshing, except in the mutant named nudoides. The differences in the terminology of Sholz and Andrés, as applied to their particular mutants, refer to differences in the normal condition of the glumes in barley and maize. The term ‘‘seminudoides”’ (half- nude) applies to the barley mutant because normally the threshed barley grain is completely covered, while ‘‘semi- vestidos”’ (half-clothed) is better for the maize mutant because normally the maize grain is exposed.

Papyrescent glumes in maize consist largely of non- specialized parenchyma-cells which, at 18 days from pol- lination, are large and fleshy, being swollen with water (Plate X, fig. 1). During final maturation of the ear, dessication causes these parenchymatous glumes to shrink to athin, almost transparent condition with the vascular bundles becoming prominent ridges (Plate LX, fig. 1). There is little contraction in length and width, however, and the glumes remain partially covering the surface of the mature grain. Finally, they become papery and brit- tle and are distinctly different from those glumes which protect the caryopses of other grasses.

The epidermal layers of the papyrescent glume ap- proach those of the normal glume in regard to cell size, but differ from normal in being less lignified and in lack- ing stomata. Pubescence is usually confined to the mar-

[ 65 ]

ginal part of the glume. The vascular bundles are es- sentially normal in structure, although they are not so closely spaced.

The succulent nature of immature papyrescent glumes often encourages the destructive activities of ear-rotting fungi. This suggests that a hardening of these glumes, such as is caused by teosinte introgression, may promote resistance to pathogenic fungi.

The gene which produces this defective character has been previously designated as pseudopod (Pp) (Galinat and Mangelsdorf, 1955), but it now seems more desirable to change its name and symbol to papyrescent (Pn) (Galinat and Mangelsdorf, 1957). This change will call attention to its final papery character and its similarity to the “‘papyrascens’’ character of Sorghum, and will avoid confusion with the symbols for heterozygous peri- ‘arp color (Pp) as well. Our Px gene was originally iso- lated from a Peruvian variety of maize (Mangelsdorf, 1948). It is located close to the bd (branched-silkless) gene near the end of the long arm of chromosome-7. Its linkage relations will be discussed in more detail in a future paper.

Tunicate glumes. The highest tunicate allele (7'w) from the series of multiple-alleles at the T’u-tu locus was chosen in order to study, in accentuated form, the effects of genes at this locus. Weaker alleles at the tunicate locus have intermediate effects on the length, shape and texture of the glume (Mangelsdorf, 1948). Further dis- cussion of tunicate glumes will refer to the phenotype of the strongest tunicate allele (Plate LX, fig. 2).

The glumes of tunicate maize are like those of most other grasses in being long enough to enclose the grain and in being foliaceous. They differ therefore, in length and texture from the pistillate glumes characteristic of the American Maydeae, including normal maize, and of

| 66

Manisuris spp. in the Andropogoneae. In the evolution- ary sequence leading to formation of the cupulate fruit case, tunicate glumes of maize are similar to the glumes of Mlyonurus tripsacoides (Galinat, 1956). Anatomically, tunicate glumes have all of the charac- teristics of a typical leaf-sheath, but on a reduced scale (Plate X, fig. 2). The close similarities of these organs may be enumerated as follows: (1) The inner (adaxial) and outer (abaxial) surfaces of tunicate glumes and leaf- sheaths are about equally pubescent with soft trichomes. This condition differs from that of the leaf-blade where pubescence is located chiefly on the upper (adaxial) epi- dermis and from that of the normal glume where the trichomes are confined to the lateral wings. (2) Both tunicate glumes and leaves (including sheath and blade) are herbaceous in texture because their mesophyll lacks the region of lignified cells which characterizes the nor- mal glume. (3) Stomatal development in tunicate glumes resembles that of the leaf-sheath and differs from the condition found in the normal glume. ‘Tunicate glumes and leaf-sheaths have well-developed stomata on both surfaces although they are more frequent and functional on the outer epidermis. As noted previously, if well- developed stomata occur in the normal glume, they are located on the inner surface. (4) The vascular system of tunicate glumes is more similar to the venation of leaves than it is to that of normal glumes. In both tunicate glumes and leaves, the anastomosing venation connects the principle bundles at irregular intervals along their entire length rather than just in the distal region as in the normal glume. (5) The structure of tunicate bundles is frequently identical to that of leaf bundles in contrast to the more reduced veins of normal glumes (Plate X, figs. 2 and 3). All of these anatomical comparisons sup- port the contention that the tunicate locus controls evo-

[ 67]

lution from the primitive foliaceous state which is typi- cal of glumes in general to the highly specialized glumes of modern maize.

Vestigial glumes. The vestigial glume (Vg) mutant re- ported by Sprague (1989) is of particular morphological and agronomic significance. Originally this naked- flowered character was thought to be of no economic value because it was associated with pollen-blasting tas- sels. The discovery of modifying genes which usually permit abundant pollen production in Vg tassels, aroused interest in the utilization of the Ve character to improve the structure of the sweet corn (maize) ear (Galinat, 1951, 1953). It became apparent, however, that in order to insure pollen-production in inbred lines grown under various environmental conditions, it would be necessary to have tassels with glumes of normal length on Vg plants bearing glumeless ears. This desired Vg pheno- type became a reality after a certain weak tunicate allele derived from the race ‘‘Chapalote’’ was combined with the Vg gene (Galinat, 1955, 1957). As a result, the agronomic development and testing of sweet corn hybrids with this new Vg phenotype is now in progress.

Vg glumes deviate from normal in the opposite direc- tion from the deviation of tunicate glumes (Plates [X and X, figs. 2, 3, 4). In comparison to normal ones, Vg glumes are thicker, shorter and have increased lignifica- tion, while 7’ glumes are thinner, longer and _ have re- duced lignification (lacking the lignified region in the mesophyll). The progressive thickening of glumes in this series of three types (7'u; normal; Vg) appears to result largely from the development and proliferation of many small lignified cells in a region just under the ab- axial epidermis. The smaller the cell in this region, the thicker the deposit of lignin in the secondary wall (Plate X1).

[ 68 |

The accumulation of this lignin is known to be second- ary, in that it occurs after the glumes have already at- tained optimum length and during the development of the caryopsis. If fertilization does not occur, then the glumes remain non-lignified and finally become papery. If only a portion of the ear is fertilized, then only those glumes which are either directly associated with fertilized ovaries or indirectly associated by being adjacent to spike- lets with fertilized ovaries will become lignified. This im- mediate effect of fertilization in stimulating lignification of maternal tissue occurs also in teosinte and T’ripsacum. Also the cupule of maize, as well as its counterpart in these close relatives, is included in the same metaxenial phenomenon.

Certain features of the vestigial glume syndrome seem to be correlated with morphological homologies. The effect of the Vg gene in reducing glume size appears to be accomplished largely by suppression of the lateral wings (probably sheath-auricles), with the remaining por- tion representing a small, highly lignified sheath (com- pare figs. 8 and 4in Plate IX). In addition to the lateral wings of the glumes, the /’2 gene also causes, under cer- tain conditions, a reduction of the lateral wings of the cupules and the ligules of the leaves; this last-named effect was noted first by Laughnan (1956). Glume wings and ligules have at least two features in common in that both of them are papyraceous in texture and are situated as erect prolongations of the sheath, or its homologue, adjacent to the actual or theoretical insertion-point of the blade. Furthermore, these structures appear to be homologous, as is suggested by the vegetative leaves of certain grasses (4 mmophila arundinacea) with blades that are narrower than their sheaths. In such grasses, the lateral portions of the ‘‘ligule’’ elongate and develop as sheath-auricles with the same textural characters as the

| G9 |

EXPLANATION OF THE ILLUSTRATION

PLare XI, A few contrasting cell types from a cross-sectional view of the outer pistillate glume of an ear with the vestigial glume character. Note that the smaller cells have a higher degree of lig- nification in the secondary walls (the stippled area). About 500 times natural size.

Drawn by Warron C. GaLinat

[ 70 ]

| EG

PLATE

sheath, whereas the portion which is adjacent to the blade is reduced and papery. A similar phenomenon occurs in the lemmas of many grasses (4vena and Bromus) where the blade is modified into a narrow awn.

The effect of the /¥@ gene in reducing the size of the cupule wings is not necessarily comparable to its effects on either glume wings or ligules. Rather the small size of Ie spikelets may decrease their capacity to depress the adjacent rachis-internode during youth and thereby result in shallow cupules.

Discussion

The features of Tu and V2 glumes differ from normal in Opposite ways and in doing so they reflect glume-ty pes involved in the early evolution of maize and T'ripsacum in the American Jaydeae. Teosinte is thought to have developed later from the hybridization of maize and Tripsacum (Mangelsdort and Reeves, 1939). The folia- ceous character of tunicate glumes is typical of the An- dropogonaceous grasses (as in HNlyonurus tripsacoides). Also it is a starting point from which the glumes of maize, Tripsacum, and possibly Manisuris could have evolved (Galinat, 1956). In the differentiation of T?ip- sacum and Manisuris, the glumes underwent reduction and lignification or sculpturing while the paired grain- bearing spikelets were reduced to singleness. Maize evolved in a separate direction which was controlled largely by mutation at the Z'w-tw locus (Mangelsdorf, 1948). Nevertheless, a latent ability to evolve a Tripsa- coid-type of outer glume might have been retained in modern maize and finally expressed as the /g mutant. We conclude that in regard to this series (7; normal; Vg), the structure of tunicate glumes indicates that the Tu-tu locus may have been involved in previous evolu- tion of the maize glume and that the mutation to the

vestigial glume condition demonstrates a potential in maize to evolve a Tripsacoid-type of glume.

The evolutionary changes resulting in shorter glumes during the domestication of maize could hardly have in- volved mutation at the ’g—vg locus. The Vg gene is a rare type of dominant mutation which seems to be of recent origin. Furthermore, it is not known to be part of a multiple-allelic series such as has been reported for the Tu-tu locus. There is also strong morphological evi- dence that the Vg gene has not been active in maize evolution. If such V’g activity had occurred, then it would be revealed by a characteristic reduction of the lateral wings of glumes. On the contrary, these wings are well-developed and papery in both modern and archaeological maize. In teosinte and Tripsacum, how- ever, the glume wings are somewhat reduced and highly lignified.

In the series Tu; normal; Vg, decreases in glume- length are accompanied by increases in the thickness of the outer lignified zone. Since lignification of this outer zone of the glume occurs at the same time as kernel- development, there may be competition for the available energy. In the tunicate series of alleles (Mangelsdorf and Mangelsdorf, 1957) some of the energy conserved from shorter glumes is apparently diverted to increased grain production in spite of an increase in glume-lignification. Experiments are now in progress to determine the effect of the Vg gene on yield of mature grain.

Pn glumes exemplify a condition which does not fit into this pattern of variation and, notwithstanding the suggestion of Andrés (1950), could scarcely have exer- cised a protective role at any stage in the evolution of maize. The only possible evolutionary counterpart of papyrescent glumes in other grasses appears to be repre- sented by a certain degenerate change found in several

[73 |

varieties of Sorghum (Rangaswami and Panduranga, 1936). A papery condition which is apparently similar to that of papyrescent maize has been produced by two X-ray induced mutations in barley (Scholz, 1956).

ACKNOWLEDGMEN'TS

During the preparation of the manuscript, many help- ful suggestions were made by Professor P. C. Mangels- dorf of Harvard University. Professor Angelo Bianchi of Pavia University helped in the preparation of some of the slides.

LITERATURE CITED

Andrés, J. M., 1950. Granos semivestidos, restos de un character ancestral del maiz. Rev. Argentina Agron. 17: 252-256.

Bonvicini, M., 1932. Sulla ereditarieta di una anomalia del maiz. L’Italia Agricola 69: 3-9.

Galinat, W. C., 1951. Glumeless hybrid sweet corn. Jour. Hered. 42: 114-116.

——, 1953. Resistance to blasting of Vg tassels. Maize Gen. Coop. News Letter 27: 76.

, 1955. Maize with vestigial-glume ears and normal tassels. Maize Gen. Coop. News Letter 29: 27.

——, 1956. Evolution leading to the formation of the cupulate fruit case in the American Maydeae. Bot. Mus. Leafl. Harvard Univ. 17: 217-239.

———, 1957. The effect of weak tunicate alleles on the expression of

the Vg gene. Maize Gen. Coop. News Letter 31: 68.

Galinat, W. C., and P. C. Mangelsdorf, 1955. Pseudopod, a possible allele of vestigial glume. Maize Gen. Coop. News Letter 29: 26.

——, 1957. Papyrescent maize. Maize Gen. Coop. News Letter 31: 67.

Galinat, W. C., P. C. Mangelsdorf, and L. Pierson, 1956. Estimates of teosinte introgression in archaeological maize. Bot. Mus. Leafl. Harvard Univ. 17: 101-124.

Laughnan, J. R., 1956. Effect of Vg on development of the ligule. Maize Gen. Coop. News Letter 30: 67.

Mangelsdorf, P. C., 1948. The role of pod corn in the origin and evo- lution of maize. Ann. Mo. Bot. Gard. 35: 377-398.

and R. G. Reeves, 1939. The origin of Indian corn and its rel- atives. Texas Agric. Exper. Sta. Bull. 574.

Mangelsdorf, P. C., and H. P. Mangelsdorf, 1957. Genotypes in- volving the J'u-tu locus compared in isogenic stocks, Maize Gen. Coop. News Letter $1: 65-67.

75 |

Rangaswami Ayyanger, G. N., and V. Panduranga Rao, 1986. Sor- ghum papyrascens Stapf. Jour. Indian Bot. Soc. 15: 139-142.

Scholz, F., 1956. Mutationsversuche an Kulturpflanzen V. Die Vere- bung zweier sich variabel manifestierender Ubergangmerkmale von bespelzter zu nackter Gerste bie réntgeninduzierten Mutanten. Die Kulturpflanze 4: 228-246.

Sprague, G. F,, 1939. Heritable characters in maize 50—Vestigial Glume. Jour. Hered. 30: 143-145,

Weatherwax, P., 1954. Indian corn in old America. The Macmillan Co., New York, 253 pp.

BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY

CAMBRIDGE, MAssACHUSETTS, JANUARY 27, 1958 VoL. 18, No. 3

ORCHIDACEAE NEOTROPICALES IV NoreEs ON ‘THE GENUS Caularthron Rar. BY Ricuarp Evans SCHULTES

CAULARTHRON was recognized by Rafinesque as a generic concept distinct from EMpidendrum as early as 1836. It was treated as aspecial section of Mpidendrum by Lind- ley in 1841 under the name Diacrium. Section Diacrium was raised to generic status by Bentham in 1881. This is the name which has been applied to the genus by most subsequent orchidologists.

Were the genus large or had it become important in horticulture, the resumption of an earlier and unfamiliar name might be unfortunate; and the best procedure might be an attempt to include Diacrium amongst the officially conserved generic names. I believe that this procedure should be reserved for names of large genera of great economic or horticultural importance, where the abandonment of a long and well established epithet would lead to extreme confusion. Therefore, | reeommend the substitution of Caularthron for Diacrium.

Our herbaria have not had abundant material of Cau/- arthron, nor does the genus appear to have become a wide favorite in horticulture. In great part owing to these circumstances, it has not been so thoroughly understood as we might wish. Recent collections of Caularthron in Middle America, northern South America and Trinidad

[ 77 ]

and ‘Tobago have added somewhat to our knowledge of the range and variability of the genus, but there still re- mains much to clarify.

In preparing the orchid section for the Fora of Trini- dad and Tobago, 1 have had to consider critically the concepts which have been known as Diacrium bicornutum and Diacrium indivisum. A study of the available her- barium specimens and of the very superior material pre- served in alcohol and sent in recently by Dr. Wilbur G. Downs and Dr. T. H. G. Aitken of Port-of-Spain, Trin- idad, made it early apparent that the fundamental prob- lems involved could not satisfactorily be handled without an examination of the generic concept as a whole. The present paper embodies the results of that examination.

I wish to thank the officials of the following herbaria for making available material entrusted to their care: Reichenbach Herbarium (in the Naturhistorisches Mu- seum in Vienna); Gray Herbarium; Royal Botanic Gar- dens at Kew; New York Botanical Garden: United States National Herbarium; Chicago Natural History Museum and Missouri Botanical Garden. Material from these herbaria has supplemented the large collection of Caularthron preserved in the Orchid Herbarium of Oakes Ames of the Botanical Museum of Harvard University. It is, furthermore, a pleasure for me to thank Mr. G. C. K. Dunsterville for kind permission to reproduce two carefully executed drawings prepared for his forthcoming book of illustrations of Venezuelan orchids.

When Bentham described Mpidendrum bicornutum in 1834 on the basis of material from Trinidad, he stated that he had consulted Lindley concerning its generic status, and had received the opinion that it ‘‘is certainly anew species; but I think it cannot be separated from Kpidendrum. The only distinction between it and that Genus consists in the labellum being distinct from the

[78 |

column: but you will find various degrees of separation between those parts. . .”’ in several species **. . . which nobody can doubt are genuine Lpidendra... Should you, however, be of opinion that it nevertheless must form a new Genus, its character will have to depend upon the large size of the peta/s and the slight adhesion of the sepals at their base.*” Seven years later, Lindley erected his subgeneric concept Diacrium on the basis of Epidendrum bicornutum. 'Then, in 1881, when he raised Diacrium to generic status, Bentham wrote that “‘the peculiar bicornute labellum, neither adnate to nor paral- lel with the column, gives the flower a very different aspect from that of true species of Hpidendrum and can not be included in them without doing violence to the generic character.’

Having now at hand material from a wide geographic range—all with certain diagnostic characters which show no variation—I believe that the best interests of orchid classification may be served by keeping the concept dis- tinct on a generic level. It is obviously very closely allied to Mpidendrum, and some of the differences used to sep- arate it may appear to be superficial. Nevertheless, it would seem that they represent perhaps a definite tan- gential evolutionary trend which ought to be recognized.

There can be no doubt that Caularthron should be used as the name for this generic concept in preference to Diacrium. There is a widespread aversion to the accept- ance of many names proposed by Rafinesque. But I am sure that most botanists will agree with Merrill (Merrill, E.D.: ‘‘Index Rafinesquianus’’ (1949) 26, 29) that “*... where a new Rafinesque name was based wholly on a previously described or illustrated species of some other author, all we have to do to understand the application of the Rafinesque name is to determine the status of the originally described form . . . Thus it seems to be a logi-

[ 79 |

cal course to follow to continue to select the few sound grains from the overwhelming amount of chaff in the Rafinesque technical botanical papers, even if, occasion- ally, some more or less universally used generic or speci- fic name might fall before those proposed by Rafinesque at earlier dates. ”’

No greater precision could be desired than that which we find in Rafinesque’s description of Caularthron as a new genus. He not only published a very adequate de- scription which makes definite references to key morpho- logical characters separating the concept from Lpiden- drum, but, in a day when few botanists even mentioned what we now call types, he named the concept on which he was basing Cauwlarthron by citing Hooker’s Epiden- drum bicornutum and referring to its place of publication. The later name for this same generic concept, Diacrium, was likewise based on H’pidendrum bicornutum. The fact that, in second place under his generic description, Rafinesque made the new name Caularthron umbellatum, citing as basis for it Mpidendrum stenopetalum Hook., does not militate against the wisdom or the necessity of accepting as valid his generic name, especially so since his generic description is obviously based on Kpidendrum bicornutum and not on the very distinet 22. stenopetalum.

Caularthron Rafinesque F 1. Tellur. 2 (1836 [1837] ) 40, pro parte.

E’pidendrum Linnaeus sect. Diacrium Lindley in Hooker Journ. Bot. 3 (1841) 81; Bot. Reg. 81 (1845) Mise. 23; Fol. Orch. (1853) Epidendrum 8; Reichen- bach fil. in Walpers Ann. Bot. 6 (1862) 345.

Miacrium (Lind).) Bentham in Journ. Linn. Soe. 18 (1881) 812; Bentham & Hooker fil. Gen. Plant. 3 (1883) 526; Hemsley Biol. Centr.-Am. Bot. 3 (1883) 221; Warner & Williams, Orch. Alb. 4 (1885) t. 157;

[ 80 |

Rolfe in Gard. Chron. 2, ser. 8 (1887) 44; Pfitzer in Engler & Prantl Nat. Pflanzenfam. II, 6 (1888) 146; Veitch, Man. Orch. Plant. 6 (1890) 78: L’Orchidoph. (1891) 878; Rolfe in Lindenia 7 (1891) 19; Cogniaux in Martius Fl. Bras. 8, pt. 5 (1901) 186; Stein, Orch- ideenb. (1892) 214; Bois, Les Orch. (1898) 74: Ner- chove, Livre des Orch. (1894) 264; Linden, Orch.

Exot. (1894) 751; Cogniaux in Urban Symb. Antill.

6 (1910) 588: Schlechter, Die Orchideen (1915) 214:

Ames ex Standley in Field Mus. Nat. Hist. Publ. 391

(1937) 210: L. O. Williams in Ann. Mo. Bot. Gard.

33 (1946) 878; Hoehne, Icon. Orch. Bras. (1949) 208 ;

P. H. Allen in Orch. Journ. 2 (1953) 185; Ames &

Correll in Fieldiana: Bot. 26 (1953) 405; Foldats in

An. Univ. Centr. Venez. 34 (1953) 276.

Caularthron may be distinguished from Lpidendrum on the basis of the characters set forth in the following key.

A. Labellum vulgo ad columnam variabiliter adnatum et cum ea parallelum, supra numquam protuberantibus elevatis subtus ex- cavatis ornatum, Epidendrum

Aa. Labellum a columna liberum, a columnae basi angulatim patens,

supra protuberantibus duabus elevatis subtus excavatis ornatum Caularthron

E\piphytic or semi-epiphytic herbs with fleshy pseudo- bulbose, solid or frequently insect-hollowed stems. Leaves few, borne at apex of pseudobulbs, rigidly sub- ‘arnose-coriaceous. Inflorescence terminal, simple, race- mose. Flowers few to numerous, showy, short-pedicel- late. Sepals free, spreading, subequal. Petals rather similar to sepals. Lip free and spreading from base of column, 3-lobed; lateral lobes conspicuously tooth-like ; mid-lobe triangular or triangular-lanceolate; disk raised between lateral lobes into 2 hollow, often horn-like cal- luses opening from below to form 2 conspicuous inden-

[ 81 ]

EXPLANATION OF THE ILLUSTRATION

Pirate XIT. Cautarruron psicornutum (Hook. )

Rafinesque. Grown in the greenhouse of F. W. Hunnewell, Wellesley, Massachusetts.

Photograph by Ross W. Baker

Courtesy of American Orchid Society, Inc.

[ 82 |

PLATE NII

EXPLANATION OF THE ILLUSTRATION

Prare XII. Caurarruron prcornurum (Hook. ) Rafinesque. Drawing of Dunsterville 399 trom Ven- ezuela.

Drawn by G. C.K. Dunstrervitir

[ 84 |

XIII

PLATE

tations or pits on lower surface of lip. Column short, with conspicuous fleshy wings; clinandrium oblique, obtuse. Anther terminal, operculate, incumbent, more or less globose, 2-celled; cells divided longitudinally. Pollinia 4, waxy, each with a granular-viscid appendage. Capsule ellipsoidal.

Caularthron: from the Greek, meaning ‘‘jointed stem,’’ undoubtedly in reference to the persistent leaf- bases which lend to the elongated pseudobulbs the ap- pearance of being jointed.

A genus occurring from Guatemala through Middle America, northern South America and Trinidad and Tobago. Two rather variable species are known.

Key to the species of Caularthron

A. Labellum 24-28 mm. longum, profunde trilobatum, lobis usuali- ter sinu conspicuo separatis; lobo mediano propie lanceolato vel elongato-lanceolato; lobis lateralibus ovato-oblongis. Folia ob- longa vel oblongo ligulata, 7-25 (plerumque 12-20) em. longa <1,5-4.2 (plerumque 3-4) em. lata.

1. Caularthron bicornutum

Aa. Labellum 11-14 mm. longum, saepissime inconspicue trilobula- tum vel auriculatum vel aliquando subintegrum, lobis usualiter non sinu separatis; lobo mediano propie triangulari-ovato; lobis lateralibus parvis, vulgo auriformibus. Folia ligulato-lanceolata, 7-19 (plerumque 14-15) em. longa x 0.8-2.8 (plerumque 1-1.8) em. lata. 2. Caularthron bilamellatum

ORIGINAL DESCRIPTION !

138. Caularthron R. (Stem jointed) Diff. Epidendron. Label. libero, ad basi alato glanduloso. Col. libera bialata dentata, anthera termi- nalis 4 pollen. Caul. articulato, vaginato, bifolio, paucifloro—Habit very irregular. Types 2 Sp. 1. Caul. bicornutum. Epid. do Hook. b. m. 3332. Bulbo cauliformis, fol. rad. ligul. retusis, scapo paucifl. label. trilobo, medio lane. ae basi bicorne, petalis ellipt. acutis albis. Trin-

idad...

Caularthron bicornutum (Hook.) Rafinesque F. Tellur. 2 (1836 [1837] ) 41. Hpidendrum bicornutum Hooker in Bot. Mag. (1834) t. 833882. [ 86 |

Diacrium bicornutum (Hook.) Bentham in Journ. Linn. Soc. 18 (1881) 3812.

Diacrium amazonicum Schlechter in Beih. Bot. Cen- tralbl. 42, Abt. 2 (1925) 108; Pabst in Arqu. Bot. Est. Sado Paulo 3 (1955) 125, t. 316.

CoMMON NAME: Virgin Mary; Virgin Orchid (Trin- idad and Tobago).

Pseudobulbs long-cylindric, terete, 10-80 cm. long, 2-6 em. in diameter. Leaves 3-4, thick, oblong, usually quite obtuse to rounded, 6-20 (mostly 15-17) cm. long, 20-50 mm. wide. Flowers few to 20, 5.5-6 ecm. wide; pedicel (with ovary) 8-5 cm. long. Sepals broadly ovate- lanceolate, bluntly short-acuminate, 25-82 mm. long, 15-18 mm. wide. Petals ovate, usually clawed, acutish, 22-28 mm. long, 20-23 mm. wide, upper margin usually with a conspicuous notch. Lip fleshy, as long as petals but narrower, deeply 3-lobate, basally with triangular tooth on each side; lateral lobes elliptic-ovate, rounded; mid-lobe oblong, obtuse-acuminate; disk above with 2 erect, triangular, plate-like, obtuse, hollow projections near middle. Column 14-15 mm. long.

The type of Caularthron bicornutum was collected in Trinidad by Bradford, who wrote on the label: ‘*This most beautiful species is found in the greatest abundance on the coast and on the adjacent islands at the Boca de Moros, Trinidad.—The rocks and trunks of decaying trees are in some places covered with it. This specimen was gathered on Gaspare Island March 12, 1846 on my return from an expedition to the coast of Venezuela. It flowers especially in the early part of the year from Jan- uary to April.”

The culture of Caularthron bicornutum requires a rather warm greenhouse with high humidity (Warner, R. & B.S. Williams: Orch. Alb. 4 (1885) t. 157; Rolfe,

| 87 |

R. A.: Gard. Chron. 2 (1887) 44; Lindenia 7 (1891) 19; Linden, L.; Orch. Exot. (1894) 752). Much difficulty has sometimes been experienced in establishing the plant, as the pseudobulbs, although large and apparently rather tough, seem to damage easily. Once established, it seems to thrive well under cultivation, especially if not dis- turbed. ‘The most recent article dealing with its culture (Schairer, J. Fo: Bull. Am. Orch. Soc. 24 (1955) 106, t. p. 107) states: ‘‘Plants are grown potted firmly in brown osmunda. During autumn and winter, when the plants are in active growth, they require a warm spot (night minimum 63° F if possible), high humidity and good light (as much or more than most Cattleyas) with plenty of water at the roots, and they appreciate supple- mental feeding. At blooming time, they prefer a cooler spot and less light and moisture. They never require a severe rest period after blooming but are kept somewhat dry for a few months before new growth begins. They are native of the West Indies and the Guianas, where they often grow on bare rocks and tree trunks near the sea where they get plenty of moisture and cooling breezes during hot weather. Propagation is by division and they never propagate fast enough to supply the insistent de- mands of your friends. ”’

Caularthron bicornutum and C. bilamellatum occupy rather clearly defined geographic areas: the former is native to South America north of the Amazon River and to Trinidad and ‘Tobago, whereas the latter occurs in Middle America, along the northernmost rim of South America and in Trinidad. It is in Trinidad chiefly that the two species are contingent, but here there appears to be an ecological delimitation of the two species— Cau- larthron bicornutum forms a conspicuous element along and near the sea coast, while C. bilamellatum is known

only from inland districts.

[ 88 |

There are anumber of minor morphological characters which serve to distinguish Caularthron bicornutum from C. bilamellatum. The former is, in general, a much larger and more robust plant than the latter and has flowers approximately twice as large. This difference is variable, and the specimens of Caularthron bicornutum from Brit- ish Guiana are notably smaller in habit and flowers than those from other parts of the range of the species.

Caularthron bicornutum may be recognized at once through its deeply trilobate lip. The mid-lobe is charac- teristically lanceolate or elongate-lanceolate with the api- cal portion either long- or short-acuminate. The lateral lobes, which are ovate-oblong and either obtuse or sub- acute, are usually separated from the mid-lobe by a deep sinus. Caularthron bilamellatum, on the contrary, usually has an inconspicuously trilobulate or auriculate lip, the mid-lobe of which, characteristically triangular-ovate and acute, is not separated from the small, auriform lateral lobes by a recognizable sinus. ‘To be sure, several collec- tions are known from Trinidad which show a somewhat intermediate lobation of the lip, and here there may be evidence of hybridization. When the available material of these two species from their entire geographic ranges is taken into consideration, however, the significance of the shape of the lip as indicative of a possible evolutionary trend may be appreciated.

A convenient character for separating Caularthron bicornutum trom C. bilamellatum—and a character which seems to have been overlooked—is found in the peculiar notching of the upper margin of the petals of C. becornu- tum. This margin, which faces the dorsal sepal, usually has one conspicuous notch situated one-third or one-half of the distance from the base of the petal. In some of the specimens, the notch is sharp and triangular; in others, it is less clearly defined. In all cases, however,

[ 89 |

the presence of this notch causes the marginal area of the petal to be somewhat ruffled or crumpled. I have very rarely observed anything similar to this condition in Caularthron bilamellatum.

In the general shape of the hollow processes or ‘‘horn- like calluses*’ which arise from the disk of the lip and which have been used as the outstanding generic charac- ter, there is evident little variability from specimen to specimen, albeit some variation in relative size may be seen. Similarly, there is no significant morphological difference between these processes in Caularthron bicor- nutum and C. bilamellatum.

Unlike the concept Caularthron bilamellatum, C. bicor- nutum does not have a large synonymy. This is due partly, perhaps, to the greater variability in the former than in the latter species.

In 1925, Schlechter described Diacrium amazonicum from material collected in Brazil. The type material of Diacrium amazonicum is no longer extant, but a study of the type description convinces me that Schlechter specified no differences which, with the material at hand today, we could not easily accommodate in Caularthron bicornutum. Pabst (loc. cit.) determined a specimen (Froes 21541) from the Rio Negro in Amazonian Brazil as representing Diacrium amazonicum and published a diagnostic drawing of the floral parts. This drawing like- wise shows no character which could serve to distinguish the plant from Caularthron bicornutum.

Type cottection: Cult. Hort. Wentwlorth], from Trinidad (Herb. Kew).

Curtivarep: (Herb. Rehb. 898); Hort. Rucker, May 27, 1840 (Herb. Rehb. 899; Herb. Ames 69096a); ‘“‘from the type plant of Gard. Chron. 1887, pt. 2, p. 45, fig. 11,’ Hort. Kew, May 1887, R.A. Rolfe s.n. (Herb. Kew); Botanic Station, Grenada, W.I. ‘‘Orig- inally from Trinidad,’’ April 14, 1906, W. EH. Broadway s.n. (Herb. Ames 7906, 7907).

| 90 |

Brazit: Estado do Amazonas, Mandos, Spengler s.n. (Herb. Rehb. 896); Estado do Amazonas, Rio Negro, December 21, 1945, R. L. Frées 21541 (Record in Herb. Ames of specimen in Herb. Inst. Agron. Norte).

Bririsn Guiana: *‘Flowers white, pink towards the apex, labellum pink with a yellow disk.’’ 1837, R. Schomburgk 429 (Herb. Kew; Herb. Field Mus. 1025283); Banks of Corentyne River near Crealla, September 1879, E. F. im Thurn s.n. (Herb. Kew); Essequibo River, December 1886, Jenman 3590 (Herb. Kew); Jenman 7761 (Herb. Kew); Hort. Kew, May 1889 (Herb. Kew); Mount Roraima, Autumn 1894, J.J. Quelch & F. McConnell 280 (Herb. Kew); C. F. Appun 657 (Herb. Kew); Cuyuni River, islet at the Akaio Falls. “‘Epiphyte at about 15 feet; roots in dense clusters. Fl. shallow, cup-shaped, like an Anthericum, pedicels mauve. Fl]. resupinate. Pets. and sep. pure white, with striations. Stele at base and all labellum spotted with purple. Humps of label. and depression between them and stele yel- low, spotted with purple. Label. otherwise white.’’ November 25, 1929, N. Y. Sandwith 685 (Herb. Kew).

Cotompr1a: Comisaria del Vaupés, Rio Negro, El Castillo (San Fe- lipe). ‘‘Flowers white, sepals delicately pink-tinged. Lip yellow, spotted brown. Very fragrant.’? December 12, 1947, R. EF. Schultes & F. Lopes 9355a (Herb. Ames 67526, 67527).

Tonaco: G. W. Meyer (?) s.n., December 1879 (Herb. Kew); Rox- borough Bay, Military Road, “growing on the stems of logwood tree,’”’ February 2, 1879, G. W. Meyer s.n. (Herb. Kew); Bacolet, ‘*on rocks and trees near the sea, flowers white with a few purplish spots, very fragrant,’ January 20, 1910, W. E. Broadway s.n. (Herb. Gray 4278); Rockley Vale, ““on trees, Virgin Mary,’’ February 20, 1913, W. E. Broadway s.n. (Herb. Gray 4277; U.S. Nat. Herb. 759445); Feb- ruary 19, 1932, D. Fairchild 2930 (U.S. Nat. Herb. 1625959; Herb. Ames 69057).

Trinwwap: Borroughs s.n. (2) (Herb. Rehb. 896); Bradford 1846 Herb. [Hance 5332] Herb. Kew); Gasparee Island, ‘on rocks and trees overhanging the sea,’’ December 30, 1906, W. EF. Broadway s.n. (Herb. Ames 10085, 10086, 10087, 10088, 10089, 68642; Herb. Field Mus. 464601); Maraval, ‘‘on rocks and trees,’” February 8, 1911, W. E. Broadway s.n. (Herb. Kew; Herb. Ames 69058); Moruga sea shore, Lance Mettan. “‘On shrubs, trees and rocks.’* February 9, 1916, W. E. Broadway s.n. (Herb. Trin. 7595; Ames 69924); “‘On a tree,’ May 26, 1918, W. EF. Broadway s.n. (Herb. Ames 22062; Herb N. Y. Bot. Gard.); Manzanilla, ‘‘Flowers white. On a fallen tree,’? March 9, 1921, N. L. Britton & E. G. Britton 2173 (Herb. N.Y. Bot. Gard.; U.S. Nat. Herb. 1198102); Little Gasparee, April 4, 1921, N. L. Britton 2659 (Herb. N.Y. Bot. Gard.); Chacachacare,

[ 91 |

January 5, 1922, W. E. Broadway s.n. (Herb. Trin. 10566, 2 sheets); Herb. Ames 69925); St. Ann’s (Cult.) ‘‘wild on rocks and trees along sea shores,’’ March 8, 1923, W. E. Broadway s.n. (Herb. Field Mus. 549522); Between Balandra and Toco, “‘rocks by shore,’? February 2, 1926, W.G. Freeman s.n. (Herb. Trin. 11519); Balandra Bay, ‘‘on rocks and low trees near the sea shore. Virgin Orchid,’’? February 22, 1931, W. E. Broadway s.n. (U.S. Nat. Herb. 1519971: Herb. Mo. Bot. Gard. 1005390); “*On trees and rocks, sea shore districts: Vir- gin Orchid,’’ February 23, 1934, W. EF. Broadway s.n. (Herb. Ames 40385; Herb. Kew); Quinam Bay, St. Patrick, “‘On trunk of tree, fls. white with faintly speckled throat,’’? January 80, 1946, L. H. Bailey 121 (Herb. Ames 62541); Chacachacare Island, January 20, 1956 [Capt. Mendez] W. G. Downs & T. H. G. Aitken 15f (Herb. Ames Ale. Coll, 3187); Sangre Grande, Rio Grande Forest Tree Station, 14 mile from coast, March 7, 1956, W. G. Downs & T. H. G. Aitken 15g (Herb. Ames Ale. Coll. 3183); Chachachacare Island, scacoast, January 14, 1957 [A. S. Fenwick| W. G. Downs & T'. H. G. Aitken 15c (Herb. Ames Ale. Coll. $204a); Vega de Oropouche, 1 mile from sea, W.G. Downs & T. H. G. Aitken 15a (Herb. Ames Alc. Coll. 3206a).

Venezurta: [Drawing of a flowering specimen] Carabobo, alt. 2500 ft., 1851 (Herb. Rehb. 890; Herb. Ames 69098); Paria Peninsula, Cariaquita, January 16-21, 1911, F. FE. Bond, T. 8. Gillin & S. Brown 40 (Herb. N.Y. Bot. Gard., U.S. Nat. Herb. 1189830); vicinity of Cristobal Colon, January 5—February 22, 1923, W. E. Broadway 337 (Herb. Gray 4275; Herb. N.Y. Bot. Gard. ; U.S. Nat. Herb. 1187925, 1197666); vicinity of Crist6ébal Colon, Avicana, January 5—February 22, 1923, W. FE. Broadway 616 (Herb. Gray 4276; U.S. Nat. Herb. 1197675); April 29, 1941, H. Pittier s.n. (Herb. Ames 68209); [Draw- ing of a flowering specimen] G. C. K. Dunsterville 399 | Arrigo R. s.n. |, Puerto Ayacucho (Herb. Garay 6358).

Caularthron bilamellatum (?chb.f.) R.H. Schultes comb, nov.

E’pidendrum bilamellatum Reichenbach fil. in Walpers

Ann. Bot. 6 (1862) 845.

E’pidendrum bigibberosum Reichenbach fil. loc. cit. 8

(1862) 346.

Epidendrum indivisum Bradford ex Grisebach FI. Brit.

W. Ind. Isl. (1864) 614.

Diacrium bigibberosum (Rehb.f.) Hemsley in Godman

& Salvin Biol. Centr.-Am., Bot. 8 (1883) 222.

[ 92 |

Diacrium bilamellatum (Rehb.f.) Hemsley in Godman

Diacrium indwisum (Bradf. ex Griseb.) Broadway in

Bull. Mise. Inform. Trinidad 2 (1895) 79.

Diacrium bicornutum (Hook.) Bentham var. indivisum

(Bradf. ex Griseb.) Cogniaux in Martius FI]. Bras. 38,

pt. 5 (1898) 188.

Diacrium venezuelanum Schlechter in Fedde Repert.

Sp. Nov. Beih. 6 (1919) 41.

Diacrium bilamellatum (Rchb.f) Hemsley var. Reich-

enbachianum Schlechter loc. cit. 17 (1922) 47, in textu.

Diacrium bwalvatulum Schlechter loc. cit. 19 (1923)

128.

Pseudobulbs subcylindric to long-fusiform, terete, 5-23 cm. long, up to 4 cm. in diameter. Leaves 2-3, ligulate-lanceolate to linear-oblong, obtuse, 5-22 cm. (mostly more or less 15) em. long, 6-25 mm. wide. In- florescence erect, up to 15 em. long. Flowers white or white tinged with pink or lavender, few to numerous, up to about 8 em. wide; pedicel stout, (with ovary) 1.5-2.5 em. long. Bracts triangular, cucullate, acute, up to 5.5 mim. long. Sepals concave, elliptic-ovate, acute or some- times subobtuse and apiculate; dorsal sepal 12-17.5 mm. long, 6 mm. wide; lateral sepals 10-16 mm. long, 6-8 mm. wide. Petals short-clawed, elliptic-obovate to broadly obovate, acute or subobtuse, 10-16 mm. long, 6-9 mm. wide. Lip fleshy, as long as petals, subentire to more or less 38-lobulate; lateral lobes small and auricu- late, sometimes only tooth-like, usually without a sinus; mid-lobe triangular-ovate, apically short- or long-acute ; disk above with 2 large, erect, triangular, plate-like, ob- tuse, hollow projections. Column 8-11 mm. long. Cap- sule up to 2.8 em. long.

The type of Hpidendrum bilamellatum was collected in ‘*Caracas’’ (referring probably to a very extensive area

[ 93 ]

EXPLANATION OF THE ILLUSTRATION

Pirate XIV. CavuiarTuron pirametiatum (Rchbf.) R.E. Schultes. 1, flowering and fruiting plant, one third natural size. 2, flower, approximately natural size. 3, column and lip, side view, almost twice natural size.

Drawn by Gorvon W. DiLLon

[ 94 ]

PuateE XIV

in Venezuela and not specifically to the city), Venezuela, by Wagener and is preserved in the Reichenbach Her- barium on sheet No. 891. This sheet, which is labelled “Epid. bicornutum’’ has, pasted on a card attached to the sheet, three flowers from Fendler 2436, collected near Colonia ‘Tovar in Venezuela in 1856-1857. This Fendler material, representing Caularthron bilamellatum, has been given the number of Herb. Reichenbach 891a in order to distinguish it from the type or Wagener col- lection. There are also, pasted on the sheet with the type, two labels indicating that the collection was made by Linden in ‘‘Nouvelle Granade,’’ but since the original handwritten label is obviously the correct one, we must assume that, as so often happened, the printed Linden labels were glued to the sheet at a subsequent date and undoubtedly in error.

An examination of the type material of Hpidendrum bigibberosum (Herb. Reichenbach 893, 894) and of Reich- enbach’s diagnostic sketches of the floral parts fails to uncover a single character which might serve to distin- guish this collection from the type of 2. bilamellatum. In his original description of the concept Hpidendrum bigibberosum, Reichenbach likewise failed to point out any differences. We, therefore, must reduce Mpidendrum bigibberosum to synonymy under Caularthron bilamella- tum.

For some time, I have been undecided as to what the concept which has been known as Diacrium indivisum (Epidendrum indivisum) really represented. Bradford drew up a description based on one of his Trinidad col- lections, and this was published as Mpidendrum indivisum by Grisebach. It was transferred to Diacrium by Broad- way in 1895. Cogniaux maintained that the concept represented a variety of Diacrium bicornutum and made the necessary nomenclatural adjustment. In 1956, I pub-

[ 96 ]

lished a note indicating my belief that there seemed to be sufficient morphological evidence to maintain it [Dr- aerium indivisum| as specifically distinct from the only other Trinidad representative of the genus. Subsequent- ly, when Dr. Wilbur G. Downs sent me from Trinidad photographs and additional material of the two concepts of Caularthron known to grow on the island, I began to realize that a revision of the genus Diacrium was neces- sary before a clear understanding of the Trinidad material could be expected.

The type specimen of EHpidendrum imdivisum is pre- served at Kew, together with Bradford’s handwritten description of the concept. The type bears the annota- tion ‘‘Herb. Hance 5334.’” On the same sheet with the type there are pasted two inflorescences from plants col- lected in Trinidad and flowered at Kew in May 1889. Rolfe has annotated this collection as ‘‘E/pidendrum bi- cornutum var. cleistogamic flowers=D. indivisuwm.””

According to Bradford’s manuscript description, the type has an undivided lip. In the published description, the lip of the type was stated to be ‘undivided or mi- nutely auricled above the base.”’ The apical part of the lip was described as ‘tacuminatum’”’ in the manuscript and ‘‘subulate-lanceolate’’ in Grisebach’s Fora. Untor- tunately, the type has, at the present time, only two buds and one imperfect flower. We know from later ma- terial, however, that, in Trinidad, this concept is often cleistogamous. It is possible that the type flowers may have been peloric. At least, we do know from the ma- terial now available from Trinidad that the lip is very rarely entire but is most often laterally auriculate at the base or inconspicuously bilobulate. In this, as in other respects, the concept does not depart from Caularthron bilamellatum, of which it is, consequently, here desig- nated as a synonym.

[ 97 |

EXPLANATION OF THE ILLUSTRATION

Prare XV, CauLarruron BILAMELLATUM (Rehb f.) R. EF. Schultes. Drawing of Dunsterville 181 from Venezuela.

Drawn by G. C. K. Dunsrervitte

[ 98 |

PLATE XV

Schlechter’s Diacrium venezuelanum, a record of the type of which is preserved in the Ames Herbarium, shows no character which cannot fall easily within the variability of Caularthron bilamellatum. Diacrium bival- vatulum, likewise, presents, according to Schlechter’s original description, no differences of sufficient impor- tance for the maintenance of it as a distinct species. Dia- cerium venezuelanum and D. bivalvatulum are, therefore, placed in synonymy under Caularthron bilamellatum.

Type cottection: Venezuela, ‘‘Caracas,”’ Wagener s.n. (Herb. Rehb. 891; Herb. Ames 69055).

CoLompia: 1842, Sinclair s.n. (Herb. Kew).

British Honpuras: W. A. Schipp S-482 (Herb. Ames 40443).

Cosra Rica: [Drawing of a flowering specimen and diagnosis of flower] Punta Arenas, February 1909, 4. & C. Brade 1265 (Type of D. bivalvatulum) ; Provincia de Alajuela, El Coyolar, alt. about 240 m. “On tree; bulbs elongate, full of ants.’’ April 1-3, 1924, P. C. Standley 39982 (Herb. Ames 29865); Golfo Dulce, Playa Blanca, sea level, February 25, 1938, M. Valerio 553 (Herb. Field Mus. 893155).

GuatemaLa: Cult. Hort. Shiller, from Guatemala, Skinner 1458 (Typr of FE. bigibberosum, Herb. Rehb. 893, 894; Herb. Ames 24081, 69091); Montufor Flats, February 27, 1939, M. W. Lewis 217 (Herb. Ames 69926); Departamento de Izabal, between Milla 49.5 and Cris- tina, alt. 65-70 m. ‘‘On limbs of tree along wooded margins of prairie. Pseudobulb terete, pale green. Leaves thick, coriaceous, rich olive- green. Stem olive-green with purple suffused above and at nodes. Pedicels purplish. Buds orchid-colored. 8 outer sepals delicate orchid without, white within. 2 petals white. Lip white on upper petal and stigma and white on beak with pale orchid color along edge or the column with more lavendar.’’ J. A. Steyermark 38389 (Herb. Field Mus. 1043049); Departamento de Izabal, Bay of Santo Tomas, be- tween Escobas and Santo Tomas, alt. sea level to2 m. April 13, 1940, J. A, Steyermark 39341 (Herb. Field Mus. 1041145; Herb. Ames 63752); Departamento de Izabal, shores of Lago Izab4l, opposite San Felipe; between San Felipe and mouth of Rio Juan Vicente, alt. 50 m. April 19, 1940, J. 4. Steyermark 39692 (Herb. Field Mus. 1035306).

Panama: Santa Rita Trail, February 27, 1905, J. F. Cowell 160 (Herb. N.Y. Bot. Gard.); Cultivated from Canal Zone, Culebra, alt. 50-100 m., April 8, 1911, H. Pittier 3406 (Herb, Ames 21797); Pa- tifio, southern Darien, ‘‘on cliffs along the beach,’’ February 13, 1912, H. Pittier 5706 (Herb. N.Y. Bot. Gard.; U.S. Nat. Herb.

[ 100 ]

715996); Canal Zone, Balboa. ‘*‘Never flowers a full spray. It is necessary to pick a flower at a time as they show.”’ [rec’d] May 4, 1923, C. W. Powell 67 (Herb. Ames 23965, 23966, 69927; Herb. Ames Ale. Coll. 438; Herb. Mo. Bot. Gard. 955922); Canal Zone, Fort Sherman, January 15, 1924, P. C. Standley 31231 (U.S. Nat. Herb. 1225409); Provincia de Panama, between Matias Hernandez and Juan Diaz, January 21, 1924, P. C. Standley 31944 (U.S. Nat. Herb. 1225418); Provincia de Panama, swamp between El] Jagua Hunting Club on Rio Jagua and El Congor Hill, alt. 2 m., February 10, 1935, A. A. Hunter & P. H. Allen 473 (Herb. Ames 42248); Pearl Islands, Trapiche Island, March 15, 1937, G. 8. Miller 1908 (U.S. Nat. Herb. 1688731); Pearl Islands, San José Island, March 16, 1937, G. S. Miller 1909 (U.S. Nat. Herb. 1688746) Isla Colon, April 1, 1940, HH. von Wedel s.n. (Herb. Mo. Bot. Gard. 1227010); Provincia de Bocas del Toro, vicinity of Chiriqui Lagoon, Old Bank Island, ‘flowers purplish,’? February 15, 1941, H. von Wedel 2100 (Herb. Ames 61530; U.S. Nat. Herb. 1863094); Perlas Archipelago, Gulf of Panama, San José Island (mouth of Mata Puerco), about 55 miles southeast of Balboa, April 12, 1945, J. M. Johnston 703 (Herb. Ames 64953).

Trintpap: Cult. Hort. Trin. from ‘‘Inland Districts’? J. H. H[art] s.n. (Herb. Trin. 5512); Inland Woods, 1896, J. H. Hart] s.n. (Herb. Trin. 5983; Herb. Kew; Herb. Ames 66910, 68215); Brad- ford s.n. [Herb. Hance 5334] (Tyrer of EF. indivisum, Herb. Kew); Cult. Hort. Kew from Trinidad, May 1889 (Herb. Kew); Government House Grounds, June 3, 1907, W.E. Broadway s.n. (Herb. Ames 10736); Government House Grounds. “‘Flowers white,’’ June 22, 1907, W. E. Broadway s.n. (Herb. Ames 10727); Erin, March 27, 1908, W. E. Broadway s.n. (Herb. Ames 10640); Santa Cruz, Feb- ruary 23, 1912, W.E. Broadway s.n. (Herb. Kew; Herb. Mo. Bot. Gard. 918485); St. Augustine, Imperial College of Tropical Agricul- ture, April 15, 1949, N. W. Simmonds 351 (Herb. Trin. 14438; Herb. Ames 66981); St. Augustine, 6 miles from sea, W. G. Downs & T. H. G. Aitken 156 (Herb. Ames Ale. Coll, 3205a); Caigual, about 4 miles from coast, January 24, 1955, W.G. Downs & T. H. G. Aitken 15e (Herb. Ames 67831, 67783); Plain Road, 3-4 miles from sea, January 14, 1957, W. G. Downs & T. H. G. Aitken 15d (Herb. Ames Ale. Coll. 3203a).

Venezurta : Near Colonia Tovar, 1856-1857, A. Fendler 2436 (Herb. Rehb. 891a; Herb. Kew; Herb. Gray 4077); [Drawing of flowering specimen and floral diagnosis] Cult. Hort. K. W. John, flowered June 1904, from Venezuela (Typr of D. venezuelanum, Herb. Ames 69928); Between Valencia and Maracay, January 31, 1918, H. Pittier 7748 (U.S. Nat. Herb. 987846); Rastrojos, near Cabudare, Lara, Decem-

[ 101 |]

ber 1923, J. Saer 116 (U.S. Nat. Herb. 1193211); June 1939, V. Barnes 5923 (Herb. Ames 58216).

ExcLUDED OR UNCERTAIN CONCEPTS

Diacrium bidentatum (Lind/.) Hemsley in God- man & Salvin Biol. Centr.-Am., Bot. 8 (1888) 221.

Hpidendrum bidentatum Lindley Gen. and Sp. Orch.

Pl. (1831) 98.

The type of Mpidendrum bidentatum is preserved in the British Museum. Study of a photograph of the type and diagnostic sketches in the Ames Herbarium indi- cates that this concept cannot be referred to Caularthron, but that, without a doubt, it represents, as has previ- ously been suggested (Williams, L. O. in Ceiba 2 (1951) 174), Hpidendrum Boothianum dl.

Diacrium Ulmckei Arinz/in Mitteil. Inst. Allg. Bot. Hamb. 6 (1927) 419.

This concept was described on the basis of material which flowered in the Hamburg Botanical Garden. The plant had presumably been collected in Guatemala. Since the Hamburg Herbarium has apparently disappeared, we are unable to trace a type, if indeed an herbarium specimen were ever preserved there when the concept was described. After an examination of the description, I rather doubt that the concept can be accommodated in the genus Caularthron.

[ 102 ]

STUDIES IN AMERICAN ORCHIDS Il’

A NEW GENUS FROM THE COLOMBIAN AMAZONIA BY Lestige A. Garay?

TIME and again, the interpretation of orchid genera has caused grave problems, since we often find that many of the species which are assigned to a particular genus are either atypical or obviously do not belong to the concept in question. Recently, I encountered a problem of this nature, which I wish to discuss here.

In 1934, Mansteld described Hybochilus Huebnert, which he based on a collection sent in by Mr. Huebner from the Amazon region of Colombia. At first, it ap- peared to him that this collection might represent a new generic concept, but for some unknown reason he finally decided to include it inthe genus Hybochilus. This spe- cies has subsequently been all but unknown, and its nomenclatorial type was destroyed during the Second World War.

While studying a number of collections from Colom- bia in the Ames Herbarium, I found two undetermined specimens which appeared to be quite peculiar in their

1 Previous numbers of this series were published in the Canadian Journal of Botany, vol. 34: (1956), pp. 241-260, 721-748.

? Assistant Curator of the Herbarium, Department of Botany, Uni- versity of Toronto, Toronto, Canada; at present, a Guggenheim Fel- low at the Orchid Herbarium of Oakes Ames, Harvard University.

[ 103 |

general habit and which immediately suggested Queket- tia, another genus of the subtribe Capanemieae and known only from the Guianas. I consequently investi- gated all of the genera and species of the Capanemieae, and I am now convinced of the identity of my material with that of the concept Hybochilus Huebneri.

According to Schlechter’s system, such features as an incumbent anther and auricles which are a continuation of the clinandrium characterize the Capanemieae. A com- pletely dorsal anther and the presence of auricles on the column which enfold the stigmatic cavity are characters of the monogeneric subtribe Papperitzieae; these char- acters are clearly observable in Hybochilus Huebneri. The structure of the column, however, is much more complicated than in Papperitzia. There are two pairs of lobes: one pair is located in front of the column, with their parallel sides enclosing the stigma, the lower mar- gins being completely connate to form a shallow, cup- like cavity immediately beneath the stigmatic surface; the other pair of lobes terminate the column. At first, these are parallel and enclose the lower portion of the anther. When the anther is removed they open up slight- ly and become spirally twisted. It is hard to state the real function of these lobes, since we have no field obser- vations. It appears, however, that so long as the flowers are not visited by pollinating insects they remain fairly connivent, but after the removal of pollinia they open up slightly. I believe that these auricles protect the vis- cosity of the stigma and pollen-gland from rapid desic- cation.

In view of these dissimilarities, which I consider to be of basic importance in the differentiation of subtribes as well as of genera within the subtribes, I believe that Hy- bochilus Huebneri represents a distinct genus which should be placed in the Papperitzieae.

| 104 |

Polyotidium Garay gen. nov.’

Sepalum posticum asepalis lateralibus omnino liberum, valde concavum; sepala lateralia usque ad medium con- nata, basi leviter saccata. Petala plana, sepalo postico subsimilia. Labellum basi columnae adnatum, non articu- latum; hypochilum late excavatum, epichilum cuneato- flabellatum, planum. Columna apoda, brevis, paulo arcu- ata, apicem versus leviter dilatata, alis quaternis ornata; alae duae inferiores in facie columnae juxta orificium stigmaticum patentes, superiores duae in apice columnae erectae. Orificium stigmae ellipticum, valde concavum. Rostellum valde conspicuum, alte bifidum. Clinandrium posticum, vix rite evolutum, integerrimum. <Anthera dorsalis, dehiscens, in medio columnae inserta, erecta, ovato-oblonga, unilocularis. Pollinia 2, subglobosa, stip- ites longi angustissimi, glandula obovato-triangula. Her- bae epiphyticae, habitu). Quekettiam microscopicam in mentem revocans. Pseudobulbi parvi, unifoliati. Folia lineari-subulata, cornoso-coriacea. Inflorescentiae late- rales, simplices vel pauciramosae. Flores parvi, distincte pedicellati.

Genus monotypica e regione Amazonicae Colombi- anae.

Polyotidium Huebneri (Mansf:) Garay comb. nov.

HAybochilus Huebneri Mansf. in Fedde Repert 36: (1984) 61.

Epiphytica, caespitosa, usque ad 20 cm. alta; radici- bus filiformibus, leviter flexuosis, glabris; pseudobulbis anguste cylindraceis, unifoliatis vaginis 8, scariosis, acutis vel brevissime acuminatis omnino vestitis; folio subuli- formi, carnoso-coriaceo, erecto vel leviter arcuato, 12-19 em. longo, 2-8 mm. in diametro; inflorescentia e basi

* Polyotidium: poly = many; otion =small ear, in allusion to the several auricles present on the column.

[ 105 |

EXPLANATION OF THE ILLUSTRATION

Piate XVI. Potyoriptum Hursnert (Mansf.) Garay. 1, plant, three quarters natural size. 2, lateral sepals, four times natural size. 3, petal, five times natural size. 4, lip, five times natural size. 5, column, lateral view, nine times natural size. 6, column, front view, eight times natural size. 7, pollinia, greatly enlarged. 8, anther, greatly enlarged.

[ 106 ]

PLaTE XVI

pseubobulborum simplici vel pauciramosa, laxe pluri- flora, 9-16 cm. longa; bracteis lanceolatis, ovariis pedi- cellatis multoties brevioribus, 2 mm. longis ; sepalo postico elliptico, valde concavo, 3-nervio, apice acuto, 6-8 mm. longo, 2.5-3 mm. lato; sepalis lateralibus usque ad me- dium connatis, lanceolato-ellipticis, apice acutis et inter se divergentibus, basi leviter saccatis, 3-nerviis, ca. 7-8 mm. longis, supra basin 8.5-4 mm. latis; petalis late ellipticis, acutis, 5-nerviis, 7-8 mm. longis, 3-4 mm. latis; labello sessili, basi columnae adnato, in medio con- stricto, 9-10 mm. longo; parte basilari (i.e. hypochilo) subglobosa, valde concava, ad basin utrinque callo mi- nuto, globoso ornata, epichilo plano, cuneato-flabellato, antice emarginato cum apiculo; disco in medio ad con- structionem callis 2 approximatis donato; columna apoda, 4-auriculata, 5-6 mm. alta; ovario cylindaraceo, pedicello incluso 4-6 mm. longo.

Huebner 226 was originally designated as the type of Hybochilus Huebneri, but it was destroyed during the Second World War. ‘To the best of my knowledge, the two collections in the Ames Herbarium cited below con- stitute our only representation of this concept. Since this species is the basis of a distinct genus, it is essential to select a new nomenclatorial type, for which I hereby designate the excellent collection Schultes & Lopez 9300 (Tyre in Ames Herbarium No. 68524).

Cotomp1A: Comisaria del Vaupés, Rio Negro basin, El Castillo, near confluence of Guainia and Casiquiare. Dec. 12, 1947. Richard Evans Schultes §& Francisco Lépez 9300. Neoryer: (Ames Herbarium No. 68524).—Comisarias del Amazonas y Vaupés: Rio Apaporis, Raudal Jirijirimo (below mouth of Kananari). Quartzite base, 900 ft. alt. Flowers vermillion. Jan. 21, 1952. Richard Evans Schultes & Isidoro

Cabrera 14960 (Ames Herbarium No. 68495).—Comisaria del Am- azonas: La Pedrera. Feb. 1926. G. Huebner 226 (ex Mansfeld).

[ 108 ]

A NEW LEPANTHOPSIS FROM VENEZUELA

BY CHARLES SCHWEINFURTH

LikE many other small, recently segregated genera, Le- panthopsis is now recorded as rather widespread in the American tropics from southern Florida, Central Amer- ica and the West Indies into most of northern South America as far as Brazil and Peru. The following species is the first of its genus, however, to be described from Venezuela.

Lepanthopsis vinacea C. Schweinfurth sp. nov.

Herba gracilis, epiphytica, caespitosa. Caules tenues, vaginis tubulatis, lepanthiformibus, arctis omnino celati. Folium erectum, breviter petiolatum; lamina ovalis vel oblongo-elliptica, obtusa. Inflorescentiae axillares, singu- lae vel perpaucae, quam folium multo longiores, graciles, prope apicem densiflorae, cum floribus secundis, vinaceis. Flores parvi, membranacei, horizontales. Sepalum dor- sale lanceolato-ovatum, superne abruptius angustatum. Sepala lateralia oblique ovato-lanceolata, basi connata, in partem superiorem acuminatam curvata. Petala sepalis multo minora, elliptico-ovata, acuta. Labellum quam petala brevius sed latius, in circuitu orbiculare cum auri- culis retrorsis, triangulari-incurvis. Columna generis.

Plant small, slender, epiphytic. Roots fibrous, flexu- ous, glabrous, numerous. Stems simple, caespitose, slen- der, apically unifoliate, up to 5.7 cm. long, entirely con- cealed by several (up to 9) close, tubular, lepanthiform

[ 109 ]

EXPLANATION OF THE ILLUSTRATION

Prate XVIL. Lepanruopsis vinacea C. Schweinf. Drawing of Dunsterville 425 from Venezuela.

Drawn by G. C. K. Dunsrervitie

[ 110 |

PLatE XVII

A nen

Column & Lp, spread

Aimm

Column & Lap

4e5

GCKD. 7-57

sheaths which are very minutely muriculate on the angles and terminate in a spreading, ovate, marginate, hispid mouth. Leaf erect, very shortly petioled, rather thick and fleshy; lamina oval to oblong-elliptic, obtuse, cune- ate below, marginate, 15-21 mm. long, up to 9 mm. wide; petiole about 3 mm. long. Inflorescences axillary, 1-4 (rarely 6) to astem, much surpassing the leaves, a single one floriferous at a time, suberect to flexuous or spread- ing, about 4.5-5 em. long: peduncle filiform, with sev- eral, inconspicuous, remote, tubular sheaths; raceme densely secund-flowered, about 1-2 cm. long. Floral bracts minute, infundibuliform. Flowers few to numer- ous, horizontal, with widely spreading segments, deep wine-color or bright purple. Dorsal sepal lanceolate- ovate, rather abruptly narrowed to a subacute tip, 1- nerved, about 2-8 mm. long and 1-1.5 mm. wide. Lat- eral sepals similar, obliquely ovate-lanceolate, connate near the base, with outcurved, acuminate tips, about 2-3.2 mm. long, up to 1.2 mm. wide. Petals much smaller than the sepals, elliptic-ovate, acute, 1-nerved, about 0.9-1.2 mm. long and 0.5-0.9 mm. wide. Lip shorter than the petals but broader, orbicular in outline, about 0.8-1.1 mm. in greatest length and equally broad, with a pair of basal, retrorse obliquely triangular-incurved auricles, 8-nerved, rather fleshy. Column minute, yel- lowish or creamy white.

This species, which has the characteristic appearance of the genus, appears to be most closely allied to the Colombian Lepanthopsis acuminata Ames, but is a larger plant, with less acuminate sepals, dissimilar lip and dif- ferently colored flowers.

Venezurta: Bolivar [Mount], Ptari-tepui, at 1600-2000 meters al- titude, epiphytic, occasional in the southwest slope forest, flowers deep wine-color with yellowish column, December 17, 1952, Bassett Maguire & John J. Wurdack 33866 (Vyrr in Herb. Ames No. 68948). Bolivar, El Dorado road, 22 km. south of camp km. 88, “‘sepals and petals bright purple with smooth-glandular surface.’’ G. C. K. Dun- sterville 425 (Herb. Garay No. 6956).

[ 112 ]

BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY

VoL. 18, No. 4

CamsripGE, Massacuusretts, APRIL 4, 1958

PLANTAE AUSTRO-AMERICANAE X AMERICAE AUSTRALIS PLANTAE NOVAE VEL ALIA RATIONE SIGNIFICANTES BY Ricuarp Evans SCHULTES

One hundred years ago, Richard Spruce predicted that the northwestern parts of the Amazon Valley, especially that part lying in Colombia, would be found to be an area peculiarly rich in plants still unknown to science. Continued studies of collections made from this region during the past ten or fifteen years convince us of the soundness and wisdom of Spruce’s observation.

The present paper, a miscellany of notes treating chiefly of new species and varieties, furthers a series di- rected towards our understanding of the flora of South America, with primary reference to the northwestern Amazonia.

It is with appreciation that I acknowledge the collabo- ration of the following specialists: Dr. John Wurdack (Melastomaceae) and Dr. Bassett Maguire (Guttiferae) of the New York Botanical Garden; Dr. José Cuatre- casas (Paraprotium) and Dr. Richard S. Cowan (Callian- dra, Cassia, Inga, Jacqueshuberia, Macrolobium) of the Smithsonian Institution; Dr. John D. Dwyer (T'achi- galia) of St. Louis University; Dr. Harold N. Moldenke (Hriocaulon) of the Trailside Museum; and Dr. Robert W oodson (4 pocynaceae) of the Missouri Botanical Gar- den. I likewise thank Miss Ruth Barton for the careful

| 118 |

line drawings which she has prepared of several of the novelties. These drawings and their publication were made possible through a grant from the National Science Foundation.

The families are arranged in accordance with the Engler-Gilg system, and the genera are enumerated alphabetically under the families.

CYCADACEAE

Zamia madida PR. L. Schultes sp. nov.

Truncus grandis, elongato-obovoideus, valde amyla- ceus, 10-18 cm. longus, basi 5-7 cm. in diametro, per- fecte subterraneus, apice multis bracteis fibroso-coriaceis, siccis, fuscis, elongato-triangularibus, usque ad 6 cm. longis et basi 0.5-1 em. latis coronatus. Folia una ad tres, cum petiolo usque ad sex ped. longa; petiolo ro- busto, valde sulcato, basi rigidissimo sed apicem versus aliquid flexuoso, basi spinulis lignosis brevibus armato, basi 1-1.5 em. in diametro; rhachide 1.3-1.5 m. longa; segmentis leviter coriaceis, supra nitida, infra pallidiora, utrinque duodecim ad quattuordecim, oppositis vel sub- oppositis, plusminusve 2.5 cm. distantibus, ellipticis vel elliptico-lanceolatis sed vulgo obliquuis, margine leviter revolutis atque basim versus integris sed apicem versus remote et irregulariter dentato-serratis, apice oblique acuminatis, basi in petiolulo plerumque 2—2.5 (sed saepe usque ad 5) em. longo productis et Jaminarum base ipsa collo conspicuo coronatis, plerumque 80-40 cm. longis, 5-7 cm. latis, nervis numerosis, parallelis, plusminusve triginti quattuor. Strobilus femineus inter foliis solitar- ius, in pedunculo erecto, robusto, subterete, apicem ver- sus valde carnoso-inflato, stramineo, dense sed minute lanato, usque ad 80 cm. longo, 1-1.5 cm. in diameter, cylindricus, apice apiculatus, 7-8 cm. longus, densissime brunneo-tomentellus, peltis in seriebus plusminusve sex

[114 ]

verticalibus, hexagonis, majoribus 8-9 mm. longis, 5—6 mm. latis; semina non visa. Strobilus masculus parvus, in pedunculo gracillimo debileque, plusminusve 15 cm. longo, 9-10 mm. in diametro, densissime brunneo- tomentellus, peltis in seriebus plusminusve octo, 4 mm. longis, 2 mm. latis.

Zamia madida is so named (Latin for ‘‘drenched”’) in allusion to the extraordinarily humid forest habitat which it prefers. The forests of the Golfo de Uraba area of northwestern Colombia have one of the heaviest rain- falls in all tropical America, receiving between five and six yards of rain a year. The plant is not uncommon in the region of Mutata, but it does not occur in colonies, as do many other species of Zamia. It is a majestic plant, sometimes as tall as a man, and grows erect, not spraw!l- ing, with the heavy ends of the leaves, on which the large leaflets are borne, slightly flexuous.

Zamia madida is well characterized by the very con- spicuous swelling at the base of the leaflets. This swell- ing forms a strong collar which almost completely sur- rounds the apical part of the petiolule.

CotomBia: Departamento de Antioquia, Municipio de Mutata, Villa Arteaga. Altitude about 150 feet. “‘In forest.’? February 16-20, 19538. Richard Evans Schultes & Isidoro Cabrera 18694 (Tyrr in Herb. Gray),—Same locality and date. Schultes §& Cabrera 18640. (Tyrr of male inflorescence in Herb. Gray ).—Same locality and date. Schultes & Cabrera 18679.

ARACEAE

Anthurium apaporanum PF. E. Schultes sp. nov.

Caudex abbreviatus, circiter 1.5 em. in diametro. Pet- iolus robustior, crassus, canaliculatus, geniculo brevi in- structus, usque ad plusminusve 80 cm. longus, 6 mm. in diametro. Folia firme chartacea, pallide viridia, late lan- ceolata, apice breviter acuminata, basi cuneata, usque ad 40 cm. longa, 12 cm. lata, nervis lateralibus primariis

[ 115 ]

EXPLANATION OF THE ILLUSTRATION

Pirate XVIII. Zamra mapipa R. E. Schultes. 1, habit sketch. 2, portion of leaf, about two fifths natural size. 3, pistillate inflorescence, about two fifths nat- ural size. 4, basal portion of leaflet, showing collar. 5, segment of staminate inflorescence, about two and one half times natural size. 6, staminate in- florescence, about two fifths natural size.

Drawn by Ruru Barron

[ 116 |

PuatTe XVIII

ZAMIA madida

RE. Schuller

EXPLANATION OF THE ILLUSTRATION

Pirate XIX. Antruurtum apAporanum RR. EF, Schultes. Habit, slightly less than one quarter natural size ; and fertilized pistillate flower, slightly less than five times

natural size. Drawn by Ruru Barron

[ 118 ]

PLatTE XIX

} ANTHURIUM

dpapora ALUM

mb

angulo 60° a costa abeuntibus, inter se 8-12 mm. distan- tibus et in nervum collectivum medio a margine 2-3 mm. remotum. Pedunculus ad 42 cm. longus, 8 mm. in dia- metro. Spatha lanceolata, 8 cm. longa, 1.5 cm. lata, apice acutiuscula, viridis. Spadix viridis, sessilis, usque ad 16 cm. longis, 15-17 mm. in diametro. ‘Tepala omnino albido-pulverulenta, circiter 2 mm. lata, 4 mm. longa. Pistillum rufum, ovoideum, 4 mm. longum, 8 mm. in diametro.

Anthurium apaporanum appears to be nearest to the complex of 4. Harrisit (Grah.) Endl., but several differ- ences in floral structure and in relative size of the vege- tative parts would suggest that the two concepts are best considered as distinct. In many of its vegetative charac- ters, Anthurium apaporanum likewise resembles A. na- paeum Kingl., but the two are easily separated by floral differences.

CotompiA: Comisaria del Vaupés, Rio Kananari, Cerro Isibukuri.

On sandstone cliffs. ‘‘Spadix and spathe green.’’ August 4, 1951. Richard Evans Schultes & Isidoro Cabrera 13335 (Tyre in Herb. Gray).

Anthurium nanum PR. EF. Schultes sp. nov.

Herba epiphytica. Caudiculus crassus, internodiis ab- breviatis. Folia subcoriacea, petioli strictissimis, crassis, teretibusque, sulcatis, usque ad 2.5-3.5 cm. longis, 0.5 mm. in diametro, lamina pallide viridis, glabra lineari- lanceolata, apice acuminata, basi longe cuneata, margine inconspicue et leviter revolta, 6.5—7 cm. longa, 8-10 mm. lata, nervis crassioribus, utrinque prominentibus. Spatha membranacea, triangularis, apice crasso-apiculata, utrin- que albido-pulverulenta, rosea, 9 mm. longa, basi 8-8.5 mm. lata, vivo reflexa. Spadix sessilis, cylindricus, apice truncatus, 11-12 mm. longus, plusminusve 2 mm. in diametro, rufo-purpureus.

It would appear that Anthurium nanum, set apart be-

[ 120 |

cause of its small stature, as its name implies, is allied to the complex of A. gracile (Rudge) Engl. and to 4. font- oides R. E. Schultes. The type specimen, while a fully mature individual, is of note because of its very diminu- tive size.

Cotomspra: Comisaria del Vaupés, Rio Kananari, Cachivera Palito. Altitude about 250 m. “‘Epiphyte. Spathe pink. Spadix red.’’ July

25, 1951. Richard Evans Schultes & Isidoro Cabrera 18183 (Tyrer in Herb. Gray).

Philodendron colombianum PF. E. Schultes sp. nov.

Herba in locis humidis crescens, usque ad 15 pedes alta. Folia ovata, apice sensim cuspidato-acuminata, basi cordata, usque ad 20 cm. longa, 10 cm. lata, nervis later- alibus angulo circiter 80° a costa adscendentibus, margi- nem versus conjunctis; petiolus apicem versus callosus, plusminusve 18 em. longus, teres, inferne vagina obtec- tus. Pedunculus basi in bracteis fibrosis inclusus, plus- minusve 9 em. longus, 2 mm. in diametro. Spatha apice subacuta, intus alba sed extus purpurascens, 5 cm. longa, 3.5 em. lata, tubus ventricosus fauce valde contractus, 3.5 cm. longus, 1-1.5 cm. in diametro. Spadix stipitate, 5-6 mm. longo suffultus, usque ad 6 cm. longus. In- florescentiae pars feminea 2.5 cm. longa, plusminusve 11 mm. in diametro, pars mascula 2.4 cm. longa. Stamina 1 mm. longa, 1.2 mm. lata. Pistillum 1-1.2 mm. longum, 0.8 mm. in diametro. Ovarium quattuor- vel quinque- loculare.

Cotomp1a: Comisarias del Amazonas y Vaupés, Rio Apaporis, Sora- tama (above mouth of Rio Kananari) and vicinity. Altitude about 900 feet. March 26, 1952. Richard Evans Schultes & Isidoro Cabrera 16087 (Type in Herb. Gray).

Stenospermatium verecundum PR. LE. Schultes sp. nov. Suffrutex parvus, in savannarum arena crescens, Cau-

[ 121 ]

dex valde abbreviatus, circiter 2 cm. crassus. Folia membranacea, atroviridia irregulariter albido-maculata, lanceolato-linguiformia, utrinque glabra, apice acutissima cum apiculo flexuoso, 13-17 mm. longo, basi inaequaliter rotundata, in petiolo 5-9 cm. longo, basi purpureo- et roseo-maculato, ad vaginam latiusculo, 7-80 mm. longo producta; lamina 11-15 cm. longa, 2.5—-4 em. lata; ner- vis lateralibus angulo acuto adscendentibus. Pedunculus crassus, usque ad 20-22 cm. longus sed vulgo brevior, in vagina foliacea saepe inclusus, apicem versus, 2 mm. in diametro, Spatha aperta lanceolato-elliptica, longe acu- minata, usque ad 9 cm. longa, 1.5—1.8 cm. lata, utrinque viridia, Spadix cylindricus, apicem versus tenuior, stipi- tate robusto, brevi, minute albo-granuloso, 4mm. longo, 4—5 mm. in diametro suffultus. Staminum filamenta late linearia, 0.5 mm. longa, antherae minutissimae. Pistilla 1 mm. longa, vertice tetragono stigmate coronata.

The long and flexuous apicules on the leaves of this new species seem to set it apart from other concepts of Stenospermatium. Stenospermatium verecundum would appear to be nearest to S. Ulei, at least vegetatively, but the former has a much shorter and stouter peduncle than the latter species.

Stenospermatium verecundum is apparently endemic to the quartzitic mountains of Amazonian Colombia, where it grows in white sandy savannahs, under conditions of extreme xerophytism. It is often associated in the field with the lichen Cladonia, as illustrated in the accompa- nying photograph (Plate X XII).

The specific name, meaning ‘‘modest,’’ refers to its inconspicuous habit of growth.

Cotompia : Comisarias del Amazonas y Vaupés, Rio Piraparana, Cerro E-ree-eé-k6-mee-b-kee. ““Terrestrial. On sterile sand. Spathe green on both sides, spadix white. Leaves green on both sides. Base of stem mottled purple-brown and pink.’’ September 18, 1952. Richard Evans Schultes & Isidoro Cabrera 17496 (Tyew in Herb. Gray).

[ 122 ]

Urospatha somnolenta PR. LH. Schultes sp. nov. Herba paludosa saxicola, usque ad 2.5-8 pedes alta. Rhizoma spongiosum. Folia erecta, papyracea, hastato- sagittata, latiloba, supra atroviridia, infra vix pallidiora, utringue glabra, nervis utrinque prominentibus; lobo antico lanceoloto-triangulari, apice acuminato, usque ad 25 cm. longo, parte inferna 10 cm. lato; lobis lateralibus paullo longioribus, oblongo-lanceolatis, sinu profunde parabolico distantibus, costis posticis in sinu 2 cm. denu- datis, nervis lateralibus primariis lobi antici duo superi- oribus angulo acuto adscendentibus in apice exeuntibus, infimo juxta margine fere totius lobi procurrente, nervis loborum lateralium superioribus adscendentibus margi- nem versus conjunctis. Petiolus apicem versus callosus, parte callosa geniculata, plusminusve 8 cm. longa, usque ad 85-90 cm. longus, aliquid turgido-carnosus, atroviridis et conspicue purpureo-maculatus. Pedunculus robustior, turgido-carnosus, colore petiolo similis, usque ad 55-60 em. longus, basi 1 em. in diametro. Spatha anguste lan- ceolata, longe acuminata, statu juvenili valde spiraliter contorta adulta apicem versus contorta, extus atropur- purea vel nigro-purpurea, intus atroviridis, 15-18 cm. longa, basi 8 cm. lata. Spadix purpureus, apparenter non stipitatus, usque ad 5-6 cm. longus, 8-3.5 cm. in diametro, maturitate valde reflexus. Flores usualiter quinque-tepali, purpurei. Ovarium obovoideum, vulgo biloculare. Bacca obovoidea, 6-7 mm. longo, 4.5 mm. crassa. Semina subreniformia, laeviter rufo-punctulata. Urospatha somnolenta resembles U. sagittifolia (Ro- dach.) Schott, U. Spruceana Schott and U. decipiens Schott, both of which occur in the northwest Amazon. Urospatha somnolenta appears to be amore robust plant than is generally reported for U. sagittifolia and its vari- eties. It is rather outstanding in its choice of habitat — sandy and boggy holes and crevasses in rocks in the vicin-

[ 128 ]

ity of waterfalls. It is ikewise noteworthy in the strongly drooping position which the ripening fruiting head grad- ually assumes. The very young spadix stands perfectly erect, but, as the fertilized ovaries enlarge, the spadix bends and finally droops in a deeply nodding position, as shown in the accompanying habit photographs. I have not seen this peculiarity reported for the other species of Urospatha; none of the herbarium specimens which I have studied give indication of it, nor have I observed it in any other species of the genus which I met in the field. The specific epithet somnolenta (Latin meaning ‘*full of sleep’’) refers to this apparently unusual nodding position of the mature spadix.

CotomsiaA : Comisarias del Amazonas y Vaupés, Rio Apaporis, Raudal Yayacopi (La Playa) and vicinity. Altitude about 800 feet. **In damp crevasses in rocks, in sand and stagnant water. Stem mottled green- purple. Spathe dark green inside, dark purple-black outside. Fruit dark purple-green.’? February 16, 1952. Richard Evans Schultes & Isidoro Cabrera 15402 (Tyrw in Herb. Gray).

ERIOCAULACEAE

Eriocaulon vaupesense Moldenke sp. nov.

Herba parva annua; foliis caespitosis tenuiter mem- branaceis fragilibus graminodeis 3.5—5.5 cm. longis fene- stratis glabris; pedunculis 7-14 cm. longis solitariis gra- cillimis stramineis tricostatis; vaginis arcte adpressis 2-2.5 cm. longis glabris; capitulis parvis hemisphaericis albis 8-5 mm. latis; bracteolis late obovatis crassis stra- mineis imbricatis nitidis ca. 1.6 mm. longis ad apicem rotundatis.

Small, probably annual, herb; stem much abbreviated or obsolete; leaves tufted, very thin-membranous and fragile, apparently erect when fresh, spreading or reflexed in age, linear and grass-like, uniformly pale-green on both surfaces, 3.5-5.5 cm. long, 1.5-2 mm. wide below the middle, attenuate-subulate at the apex, fenestrate, gla-

[ 124 ]

brous on both surfaces; peduncles usually one per plant, 7-14 cm. long, very slender, stramineous, 8-costate, gla- brous; sheath slender, closely appressed, 2—2.5 cm. long, glabrous throughout, split at the apex, the blade about 5 mm. long, erect or the tip very slightly spreading, subulate-tipped, glabrous; heads small, hemispheric, white, 83-5 mm. wide; involucral bractlets broadly obo- vate, thick-textured, stramineous, closely appressed and imbricate, about 1.6 mm. long and wide, shiny, rounded at the apex, glabrous; receptacular bractlets similar but appressed, white-pilose on the back and shortly white- barbate towards and at the apex with stiffly erect hairs; staminate florets: sepals 3, connate at the base, cuneate- spathulate, about 0.8 mm. long and 0.4 mm. wide, rounded and white-barbate at the apex with stiffly erect hairs; petals 8, connate into a slender stramineous tube about 0.8 mm. long, the free tips densely white-barbate like the sepals; stamens 6; filaments very short, about 0.16 mm. long, glabrous; anthers oblong, dark-brown, about 0.2 mm. long and 0.12 mm. wide; pistillate florets not seen.

Since the type locality of Hriocaulon vaupesense lies on the boundary between Colombia and Venezuela, this species probably occurs also in Venezuela.

Cotomsra: Comisaria del Vaupés, Rio Guainia, Cafio del Caribe (be- tween Isla del Venado and San José) and vicinity. Altitude 850-900 feet. Lat. 2°45’ N, Long. 67°50’ W. November 2, 1952. Richard Evans Schultes, Richard E. D, Baker & Isidoro Cabrera 18274 (Tyre in U.S. Nat. Herb.).

ARISTOLOCHIACEAE Aristolochia georgica R. 17. Schultes sp. nov. KFrutex scandens, gracilis. Caulis volubilis; rami vol- ubilis, teretes, striolati, cum cortice fusco-brunneo, tenui,

minutissime puberulente. Folia exstipulata, coriacea vel firme chartacea, oblonga, apice breviter apiculata, basi

[ 125 ]

EXPLANATION OF THE ILLUSTRATION

Pirate XX. PuitopenprRon coLomBranuMm R, EF, Schulltes. Habit, approximately two fifths natural size; and fer- tilized pistillate flower, approximately nine times natu-

ral size. Drawn by Rutu Barron

[ 126 ]

PLaTE XX

NY

M// PHILODENDRON = cotombianum | RA. & Sckaltes

EXPLANATION OF THE ILLUSTRATION

Pirate XXI. SrenosperMATIUM verEcUNDUM R. £, Schultes. 1, habit, approximately one third natural size. 2, inflorescence, approximately two thirds

natural size. Drawn by Ruru Barton

[128 |

PLATE

XX!

Se PERMATIUM VEPTECUNAUM RE Schulles

EXPLANATION OF THE ILLUSTRATION

Prare XXIT. SrenosperMaAtTiuM verecuNDUM RR, EF, Schultes. Habit photograph of the colony from which the type collection was made.

Photograph by Ricuary Evans Scuures

[ 130 |

PLaTE XXII

EXPLANATION OF THE ILLUSTRATION

Prare XXIII. Urosparua somnotenta R. EF. Schultes. 1, leaf, approximately one sixth natural size. 2, ripen- ing spadix with spathe, approximately one sixth nat- ural size. 3, position of fully ripened spadix. 4, fer- tilized pistillate flower, approximately three and one half times natural size.

Drawn by Ruru Barron

[ 132 ]

Piate XXIII

UROSPATHA somno/enta , RE. Schultes

Hy

EXPLANATION OF THE ILLUSTRATION

Pirate XXIV. Urosparua somnoventa PR, EB. Schultes. Habit photograph of the colony from which the type collection was made.

Photograph by Ricuarp Evans ScuuiTes

[ 1384 ]

XALV

PLATE

EXPLANATION OF THE ILLUSTRATION

Pruare XXV. Urospatrua somnotenta I. Eb, Scehultes. Ripening spadix of the type plant.

Photograph by Pacuarp Evans Scuuitres

[ 136 ]

riATeE ASV

-

valde contorta, petiolata (petiolo plusminusve 1.5—2 ¢m. longo, pulverulenta-puberulente), nervis lateralibus pler- umque quattuor, supra vivo atroviridia, siccitate stram- inea, nitidissima, nervis non conspicue elevatis, glaber- rima, subtus vivo pallide viridia, nervis omnibus promi- nenter elevatis, minutissime albido-tomentulosa, cum petiolo usque ad 12-14 em. longa, plerumaque plusmin- usve fem, lata. Flores in racemis congestis, multifloribus et axillaribus, plusminusve 5.5 em. longis, pedunculo minutissime pulverulento, gracile, usque ad 8-8.5 em. longo. Perianthit pars inferna valde ovoideo-dilatata, usque ad 20 mm. longa, 5 mm. in diametro, extus mi- nutissime pulverulens, intus carnoso-verrucosus; tubus eylindricus, 8-12 mm. longus, plusminusve 2 mm. in diametro, apicem versus saepe paullo dilatatus; limbus altero latere expansus, altero truncatus, membranaceus, intus distanter hirtellus, brunneo-viridis, extus minutis- sime pulverulentus, brunneo-purpureus, oblongus, apice rotundatus et saepe minute apiculatus, apiculo dense aureo-tomentoso, usque ad 22 mim. longus, 10 mm. latus. Columna genitalis 8 mm. longa, breviter stipitata, per dimidium sex-divisa; lobis pseudo-stylinis acutis. Stam- ina sex in serie unica, colummae adnata, antheris lineari- bus, usque ad 0.8 mm. longis, longitudinaliter dehiscen- tibus. Fructus late ellipsoideus, usque ad 8 cm. longus, 3.5 em. in diametro, pedunculo usque ad 2.5 ¢m. longo. Semina cuneato-subquadrata, fusco-cinerea, plusminusve 10 min. Jata, 7 mm. longa, altero latere aliquid nitida, altero dense papillosa.

Aristolochia: georgica, so named because it prospers most successfully in disturbed agricultural plots near human habitation, belongs to Hoehne’s section [vwstip- ulatae. It does not appear to resemble closely any species hitherto described. Vegetatively, it suggests Lristolochia cornuta Mast. of Amazonian Brazil and Peru, but there

are many major floral differences to separate the two con- cepts. It also looks vegetatively rather like Aristolochia mutensis O. C. Schmidt, but the lip of the flower is very different.

Of interest as a peculiarity of -fristolochia georgica is the curious dense, golden, tomentose apicule terminating the broad, oblong, rounded lip. The species is unusual in the Mustipulatae in having amany-flowered inflorescence.

Cotompra: Comisarias del Amazonas y Vaupés, Rio Apaporis, Sora- tama (between Rio Kananari and Rio Pacoa). Altitude about 250 m. “Vine. Flowers mottled green and brownish purple. Puinave name = too-cheé-ka-pee. On flood-bank.** August 27, 1951. Richard Evans Schultes & Isidoro Cabrera 13727 (Tyrer in Herb. Gray).

Aristolochia xerophytica PR. 1’. Schultes sp. nov.

Suffrutex scandens, in locis arenosis maxime calidis crescens. Caulis volubilis: rami volubiles, teretes, sub- striolati, cum cortice fusco-brunneo, tenul, minutissime puberulente. Pseudostipulae foliis similes sed) multo parviores, usque ad 1.5 cm. longae. Folia exstipulata, coriacea vel firme chartacea, oblongo-cordata, basi cordi- forme incisa, apice subacuta, petiolata (petiolo 1-8 em. longo, pulverulento-puberulente), nervis lateralibus pler- umque duo, supra apparenter atroviridia, minutissime et irregulariter scrobiculata, nervis non conspicue elevatis, subtus pallide viridia vel cinereo-viridia, minute et dense hirsuta, cum petiolo usque ad 8-9 cm. longa, parte me- diana 1.5-8 em. lata. Flores apparenter semper solitari: pedunculo gracile, pulverulento, usque ad 8 em. longo. Perianthilt pars inferna valde subglobosa vel ovoideo- dilatata, 0.7-10 mm. in diametro, extus fusca, glabra, intus aliquid verrucosa; tubus cylindricus, 9-10 mm. longa, 4 mm. in diametro, apicem versus valde infundi- bulitormis: limbus altero Jatere valdissime expansus, altero truncatus, membranaceus, intus subglabrus sed apicem versus albido-hirsutissimus, apparenter flavo-

[ 139 |

EXPLANATION OF THE ILLUSTRATION

Prare XXVI. Artsrotocuta Groraica PR. EB. Schultes, Flowering and fruiting branch, approximately one

half natural size. Drawn by Ruru Barton

[ 140 |

PLateE XXVI

ARIS TOLOCHIA COTCICE [rg n.¢Lchutles Nl’) f

brunneus, extus subglabrus, lanceolatus, apicem versus paullo dilatatus, apice ipso abrupte et breviter acumina- tus, usque ad 32 mm. longus, basi 10 mm. latus. Col- umna genitalis 2.5-8 mm. longa, non stipitata, apice sex-divisa: lobis pseudostylinis obtusis, 0.7 mm. longis. Stamina sex in serie unica, columnae valde adnata, anth- eris linearibus, plerumque 2 mm. longis, longitudinaliter dehiscentibus. Fructus adhuc ignotus.

Aristolochia verophytica does not seem to match closely any of the species included in Hoehne’s Pseudostipulosae. Except for its foliaceous stipules, it would seem vegeta- tively to resemble most closely sfristolochia rumicifolia Mart. & Zuce., but the flowers of the two species are very dissimilar,

As the specific name implies, -fristolochia werophytica grows ina hot, dry and sandy region, a rather unusual habitat for members of this genus.

Peru: Provineia de Jaen, Departamento de Cajamarca. Near Junta on Huancabamba River. *‘ Herbaceous vine, growing in very hot, dry, sandy soil. Flower pelican-shaped, brown.’’ August 3, 1943. EF. L. Evinger 540 (Tyrr in U.S. Nat. Herb. 1953576; 2105024),

LEGUMINOSAE

The descriptions and discussions of Calliandra, Cassia, Inga, Jacqueshuberia and Macrolobium have been con- tributed by Dr. Richard S. Cowan; of Tachigalia by Dr. John Dwyer.

Calliandra vaupesiana Cowan sp. nov.

Arbuscula, ramulis albo-pilosulis et brunneo-puberulis ; stipulae persistentes, 4-5 mm. longae, 1.5 mm. latae, lanceolatae, acutae, externe minuto-strigulosae, intus glabrae; petioli 5-7 mm. longi, eglandulares, albo- strigulosi vel -puberuli et brunneo-puberuli, rhachibus 13— 38 mim. longis, albo-strigulosis vel -puberulis et brunneo- puberulis: pinnae 8-G6-jugatae, anguste oblongae, petiolis

[ 142 ]

secundaris I mm. longis, minute albo-strigulosis, rhachi- bus secondarts 4.5-9 cm, longis, minute albo-strigulosis : foliola sessilia, 56-70-jugata, approximata, anguste ob- longa, haud falcata, apice obtusa ad subacuta, basaliter excentrico-rotundata, glabra, margine minute parceque cillolato, obscure 2-8-nervia infra, supra laevia, pinnarum toliolis mediis, 7-12 mm. longis, 1-1.5 mm. latis: in- florescentiae axillares in axillis foliorum superiorum et terminales, capitatae, pedunculo 18-22 mm. longo, mi- nute albo-striguloso, floribus sessilibus, rubris, perianthio externe albo-sericeo (corolla densissime sericea), intus glabro; calyx cupularis, tubo ca. 3.5 mm. longo, lobis 0.6-1 mm. longis, ovatis, acutis; corolla infundibulitor- mis, tubo ca. 5 mm. longo, lobis ovatis, 8 mm. longis, 2 mm. latis, acutis: filamenta 5 em. longa, glabra; ova- rium densissime sericeum, stylo glabro; fructus 90 mim. longus, 12 mim. latus, oblongus, ad basim angustior, velutinus.

The specific epithet, derived from the name of the geo- graphic region from which these four collections were taken, seems appropriate in light of the apparent re- stricted distribution of many species of Calliandra.

The new species does not seem to have any very close relatives; it may be recognized by the large number of small, subglabrous leaflets, the few pairs of pinnae, and the rather large, sericeous flowers.

Cotomspra: Comisaria del Vaupés, Rio Kuduyari, Cerro Yapoboda.

Sabanas sobre piedras areniscas. Altitude about 450 m. October 3-4, 1951. Richard Evans Schultes & Isidoro Cabrera 14203 (Tyrer in U.S. Nat. Herb. 2171380).—Comisaria del Vaupés, Rio Parana Pichuna. Altitude about 700 ft. June 1953. Schultes & Cabrera 19915.—Com- isaria del Vaupés, Rio Karurti, Mesa de Yambi, Savannah de Goo- ran-hoo-da, Altitude 950-1000 ft. April 1953. Schultes & Cabrera 19152,—Comisaria del Vaupés, Rio Kuduyari, Yapoboda. Quartzite savannah near headwaters. Altitude 900-1000 ft. April 1953. Schultes

& Cabrera 19985.

[ 143 ]

EXPLANATION OF THE ILLUSTRATION

Prate XXVIT. Artsrotocuia xeropuytica R. EF. Schultes. Flowering branch, approximately one half natural size.

Drawn by Ruru Barron

[ 144 ]

PLaTE XXVII

ARISTOLOCHIA xerophyl (Ca fo

A. & Sekultes cae Ne

Cassia gigantifolia (Britt. & Killip) Cowan comb. nov.

Chamaefistula gigantifolia Britton & Killip, N. Y.

Acad. Sci. 35 (1986) 171.

This collection has been compared with the type (G. Klug 1622) at the New York Botanical Garden, and there is little doubt of their conspecificity, although the petals of the Schu/tes collection are larger; those on the type are up to 2 cm. long, and those in the present col- lection are up to 8.5 cm. long.

CotomsBiA: Comisaria del Vaupés, Rio Negro, San Felipe and vicin- ity. Richard Evans Schultes, Richard FE. D. Baker & Isidoro Cabrera 17964.

Inga involucrata Cowan sp. nov.

Arbuscula, ramulis juvenilibus strigosis sed mox glab- rescentibus; stipulae persistentes, subcoriaceae, 15-23 mm. longae, 10-12 mm. latae, ovatae, acuminatae, glab- rae; folia glabra, 3-jugata, petiolo 47 mm. longo, alis oblongis, 41 mm. longis, 6 mm. latis, glandula disciformi, sessili, 8.5 mm. diametro, rhache 18 cm. longa, alis ob- lanceolatis, 75-85 mm. longis, 6 mm. latis; foliola sub- coriacea, discolora, elliptica, base obtusa ad subobtusa, apice acuta, faciebus ambobus costa salienti sed infra validiore, venis primariis 10-15 costae utroque latere, supra leviter impressis, infra valde salientibus, venulis obscuris, foliolorum jugo inferiore 16.5 cm. longo, 7.5 cm. lato, jugis superioribus ca. 25-80 cm. longis, 9.5-12 cm. latis; inflorescentiae axillares, capitatae, pedunculo 7.5-9 em. longo, sparse strigoso, glabrescenti, floribus subsessilibus, viridibus, involucratis (bracteis imbricatis, multis, sparse strigosis ad glabris subtendentibus), brac- teis exterioribus ovatis, acutis, 7-10 mm. longis, 6.5—-8.5 mim. latis, bracteis interioribus ellipticis, acutis, ca. 8-10 mim. longis, 5 mm. latis, pedicellis 1.5 mm. longis, stri- gosis; calycorum tubus glaber, ecostatus, subeylindricus,

[ 146 ]

sparse strigosus, 21-23 mm. longus, apice 8-9 mm. dia- metro, calycorum dentibus triangularibus, 8 mm. longis, strigosis; corolla immatura, apice strigosa; gynoecium glabrum.

Inga involucrata is not closely related to any other known species. It is abundantly distinct by virtue of its persistent, foliaceous stipules and bracts; the latter form a conspicuous involucre (hence the specific epithet) at the apex of the peduncle, subtending the subsessile flowers. Another very striking feature of the new species is the very large calyx; the only species which shares this char- acter is Inga inflata Ducke, an Amazonian species, but this probably does not indicate very close relationship, for in all other respects the two species are quite dissimilar.

CoLombia: Departamento de Antioquia, Golfo de Uraba, Municipio de Mutata, Villa Arteaga. Altitude about 150 ft. February 16-20, 1953. Richard Evans Schultes & Isidoro Cabrera 18651 (Typr in U.S. Nat. Herb. 2172228).

Inga longifoliola Cowan sp. nov.

Arbuscula vel arbor ad 3 m. alta, ramulis minute strigu- losis, albo-lenticellatis; stipulae caducae, ca. 2.5 mm. longae, 0.7 mm. latae, acutae, anguste ellipticae; folia parce minuto-strigulosa, petiolis teretibus, 20-22 mm. longis, rhachibus 24-33 mm. longis, teretibus ad margina- tis, glandulis sessilibus et ca. 2 mm. diametro; foliola 2-jugata, raro unijugata, petiolulis 2-2.5 mm. longis, laminis 13-17.5 cm. longis, 8-5 cm. latis, anguste ellip- ticis ad lanceolato-ellipticis, base acutis, apice longo- acuminatis, faciebus ambobus parce strigulosis, supra metallicis et nitidis, costa salienti, venis primariis ca. 15- jugatis, supra planis, infra salientibus; inflorescentiae axillares, pedunculo (80—) 45-55 (-62) mm. longo, mi- nute striguloso, axe ca. 12-15 mm. longo, minute strig- uloso, bracteis caducis, lanceolatis, acutis, 2 mm. longis et 1 mm. latis, externe minute strigulosis, intus glabris,

[ 147 ]

EXPLANATION OF THE ILLUSTRATION

Pirate XXVIII. a-d, Inca invoLtucrata Cowan. a, habit, one half natural size. b, calyx, one and one half times natural size. c, outer bract of in- florescence, one and one half times natural size.

d, inner bract, one and one half times natural size.

e-f, INGA LonGIFOLIOLA Cowan. e, habit, one half natural size. f, one flower, one and one half times natural size.

Drawn by Ricuarp S$. Cowan

[148 ]

Piate XXVIII

rad Ea \

SSS

mt ee : SESS ey / oN

“s oN SSE ee

pedicellis 1.5—2 mm. longis, minute strigulosis; calyx brevi-cylindricus, 7-9 mm. longus, strigulosus, tubo 5-6 mm. longo, lobis triangularibus, 2-8.5 mm. longis, acutis ; corolla villoso-sericea, 14-18 mm. longa, tubo 11-14 mm. longo, ad apicem dilatato, lobis lanceolatis ad ovatis, acutis, 8-4 mm. longis; filamenta ca. 85-40 mm. longa, tubo 15-20 mm. longo, incluso vel parce exserto: gyn- oecium glabrum.

The long, narrow leaflets of the new species provide the basis for the specific epithet. It is probably not closely related to Inga strigillosa, but it has many characters in common with this species. Both have terete petioles and rhachis (terete to marginate rhachis in the new species), the leaflets of both are in two pairs and they have a me- tallic lustre on the upper surface, the pubescence is alike, and the inflorescence is very similar. The two species differ in the size of all the flower parts and the width- length ratio and the shape of the leaflets.

CoLompra: Comisaria del Vaupés, Rio Apaporis, Jinogojé (at mouth of Rio Piraparand) and vicinity. “Bush. Flowers white.’’ September 25,1952. Richard Evans Schultes & Isidoro Cabrera 17609 (Typrin U.S. Nat. Herb. 2172944).—Same locality. June 1952. Schultes § Cabrera 16677.—Same locality. July 1952. Schultes & Cabrera 19807,—Com-

isaria del Amazonas, Rio Caqueta, La Pedrera and vicinity. May 1952, Schultes & Cabrera 16382.

Jacqueshuberia amplifoliola Cowan sp. nov.

Arbor probabiliter, ramulis Juvenilibus parce strigulo- sis; petioli 8.5-5 cm. longi, parce strigulosi, rhachibus 5-7 cm. longis, marginato-canaliculatis, parce strigulosis ; pinnae 8-jugatae, petiolis secundariis 10 mm. longis, rhachillibus 5.5-8 cm. longis, marginato-canaliculatis, parce strigulosis; foliola glabra, coriacea, supra nitida, 6-8-jugata, sessilia, 18-28 mm. longa, 8-15 mm. lata, oblonga, base inaequilateralia et acuta, apice rotundata retusaque, costa plana, venulis obscuris; inflorescentiae

[ 150 |

terminales, composito-racemosae, ca. 85-40 cm. longae, parce strigulosa, bracteis deciduis, lanceolatis, acumina- tis, 8.5 mm. longis; pedicelli 17-20 mm. longi, graciles, parce strigulosi, hypanthio 4—4.5 mm. longo, cupulare ; sepala 5, externe strigulosa, ciliolata, duo late triangu- laria, 8.5-4 mm. longa, 4.5 mm. lata, cetera sepala ovalia, 5-5.5 mm. longa, 4-4.5 mm. lata, apice rotundata; pe- tala 5, ovalia, ciliolata, externe ad basem parce strigulosa, intus glabra, 6.5—7 mm. longa, 4.5 mm. lata; filamenta ca, 80 mm. longa, base villosa, tubo inaequale 10 mm. longo cohaerenti, antheris oblongis, 5.5-6 mm. longis, mucro 0.5 mm. longo praedito; stigma capitellatum, stylo glabro, ca. 10 mm. longo, gracili; ovarium ad mar- ginem sparse strigulosum vel glabrum, lateraliter pro- funde unisulcatum.

There certainly is no difficulty in distinguishing this species from the two previously described ones. ‘Those have very numerous, small leaflets and Jacqueshuberia amplifoliola (as the epithet implies) has relatively very large and fewer leaflets; there are also abundant floral characteristics to separate the species. The nearest rela- tive of the new one is Jacqueshuberia quinquangulata which has pubescent ovaries, deeply sulcate branchlets, etc., in addition to the very different leaves.

CotomsBiA: Comisaria del Vaupés, Rio Kubiyti, Cerro Kafienda, Savannah about 15 miles upstream from mouth. Quartzite base. Alti- tude about 800-900 ft. November 10, 1952, Richard Evans Schultes & Isidoro Cabrera 18402 (Typnin U.S. Nat. Herb. 2195820).—Same lo- cality and date. Schultes §& Cabrera 18376.

Macrolobium microcalyx Ducke var. minimum Cowan var. nov.

A var. microcalyce ramulis puberulis, foliis 2-3-jugatis, rachibus 7-17 mm. longis, canaliculatis, foliolis ad basim inaequilateralibus, latere superiore acuto, latere inferiore obtuso, infra glaucis, bracteolis glabris intus, extus pu-

[ 151 ]

berulis, 2 sepalis adaxialibus saepe ad apicem cohaerenti- bus, petalarum lamina glabra, ovario marginaliter villoso differt.

The differences separating this variety from the typi- cal one are not strikingly obvious, but they are sufficiently important to justify recognition. The epithet is intended to convey the impression that the new variety is a smaller form of Macrolobium microcalyx. The leaflets of this new variety are mostly in two or three pairs, and most of the leaflets are smaller than those characteristic of the typical variety. The under surfaces of the leaflets are glaucous, the bracteoles are glabrous on the inner surface and the ovary is marginally villose instead of completely villose as in the typical variety.

Cotomsra: Comisaria del Vaupés, Rio Kananari, Cerro Isibukuri.

Piedra de arenisea. Altitude 250-700 m. October 28, 1951. Richard Evans Schultes & Isidoro Cabrera 14456 (Tyrr in N.Y. Bot. Gard.).

Tachigalia Schultesiana Dwyer sp. nov.

Arbores magnae; virgae teretes minute velutinae; stipulae non visae; foliaad 20 cm. longa; petioli graciles, 0.2—-0.4 em. lati, dense pubescentes vix canaliculati myr- mecodomatio oblongo, ad 4 cm. longo, ad 1.5 cm. lato, saepe infra primum foliolorum jugum constructo; rha- chides graciles ad 0.15 cm. lati, dense pubescentes; pet- ioluli dense pubescentes ad 0.3 cm. longi; foliola 8—10- jugata, tenul-coriacea minute puberula supra dense velu- tina infra (praecipue in costa venisque) inaequilateraliter angusto-oblonga, 2.5-8 cm. longa, 1—2.3 cm. lata, apice obtusa ultime distinecte acuminata, costa prominula venis principalibus 6-10 supra prominulis subevanescentibus infra subprominentibus subrigidis arcuato-ascendentibus, marginibus revolutis; inflorescentiae dense velutinae, racemis ad 18 em. longis rhachidibus cire. 0.2 cm. in medio latis, inferioribus evidenter arcuato-ascendentibus

[ 152 ]

superioribus laxioribus et gradatim brevioribus, floribus multis mox deciduis; bracteolae mox deciduae lanceola- tae, ad 6 mm. longae, ad 2 mm. latae; pedicelli ad 1.5 mm. longi; calyx receptaculi inaequilateraliter cupuli- formis pariete circ. 0.85 mm. crasso extus dense pubes- cente; sepala tenui-carnosa oblonga vel ovata-oblonga, ad 4 mm. longa, 8-3.2 mm. lata, utrinque dense lanulosa comis paucis parvis rubescentibus corporibus mistis, mar- ginibus villosulis; petala in vivo lutea oblonga, 4—5 mm. longa, basi unguiculata villosa intus praeter margines (extus comis paucis basi) venis evidentibus; filamenta subulata, 4-7.5 mm. longa, 0.3-0.35 mm. basi lata, vil- losa basi comis solum intus dispositis, antheris cire. 1.3 mm. longis; stipes ovaril 1.3-2.7 mm. longus, aureo- villosus, ovario ad 8 mm. longo, cire. 1.5 mm. lato, dense aureo-villoso stylo ad 4 mm. longo, ovulis 6-8; fructus non Visi.

Tachigaha Schultesiana, named in honor of one of the collectors of the type material, is marked by multijugate leaves. Only two other species of the genus, T'achigalia polyphylla Poepp. & Endl. and 7". multijuga Benth., have numerous pairs of leaflets. The smaller flowers of the new species, with uniformly thick filaments of the stamens, suggest immediately a close relationship with Tachigalia polyphylla of Amazonian Brazil.

A number of floral characters distinguish T'achigalia Schultesiana trom T. polyphylla: the receptacle-cup is cupuliform in outline rather than oblong-rotund, the hairs of the petals are subappresed, fewer, and do not extend to the margins, the anthers are longer, the stipe of the pistil is thicker and much longer at maturity, and the ovarian hairs are frequently fewer along the two longi- tudinal areas of the valves, thus giving the ovary a stri- ate appearance.

Occasional petals show small and scattered glandular

[ 153 ]

EXPLANATION OF THE ILLUSTRATION

PLrare XXIX. a-c, JACQUESHUBERIA AMPLIFOLIOLA Cowan. a, habit, one half natural size. b, one flower, one and one half times natural size. c, por- tion of inflorescence, one half natural size.

d—-f, CALLIANDRA VAUPESIANA Cowan. d, habit, one half natural size. e, two leaflets, illustrating vari- ation in size and shape, three times natural size. f, one flower, one and one half times natural size,

Drawn by Ricuarp S. Cowan

[ 154 ]

Pirate X XTX

bodies among the hairs. These have been observed pre- viously in petals of Vachigalia, although occurring on the inner surface of sepals of several species.

Apparently individual flowers or small clusters of flowers are subjected to insect attack. Gall-like forma- tions, somewhat rounded and measuring up to 1.5 cm. length, result. The pedicel of a gall-bearing flower is ob- vious; the receptacle-cup, while tumescent and abnor- mally large, maintains much of its symmetry, and the sepals and petals are larger, more crassate, and appar- ently more pubescent. The stamens and pistil form an amorphous mass within. Dissection of the mass reveals that it is somewhat coriaceous in texture; pollen grains and ovules were not found. These gall-like structures are soon deciduous.

Cotomp1a: Comisarias del Amazonas y Vaupés, Rio Apaporis, Cachi- vera de Jirijirimo and vicinity. Altitude about 250 m. “‘Large tree. Flowers yellow.’’ September 16, 1951. Richard Evans Schultes & Isi- doro Cabrera 14045 (Type in Herb. Mo. Bot. Gard.).

BURSERACEAE

Paraprotium amazonicum Cuwatrecasas sp. nov.

Arbor magna, ramis extremis minute tomentulosis brunneis angulosis. Folia alterna composita, in specimine paripinnata 8-juga, 90-100 cm. longa, foliolis petiolula- tis oppositis. Petioles 80 cm. longus robustus rigidus striatus basi incrassato-dilatatus, minute puberulus vel glabratus in rhachidem productus. Rhachis 40 em. longa internodiis 5—6 cm. longis striata minute puberula. Peti- oluli circa 12 mm. longi robusti erecti vel striati minute puberuli. Lamina foliolorum oblonga, basi rotundata subsymmetrica vel leviter asymmetrica (in foliolis duobus terminalibus basi cuneata asymmetrica excepto), apicem versus paulo attenuata apice subite cuspidato-caudata, margine integerrima, 14-26 cm. longa, 5.5—7 cm. lata,

[ 156 |

plus caudicula 2—2.5 em. longa; supra in sicco pallide brunnescenti nitida glabra costa angusta notata, nervis secundariis filiformibus prominulisque, nervulis minori- bus reticulatis plusminusve prominulis; subtus pilis mi- nutis conicis rubellis copiosis praedita, costa crassa elevata striata puberula, nervis secundariis 12-16 utroque latere prominentibus patulis prope marginem subite arcuatis anastomosatisque, ceteris nervis laxum reticulum prom- inulum formantibus. Inflorescentiae axillares, in speci- mine fructiferae ad terminationem ramuli, paniculatae ; axibus robustis striatis, basi valde incrassatis, minute tomentulosis 22-80 cm. longis, ramulis patulis 7— — cm. longis tomentellis. Pedicelli fructiferi crassi 2-4 mm. longi tomentosi. Fructus vulgo 1-pyrenus oblique ovatus basi asymmetrice rotundatus extremo acutiusculus in sicco 2.2—2.5 cm. longus, 1.5 cm. latus, vel 2-pyrenus late ovatus adpressus basi subtruncatus apice subacutus in sicco 2.2 em. altus, 2.8 cm. latus; epicarpio coriaceo extus brunneo tomentoso denique dehiscenti, mesocarpis pulposo, endocarpio corneo. Pyrenus ovoideo-oblongus zygomorphus uno latere gibbosus basi rotundatus apice angustatus triquetro-acutissimus circa 18 mm. longus, 9-10 mm. latus. Epispermum membranaceum, embry- one recto cotyledonibus plano-convexis circa 14 mm. longis (in sicco) 6 mm. latis.

Dr. Cuatrecasas writes: ‘‘Paraprotium amazonicum is closely related to P. vestitum Cuatr., from the Pacific Coast of Colombia. It differs from it by its leaves having more numerous oblong, narrower leaflets which are mi- nutely puberulent and more prominently reticulate beneath. Furthermore, Paraprotium amazonicum has smaller and more zygomorphous fruits and more acute pyrenes. ”’

Brazit: Estado do Amazonas, Rio Negro, Jucabi (near mouth of Rio Curicuriari) and vicinity. ““Tree 75 feet tall with columnar trunk above buttress and stilt roots to 4 ft. Bark outside smooth, brown,

[ 157 ]

inside red. Terminal branches with white, latex-like sap. Fruit out- side golden, inside smooth, pink. Seed olive green, surrounded by acidic white pulp. On high Jand.’’? September 21-25, 1947. Richard Evans Schultes & Francisco Lépez 8835 (Tyrer in Gray Herb. ).

JUTTIFERAE

Dr. Bassett Maguire has submitted the following criti- ‘al notes on the genus Lorostemon.

In an area near Mandos, where for more than a cen- tury the famous botanical explorers of the Amazon Basin (Martius, Spruce, Ule and others) had made headquarters for long periods of time, Ducke discovered a remarkable stand of small trees to which he gave count, “*. . . ving- taine d’arbres adultes. ..°° From specimens collected there, he described Lorostemon bombaciflorum which he proposed (Ducke, A. in Arch. Inst. Biol. Veg. 1 (1985) 210) as the type of a new monotypic subfamily, the Lorostemonoideae of the Guttiferae. Principal criteria upon which the new subfamily was based were: 1, the arrangement of the anthers, five each in five concrete un- divided phalanges; and, 2, the ovary borne on a long gynophore similar to that of many capparids. Arrange- ment of the stamens in five phalanges is, of course, not unique in the Guttiferae, being generally characteristic of the genera of the Moronobeae. But the stipitate ovary base was not otherwise known for the family.

I have before me an isotype of Lorostemon bombaci- florum: Ducke 28768 (NY)—and two additional sheets, Ducke 944 and 1200, obtained from the type locality in 1942 and 1948, respectively. All three collections are faithful to the original description.

Now, Schultes and Cabrera have made three addi- tional collections of Lorostemon, but from the region of the Rio Apaporis, an affluent of the Caqueta or Japura, some 800 miles to the west and north of Mandos. Two

[ 158 |

collections (Schultes & Cabrera 17619 of September 25, 1952, and 17638 of September 25-26, 1952) differ strongly from the isotype and topotypes of the original species, as the following table of comparison will show. The third, a fruiting collection (Schultes & Cabrera 15910 of March 2, 1952) has the dark-colored branchlets, smaller gradate and more numerous sepals of Lorostemon bombaciflorum, and, as in that species, lacks the conspicuous torus-like peduncle between the bractioles and calyx. The leaves, however, are oblong and not obovate as in Lorostemon bombaciflorum. Schultes & Cabrera 15910 is tentatively assigned to Lorostemon bombaciflorum until flowering material of it is collected.

Branchlets Leaf-blades

Calyx

Peduncle

Petals

Stamens

L. bombaciflorum dark reddish brown

10-15 4-6 em., broadly obovate

sepals gradate, 9-11, ovate,

ad

acute, 6-12 mm. long

peduncles lacking, flowers sessile

80-105 10 mm,

in 5 wholly concrete phal- anges, anthers 5 per phal- ange

L. colombianum

pale reddish brown

18-22 X 6-7 cm. ,oblong- elliptic

sepals little or not at all gradate, 5, orbicular, 18-20 mm. long

peduncle 8-12 mm.long, crassate, turbinate

80-90 X20 mm.

in 5 phalanges, filaments adnate only below the middle, above the mid- dle free, anthers 10-12 per phalange, free

Lorostemon colombianum Maguire sp. nov. Arbor mediocris 6—7 metralis alta cum latice luteo;

ramulis glabris non-angulatis, tenuis, internodiis 8-6 cm. longis; foliis oppositis aliquando suboppositis vel alter- natis; laminis firme chartaceis vel subcoriaceis, oblongis vel elliptico-oblongis, vulgo 18-22 cm. longis, 4.5—7 cm. latis, apice prominente acute et abrupte acuminato, 10—

[ 159 |

15 mm. longo; basi obtusa vel acuta; costa prominenta, venis lateralibus parallelis, prominulis 1-2 mm. distanti- bus, nervo collectivo a margine ca. 1 mm. remoto; flora solitari terminale exterior] rubro sursum maculato; brac- teis paucis aggregatis acute triangularibus vel ovato- triangularibus 2-5 mm. longis: pedunculo crasso turbin- ato 5-10 mm. longo; sepalis 5, imbricatis non-gradatis ovato-orbicularibus, 15-18 mm. latis, 15-20 mm. longis, subseariosis; petalis 8-9 cm. longis, ca. 2 cm. latis, lineari-oblanceolatis, apice acutiusculo; staminibus 50— 60 in 5 phalangibus, quisque cum 10-12 staminibus; fil- amentis ad basim ca. 2 cm. in ligulo connatis, partis liberis planis tenuis, ca. 0.6-0.7 mm. latis, 2—2.5 cm. longis; antheris introrsis linearibus 2—2.5 cm. longis, 0.6—-0.7 mm. latis, apice saepe recurvato, connectivo ali- quantum alato; pistillo longi-stipitato, ovario 5-loculari, ovulis numerosis, stylis brevi-cornutis, 5; fructu ignoto.

Cotomsia: Comisaria del Amazonas, Rio Ricapuy4, tributary of Rio Apaporis below mouth of Rio Piraparand. September 25-26, 1952. Richard Evans Schultes & Isidoro Cabrera 17638 (Type in Herb. N.Y. Bot. Gard. ).—Comisarias del Amazonas and Vaupés, Rio Apaporis at mouth of Rio Piraparand. “‘“Tree 6 m. high. Flowers red outside, greenish at base and red-spotted above; calyx green. September 25, 1952. Schultes & Cabrera 17619.

MELASTOMACEAE

The following studies in the Melastomaceae have been contributed by Dr. John Wurdack.

Centronia vaupesana Wurdack sp. nov.

Centroniae reticulatae Triana affinis sed cum foliorum laminis supra non reticulato-venosis indistincte 5-nerviis, ovario apice glabro. Rami robusti indistinete suleati cum petiolis pedunculoque primum rufo-furfuracei mox glab- rati. Petioli 2-8 em. longi; lamina 11-17 X4-7.5 cm. ovata apice hebeti-acuta paulo acuminata) basi rotun-

[ 160 ]

data, integra 5-nervia nervis duobus exterioribus sub- marginalibus debilioribus nervis primarlis secundarlisque supra leviter angusteque impressis subtus bene elevatis nervulis supra obscuris subtus paulo elevato-reticulatis, supra glabra et nitidula, subtus in superficie glabra in nervis sparse furfuraceis. Inflorescentia 4—5 cm. longa pauciflora, floribus 5-meris fasciculatis. Pedicelli 4-7 mm. longi cum hypanthio calyceque setulosi, setis crassis (ad basim 0.8 mm. diam.) usque ad 1.5 mm. longis minute tuberculatis. Hypanthium (ad torum) 74-5 mm.; calyx in alabastro clausus 9 mm. longus conicus ad an- thesim ad torum dehiscens. Petala 20X18 mm. apice oblique rotundata. Stamina isomorphica; filamenta 6 mm. longa; thecae 7 mm. longae subulatae, connectivo basi non prolongato dorsaliter calcarato calcari acuto 1.6 mm. longo. Stylus ca. 14.0.5 mm. ; stigma truncatum ; ovarium 5-loculare glabrum apice in collum 1.3 mm. altum circum stylum protractum, collo apice breviter (0.2-0.8 mm.) 10-lobato.

Spruce 4840 (KK), the type collection of Centronia reticulata, has been compared with C. vaupesana. The Peruvian species has striking upper leaf surface venule reticulation, 3-nerved leaf blades with no trace of fourth and fifth veins, slightly larger flowers, and (despite Cogniaux and Triana’s ‘glabrous’ descriptions) the ovary apex sparsely puberulous with slender brown hairs to 0.3 mm. long. Centronia laurifolia D. Don is another possi- ble relative, but its leaves are larger and 3-plinerved, the flowers ‘subsessile,’ and the ‘subacute’ petals only 1 em. long.

Cotomsra: Comisaria del Vaupés, Rio Piraparand, Raudal Na-hoo- gaw-he. ‘*Spindly tree 8 m. tall. Flowers white. Maku name: ka- nang.’? August 30, 1952. Richard Evans Schultes & Isidoro Cabrera 17100 (Tver in Herb. N.Y. Bot. Gard. ).

[ 161 |

EXPLANATION OF THE ILLUSTRATION

Pirate XXX. LorostemMon coLomBIANUM Maguire, drawn from the holotype, R. EF. Schultes & I. Cab- rera 17638 (NY). 1, flowering branch, one half natural size. 2, ventral view of phalange of sta- mens, one and one half times natural size; note dextrorse orientation of anthers. 3, cross-section diagram showing arrangement and orientation of filaments at position 3-3, 4, ventral view of ex- trorsely oriented anther, twenty times natural size. 5, cross-section of anther at positions 5-5, twenty times natural size. 6, cross-section of filament at positions 6-6, twenty times natural size. 7, lateral view of anther, twenty times natural size. 8, pis- til, one and one half times natural size.

Drawn by C. C. Clare

[ 162 ]

i run. a. @. 4

=I (O

eet Lobia Sere ne ok, Oe GUN ARIS ANI a Ca EER CNIK Bit _— SHR OAS

AO

©

Ny

colombianum Maguire

LOROSTEMON

Macairea Schultesit Wurdach sp. nov.

Macairea arirambae Hub. et M. scabrae Cogn. affinis, sed foliis breviter petiolatis 5-nerviis. Fruticosa ca. 1- metralis, ramulis pilis rufidulis eglandulosis appressis ca. 1 mm. longis densiuscule vestitis. Petiolus 2-5 mm. longus; lamina 2.5-6X2-8.5 mm. elliptica vel paulo ovato-elliptica apice late acuta bis obtusa basi obtusa bis subecordata margine integra 5-nervia vel sub—7-nervia, supra dense tuberculato-aspera setis 1— 4mm.° basi ad 0. 4 mm. expansis apice pilo deciduo ad 0.5 mm. longo arm- atis, subtus densissime strigulosa pilis gracilibus rufes- centibus eglandulosis ad 0.38 mm. longis (ad nervos ad 1.5 mm. longis) et cum guttulis flavidis modice inter- spersa. Bracteae inflorescentiae 8-5 X2-8 mm. deciduae sparse pilosulae extus modice gracili-strigulosae et sparse cum guttulis flavidis armatae; pedunculi pedicellique pubescentia ut ramulorum eadem; pedicelli 1-4 mm. longi. Hypanthium 2.7-8X2-2.8 mm. cum sepalis ex- tus modice graciliterque rufo-strigulosum et cum guttulis flavidis dense inspersum.