BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY PRINTED AND PUBLISHED AT THE BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VOLUME XVI BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS 1953-1954 TABLE OF CONTENTS NuMBER 1 (February 12, 1953) Miscellaneous Notes on Tropical American Orchids By CHARLES SCHWEINFURTH NuMBER 2 (March 16, 1953) Studies in the Genus Hevea VII By Ricwarp Evans SCHULTES . NuMBER 8 (May 15, 1958) Studies on Peruvian Pleurothallis By CHARLES SCHWEINFURTH NuMBER 4 (June 19, 1953) Plantae Austro-Americanae VIII By RicHarD EVANS SCHULTES . NuMBER 5 (July 6, 1958) Notes on the Cultivated Lulo By Ricwarp Evans SCHULTES AND JOSE CUATRECASAS Plantae Colombianae XIV By RicuHarp Evans SCHULTES . A New Saurauia from Mexico By Ricuarp Evans SCHULTES . Numser 6 (December 4, 19538) The Pre-Columbian Cultivated Plants of Mexico By Rosertr L. DreEssLeER [v] 21 57 97 . 106 . 112 . 115 Number 7 (December 31, 1953) Frederick Oliver Thompson, 1883-19538 By Exso S. BAaRGHOoRN. ........ . . 178 NuMBER 8 (February 19, 1954) Plantae Austro-Americanae IX By Ricuarp Evans SCHULTES. ..... .. .179 Number 9 (July 12, 1954) Fossil Maize from the Valley of Mexico By Enso S. BarGHoorn, MarGarer K. WoLFE AND Katuryn H. Cuispy...... . . . 229 A New Narcotic Snuff from the Northwest Amazon By Ricuarp Evans ScHULTES. ...... . 241 The Origin and Possible Evolution of Sub-tassel Ears in Maize By WaLron C. GALINAT ......... . 261 NuMBER 10 (November 26, 1954) On the Origin of Oats By Dexter R. Sampson .......... . 265 [vi ] INDEX OF ILLUSTRATIONS PLATE Anthurium atropurpureum Schultes & Maguire XXII Anthurium fontoides R..Schultes . . . . . . IX Anthurium nemoricola Schultes & Maguire . . XXII Anthurium pluviaticum R..Schultes . . . . . IX Anthurium tikunorum R.H.Schultes . 2... 2 2 2 xX Carludovica aurantiaca R.H.Schultes . . . VII, VIII Clusia chiribiquetensis Maguire . . . . . XXXIII Clusia Schultesii Maguire ........XXXIV Combretum Wandurraganum R.F.Schultes. . XIV Cuphea beneradicata Lourteg. . . . . . XXXVII Cuphea kubeorum Lourteig ....... XXXV Cuphea philombria Lourteg . ..... .XXXVI Cuphea stygialis Lourteyg ....... XXXVII Cuphea sucumbiensis Lourteg ..... .XXXVI Cuphea sunubana Lourteig . . . . . . . . XXXV Frederick Oliver Thompson. . ....... XXI Herrania breviligulata R.#.Schultes . . . . . . XII [ vii ] Herrania pulcherrima Goudot. . . . . . . XXXI Herrania purpurea (Pitt.) R.H.Schultes . . . XIII Herrania tomentella R..Schultes . . . 2. . XXXII Hevea microphylla Ule. ........... (Il Hevea rigidifolia (Spruce ex Benth.) Muell.-Arg. IV Mendoncia lasiophyta Leonard . ....... XV Navia Garcia-Barrigae L.B.Smith . . 2... . XXV Navia heliophila 1. B.Smth. . . . XXVI, XXVIII Navia Lopezi L.B.Smth ....... .XXVIII Navia myriantha L.B.Smith . . . . . . . XXVIII Navia reflexa L.B.Smith. . ....... .XXVI Nicotiana rustica Z.. . . . . . .. MX Paepalanthus fasciculatus (Rottb.) Korn. . . . . XI Paepalanthus Moldenkeanus R. E. Schu/tes XXIII, XXIV Pitcairnia macarenensis L.B.Smith . . 2. . . NXV Pleurothallis angustipetala C.Schwemf. . 2... . V Pleurothallis excisa C. Schweinf. . . . . . . . . Vi Rhytidanthera regalis R.H.Schultes . . . . XVIII Sanchezia thinophila Leonard... . .. .. XVI Saurauia Comitis-Rossei R.l.Schultes . 2. . . XIX [ viii ] Siphonia nitida Mart. ............ .THI Siphonia nitida Mart., manuscript description. . . I Solanum quitoense Lam. var. septentrionale R.E.Schultes & J.Cuatrecasas ..... . XVII Tetrapteris methystica R.H.Schultes. . . . . XXX Thurnia sphaerocephala (Rudge) Hook.f. . . . XXIV Vellozia lithophylla R.#.Schultes . . 2... . XXIV Vellozia Maudeana FR. M.Schultes . . XXIX, XXX Virola calophylla Warburg . . XX XIX, XLI, XLII Virola calophylloidea Markgraf ....... XL Zea Mays L. pollen grains ...... .XXXVIITI Zea Mays L. sub-tasselears .... XLIII, XLIV [ ix ] INDEX TO GENERA AND SPECIES AA chiogena Schltr., 2 Weberbaueri Schltr., 2 ABUTA, 257 rufescens Aublet, 67 ACACIA angico, 257 Niopo, 252 acahualli, 134 ACANTHACEAR, 57,92 achiote, 124 achiotl, 124 ACHRAS, 120 Zapota L., 138 acocotli, 132 acocoxochitl, 132 AGAVE, 120,121 atrovirens Karw., 120,153 fourcroydes Lem., 121,153 latissima Jacobi, 120,153 mapisaga 7T’rel., 120,153 sisalana Perrine, 121,153 tequilana Weber, 120 aguacate, 141 ahate, 124 ahudcatl, 141 ahuehuete, 148 ahuéhuetl, 148 alegria, 121 algodén, 132 ALTENSTEINIA chiogena (Schitr.) C. Schweinf. z Weberbaueri (Schltr. ) C. Schweinf. , 2 AMARANTHUS, 158 cruentus L., 121,154 dubius Mart., 122 hybridus L., 121 leucocarpus 8. Wats., 121,153 Powellii 8. Wats., 121 ANACARDIUM occidentale L., 122 ANANAS, 156 comosus (L.) Merr., 122,154 anil, 135 ANNONA, 122,128 Cherimolia Mill, 123,154,156 diversifolia Safford, 123,153 glabra L., 123 muricata L., 123 purpurea Moc. & Sessé, 123, 153 reticulata L., 124 sect. Atta, 124 squamosa L., 124 anona, 124 ANTHURIUM atropurpureum Schultes & Maguire ex Schultes, 60, | 179,180 var. apertum Schultes, 180 | cabrerense Engl., 183 [x ] chlorocarpum Sodero, 63 crassinervium (Jacq.) Schott., 181 fontoides R. E. Schultes, 60,61 gracile (Rudge) Engl., 61,181 Idroboanum R.E.Schultes, 181,182 macarenense Ft. EF. Schultes & Idrobo, 183 macrocephalum R. E.Schulies, 184 magnificum Linden, 182 nemoricola Schultes §- Maguire ex Schultes, 61,62,64,185 oblongo-cordatum Engl., 185 panduratum Mart. ex Schoit., 185 pangoanum Sod., 184 pentaphyllum (Aubl. ) G. Don, 186 pluviaticum R. FE. Schultes, 62, 63 popayanense Engl., 186 reticulatum Benth., 186 scolopendrium (Ham. ) Kunth, 186 § Cardiolonchium Engl., 182 § Digitinervium Engl., 184 tikunorum R. E.Schultes, 63, 64 apazote, 128 APOCYNACEAE, 89,228 ARACEAE, 60,179 ARACHIS, 156 hypogaea L., 124,154 arataciu, 71 ARCTINURUS thompsoni Miller § Unklesbay, Va.178 ARRABIDAEA Fanshawei Sandwith, 91 xanthophylla Burret & K. Schum., 91 arrayan, 145 AVENA, 265,267,270,281, 287,292,297 abyssinica Hochst., 270,278, 274,279, 280,282,283, 285- 288,298 barbata Pott, 270,272-274, 280-288 brevis Roth, 266,267,272,296 Bruhnsiana Grun., 269,271, 280,292 byzantina C. Koch, 270,273, 275,276,279, 280, 283, 284, 292-294, 298,299 ssp. byzantina, 276 prol.denudata (Hausskn.) Malz., 276 ssp. nodipubescens Malz., 276 ssp. pseudo-sativa Thell., 276 clauda Dur. ,266,268-27 1,280 eriantha Hack., 266 fatua L., 266-268,270,275- 280, 283-286, 289,295,298 ssp. cultiformis Ma/z., 268, 277,2(8 ssp. fatua, 267,268,277, 278 ssp. macrantha, 268 ssp. meridionalis Malz., 268,277 278,298 ssp. nodipilosa, 268 ssp. nuda, 267 ssp. praegravis, 268 ssp. saliva, 267,268 ssp. septentrionalis Malz., 268,277,278 [ xi ] hirsuta Moench., 266 hirtula Lag., 269,272,273, 280,282,288 longiglumis Dur., 266,268- 271,280,285,286, 288,290 nuda L., 266,273 var. biaristata Asch. & Gr., 272 nudibrevis Vav., 272 orientalis Schreb., 266,267, 279 pilosa M. Bieb., 268-271,280 sativa L., 266,270,273,278- 280,283-286,289-299 ssp. macrantha (Hack. ) Malsz., 278 prol. nudata Mals., 278 ssp. nodipilosa Malz., 278 prol. decorticata Malz., 278 ssp. seein (Krause) Mals., 2 prol. t oedianiis Malz., 279 ssp. sativa, 278 prol. chinensis (Fisch. ) Malz., 278 var. abyssinica Engl., sect. Avenastrum, 266,26 sect. Euavena, 266 ser. Biformes, 266,267,278, 291 ser. Conformes, 266,267,278 ser. Eubarbatae, 268,269, 282 ser. Inaequaliglumes, 268, 269 ser. Stipitatae, 268,269 sterilis L., 266,268,270,274- 280,283,284, 286, 290-292, 294,295,298 [ xii ] ssp. byzantina, 267,268 ssp. Ludoviciana ( Dur.) G. & M., 268,275,276,298 ssp. macrocarpa (Moench) Briq., 267,268,275 ssp. nodipubescens, 268 ssp. pseudo-sativa, 268 ssp. sterilis, 275,276 ssp. trichophylla (C. Koch) Malz., 268,275,276 strigosa Schreb., 266,268,269, 27 1-273,279, 280,282,284, 285,295-298 ssp. abyssinica (Hochst. ) Thell., 267,268,274 ssp. barbata (Pott) Thell., 267,268,274 ssp. hirtula (Lag.) Malz., 268,272,288 ssp. strigosa (Schreb. ) Thell., 267,268,272 prol. brevis (Roth) Hausskn., 272,288 prol. nuda (L..) Hausskn. O70 a(x ssp. Vaviloviana Malz., 268, 273,274 ssp. Wiestii (Steud.) Thell., 267 ,268,273,274 var. eas Thell., 273 subsect. Agrestes, 266 subsect. Aristulatae, 267-270 subsect. Denticulatae, 267- 270 subsect. Sativae, 266 Vaviloviana (Mals.) Mordv., 270,273,279,280,288 ventricosa Bal., 266,268, 269,271,280 ssp. Bruhnsiana, 268 ssp. ventricosa, 268 Wiestii Steud., 270,272,273, 280-282, 288 azi, 136 avin, 136,147 ayecote, 142 ayécotl, 143 ayotli, 130 BANISTERIOPSIS Caapi (Spruce ex Griseb.) Morton, 204 spp., 248 batata, 135 BIGNONIACEAR, 91 BIXA Orellana L., 124,154 bledos, 121 BLETIA ? ensformis Ruiz & Pavon, 11 BOMAREA edulis (Tuss.) Herb., 125 BomsBacackak, 75 BOMBAX coriaceum Mart. & Zucc., 75, 188,201 sordidum R. E.Schultes, 75 boscacao, 79 BROCCHINIA hechtioides Mez, 193 BROMELIA, 121,122 BroMELIAckAk, 179,192 BROSIMUM Alicastrum Swartz, 125,153 BUCHTIENIA boliviensis Schitr., 2 bul, 142,143 bule, 136 BYRSONIMA crassifolia (L.) DC., 125,153 caapi, 204,248 cacahuate, 124 cacaito, 207 cacao, 148 cacao de monte, 213 cacao cahouai, 211 cacao esquinado, 213 cacaoballi, 71 cacaoito, 213 cacaoilo de monte, 217 cacomite, 149 cacomitl, 149 calabasa, 130 CALLIANDRA Sp: 201 CALOCARPUM mammosum (LL.) Pierre, 125, 153 viride Pitt., 126,153 camote, 135 camolli, 135 CAMPYLOCENTRUM Ulaei Cogn., 20 var. peruvianum C. Schweinf. , 19 CANAVALIA, 131 ensiformis (L.) DC., 126,154 CAPSICUM, 158 annuum L., 126,154 frutescens L., 126,127,154 pubescens FR. & P., 127 capulin, 145 CARICA cauliflora Jacg., 127 Papaya L., 127,153 [ xiii ] CARLUDOVICA aurantiaca R. E.Schultes, 58, 96 pygmaea Gleason, 59 CASIMIROA edulis LaLlave § Lex., 128, 153 Sapota Oerst, 128,154 CASSIA fastuosa Willd., 257 CEDRELINGA catenaeformis Ducke, 257 cempoalxochitl, 147 centli, 150 CEPHALOCARPUS Dracaenula Nees, 58 ceresa, 145 CHAMAEDOREA, 128 Tepejilote Liebm., 128,153 Wendlandiana (Oerst.) Hemsl. 128,153 chan, 134 chay, 129 chaya, 129 chayole, 146 chayotli, 146 CHENOPODIUM, 155,158 ambrosioides L., 129 Berlandieri Moqg., 129 Nuttalliae Safford, 128,154, 156 Ppueblense Reed, 128 Quinoa Willd., 128,129 chia, 184,146 chia de Colima, 134 chia grande, 134 chiantzozolli, 146 chicozapote, 138 chimalacatl, 134 chimalatl, 134 chinini, 141 chipilin, 130 chirimoya, 128 CHONDODENDRON toxicoferum (Wedd.) Krukoff & Moldenke, 67 cimatl, 143 ciruela, 147 ciruela amarilla, 147 CITHAREXYLUM Ulei Moldenke, 90 CLUSIA chiribiquetensis Maguire ex Schultes, 219 Schultesii Maguire ex Schultes, 219 spp., 259 CNIDOSCULUS, 136 aconitifolia (Mill.) I. M. Johnst., 129 Chaya Lundell, 129 Chayamansa McVaugh, 129, 153 coca, 68,248 cochitsapotl, 128 coco, 129 COCOS, 158 nucifera L., 129 CoMBRETACEAR, 87 COMBRETUM laxum Jacq., 87 rotundifolium L.C. Richard,88 Wandurraganum, R. FE. Schultes, 88 ComposiITaAk, 95 [ xiv ] COPAIFERA officinalis L., 259 copal, 144 costiczapotl, 144 COUMA catingae Ducke, 228 utilis (Mart.) Muell.-Arg., 89 coyo, 141 coyolxdchitl, 125 costicxécotl, 147 CRATAEGUS mexicana Moc. & Sessé, 130 pubescens (H BK.) Steud., 130,153 stipulosa (HBK.) Steud., 130 CRESCENTIA alata HBK., 130 Cujete L., 130,153 CROTALARIA longirostrata Hook. & Arn., 130,154 cuajilote, 141 cuauhzontli, 128 cuchara-caspi, 90 CUCURBITA, 118,130,158 ficifolia Bouché, 131,154 maxima, 131 mixta Pang., 131,153 moschata Duch., 131,153 Pepo L., 131,153 texana Gray, 131 CUPHEA, 179 annulata Koehne, 227 beneradicata Lourteig, 221, 226 ciliata Ruiz & Pav., 220 fuscinervis Koehne, 220 kubeorum Lourteig, 221 microphylla HBK., 220 philombria Lourteig, 221,223, 224 sect. Brachyandra subsect. Melanium, 220 sect. Euandra subsect. Hyssopocuphea, 220,221 stygialis Lourteig, 221-223 sucumbiensis Lourteig, 221, 225,226 sunubana Lourteig, 219 CycLANTHACEAE, 58 CYMBIDIUM muricatum Sw., 18 CYNOMETRA Zamorana R.E.Schultes, 68 Cyperaceak, 58 da-ko-ree, 202 DAHLIA coccinea Cav., 132,158 excelsa Benth., 132 Lehmannii Hieron, 132,154 Mazonii Saff., 132 pinnata Cav., 132,153 rosea Cav., 132 variabilis Desf., 132 DATURA arborea L,, 91 suaveolens Humb. & Bonpl., 91 DICHAEA maculata Poepp. & Endl., 17, 18 muricata (Sw. ) Lindl., 17,18 var. maculata (Poepp. & Endl.) C. Schweinf., 17, 18 panamensis Lindl., 18 DICHAEOPSIS panamensis Schltr., 18 [ xv ] DILLENIACEAR, 81 DIOSPYROS Ebenaster Retz, 132,153 EPIDENDRUM aquaticoides C.Schweinf., var. pusillum C.Schweinf., 12,13 cyperifolium C. Schweinf. , 10 gracillimum Reichb.f., 15 inamoenum Krdnszl. var. robustum C. Schweinf. , 11 longirepens (C. Schweinf. ) C.Schweinf., 12 microtos Reichb.f. var. grandiflorum C, Schweinf., 10 pachychilum Krénszl., 12,13 refractoides C. Schweinf. var. humile C. Schweinf. , 13 Schlechterianum Ames var. longirepens C. Schweinf., 12 Schlimii Lindl., 14 Schlimii Reichbf., 14,15 var. gracillimum (Reichb,f. & Warscz.) C.Schweinf., 15 serricardium Schltr., 16,17 subreniforme C. Schweinf. var. ramosum C. Schweinf. , 15 tenellum Lindl., 14 vineentinum Lind/l., 16,17 EPITHECIA panamensis Schltr., 18 ErrocauLackak, 57,65,187,188 ERYTHROXYLUM Coca L., 248 estrellitas del sur, 191 EUCHLAENA, 230,262 mexicana, 150,262 EvupuorsBiackaAk, 22,69,205 Jlor de los muertos, 147 FOURCROYA, 121 Jryol, 142,143 girasol, 134 GODOYA, 109 GOMPHICHIS Koehleri Schltr. var. minor C. Schweinf, 1 GONGYLOLEPIS maroana Badillo, 95 GOSSYPIUM, 138,158 arboreum L., 133 barbadense L., 133 hirsutum L., 132,133,154 Raimondii Ulbrich, 133 GRAMINEAR, 57 guacamole, 137 guandbana, 123 guayaba, 145 guayabilla, 145 GUAZUMA, 206 giiisquil, 146 GUTTIFERAE, 219 ha-oom-tee-et, 248 haatie, 36 hakudifla, 245 HELIANTHUS, 158 annuus L., 134,153 var. lenticularis (Dougl.) Ckil., 134 var. macrocarpus (DC.) CkIl., 134 HELICTERES, 206 HELICTOTRICHON, 266,270 [ xvi | henequén, 120 HERRANIA, 76,205 albiflora Goudot, 75-77,207, 211 atrorubens Huber, 80 breviligulata R. EF. Schultes, 78,207 Camargoana R. E.Schultes, 207-209 Cuatrecasana, 218 kanukuensis R. EF. Schultes, 78, 79,207 laciniifolia Goudot ex Triana § Planch., 79,80,209 lemniscata (Schomb.) R.E. Schultes, 79,207 nitida (Poepp.) R. E.Schultes, 80,81,210,218 var. aspera (Karsten & Triana) R.E.Schultes, a8 lo nycterodendron R. FE. Schultes, 209,210 pulcherrima Goudol, 81,211- 213,218,219 purpurea ( Pitt.) R. E.Schultes, 81 tomentella R. FE. Schultes, 205, 206,213,217-219 HEVEA, 21-23,29,34 andenensis C.F.Jones, 30 andinensis Sperber, 30 apiculata Baill., 27 Benthamiana Muell.-Arg., 25,39-41 brasiliensis (Willd. ex A. Juss.) Muell.-Arg., 26-28,33,34, 39-41,90 confusa Hemsl., 35,37 discolor ( Benth.) Muell.-Arg., 23 guianensis J'ubl., 27-30,33, 34,43 var. lutea (Spruce ex Benth.) Ducke & Schultes, 27-30, 40 lutea (Spruce ex Benth.) Muell.-Arg., 27 microphylla Ule, 38,40,41 nitida Mart. ex Muell.-Arg., 30-32,41,42 var. toxicodendroides, 188 pauciflora (Spruce ex Benth.) Muell.-Arg., 25,32-34,36, 37,42 var. coriacea Ducke, 34-36 rigidifolia (Spruce ex Benth.) Muell.-Arg., 25,35,43,44 Spruceana ( Benth.) Muell.- Arg., 23-25,86,37 subgen. Bisiphonia, 23 subgen. Euhevea, 23,28 viridis Huber, 32,41 hé-goé-ne, 65 huauhtli, 121 huauhtzonth, 128 HYLOCEREUS undatus (Haw.) Brit. & Rose, 134,153 HYMENAEA oblongifolia Huber, 68 HYPTIS suaveolens Poit., 134,153 hwan-hwan, 228 iczotli, 150 igarapé-assu, 71 INDIGOFERA Antl L., 185 suffruticosa Mill., 135,153 INGA sp., 253 [ xvi | injerto, 126 IPOMOEA, 155,158 Batatas (L.) Poir., 135,154 fastigiata (Roxb.) Sweet, 135 tiliacea (Willd.) Choisy, 135 iscatl, 132 izote, 150 istaczsdpoll, 128 Jjalocote, 145 JATROPHA, 136,147 aconityfolia Mill., 129 Cureas L., 136,153 jicama, 140 Jicara, 130 Jiquelite, 135 jitomate, 137 Jobo, 147 Jocote, 147 JuNcACEAR, 189 Jjutai, 68 ha-pet-o6-he, 69 kah-pe-ree, 92 ku-a-vd-u, 91 kurata, 245 LAGENARIA, 130,158 siceraria (Mol.) Standl., 136, 154 vulgaris Ser., 136 LecGuMINosak, 68,257,258 LEITGEBIA, 86 colombiana R.E.Schultes, 85- 87,188,201 Gleasoniana Lasser, 86 guianensis Kichler, 86 Imthurniana Oliver, 86 LEMAIREOCEREUS, 136,140 LEPANTHES Leonii C.Schweinf., 9 tracheia Reichbf., 8 trachysepala Schitr., 8 LEPANTHOPSIS microlepanthes (Griseb. ) Ames 9,10 LIGHTIA lemniscata Schomb., 79 LoRANTHACEAR, 202 LUCUMA, 136 salicyfolia HBK., 144 lulo, 97,98, 100 lulo de perro, 101 lulo morado, 100 LYCOPERSICON, 156,158 esculentum Mill., 137,154 var. cerasiforme (Dun.) A. Gray, 137 LYTHRACEAE, 219 MACROSTACHYA Thompsonii Darrah, 178 ma-sha-kve, 99 maguey, 120 maipoelie doron doron, 79 maiz, 150 mais de teja, 134 mala mujer, 129 MALAXIS termensis (Kréinzl.) C. Schweinf. var. elata C. Schweinf. , 10 MALOUETIA Tamaquarina (4ubl.) A. DC. 90 MALPIGHIACEAR, 202 mamey colorado, 125 mamey sapole, 125,126 mameyito, 113 [ xviii ] mani, 124 MANIHOT, 155,156 Aipi Pohl, 137 dulcis (J. F.Gmel.) Pax., 137 esculenta Crantz, 137,154 utilissima Pohl, 137 MANILKARA Zapotilla (Jacq.) Gilly, 138, 153 manzanilla, 130 maranon, 122 maripoele kakaoeleo, 79 matasano, 128 maté, 99 MEDULLOSA Thompsonii Andrews, 178 MENDONCIA gigas Lindau, 93 lasiophyta Leonard, 92,93 MENISPERMACEAE, 67 MESOXYLON Thompsonii Traverse, 178 metl, 120 miltomatl, 144 MIMOSA acacioides Benth., 252-254 MONOPTERYX Uaucu Spruce ex Bentham, 69 moquillo, 83 murascaca, 71 MYRISTICA, Bicuhyba Schott, 259 calophylla Spruce, 242 fragrans Houtt., 246,247 spp., 247 MyRISTICACEAE, 246,247 nance, 125 nanche, 125 nantsinxdcotl, 125 naranjas de Quito, 99 naranjilla, 99 naranjillo, 97,99 naranjitas de Quito, 99 nardo, 144 NAVIA angustifolia (Bak. ) Mez., 194 caulescens Mart. ex Schult f. var. minor Schult. & Schult f. , 193,201 Garcia-Barrigae L. B. Smith, 194,196 heliophila L. B. Smith, 194,195 Lopezii L. B. Smith ex Schultes, 195 var. colombiana L. B. Smith, 195 myriantha L.B.Smith ex Schultes, 196 reflexa L.B.Smith, 196 xyridiflora L.B. Smith, 195 NEALCHORNIA japurensis Huber, 69,70 NICOTIANA, 118,139 rustica L., 138,139,154,155 Tabacum L., 138,139,154, LoG niopo, 252,254,255 nochtli, 140 nopal, 140 nopal nocheztli, 139 NOPALEA cochenillifera (L.) Salm.- Dyck., 139,153 nopalli, 140 oceloxéchitl, 149 OcunackAk, 85,109 [ xix ] OLMEDIA sp., 249 oom, 248 OPUNTIA, 139,140 amyclaea Tenore, 140,153 ficus-indica (L.) Miller, 139, 140,153 megacantha Salm- Dyck, 140, 153 streptacantha Lemaire, 140, 153 organo, 140 OTOBA incolor Karst. ex Warburg,242 ox, 125 pacaya, 128 PACHYCEREUS emarginatus (DC.) Brit. & Rose, 140,158 PACHYRRHIZUS erosus (L.) Urban, 140,153 PAEPALANTHUS fasciculatus (Rotth.) Kérnicke, 65 Moldenkeanus R. FE. Schultes, 187,188 Schultesii Moldenke, 65 pahua, 141 pajuil, 90 p’ak, 137 PANICUM molliculmum Swallen, 57 sonorum Beal, 141,153 papa, 147 PARAGONIA pyramidata (Rich.) Burret, 92 pa-ree-ka, 242 paricd, 242,245-247, 252-255, 257-260 parica branca, 258 parica da terra firme, 258 parica da varzea, 258 parica de cortume, 258 parica grande da terra firme, 258 paricarand, 252,258 paricazinho, 258 PARKIA spp., 257 PARMENTIERA edulis DC., 141,158 patachtli, 148 pataxte, 148 PAULLINIA searlatina Radl., 73,74 Yoco R.E.Schultes & Killip ex Schultes, 73,74 payé, 260 payes, 251 PERSEA americana Mill., 117,141, 153,154 var. drymifolia (Schlecht. & Cham.) Blake, 141 Schiedeana Nees,141,142,153 PHASEOLUS, 126,142 acutifolius 4.Gray, 142,153 coccineus L., 143,154 lunatus L., 143,154 multiflorus Willd., 143 vulgaris L., 142,143,153 PHYSALIS, 137 ixocarpa Brot., 144,153 peruviana L., 144 pubescens L., 144 picietl, 138 pitta, 122 piioncillo, 136 [ xx | PIPTADENIA, 257,258 peregrina L., 246,252,255, 201 spp., 257 pitahaya, 134 pitayo, 140 PITCAIRNIA macarenensis L.B. Smith, 192 pungens HBK., 192 patentiflora L.B.Smith, 193 PITHECOLOBIUM BPP... 257 PLEUROTHALLIS angustipetala C. Schweinf. , 45 ciliata Knowl. & Westc., 48 var. abbreviata C. Schweinf. 46 var. elongata C. Schweinf., 47 citrina Sch/tr. var. elliptica C. Schweinf. , 9 dolichopus Schlir., 50 ecuadorensis Schltr., 50,51 excisa C. Schweinf. , 48 fimbrilabia C. Schweinf. , 50 floribunda Poepp. & Endl., 51,52 juninensis Schltr. var, subaequisepala C, Schweinf. , 52 Lindenii Lind. var. longiracema C, Schweinf. , 53 macrophylla HBK., 51,52 macrophylla HBK. sensu Lindl., 51 macrophylla HBK. sensu Cogn., 51 microlepanthes Griseb., 9 pedunculata (K/.) Reichb,f. var, peruviana C. Schweinf. , 54 rubens Lindl., 46,50 tenuifolia C. Schweinf. , 56 var. longisepala C. Schweinf. 55 tunguraguae Lehm. & Krdanzl., 56 POLIANTHES tuberosa L., 144,153 po-muy, 36 pom, 144 PONTEDERIA, 190 POUTERIA campechiana (HBK.) Baehni, 144,153 hypoglauca (Standl.) Baehni, 144,153 mammosa (L.) Cronquist, 125 viridis (Pitt.) Cronquist, 126 PROTIUM Copal (Schlecht. & Cham.) Engl., 144,153 PRUNUS Capuli Cav., 145 salicifolia HBK., 145 serotina subsp. Capuli (Cav.) 7 McVaugh, 145,153 subsp. serotina, 145 PSIDIUM, 158 Friedrichsthalianum (Berg. ) Niedenzu, 146 Guajava L., 145,154 guineense Sw., 146 molle Bertol., 146 Sartorianum ( Berg.) Niedenzu 145,153 PSITTACANTHUS peronopetalus Erchler, 202 quauhcdmotl, 137 f SX | quauhtzdpotl, 123 quauhyetl, 138 quauxilotl, 141 ramon, 125 RapaTrackak, 189,190 rebalsa 90 RESTREPIA, 51 ecuadorensis Rolfe, 50 RHYTIDANTHERA, 109,110 magnifica, 109 mellifera, 109,110 regalis R. E.Schultes, 106, 108-110 splendida, 109 suleata, 109 RORAIMANTHUS, 86 SAGOTIA racemosa Baill., 70 var. brachypetala Muell.- Arg., 70 var. genuina Muell.-Arg., 70 var. ligularis Muell.-Arg., 70 var. macrocarpa Muell.- Arg., 70 SALVIA, 158 Chia Fern., 146 Chian LaLlave, 146 hispanica L., 134,146,153 polystachya Ort., 146 SAMBUCUS mexicana Presl, 146 SANCHEZIA thinophila Leonard, 94,95 SANDWITHIA guyanensis, 71 SAPINDACEAE, 73 sapole, 125 sapote blanco, 128 sauco, 146 SAURAUIA brachybotrys Turcz., 82,83 Comitis-Rossei R. E. Schultes, 112-114 pruinosa R, E. Schultes, 81-83 pseudoruiziana Busc,, 84 roseotincta R. E.Schultes, 83, 84 Ruiziana Steud., 84 var. Weberbaueri Busc., 84 Spraguiana Busc., 82,83 tomentosa Spreng., 84 SAUVAGESIA, 86 sauwi, 141 SCHIZOLOBIUM amazonicum ( Hub.) Ducke, 258 parahybum (Vell.) Blake, 258 SCHOENOCEPHALIUM arthrophyllum Seubert, 190 Martianum Seubert, 190,191 SECHIUM edule Sw., 146,154 SENEFELDERA contracta R. E.Schultes, 72,73 dodecandra, 73 inclinata Muell.-Arg., 73 macrophylla Ducke, 72 nitida Croizat, 72,73 SENEFELDEROPSIS, 205 chiribiquetensis (Schultes & Croizat) Steyerm, ex Schultes 205 Croizatii Steyerm. ex Schultes, 205 SENEGALIA spp., 258 [ xxii | seringa chicote, 39 sibisibi, 36 SIPHONIA, 34 brasiliensis Willd., 26,28 discolor Benth., 23-25 elastica Pers., 24,29 Kunthiana Baill., 33 lutea Spruce ex Benth., 27 nitida Mart., 31 pauciflora Spruce ex Benth., 25,87 rigidjfolia Spruce ex Benth., 25 Spruceana Benth., 23,37 SoLANACcEAR, 91 SOLANUM, 102,158 andigenum Juz. & Buk., 147 forma guatemalense Buk., 147 angulatum Ruiz & Pavon, 99, 103 angulosum, 105 quitense HBK., 99 quitoense Lam., 97-103,105 var. septentrionale R. FE. Schultes & J. Cuatrecasas, 98,100-104 tuberosum L., 147,156 Sorva, 89 SPHENOSTROBUS Thompsonii Levittan & Barg- hoorn, 178 SPONDIAS, 136,147 lutea L., 147 Mombin L., 147,153 purpurea L., 147,153 STELIS Endresii Feichb,f., 3 eublepharis Reichb,f. var. glabriflora C.Schweinf., 3 Herzogii Schltr., 4 Huebneri Schltr., 3 Koehleri Schitr., 5 leucopogon Reichb f., 5 Lindenii Lindl., 5 phaeomelana Schltr., 6 polycarpa Schltr., 6 santiagoensis Mansf., 7 simacoensis Schltr., 7 STERCULIA, 206 STERCULIACEAEK, 75,205 STYRACACEAR, 89 STYRAX rigidifolius forma yapobodensis, 188 Tessmannii Perkins, 89 yapobodensis (Idrobo & R.E. Schultes) Steyerm., 188 TAGETES erecta L., 147,148,153 patula L., 147,148,153 TAXODIUM mucronatum Ten., 148 tecomate, 130,136 tejocote, 130 tepejilote, 128 TETRAPTERIS discolor (G.F.W.Mey.) DC., 204 methystica R. E.Schultes, 202, 204 texocotl, 130 tezonzapotl, 125 THEOBROMA, 205-207 angustifolia DC., 148,149, Pos bicolor Humb. & Bonpl., 148, 149,153 Cacao L., 148,149,153,206 [ xxiii | leiocarpa Bernoulli, 148 Mariae (Mart.) Decn. ex Goudot, 76,79 var. lobata Pulle, 78,79 pulcherrimum (Goudot) de Wildem., 211 subincanum Mart., 249 thomé-assu, 71 THURNIA Jenmani Hook,f., 189 sphaerocephala (Rudge) Hook f., 189,190 THURNIACEAE, 189 TIGRIDIA pavonia (L.f.) Kerr., 149, 153 TILLANDSIA incurva Griseb., 198 patula Mes, 198 Tillandsiae, 198 tlaleacauatl, 124 tlaolli, 150 tlilrochitl, 149 tlilsapotl, 132 tomate, 144 tomatillo, 144 tomatl, 144 TRIPSACUM, 150,230,282, 236-238, 262 australe, 231 dactyloides, 262 fasciculatum, 262 lanceolatum, 262 maizar, 231 pilosum, 262 tu-to-a-vd-u, 91 tuna, 140 tzicosapoll, 138 uaucu, 69 vainilla, 149 VANILLA JSragrans (Salisb.) Ames, 149 planifolia Andr., 149,158 VELLOZIA, 201 Dumitiana R.E.Schultes, 201 lithophila R. E.Schultes, 188, 201 macarenensis Philipson, 200 Maudeana R. FE. Schultes, 198, 200,201 VELLOZIACEAE, 198 VERBENACEAR, 90 VIROLA, 241,242,246-248, 257,258 calophylla Warburg, 242- 244,256 calophylloidea Markgraf, 243 elongata ( Benth.) Warburg, 245 incolor Warburg, 242 lepidota A.C.Smith, 243 spp., 243 VRIESIA Barclayana ( Baker) L. B. Smith 198 Scbultesiana L. B.Smith, 197 WALTHERIA, 206 aralrocotl, 145 xicama, 140 xitomatl, 137 xviuquilitl, 135 xocoatl, 148 xocotl, 147 yaje, 248 yd-kee, 242,243,247, 248-251, 256,260 [ xxiv | ya-kee-oom, 248 yuca, 137 YUCCA, 121 aloifolia L., 150 elephantipes Regel, 150,153 ya-kee-taa, 249 yad-to, 242 yell, 138 capote prieto, 132 ea ZEA, 230 yopo, 252-255 Mays L., 150,153,230 Issued November 29, 1954 [ xxv ] ERRATA Page 2, line 8 for Schltr. read Schlechter Page 4, line 21 omit Beih. Page 5, line 27 for T’ read I Page 6, line 27 for Murayaco read Murayaca Page 65, line 2 for eretaceous read Cretaceous Page 65, line 13 for eambrian read Cambrian Page 68, line 24 for cretaceous read Cretaceous Page 69, line 6 for proterozoic read Proterozoic Page 137, line 2 for tomato read tomato Page 140, line 18 for emarginatus read marginatus Page 147, line 17 add L. after S. Mombin Page 153, line 16 for emarginatus read marginatus Page 153, line 36 for viridis read viride Page 156, line 2 for cherimolia read Cherimolia No. 8, Explanation of Plate XXX, line 4 for Tetrapteria read Tetrapleris No. 9, throughout for Belles Artes read Bellas Artes No. 9, Table I, last column, line 5 (data for Tripsacum latifolium) for 1:1.4 read 1:4.1 Beane BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Vor. 16, No. 1 CamsBripGk, MassacuuseTts, Fepruary 12, 1953 MISCELLANEOUS NOTES ON TROPICAL AMERICAN ORCHIDS BY CHARLES SCHWEINFURTH THE following miscellaneous notes regarding Tropical American Orchids include the description of new vari- eties, nomenclatural changes and amplifications of older species. The order of genera follows the system proposed by R. Schlechter in Notizblatt des Botanischen Gartens und Museums Berlin-Dahlem 9 (1926) 5638-591. Gomphichis Koehleri Sch/tr. var. minor C. Schweinfurth var. nov. Herba quam species typica multo humilior, cum foliis minoribus et vaginis caulinaribus approximatis et sepalis extus conspicue pilosis, non subglabris. Plant about 6.6 dm. high differing strikingly from the type in its much smaller vegetative size throughout. Roots (not described in the type) fascicled, tuberous, long, simple, lanuginose. Leaves five or six, basal or subbasal, oblong-ligulate, acute, slightly narrowed be- low, up to 13 em. long and 1.5 cm. wide. Stem stout, nearly concealed by about eight tubular acute sheaths. Raceme densely many-flowered, about 8 cm. long (with the apical flowers undeveloped in our specimen). Flow- ers yellowish cream-color, not whitish as in the type. he Sepals conspicuously pilose without. Parts of the peri- anth apparently closely similar to those of the type. Peru: Cuzco, Prov. of Paucartambo, Tres Cruces, at 3600 meters altitude, terrestrial in humus of forest glade, October 1941, C. Var- gas 2257 (Tyrer in Herb. Ames No. 66059). Altensteinia chiogena (Sch/tr.) C. Schweinfurth comb. nov. Aa chiogena Schltr. in Fedde Repert. Beih. 10 (1922) 35; Mansfeld in Fedde Repert. Beih. 57 (1929) t. 134, nr. 524. Altensteinia Weberbaueri (Sch/tr.) C. Schwein- Surth comb. nov. Aa Weberbaueri Schlechter in Fedde Repert. Beih. 9 (1921) 58; Mansfeld in Fedde Repert. Beih. 57 (1929) t. 105, nr. 418. Since the generic concept 4a appears to be inseparable from Altensteinia, the above nomenclatural changes are necessary. Buchtienia boliviensis Schlechter in Fedde Repert. 27 (1929) 34. This monotypic genus, originally limited to Bolivia, has lately been recognized from Peru. The collection cited is somewhat lower in stature than the type. It shows one leaf about 48.4 cm. long (in contrast to the cited maximum length of 40 cm.), while the lamina, which is oval, rather than elliptic or elliptic-oblanceolate, has a maximum length of 23.5 em. (as contrasted with 22 cm.), and a maximum width of 13 cm. (in comparison with 10 em.). Moreover, the base of the blade appears to be sometimes subcordate and not cuneate as cited. The flowers of the Peruvian collection are noted as rose- colored, whereas those of the Bolivian type are designated as greenish. The parts of the perianth are very similar in [2] both cases, but the Peruvian collection has the mid-lobe of the lip oblong-ovate and only 3 mm. long, while that of the Bolivian type is lanceolate and 4 mm. long. Prru: Cuzco, Prov. of Paucartambo, between Mistiana and Keros, at 700 meters altitude, terrestrial in dense shady forest, 1.20 meters tall, July 27, 1948, C. Vargas 7381. Stelis Endresii Reichenbach filius in Gard. Chron. (1870) 1373. Stes Huebneri Schlechter in Beih. Bot. Centralbl. 42, Abt. 2 (1925) 88. A study of the description of the Brazilian Stelis Hueb- neri and especially of several Peruvian collections shows that this concept is synonymous with the widely distrib- uted Middle American S. Hndresiu which extends from Mexico through Costa Rica to Panama and Venezuela. Peru: Hudnuco, six km. south of Tingo Maria, at 738 meters alti- tude, along road, on tree trunk in moist forested valley, flowers pale green with purplish base, December 2, 1945, R. J. Seibert 2257.— Loreto, Upper Marafion River, at mouth of the Santiago, at 160 meters altitude, epiphyte in rain-forest, September 19, 1924, G. Tess- mann 4087; Above Pongo de Manseriche, on right bank of Rio Santi- ago, at 200 meters altitude, on horizontal limb of tall tree, flowers pale green, December 6, 1931, Y. Mexia 6232; Vicinity of Iquitos, at 100 meters altitude, epiphyte in dense forest, flowers mignonette and brownish, January-February 1937, G. Klug 10069. Stelis eublepharis Reichb.f. var. glabriflora C. Schweinfurth var. nov. Planta herbae typicae similis sed floribus supra densis glabris differt. Plant small, with an ascending rhizome (not evident in the type). Secondary stems concealed by 3-4 tubular imbricating sheaths, about 5 cm. or less tall. Leaf soli- tary, long-petioled, up to 5.9 em. long; lamina oblong- elliptic or elliptic-oblanceolate, acute or subacute, grad- ually cuneate below, up to 4.5 cm. long and 1 cm. wide. [3 ] Inflorescences 1—2, shorter than to somewhat exceeding the leaves, up to 8 cm. long, commonly many-flowered, densely flowered above. Floral bracts small, but spread- ing and conspicuous. Flowers small, more or less secund, glabrous, yellow and purple or bright yellow. Sepals connate below, 8-nerved. Dorsal sepal ovate or broadly ovate, acute, about 3 mm. long and 2.4 mm. wide. Lat- eral sepals a little smaller, round-ovate, subacute. Petals much smaller than the sepals, transverse, suborbicular- oval, with a fleshy-thickened broadly rounded apex. Lip in natural position triangular-ovate, obtuse, fleshy, with a transverse, convex or subbilobed callus near the base, about 1 mm. long and wide. Peru: Cuzco, Prov. of Quispicanchis, Cachubamba, Marcapata, at 2800 meters altitude, epiphyte, perianth yellow and purple, Decem- ber 12, 1943, C. Vargas 3818 (Tyre in Herb. Ames No. 65225); Prov. of Urubamba, Machu Picchu, at 2050 meters altitude, on rocks, perianth bright yellow, January 8, 1946, C. Vargas 5544 (this col- lection consists of a single plant which is much smaller throughout than the type). Stelis Herzogii Schlechter in Fedde Repert. Beih. 12 (1918) 484; Mansfeld in Fedde Repert. Beih. 58 (1980) t. 20, nr. 80. This concept, which has been noted as occurring only in Bolivia, has now been extended to Peru by the collec- tion cited below. These specimens show some discrep- ancies from the description. The inflorescences vary from one to eight, rather than from three to five, and are either shorter or longer than the leaf, reaching a length of 9 em., rather than 5 cm. as cited. The dorsal sepal is com- monly 1.5 mm. long, instead of about 1 mm. long, and the lip is subquadrate-obovate, rather than rhombic- oval, as specified. Peru: Cuzco, Prov. of Quispicanchis, Hda. Itio, Marcapata, at 2000 meters altitude, in sunny rocky places, flowers yellow, January 27, 1943, C. Vargas 3114. [ 4] Stelis Koehleri Schlechter in Fedde Repert. 10 (1912) 386; Mansfeld in Fedde Repert. Beih. 58 (1930) ia enr.-8 2. On the basis of additional collections referable to this species, it appears advisable to point out several discrep- ancies from the original description. These specimens show stems as long as 8.5 cm., whereas the description cites amaximum length of 4 cm. The leaf, which is de- scribed as oblanceolate, appears to be rarely narrowly obovate, and it attains a length of 9.5 em., rather than the cited maximum of 7.5 cm., and a width of 2.5 em., as contrasted with 1.5 cm. The flowers are often mark- edly larger than specified, the dorsal sepal attaining a length of 6.7 mm., rather than the cited 8mm. The lip, when viewed from the front, is transversely ovate, broad- ly rounded or rounded-truncate in front commonly with a minute apicule in the middle and with the basal portion occupied by a transverse fleshy bilobed callus. As de- scribed and drawn, the lip is semiorbicular with a trun- cate apex and a transverse hollow in the middle. Peru: Hudnuco, Mito, at about 2760 meters altitude, in dense mats on dirt ledges of shrubby southwestern slopes, July 8-22, 1922, Macbride & Featherstone 1394; Same locality, altitude and habitat, April 8-18, 1923, flowers dark reddish yellow or deep yellow, scent- less, J. F. Macbride 3284. Stelis leucopogon Reichenbach filius Beitr. Orch. Centr.-Am. (1866) 95, t. 9, figs. T, 1-4. This concept, which is frequent in Central America from Guatemala to Panama, is now recorded from Peru. Peru: Huadnuco, Mufia, at about 2150 meters altitude, in dry woods, Macbride 4195.—San Martin, on ridge east of Tingo Maria, at 625— 1100 meters altitude, in jungle, Allard 22201, 22595. Stelis Lindenii Lindley Orch. Linden. (1846) 38, no. 17; Fol. Orch. Stelis (1858) 5, no. 29. This species, which was described from Venezuelan [5 ] material, has been collected in three Peruvian localities. While the type collection consists of an incomplete secondary stem without any indication of a rhizome, all three of the Peruvian specimens show a creeping rhizome with more or less remote stems. Peru: Cuzco, summit of Ceochayoc, at 1000 meters altitude, Feb- ruary 21, 1931, C. Bues s.n.; Prov. of Convencién, Hda. Potrero, Sapan Sachayoce, at 2200 meters altitude, epiphyte in forest, flowers white, March 5, 1942, C. Vargas 2555; Machu Piechu, at about 2100 meters altitude, May 22, 1915, O. F. Cook & G. B. Gilbert 866. Stelis phaeomelana Schlechter in Fedde Repert. 27 (1929) 45. Whereas the type of this species came from Bolivia, a Peruvian collection referable to this concept has re- cently appeared. The plants forming this collection show some discrep- ancies from the description and from other examples of the type. One stem is 22 cm. long, in comparison with the cited length as about 15 cm. The lamina of one leaf is 16 cm. long, as contrasted with the attributed maxi- mum length of 14 cm. The inflorescences appear to be always solitary, not only in this Peruvian material, but also in all of the plants comprising the isotype Bolivian collection in the Ames Herbarium, despite the statement that the species always has two inflorescences. The flow- ers appear to be slightly larger than those of the type. Peru: Cuzco, Prov. of Quispicanchis, region of Murayaco and Mar- capata, at 1960 meters altitude, on rocks in sunny places, flowers yellowish green, January 28, 1943, C. Vargas 3134. Stelis polycarpa Schlechter in Fedde Repert. 27 (1929) 46. Although this concept has previously been noted only as a native of Bolivia, several Peruvian collections are now available. One of these collections shows a dorsal [6 ] sepal conspicuously larger than that of the type, measur- ing 4.7 mm. long. Peru: Cuzco, Prov, of Paucartambo, S. Pedro to S. Isabel, at 1350 meters altitude, on trunks of old trees, December 6, 1947, C. Vargas 6785.—Huanuco, left bank of Rio Guallaza, above Cayumba, at 805 meters altitude, in small colonies on forest tree-trunks, October 19, 1936, Y. Mexia 8313a.—Loreto, Pumayacu, between Balsapuerto and Moyobamba, at 600-1200 meters altitude, epiphyte in forest, August— September 1933, G. Klug 0.18. Stelis santiagoensis Mansfeld in Notizbl. Bot. Gart. Berlin 10 (1928) 239. Several discrepancies from the description were noted from an examination of a photograph and a flower from the cotype in the Ames Herbarium. The leaf is some- what larger than specified—up to 18 cm. long and 1.9 cm. wide, as compared with 9-12 em. long and 1.2-1.8 cm. wide. The inflorescence appears to have a distinct and rather elongate peduncle below the raceme, instead of bearing flowers almost to the base. The dorsal sepal has a maximum length of 8 mm., rather than 2.1 mm. The lip is strictly simple, instead of shortly trilobulate, and is about 6 mm., instead of 8 mm., long. The num- ber of this cotype collection is 4086, rather than 4068. Stelis simacoensis Schlechter in Fedde Repert. 27 (1929) 47. This concept was described from Bolivian material, but a Peruvian collection is now available. In these specimens the stems are 2-8 cm. long, rather than 3.5-5 cm. long as cited. The lamina of the leaf, which is some- times oblong-elliptic rather than oblong-ligulate, has a maximum length of 8.4 em., instead of 6.5 cm., and the petiole reaches a length of 1.8 cm., rather than 1.3 ¢m. as stated. The callus of the lip appears to be near the apex, instead of nearly in the middle. ga Peru: San Martin, Zepelacio near Moyobamba, at 1200-1600 meters altitude, epiphyte in forest, flowers cream-color, January 19384, G. Klug 3520. Lepanthes tracheia Reichenbach filius in Flora 69 (1886) 557. A Peruvian collection, which has been referred to this Colombian species, is characterized as follows. Plants very small, 4 cm. or less tall, as compared with 2.6-6.2 em. high as shown by a drawing of L. tracheia from the Reichenbach Herbarium. Leaf suborbicular- ovate (rarely triangular-ovate), with a subcordate base, rather than elliptic-ovate as shown in the drawing, the size being about 1.2 cm. long and 9-10 mm. wide, as compared with 1.5—2.5 em. long and 9-17 mm. wide in the figure. The sepals appear to be very similar to those depicted in an analysis of L. tracheia from the Reichen- bach Herbarium. The petals are ciliate and pubescent (not depicted nor described) and the posterior lobe is oblong-lanceolate or linear-lanceolate and is narrower than shown in the analysis. Peru: Cuzco, Prov. of Paucartambo, San Pedro to Sta. Isabel, at 1350 meters altitude, on old tree trunks, flowers reddish wine-color, C. Vargas 006783. Lepanthes trachysepala Schlechter. in Fedde Re- pert. Beih. 7 (1920) 100; Mansfeld in Fedde Repert. Beih. 57 (1929) t. 84, nr. 1380. A Peruvian collection, consisting of a single specimen, has been referred to this Colombian species. Plant about 25 em. high, as in the type. Stem about 14 cm. long, contrasted with 5-7 cm. long as described. Leaf oblong-elliptic, acute and 4.9 cm. long, as com- pared with oblong, obtuse and 2.5-8.5 cm. long in the type. Raceme about 8-flowered, rather than 8- to 12- flowered. Flowers red with yellow spots, as contrasted with pale brown. Sepals about 9 mm. long, only 7 mm. [ 8 ] long in the description. Lateral sepals long-acuminate, rather than acute or subacute. Petals 6.5, not 5.5, mm. long. Lamina of lip not embracing the column at the base, when expanded, triangular-ovate, not lanceolate- triangular as described. Peru: Cuzco, Prov. of Urubamba, near Wenner Gren ruins, at 3400-3600 meters altitude, in dense wet dark forest with much fog and rain, Rk. D. Metcalf 30769. Pleurothallis citrina Sch/tr. var. elliptica C. Schweinfurth var. nov. Planta major, caulibus altioribus, foliorum lamina el- liptica vel oblongo-elliptica, inflorescentiis duplo longi- oribus et labello majore a specie differt. Stems more or less arcuate, up to 15 em. long, with several (up to 7), close, tubular sheaths. Leaf short- petioled, 4.1-11.5 cm. long; lamina elliptic to oblong- elliptic, 3.5-10 cm. long, up to 2.8 em. wide. Inflores- cences about 4-6 cm. long. Flowers yellow, very similar to those of the type. Lip somewhat larger than that of the species, oblong, about 3 mm. long. Peru: Puno, Prov. of Carabaya, Ollachea (abajo), at 2500 meters altitude, on rocks, December 30, 1947, C. Vargas 6936 (Tyre in Herb. Ames No. 65081). Lepanthopsis microlepanthes (Griseb.) Ames in Bot. Mus. Leafl. Harvard Univ. 1, No. 9 (1988) 24, fig. Pleurothallis microlepanthes Grisebach FI. Brit. W. Ind. Isl. (1864) 610; Cogniaux, in Urban Symb. An- till. 6 (1909) 480; Fawcett & Rendle F]. Jam. 1 (1910) 65. Lepanthes Leonn C. Schweinfurth in H. Léon. FI. de Cuba 1 (1946) 862. A recent review of the concept, Lepanthes Leoni, which was described in Dr. Hermano Leon’s Flora de Cuba, has indicated that this little orchid is referable to [9 ] the older Lepanthopsis microlepanthes (Griseb.) Ames. The complete synonymy appears above. This species occurs in Cuba and Jamaica. Malaxis termensis (A7rédnzl.) C. Schweinf. var. elata C. Schweinfurth var. nov. Herba altitudine majore, foliis late ovatis conspicue acutis, floribus minoribus a specie differt. Plant apparently witha creeping rhizome. Stems about 43 em. high including the rachis of the inflorescence, only slightly thickened below. Lower part of stem concealed by two or more tubular sheaths of which the upper are elongate and leaf-bearing. Leaves 2, subopposite, broad- ly ovate, sharply acute, appearing sessile, but abruptly contracted into sheaths, about 9 cm. long, up to 6.1 em. wide, about 14-16.5 cm. above the base. Inflorescence abbreviated, subumbellate, many-flowered. Flowers very small, membranaceous, greenish. Sepals with strongly revolute sides. Dorsal sepal oblong-lanceolate, obtuse, about 2.8 mm. long and 1 mm. wide. Lateral sepals obliquely oblong-ovate, obtuse, subequally long with the dorsal sepal, but wider (about 1.8 mm.). Petals linear, obtuse, shorter than the sepals. Lip suborbicular-ovate, abruptly narrowed to a blunt apicule, with a large 2- celled concavity below, about 1.7 mm. long and slightly broader. Column minute. Peru: Amazonas, Prov. of Bongara, Pomocochas, at about 2400 meters altitude, terrestrial in ridge forests above Quebrada Santa Rosa, northeast across Lake Pomocochas, April 4, 1944, W. H. Hodge 6152 (Tyre in U.S. Nat. Arb. No. 183663). Epidendrum cyperifolium C. Schweinfurth nom. nov. Epidendrum microtos Reichenbach filius var. grandi- florum C. Schweinfurth in Bot. Mus. Leafl. Harvard Univ. 11 (1943) 98. [10] ? Bletia ensiformis Ruiz & Pavon Syst. Veg. (1798) 230. Renewed study of the specimen previously considered to be a variant of the inadequately described piden- drum microtos shows that it should be elevated to speci- fic rank. The reasons for this course of action are: (1) the plant under consideration is represented by complete specimens (whereas J”. microtos was described only from a flowering panicle); (2) the flowers of this plant are larger with oblanceolate and acute (not spatulate and ob- tuse) petals; and (8) the lip is truly emarginate (not rounded in front). Judging from a photograph of Bletia ensiformis from the Pavon Herbarium, it seems highly probable that the latter concept is referable to this species. However, an examination of the flower is impossible, so the definite reduction is open to question. Epidendrum inamoenum A7riinz/. var. robus- tum C. Schweinfurth var. nov. Herba caulibus et foliis multo robustioribus et floribus majoribus et petalis latioribus sepalis subaequantibus a specie differt. Plant up to 2 m. high (according to the collector), simple or branched. Sheathed stems 1—1.8 cm. in diame- ter, leafy. Leaves oblong to oblong-elliptic, rounded and minutely bilobed at the apex, slightly narrowed at the semiamplexicaul base, up to 15 em. long and 8 cm. wide. Raceme recurved or nodding, densely several- to many- flowered, 3-6 cm. long, without any subtending spathe. Flowers larger than those of the type, pale yellow or pale greenish yellow. Dorsal sepal elliptic-lanceolate, acute, about 17 mm. long and 7—-8.5 mm. wide. Lateral sepals obliquely elliptic-ovate, complicate-mucronate, about 18-20 mm. long and 8-9 mm. wide. Petals ellip- E ane] tic to oval, subacute to rounded at the apex, about 14— 16 mm. long and 7-10 mm. wide. Lip adnate to the column up to its apex, suborbicular-ovate in outline, simple to lightly but distinctly trilobed at the apex, cor- date at base, about 12-14 mm. long in the middle (from the cordate base to the rounded or triangular-acute apex) and 15-17 mm. wide; disc bicallose at base, with the central longitudinal portion thickened. Column very short and stout. Peru: Cuzco, Prov. of Urubamba, “‘Kil. 97-108, F.C. C. S. A.’’ at 2200 meters altitude, on rocky slope, May 16, 1943, C. Vargas 3408 (Tyrer in Herb. Ames No. 65062); Machu Picchu, at 2040 meters altitude, rocky slope, January 9, 1946, C. Vargas 5541. Epidendrum longirepens (C. Schweinf.) C. Schweinfurth comb. nov. Epidendrum Schlechterianum Ames var. longirepens C. Schweinfurth in Bot. Mus. Leafl. Harvard Univ. 11 (1943) 115. Renewed study of the plant which I described as H/pi- dendrum Schlechterianum var. longirepens has convinced me that this concept is worthy of specific rank. While it certainly is allied to #. Schlechterianum, this entity is widely divergent by reason of its long creeping rhizome, its very small leaves, its petals which are nearly or quite as broad as the sepals (not markedly narrower) and its lip which is rounded or only slightly (not deeply) cordate at the base. Epidendrum pachychilum A7iénzlin in Fedde Repert. 1 (1905) 184. Epidendrum aquaticoides C. Schweinf. var. pusillum C. Schweinfurth in Bot. Mus. Leafl. Harvard Univ. 11 (1943) 82. Recent study of material referable to Hpidendrum pachychilum shows that this concept differs in certain re- spects from the original description. [12] Examination of a photograph of the type of . pachy- chilum shows that the floral bracts are rather conspicuous and triangular-ovate, not ‘‘subnullae’’ as described. Moreover, the sepals appear to be ovate-elliptic, rather than ‘‘oblonga’’ as noted. In the specimens examined, the somewhat immature flowers have sepals only about 8-9 mm. long, rather than 10 mm. long as described. It now appears that the concept previously described as Iipidendrum aquaticoides C. Schweinf. var. pusillum is referable to #. pachychilum. Peru: Cuzco, Prov. of Urubamba, near Wenner Gren ruins, at 3400- 3600 meters altitude, in open soaking sphagnum bog with much fog and rain, perianth fleshy, orange with faint red on outside of sepals, August 5-6, 1942, R. D. Metcalf 30575; On trail from Puyupata to Y Sayacmarca, epiphytic, at 3600 meters altitude, August 5, 1942, C. Vargas 2894 (Tyrer of E. aquaticoides var. pusillum); Same data, at 3650 meters altitude, June 23, 1948, C. Vargas 7257. Epidendrum refractoides C. Schweinf. var. hu- mile C. Schweinfurth var. nov. Herba statura minore et pedunculi spatha unica vel nulla, et inflorescentia racemosa (non paniculata) et la- belli lobo medio magis separato a specie differt. Plant small, stout, epiphytic, up to about 19 em. high. Stems (sometimes decumbent and rooting at base), stout, entirely concealed by imbricating leat-sheaths or the fi- bers of sheaths, about 10 em. long. Leaves apparently eleven or less, approximate, elliptic-ovate, obtuse to rounded at the apex, amplexicaul at the sessile base, up to 5.1 cm. long and 2.5 cm. wide (next to the upper- most blade largest), ascending-spreading. Inflorescence solitary, terminal, erect, about 9.5 cm. or less tall; pe- duncle short, about 4 cm. or less long, usually bearing a conspicuous, conduplicate, erect spathe about 2.5 cm. long; raceme rather loosely several- (7- or more) flow- ered, with the rachis about 6 cm. or less long. Floral 13] bracts lanceolate, acuminate, somewhat exceeding half of the glabrous pedicellate ovary. Flower rather small, subfleshy, greenish yellow. Dorsal sepal lanceolate- elliptic or elliptic-oblanceolate, complicate and apiculate at the apex, about 11—12.2 mm. long and 4.9-5 mm. wide, 3-nerved. Lateral sepals obliquely and broadly lanceolate-elliptic, acute, dorsally carinate, obliquely in- serted at the base of the column, about 11—-12.5 mm. long on the posterior margin and 5.1 mm. wide. Petals linear-oblanceolate or oblanceolate, obtuse or subacute, slightly oblique, 1- or obscurely 3-nerved, about 10.1-11 mm. long and 8 mm. wide. Lip adnate to the column up to its apex; lamina 3-lobed with the lateral lobes up- curved, cordate at base, about 6—6.5 mm. long in the center and 8-10.2 mm. wide across the lateral lobes when expanded; lateral lobes suborbicular-oval, with irregular or lobulate margins; mid-lobe distinct, elliptic-subquad- rate, truncate and lightly retuse in front, with the center longitudinally thickened; disc at base with a pair of prominent complanate obliquely semiobovate calli. Col- umn short and stout, dilated above in front, about 6.5— 7 mm. high at the back. Peru: Apurimac, Prov. of Abancay, ‘“bosques de Ampay,’’ epi- phyte, July 1988, C. Vargas 2050 (Tyrr in Herb. Vargas).—Cuzco, Prov. of Calea, in the vicinity of the town of Lares, on rocks at 3200 meters altitude, ‘‘leaves purple,’’ August 30, 1934, C. Vargas 3618. Epidendrum Schlimii Reichenbach filius in Lin- naea 22 (1849) 888; Lindley Fol. Orch. Epidendrum (1853) 64, no. 201, non #. Schima Lindley Fol. Orch. Epidendrum (1858) 74, no. 234. Epidendrum tenellum Lindley Fol. Orch. Epidendrum (1853) 44, no. 188. The identity of the concept Hpidendrum tenellum with the earlier 7. Schlimi has already been pointed out by Reichenbach f. in Walp. Ann. 6 (1862) 361. [ 14 ] The Peruvian collection cited below agrees well with Ei. Schlimii and thus extends the known range of this species from Venezuela. In this collection the flexuous inflorescence consists of remote abbreviated racemes and is thus to be described as a panicle. Peru: Hudnuco, Mufia, at about 2150 meters altitude, epiphyte, May 23-June 4, 1923, Macbride 3966. Epidendrum Schlimii Reichb.f. var. gracillimum (Reichb.f. & Warse.) C. Schweinfurth comb. nov. ipidendrum gracillimum Reichenbach filius & Wars- cewicz in All. Gartenz. 22 (1854) 314. Judging by arecord of Mpidendrum gracillimum from the Reichenbach Herbarium in Vienna and by a Peru- vian collection (Vargas 3268) which is referable to this concept, this entity appears to be merely a variety of Ei. Schlimu. It has stems (including the rachis of the inflorescence) about 10-19 cm. tall, while those of EF. Schlimi are about 10 cm. or less tall. The leaves are commonly about 4.5 cm. or more long, whereas those of EE. Schlimu appear to be about 8.4 em. or less in length. The inflorescence of J. gracillimum has distinct and rela- tively elongate branches, contrasted with the abbreviated and congested branches of 7. Sehlimii. A more signifi- cant discrepancy resides in the lip which is strictly simple in £, Schlimi and is distinctly 8-lobed with a short trans- verse bilobed mid-lobe in typical 2. gracillimum. Var- gas 3268, however, while otherwise coinciding with the type of the latter concept, has a lip which is simple with irregular margins or is very obscurely 8-lobed. Peru: Cuzco, Prov. of Convencion, ‘‘alturas de Pintobamba,”’ at 2300 meters altitude, epiphyte, in small damp forest, perianth pink, March 3-4, 1943, C. Vargas 3268. Epidendrum subreniforme C. Schweinf. var. ra- mosum C. Schweinfurth var. nov. Herba caule pluribus cum ramis strictis, foliis longiori- [ 15 ] bus, floribus minoribus roseo-purpureis cum labello ex- apiculato a specie differt. Plant with stem consisting of several short strict branches, entirely concealed by tubular sheaths which bear leaves except at the base of the branches. Leaves elliptic or oblanceolate-oblong, acute, sessile at the cune- ate base, up to 6.5 cm. long and 1.9 cm. wide. Inflores- cence about 8.5 cm. long, erect and shortly recurved at the apex; peduncle about 2 cm. long, bialate, with one small infundibuliform bract; raceme loose, 5-flowered, with a fractiflex rachis. Flowers rather large, rose-purple, with spreading segments. Dorsal sepal elliptic, acute, minutely apiculate, 5-nerved, 1.5 cm. long, about 5.7 mm. wide. Lateral sepals semielliptic or obliquely ellip- tic-lanceolate, acute, 5-nerved, with a deep keel near the apex, about 1.53 cm. long and 6.1 mm. wide near the middle. Petals linear-oblanceolate, oblique, subacute to obtuse, about 1.43 cm. long and 8.2 mm. wide above. Lip much larger than the other segments, adnate to the column up to the apex of the latter; lamina simple or nearly so, semiorbicular-reniform, very slightly indented or subretuse in the middle of each side, narrowed above to a broad subtruncate and lightly retuse apex, without any apicule, ecallose, deeply cordate at the base, about 1 cm. long in the middle and 2.3 cm. wide. Column short, stout, much dilated above in front, about 8 mm. long at the back, extended on each side into an obliquely semiorbicular auricle. Peru: Cuzco, Prov. of Convencién, Sahuayaco, epiphyte at 1600 meters altitude, January 17, 1947, C. Vargas 6295 (Type in Herb. Ames No. 63449). Epidendrum vincentinum Lindley in Hook. Journ. Bot. 8 (1841) 88, 809, t. 11; Cogniaux in Urban Symb. Antill. 6 (1910) 526. Epidendrum serricardium Schlechter in Fedde Re- pert. Beih. 19 (1928) 218. [ 16 ] This concept, which was originally described from St. Vincent, was reported by Cogniaux (l.c.) from Guade- loupe, Dominica and Martinique. It was subsequently found on Trinidad by W. E. Broadway and very recently was collected on the lower slopes of Mt. Britton, To- quillo, Puerto Rico by D.S. Correll and H. F. Winters. A surprising extension of range is shown by a Peru- vian collection of #. vincentinum from Cuzco, Province of Paucartambo (C. Vargas 006793). The Costa Rican Hpidendrum serricardium is obvious- ly reducible to this concept, the description being almost an exact counterpart of that of 7. vincentinum. The flowers of all the material examined, whether from the West Indies, Costa Rica or Peru, appear to be some- what larger than those described for E. vincentinum and the disc of the lip appears to have commonly one to three more or less distinct keels. The color of the flowers varies from pale green or greenish yellow, often tinged with bronze, to russet or mahogany-red and pale yellow, and the leaves are sometimes noted as violet or purple. Dichaea muricata (Sw.) Lind/. var. maculata (Poepp. & Endl.) C. Schweinfurth comb. nov. Dichaea maculata Poeppig & Endlicher Nov. Gen. ac Sp. 2 (1838) 3, t. 105, figs. 1-6; Cogniaux in Martius Fl. Bras. 8, pt. 6 (1906) 489. The concept Dichaea maculata, which is limited to Peru, has been considered to be amply distinct from D. muricata, Which was originally described from the West Indies, for two reasons. First the leaves of D. maculata are seen to be distinctly ciliolate near the apex, whereas those of D. muricata appear to be entire or nearly so at the apex. Secondly, the lip of D. maculata was described and figured as provided with a relatively long linear claw with the lamina rounded in front, while the lip of D. Lae | muricata is only shortly clawed or has a cuneate base with the lamina distinctly acute or apiculate in front. The first character (that of the ciliation of the leaves) is apparently a stable feature, the leaves of the Peruvian material referred to D. maculata being always more or less ciliolate on close examination, whereas those of the specimens referred to D. muricata appear to be almost invariably entire. On the other hand, the Peruvian specimens which have these ciliolate leaves have a lip which varies from short- clawed to elongate-cuneate at the base and the apex of the lamina is invariably apiculate (sometimes with a rounded forward margin) or even sharply acute. It seems, therefore, that the lip of D. maculata is rather polymorphic (as it is in typical D. muricata), but it appears probable that it is always more or less strongly apiculate—a character that may readily have been over- looked in the typical specimen. Since, however, this concept was described and figured with a linear-clawed lip, it may be well to recognize this tendency. For purposes of simplification in these closely allied concepts, therefore, it seems reasonable to consider D. maculata as a variety of the older concept D. muricata (based on Cymbidium muricatum Sw.) of which an excel- lent description and floral analysis is given in Fawe. & Rendle Fl. Jam. 1 (1910) 137, t. 80, figs. 22-25. Thus we segregate var. maculata as having leaves with a cilio- late apical portion and a lip which is either long- and nar- rowly clawed or with a narrowly cuneate base. Dichaea panamensis Lindley Gen. & Sp. Orch. (1833) 209; Kriinzlin in Engler Pflanzenr. LV, 50 (Heft 83)(1928) 51. Epithecia panamensis Schlechter in Orchis 9 (1915) 25. Dichaeopsis panamensis Schlechter in Beih. Bot. Cen- tralbl. 86, Abt. 2 (1918) 519. [ 18 ] This variable species, heretofore recorded only from Mexico and Central America to Panama, has recently been identified from Colombia and Peru. The latter specimen consists of a caespitose plant of eight highly variable stems, the longest of which is about S > Py 5 24 cm. in length—longer than any before noted by me. The leaves, also, are longer than usual, being up to 4 em. long. The blades seem to be indistinctly, if at all, glau- cous (as in the usual form). The flower, however, is much smaller than usual, but has segments closely sim- g Ma ilar to those of the typical plant. Peru: Junin, east of Quimiri Bridge, near La Merced, at 800-1300 meters altitude, epiphyte in dense forest, sepals green, June 1-3, 1929, E.P. Killip & A.C.Smith 25412 (U.S. Nat. Herb. No. 1359631). Campylocentrum Ulaei Cogn. var. peruvianum C. Schweinfurth var. nov. Herba foliis minoribus, labello paullo diverso atque calcari majore U-formi a specie differt. Plant small, slender, epiphytic. Stem slender, about 4.5 cm. long in the specimen, entirely concealed by loose conduplicate leaf-sheaths, lightly arcuate, producing be- low elongate fibrous roots. Leaves several, distichous, widespreading, small, narrowly lanceolate to narrowly elliptic-oblong, narrowed to a minutely and commonly obliquely bilobed apex, slightly contracted to a sessile base, up to 2.1 em. long and 5 mm. wide. Inflorescences lateral, racemose, much exceeding the leaves, puncturing the middle of the leaf-sheaths, densely many-flowered, up to about 6.5 cm. long, glabrous; peduncle filiform, up to 1.6 cm. long. Flowers very small, with membranace- ous subparallel segments. Dorsal sepal linear-lanceolate, subacute, l-nerved, about 2.8 mm. long and 0.8 mm. wide. Lateral sepals subequaling the dorsal sepal, narrow- ly and obliquely triangular-lanceolate, acute or acuminate [ 19 ] with a complicate apex. Petals triangular-lanceolate, acute or acuminate, slightly shorter than the sepals. Lip conspicuously spurred ; lamina tubular-involute in natu- ral position, ovate-lanceolate in outline when expanded, about 2 mm. long, deeply 3-lobed below the middle; lateral lobes relatively short, obliquely subquadrate and slightly dilated above with an obtuse outer tip; mid-lobe much protuberant, narrowly lanceolate-triangular, long- acuminate; spur relatively large, U-shaped when viewed from the side, slender and cylindric below, laterally flat- tened and dilated above, about 1.8 mm. long from the rounded apex to the base of the curve. Column minute. Ovary glabrous. This plant is very similar to typical Campylocentrum Ulaet Cogn., but differs in having much smaller leaves, a differently proportioned lip and a larger U-shaped spur. Peru: Junin, Chanchamayo Valley, at 1800 meters altitude, Octo- ber ‘£1924-1927,’’ Carlos Schunke 520 (Tyrer in Herb. Field Mus. 571579). [ 20 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CampripGr, Massacnuusetrts, Marcu 16, 1953 Vou, 16, No, 2 STUDIES IN THE GENUS HEVEA VII BY RicHarD Evans Scuuures' In the course of my studies towards a monograph of the genus Hevea, isolated but significant observations fre- quently accumulate. In order that these data may be available before the completion of a final monograph, I have initiated a series of articles in which the results of field and herbarium investigations may be published. This paper continues the series and consists of miscel- laneous taxonomic, nomenclatorial, phytogeographical, historical and chemical notes. The herbarium studies herein reported were carried out in 1950, during my visits to important European bo- tanical centers. The chemical examinations were made by chemists in the United States Department of Agriculture and at the National Bureau of Standards on rubber samples which I secured in the Amazon from trees the identity of which was established and has been authenticated through her- barium specimens. ' Botanist, Division of Rubber Plant Investigations, Bureau of Plant Industry, Soils, and Agricultural Engineering, Agricultural Research Administration, U. S. Department of Agriculture; Research Fellow, Botanical Museum of Harvard University. [ 21 ] 1. Notes on the specimens of Hevea in the De Candolle Herbarium Although there are larger and more comprehensive collections of Hevea in several European and American herbaria, it is true I think, that one of the most uniquely significant is contained in the De Candolle Herbarium in the Conservatoire Botanique in Geneva. The intensive and extensive field studies and collec- tions which Richard Spruce carried out a century ago in the Amazon Valley laid the first solid foundation for our understanding of the genus of the commercial rubber tree. Bentham’s critical treatment of Spruce’s material set the pace for later taxonomic work in the group. But the first attempt at a monographic synopsis of Hevea was that of Mueller of Aargau, working in Geneva. Notwithstanding the fact that Mueller had access to a number of collections of HZevea in the Delessert Her- barium in Geneva and in other European institutions, we may regard the specimens preserved in the De Can- dolle Herbarium as representing the core of his study material. ‘These were, in large part, the basis of his treat- ment of Hevea in De Candolle’s Prodromus 15, pt. 2 (1866) 717-719. Partly because of this association, the specimens and Mueller’s handwritten annotations which some of them bear are worthy of special attention. Few of the specimens are rare; on the contrary, most of them are Spruce collections and are rather well distributed amongst the major herbaria of the world. This in itself is an additional reason for a close examination of those sets which have been, in a way, authenticated by the work of that great master of the 7uphorbiaceae. The arrangement of the species in the De Candolle Herbarium follows the order of their publication in the Prodromus. Thus, the material of Hevea can be found [ 22 in the order in which the species are enumerated in the Prodromus 15, pt. 2 (1866) 717-719. In this article, I have followed Mueller’s subgeneric division of FZevea into Bisiphonia and Muhevea, now no longer accepted, and T have emploved the binomials used by Mueller re- gardless of the modern status of these names. This | have done so that the following notes will correspond with the arrangement of the material in the De Candolle Herbarium. In each case, however, I have indicated the annotation which [made in June 1950, so that there should be no difficulty in finding the present-day equiva- lent of the older binomials in those few cases where there has been some change. L wish to thank Dr. Charles Baechni, Director of the Conservatoire Botanique and other members of this institution for their kind help dur- ing my visit in June 1950. BIsSIPHONTA Hevea Spruceana ( Benth.) Mucller-Argoviensis in Linnie 34 (1865) 2ot. Siphonia Spruccana Bentham in Hooker's Journ. Bot. G (1854) 370. There is one specimen under Hevea Spruceana, a du- plicate type. Brazit: Estado do Amazonas, Rio Amazonas, ““In viecinibus Santa- rem, Prov. Para. Coll. R. Spruce, Jul. 1850.”’ [This date, printed, has been altered to read “*1851.7°] Hevea discolor ( Benth.) Mucller-Argoviensis in De Candolle Prodr. 15, pt. 2 (1866) 717. Siphonta discolor Bentham in Hooker's Journ. Bot. 6 (1854) 869. There are three specimens under this name, including a duplicate type of the species. [have annotated all three us Hevea Spruceana, Brazit: Estado do Amazonas, Rio Solimoes, near Kea [now called Teffé| [fide Mueller in Prodr. 717, no. 2] 1834, Poeppig 2595. Consisting of several leaves and flowering inflorescen- ces, this specimen ts labelled **Perou/ MW. Poeppig 1884." In a small envelope containing flowers, there is a label "2595.° For our information that the specimen was col- lected “*prope Ega,” we are indebted to Mueller, for there is no indication on the sheet that this was its local- itv. In fact, it is very probable that the concept repre- sented by Poeppig 2595 does not occur in Peru, tor it has apparently never been found in that country (ef. Seibert in Ann. Mo. Bot. Gard. 84 (1947) 261). Teffé (or Nga) represents almost the westernmost extent of Hevea Spruceana. Brazi: Estado do Amazonas, Rio Negro, ““de vicinibus Barra | now called Mansos|, Prov. Rio Negro. Coll. R. Spruce. Dec.-Mart. 1850— 51.” Originally determined as “*Siphonia elastica Pers. (°° this collection represents that expression of HZevea Spruceana which is most abundant near the mouth of the Rio Negro. It is in fruit and has several beautiful seeds very typical of the ALevea Sprucecana of the Mandos area: long, considerably flattened, with two very con- spicuous flat surfaces ventrally, almost diamond-shaped in cross section, measuring 35 mm. long, 12 mm. thick, 17 18 mm. wide. There are also a number of valves of the capsule. Brazi: Estado do Amazonas, Rio Amazonas, *‘ad oram septentrio- nalem flum. Amazonum, ad ostium Rio Negro. Coll. R. Spruce 1/77. Aug. 1851.7" Spruce 117 11s the type collection of Stphonia discolor (cf. Schultes in Bot. Mus. Leafl). Harvard Univ. 15 (1952) 253. It represents the same expression of Hevea Spruce- ana aus the collection previously discussed. [ 24 ] Hevea pauciflora (Spruce ex Benth.) Mueller- Argoviensis in Linnaea 84 (1865) 208. Siphonia paueifiora Spruce ex Bentham. in Hooker's Journ. Bot. 6 (1854) 870. There is apparently no material of this concept in the De Candolle Herbarium. Hevea rigidifolia (Spruec ev Benth.) Mueller- Argoviensis in Linnaea B+ (1865) 2038. Siphonia rigidifolia Spruce ex Bentham in Hooker's Journ. Bot. 6 (1854) 871. ‘There is one specimen of Hevea rigidifolia, a duplicate type. Brazit: Estado do Amazonas, Rio Uaupés, ‘‘prope Panuré [Ipan- oré] ad Rio Uaupés. Coll. R. Spruce 2527. Oct. 1852-Jan. 1853.77 Spruce 2527 inthe De Candolle Herbarium comprises uw branch with several adult and young leaves and abun- dant flowering material. An examination of one stami- nate and one pistillate flower trom the collection indicates agreement with the descriptions of this concept prepared on the basis of a recent study of the type and new ma- terial (Schultes in Bot. Mus. Leaf. Harvard Univ. 13 (1948) 101, t. vill). Hevea Benthamiana Mueller-Argoviensis in Lin- naea B+ (1865) 204. The specimen of Hevea Benthamiana in the De Can- dolle Herbarium is apparently the type of the concept. Brazit: Estado do Amazonas, Rio Uaupés, “prope Panuré [ Ipan- oré] ad Rio Uaupés. Coll. R. Spruce 2560, This material comprises four or five leaves and two uxes of the inflorescence. It was formerly confused with Hevea Spruceana and was distributed as Siphonia dis- color, but Mueller, recognizing it as a distinct concept, described it on the basis of this specimen. [25 | Hevea brasiliensis (W7//d. ev A. Juss.) Mueller- Argoviensis in Linnaea 84 (1865) 204. Siphonia brasiliensis Willdenow ex Adr. de Jussieu Kuphorb. Gen. (1824) t. 12, pl. 88b, fig. 1-6. The De Candolle Herbarium has two collections of this species, one of which is a fragment of the type. = 7 7 6é a . oe Brazit: Estado do Para, Rio Amazonas, © Para Hoffmannsegg. The Hoffmannsegg material of this concept, collected probably at the mouth of the Amazon by Sieber, is that on which the earliest publication of the binomial S7phonia brasiliensis and Willdenow’s accompanying diagnostic plate were based (cf. Schultes in Bot. Mus. Leafl. Har- vard Univ. 14 (1950) 79). The type is in the Willdenow Herbarium in Berlin; there is a duplicate type in Paris (Schultes le. Pl. xix). In an envelope on the sheet la- belled Hevea brasiliensis in the De Candolle Herbarium, there are two leaflets of this Sieber collection; the en- velope is marked, in) Mueller’s hand: ‘‘Folia: Para: Hoffmannsegg. ’ , . . . ee Braziz: Estado do Paré, Rio Amazonas, “Para, Spruce, 1849. This second collection of Hevea brasiliensis consists of two complete leaves and several inflorescence axes in good flower. It is the widely distributed collection which Spruce made in the region of Belém do Para shortly after his arrival in South America in 1849; since, in Para, Hevea brasiliensis flowers in August and early Septem- ber, we may assume that this collection was one of the first which Spruce, who arrived in mid-July, 1849, made in South America. It can be considered topotypical: and, indeed, it matches the type extremely well. Mueller has written an annotation to the effect that this Spruce specimen was acquired ‘‘ex hb. Van Huerck.” The Van Huerck Herbarium is incorporated in the col- [ 26 ] lection at the Natuurwetenschappelijk Museum in Ant- werp, Belgium, where there is an excellent specimen of this Spruce collection of Hevea brasiliensis and where, ona Sagot collection of H. guianensis Aubl. from French Guiana, [ found the following interesting annotation: [ex herb. DC contre un fragment de H. brasiliensis Muell.-Arg. ]. Hevea lutea (Spruce ex Benth.) Mueller-Argoviensis in Linnaea 84 (1865) 204. Stphonia lutea Spruce ex Bentham in Hooker’s Journ. Bot. 6 (1854) 370. The De Candolle Herbarium possesses two specimens which Mueller referred to Hevea lutea. I have annotated both as Hevea guianensis Aublet var. lutea (Spruce ex Benth.) Ducke & Schultes. Venezur.a: Territorio del Amazonas, Rio Negro, ‘‘prope San Car- los, ad Rio Negro, Brasiliae borealis. Coll. R. Spruce 3139, 1853-54.’’ Spruce 3139 is widely distributed in the principal her- baria. The De Candolle specimen, a duplicate type of Hevea apiculata Baillon, is especially complete, compris- ing several leaves, a few loose leaflets and abundant flow- ering material. Mueller, who, in the Prodromus (l.c. 719), reduced Hevea apiculata to synonymy under H. lutea and who later (in Martius FI. Bras. 11, pt. 2 (1874) 302) made it a variety of Hi. lutea, wrote on the speci- men: ‘‘Non differt a Hevea lutea Muell.-Arg. 18638. B apiculata Muell.-Arg. in Flor. bras.” Brazit: Estado do Amazonas, Rio Uaupés, “prope Panuré [Ipan- uré| ad Rio Uaupés. Coll. R. Spruce 2088. Oct. 1852-Jan. 1853."° The De Candolle Herbarium material of Spruce 2088, a duplicate type of Siphonia lutea, is an especially com- plete flowering specimen of a widely distributed number. [27] EUHEVEA Hevea guianensis 4 ub/et Hist. Pl. Guyan. 2 (1775) 871. There are two collections in the De Candolle Her- barium which Mueller refers to this concept. I have an- notated them both as Hevea guianensis. Frencw Guyana: 1840, Leprieur. The Leprieur collection, represented also at Paris, seems to be the earliest flowering material of Hevea guianensis. Mueller has left a label in his handwriting, which reads: ‘‘Euphorbiae. Calyx ad mediam usque 5- partitus, petala nulla, stam 5! circa rudimentum ovarii in columnam coalita, filamenta subnulla, fol. stipulata. ”’ Frencu Guyana: Maroni, 1857, P. Sagot 510. The Sagot collection is represented in several herbaria. The specimen in the De Candolle Herbarium is in abun- dant flower. 2. Miscellaneous notes, chiefly on specimens of Hevea in various European herbaria Hevea brasiliensis (Willd. ex A. Juss.) Mueller- Argoviensis in Linnaea 84 (1865) 204. Brazit: [Near mouth of Rio Amazonas] Sieber s.n. [?] The Humboldt Herbarium in Paris has a collection referable to Hevea brasiliensis and consisting of one leaf- let and several flowers in a little packet. The packet is labelled ‘“Stphonia brasiliensis W. (e specim authent. ab ipso Willdenow misso),’’ and is evidently a fragment from the type specimen in the Willdenow Herbarium collected by Sieber, which it matches perfectly (ef. Schultes in Bot. Mus. Leafl. Harvard Univ. 14 (1950) 79). In this same herbarium, there is a full specimen which [ 28 ] likewise matches the type. Unfortunately, it bears no data concerning the locality or date of collection nor a collector’s name, but I believe it to be a duplicate type. The only annotation it bears is the following: ‘‘dedit Willdenowius, 1811.”’ Hevea guianensis 4ub/et Hist. Pl. Guian. 2 (1775) 871. In the Paris Herbarium, there are two sheets of Hevea guianensis upon which is written: ‘‘Leg. A. Richard. Sta. Martha Antilles. Siphonia elastica ex hb. de Fran- queville. Herb. E. Cosson 18.°’ The special interest at- tending these particular specimens centers on the locality data. The only ‘‘Sta. Martha’’ which I have been able to find registered for the entire Antillean area is the very old city of that name on the Caribbean coast of Colom- bia. The genus Hevea, of course, is unknown from that region, and we may very safely assume that it does not exist there in a natural state. I believe this to be an erroneous annotation. The speci- men corresponds so very closely to other material of Hevea guianensis from French Guiana (including speci- mens also collected by Richard) that I am convinced that it was collected in that colony. Louis Claude Richard, who was commissioned in 1781 to carry out explorations in French Guiana and the Antilles, spent much of his time in French Guiana, later travelling in Martinique, Guadeloupe, Jamaica, St. Thomas, and some of the islands in the Gulf of Mexico (cf. Laségue ‘‘Musée Botanique de M. Benjamin Delessert’’ (1845) 474). Hevea is known in a native state in none of these areas except French Guiana. Hevea guianensis Auwb/et var. lutea (Spruce ex Benth.) Ducke & Schultes in Caldasia 3 (1945) 249. [ 29 ] EXPLANATION OF THE ILLUSTRATION Pirate I. Reproduction of a page from Martius’ notes, preserved in the Munich herbarium. The manuscript description of Siphonia nitida Martius may be seen at the top of the page. I wish to thank the officials — especially Dr. Otto Renner and Dr. Th. Suessengurth — for making available for publication this interesting historical manu- script. PLATE | ‘ eA - rT? Ca, ar ny Se a Ya i pela ee bwergnecaccn “re Liens PAY. ehlonga/, Fan fru’ moetuchil ee Py heigl eye orrce CurmlCarper agu- tm teecens C01 sete E sens ce ae Mapwla su bpara cle (Zed) (kta, ) Leaves—‘‘Resin (acetone extract) 10.52%; Rubber hydrocarbon (benzene extract) 0.77%. Typical leaf rubber. ’”’ Hevea pauciflora (Spruce ex Benth.) Mueller- Argoviensis in Linnaea 84 (1865) 208. Brazit: Estado do Amazonas, Rio Negro, Carapana (above conflu- ence of Rios Negro and Uaupés). ““Tree 55 feet tall, columnar, 18 inches in diameter. Latex light yellow, thick, coagulating to a poor rubber. Bark externally smooth, light tan-brown, inside whitish, thick, hard. Flowers borne before old leaves fall. Staminate flowers open cupuliformly, yellow but red line running up centre of sepals internally and short red lines at junctions of sepals. Leaflets thick chartaceous, horizontal. Seed valves very large.’’ November 30, 1947, Richard Evans Schultes & Francisco Lépez 9194. At the little hamlet of Carapand, a few kilometers up- stream from the confluence of the Negro and the Uaupés, there is a small planting of trees of Hevea pauciflora. The collection cited above was made from this colony. These trees were tapped and an air-dried sample of the rubber was submitted to the National Bureau of Standards in Washington for analysis. The results of this analysis are significant. Dr. Lawrence A. Wood reported (Letter and report, Dr. Lawrence A. Wood to Dr. R.D. Rands, November 7, 1949) that: ‘*The sample was given [ 42 ] the rating of Grade III (fair) based on the system used at the National Bureau of Standards for grading wild rubber.’” (or details of this system, reference is made to Rubber Age 62 (1947) 173). Further notes in the analysis are: ‘‘Shrinkage on washing and drying 13.5%. Mooney viscosity 17.5. Resins (acetone extract) 17.1% : rubber hydrocarbon 80.4% ; insoluble materials 2%.” The sample, after vulcanization for 45 minutes (optimum cure) at 141°C according to the ACS II test formula, had the following tensile physical characteristics: ulti- mate elongation 685% ; tensile strength 1885. The elon- gation at 200 psi was 251%. Hevea rigidifolia (Spruce ev Benth.) Mueller- Argoviensis in Linnaea 84 (1865) 203. Brazit: Estado do Amazonas, Rio Uaupés, Taracua. Caatinga. Feb- ruary 3-6, 1948, Richard Evans Schultes 9673, 9678, 96SO. The rubber from Hevea rigidifolia has the reputation of being of a very poor quality. Indeed, a rapid exami- nation in the field can convince the investigator of the soundness of this reputation. here seems, however, to be something more complex than would appear on super- ficial examination. This is brought out by the unexpected results of Mr. McMullan’s analysis of an air-dried sample of rubber which I obtained by tapping twenty-six trees of Hevea rigidifolia in the extensive caatinga behind the settlement of Taracud on the Rio Uaupés, near the mouth of the Rio Tikié. The collections cited above are from three of the individuals which were tapped. Mr. MecMullan’s analysis follows: ‘‘Resin (acetone extract) 3.369% ; Rubber hydrocarbon (benzene extract) 95.01% ; Insolubles 1.68%. Appeared to be somewhat stronger and harder than gwanensis.”” Brazit: Estado do Amazonas, Rio Uaupés, Ipanoré. Caatinga. February 2, 1948, Richard Evans Schultes & Francisco Lopez 9665. [ 43 | Samples of the bark and leaves of an individual of Hevea rigidifola from Ipanoré were gathered for chem- ical examination. Ipanoré is the type locality of this species. It is very interesting to compare the analysis of the bark and leaf of this plant with that of the air-dried rubber specimen from 'Taracud, a locality only a few miles east of Ipanoré. Schultes & Lopez 9665 gave the following analysis: ‘‘Bark—Resin (acetone extract) 7.8590; Rubber hydrocarbon (benzene extract) 0.00 Leaves— Resin (acetone extract) 27.16% ; Rubber hy- drocarbon (benzene extract) 1.28%. Poor, typical leaf rubber. ”’ It is probable that more analyses of rubber samples taken in the field from wild populations and authenti- cated by identifiable herbarium material might furnish us with extremely significant information relative to such variations within a species from the same general region. [ 44 ] BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 16, No. 3 CampripGcre, Massacuusetts, May 15, 1953 STUDIES ON PERUVIAN PLEUROTHALLIS BY CHARLES SCHWEINFURTH THe following notes on Pleurothallis in Peru include two new species and six new varieties, as well as other nomenclatural discussions. Pleurothallis angustipetala C. Schweinfurth sp. nov. Herba mediocris, epiphytica, cum rhizomate repenti. Caules graciles, unifoliati, vaginis pluribus ornati. Folium breviter petiolatum ; lamina oblongo-elliptica, obtusa vel acuta, basi cuneata. Inflorescentia unica, folio multo longior, supra subdense multiflora. Flores minores, se- cundi, cum segmentis patentibus. Sepala similia, lanceo- lata vel anguste lanceolata, supra longe angustata, later- alia leviter recurva. Petala circiter duplo breviora, line- aria, acuta. Labellum petalis paulo longius, lateraliter visum sigmoideum, pandurato-oblongum, obtusum vel subacutum; pars basalis carinis binis humilibus ornata; pars anterior paulo major, oblongo-ovata. Columna par- va, cum ala terminali tridentata. Plant medium-sized, epiphytic, slender. Rhizome creeping, concealed by blackish sheaths. Roots fibrous, glabrous. Stems apparently in pairs, slender, about 4-11 em. long, unifoliate, bearing three or four close, tubular sheaths, the uppermost being longest and commonly separated. Leaf petioled, up to 7.2 em. long; lamina [ 45 ] oblong-elliptic, acute or obtuse, cuneate below, up to 6.5 cm. long and 1.8 cm. wide. Inflorescence solitary, nearly twice as long as the leaf, about 12.2 cm. or less long, suberect, densely or subdensely many-flowered above, looser below. Flowers commonly secund, rather small, yellow-green, with spreading segments. Sepals minutely cellular-papillose within. Dorsal sepal narrowly lanceolate, long-narrowed to an acute apex, concave be- low, up to 9.8 mm. long and 2.6 mm. wide near the base. Lateral sepals lanceolate, oblique and gently recurved above, up to 9.9 mm. long and 2.9 mm. wide. Petals much smaller than the sepals, linear, acute, prominently 1-nerved throughout, about 4.6 mm. long, up to 1 mm. wide. Lip somewhat longer than the petals, sigmoid when viewed from the side in natural position, pandurate- oblong, about 5.9 mm. long when expanded, up to 2.1 mm. wide above the middle, obtuse or subacute; basal portion very shortly clawed, subquadrate-oblong, with a pair of low keels that extend onto the basal part of the anterior portion; anterior portion slightly longer and broader, oblong-ovate. Column small, shorter than the petals, about 3 mm. long, narrowly winged on each side, with a short trilobulate wing at the apex, produced into a short foot. This species superficially resembles Pleurothallis rubens Lindl., but commonly shows a creeping rhizome produc- ing stems in pairs, has very narrow petals and a lip which differs in several details. Cuzco: Prov. of Paucartambo, Pillahuata, at 2800 meters altitude, epiphyte, January 25, 1945, C. Vargas 4956 (Tyee in Herb. Ames No. 65385); Same locality, at 3200 meters altitude, epiphytic in rain forest, ‘‘ceja de montafia,’? December 11, 1942, C. Vargas 3006a. Pleurothallis ciliata Know/. & Westc. var. abbre- viata C. Schweinfurth var. nov. Herba variabilis, inflorescentiis abbreviatis et floribus [ 46 ] extus dense pubescentibus et labelli basi cum lobulis parvis dentiformibus a specie differt. Plant epiphytic, medium-sized to large, apparently with a creeping rhizome. Stems unifoliate, 6-27 cm. long, with two or three close, tubular, evanescent sheaths near the base. Leaf erect, oblong-elliptic, subacute, cu- neate at the sessile base, 9-14 cm. long, up to 3.5 cm. wide. Inflorescences one or two (perhaps more), abbre- viated, nodding, densely few-flowered. Flowers small, finely short-pubescent without, bilabiate, brownish yel- low somewhat ‘“‘striped with darker brown-yellow.”’ Dorsal sepal narrowly oblong-lanceolate, acute and mu- cronate, up to 9.4 mm. long and 2.2 mm. wide, 3- to 5-nerved. Lateral sepals connate into an elliptic-lanceo- late, sharply bidentate lamina which is about as long as the dorsal sepal but nearly twice as wide. Petals much smaller than the sepals, elliptic-lanceolate, short-acumi- nate, serrate above, about 4 mm. long and 1.2 mm. wide near the middle. Lip smaller than the petals, lightly arcuate below, ovate-oblong, rounded at the apex, tri- lobulate above the cuneate base with a pair of small, erect, dentiform, obtuse lateral lobes, bicarinate through the middle, about 2.2 mm. long. Column about as long as the lip, abruptly dilated above. This variety of a widespread species is commonly dis- tinctly larger than the type and has very short inflores- cences. Moreover, the flowers are densely pubescent without, and the lip has short but distinct lateral lobules. Hvuanuco: Yanano, at about 1800 meters altitude, on mossy tree, May 13-16, 1923, J. Francis Macbride 3837 (Tyrr in Herb. Ames No. 37529). Junin: Chanchamayo Valley, at 1200 meters altitude, September 1930, Carlos Schunke 1127 (whole plant very small), Pleurothallis ciliata Anowl. & Westc. var. elon- gata C. Schweinfurth var. nov. [ 47 ] EXPLANATION OF THE ILLUSTRATION Pirate V. PLEUROTHALLIS ANGUSTIPETALA C. Schwein- Jurth. 1, plant, one half natural size. 2, flower ex- panded, three and one half times natural size. 8, column and lip from side, natural position, three and one half times natural size. Drawn by E.mer W. Siri PLate V PLEUROTHALLIS angqustipe tala C.S. ity i Herba mediocris, inflorescentiis folio multo longioribus et petalis caudatis a specie differt. Plant epiphytic, with a creeping rhizome. Stems slen- der, unifoliate, 5-8 em. long, with two close, tubular, evanescent sheaths below, about 1 cm. or more apart. Leaf erect, oblong-elliptic to oblong-oblanceolate, acute, cuneate at the sessile base, up to 7.7 cm. long and 2 cm. wide. Inflorescences one or two, much exceeding the leaf, arcuate to lax, loosely several-flowered above, up to about 18 cm. long. Flowers rather small, bilabiate, greenish yellow striped with red-brown, glabrous. Dor- sal sepal narrowly lanceolate-oblong, acute and apiculate, about 14 mm. long and 2.8 mm. wide. Lateral sepals connate into a broadly oblong, apically bidentate lamina which is about as long as the dorsal sepal but twice as broad. Petals much smaller than the sepals, oblong- lanceolate, lacerate-dentate above the base (especially on the anterior margin), long-caudate above, about 7 mm. long. Lip shortly clawed, about 4 mm. long, lightly re- curved below; lamina oblong-ovate, obtuse, subtruncate at base, erose-denticulate, with a pair of remote obscure carinate thickenings below, about 2 mm. wide; claw short, with a fleshy U-shaped callus. Column small, arcu- ate, wing-dilated above, about 8 mm. high at the back. This concept differs from the variable Pleurothallis ciliata in having the inflorescences much surpassing the leaf and in having the petals long-caudate. Lorero: Vicinity of Iquitos, at 100 meters altitude, on living tree in clearing, November to December 1936, G. Klug 10013 (Tyrr in Herb. Ames No. 60889). Pleurothallis excisa C. Schweinfurth sp. nov. Herba major, epiphytica, caespitosa. Caules approxi- mati, unifoliati, vaginis pluribus tubulatis ornati. Folium breviter petiolatum, erectum; lamina oblongo-elliptica, apice obtusa, basi cuneata. Inflorescentiae saepissime [ 48 ] unicae, folium multo superantes, supra dense multiflorae. Flores secundi, aurei. Sepala similia, lineari-lanceolata, longe acuminata, tubulari-coneava, lateralia valde obli- qua, quam sepalum dorsale latiora. Petala multo minora, oblique oblonga, apice rotundata vel subtruncata. Label- lum petalis paulo majus, pandurato-oblongum, dimidio basali subelliptico cum carinis binis conspicuis percurso, dimidio apicali ovato-elliptico, exciso. Columna parva, gracilis, arcuata. Plant rather large, epiphytic, up to about 80 cm. or more tall. Roots fibrous, numerous, elongate, glabrous. Stems approximate, unifoliate, up to 15.8 em. long, pro- vided with three or four close, tubular sheaths of which the uppermost is separated and elongate. Leaf short- petioled, up to 10.2 cm. long; lamina oblong-elliptic, minutely apiculate at the obtuse apex, cuneate at the base, up to 9 cm. long and 2.5 cm. wide, coriaceous. In- florescences one or two, much surpassing the leaf, many- flowered, dense above, loose below, 17—-18.7 cm. long, erect or spreading, in the axil of a small conduplicate spathe up to 11 mm. long. Flowers secund, yellow, medium-sized, with spreading segments. Sepals similar, narrow, long-acuminate, tubular-concave. Dorsal sepal linear-lanceolate, about 12 mm. long and 2.8 mm. wide near the base when expanded. Lateral sepals lanceolate, very oblique below, about as long as the dorsal sepal, 2.9 mm. wide, long-decurrent on the column-foot. Petals much smaller than the sepals, obliquely oblong, rounded or subtruncate at the apex, about 4.6 mm. long and 1.8 mim. wide. Lip slightly larger than the petals, pandurate- oblong, abruptly recurved at the base, about 5.6 mm. long; basal half subelliptic, traversed by a pair of high fleshy keels; anterior half elliptic-ovate, bidentate at the apex, concave. Column small, slender, arcuate, extended into a long ligulate apical wing which is incurved at the [ 49 ] EXPLANATION OF THE ILLUSTRATION Pirate VI. PLeurotrHauiis excisa C. Schweinfurth. 1, plants, one half natural size. 2, flower expand- ed, three times natural size. 3, column and lip from side, natural position, three times natural size. Drawn by Ev_mer W. Smitru Puate VI 2 \ tM PLEUROTHALLIS He ff ¢xcisa / i Ch chweinf: tip, about 5.5 mm. high including the wing, prolonged into a prominent curved foot. Among the South American species, this plant recalls Pleurothallis rubens Lind|., but has a dissimilar lip. It appears to be more closely allied to the Central Ameri- ean P. dolichopus Schitr., but is unique in having an excised lip. Cuzco: Prov. of Paucartambo, Pillahuata, at 3200 meters altitude, epiphytic in rain forest, ““ceja de montafia,’’ December 11, 1942, C. Vargas 3006 (Tyrer in Herb. Ames No. 65388). Pleurothallis fimbrilabia C. Schweinfurth nom. nov. Restrepia ecuadorensis Rolfe in Kew Bull. (1892) 188, non Pleurothallis ecuadorensis Schltr. Plant rather small, epiphytic, with a creeping rhizome. Stems approximate, slender, several-jointed, about 4-10 em. high, entirely or mostly concealed by several (up to 7) sheaths which are loose, tubular, scarious and finely red-purple maculate (the upper ones maculate only at their base). Leaf solitary, erect, oblong-elliptic to oval, obtuse or acute, cuneate at the sessile base, up to 6 cm. long and 2.9 cm. wide. Inflorescences fascicled, one to several, 1-flowered, issuing from the uppermost sheath just below the leaf, with the pedicel about 5 cm. long. Flower rather large for the genus, bilabiate, membrana- ceous. Dorsal sepal linear-lanceolate, about 2.2 cm. long, from a concave oblong-lanceolate lower portion gradually narrowed to a filiform apex which is clavate-thickened near the tip. Lateral sepals connate into an elliptic-oblong bidentate lamina which is subequally long with the dor- sal sepal but more than twice as wide. Petals similar to the dorsal sepal but much smaller, the oblong-lanceolate lower portion sometimes with a more or less distinct ap- pressed tooth. Lip about 9 mm. long, abruptly dilated from a small claw into a relatively broad suborbicular [ 50 | basal portion provided on each side with a slender seti- form lobule, then contracted into a linear-oblong, sharply bidentate lamina which is erose-fimbriate and muriculate through the center. Column slender, clavate above, about half as long as the lip. This description is mostly based on the Peruvian col- lection cited below which differs in some respects from Rolfe’s Ecuadorian material. The dorsal sepal and petals are slightly shorter. The sepals and petals are noted as translucent white lined with red or maroon, the lateral sepals being yellow thickly spotted with maroon or dark red. Since we consider that the concept Restrepia HBK. is referable to the variable genus Pleurothallis R.Br., I have made the above transfer. A new specific epithet is necessary, since the name Pleurothallis ecuadorensis was given to another species by Schlechter in 1915. Huanuco: Mufia, at about 2100 meters altitude, in dense forest, J. Francis Macbride 4018. Pleurothallis floribunda Poeppig & Endlicher Nov. Gen. ac Sp. 1 (1886) 48, t. 84. Pleurothallis macrophylla HBK. sensu Lindley in Bot. Reg. 28 (1842) Misc. p. 74, no. 44; Fol. Orch. Pleuro- thallis (1859) p. 8, no. 2. Pleurothallis macrophylla HBK. sensu Cogniaux in Martius FI. Bras. 8, pt. 4 (1896) 396. Several authors of note, especially John Lindley and Alfred Cogniaux, have treated Pleurothallis floribunda Poepp. & Endl. as referable to the older P. macro- phylla HBK. However, a careful consideration of the available evidence, including a photograph of P. macro- phylla from the Muséum d’ Histoire Naturelle at Paris, has convinced me that the two concepts are amply dis- tinct. The differences are shown in the following table. [ 51 | P. floribunda Leaf distinctly petioled Flowers subsessile Lateral sepals free to the base Petals half as long as the sepals, 1-nerved Lip subrotund in the middle and narrowed toward the base and apex,concave, scarcely shorter than the petals Habitat: Peru P. macrophylla Leaf sessile Flowers with a rather elongate slender pedicel Lateral sepals connate up to the apex Petals a little shorter than the sepals, 3-nerved Lip ligulate with revolute mar- gins, a little shorter than the sepals Habitat: Colombia Pleurothallis juninensis Sch/tr. var. subaequi- sepala C. Schweinfurth var. nov. Herba foliis magis oblongis acutis et sepalis subae- quilatis et sepalo dorsali ovato vel ovato-lanceolato a specie differt. Plant large, caespitose, up to 87 cm. tall from the base of the stem to the tip of an erect leaf. Stems 14— 27.5 em. long, often stouter than in the type. Leaf ovate-oblong or elliptic-oblong, erect or widely spread- ing, shortly cordate at base, acute or short-acuminate at the apex, 10-11 cm. long, up to 3.5 cm. wide. Flowers bilabiate, brown or reddish. Dorsal sepal oblong-ovate or ovate-lanceolate, acute or acuminate, about 9.5-16 mm. long, up to 5.4 mm. wide. Lateral sepals connate into an oblong-ovate or oblong-lanceolate lamina which is subequally long with the dorsal sepal and slightly nar- rower, acute or acuminate. Petals similar to those of the type. Lip similar to the typical form, oblong, rounded and minutely apiculate at the apex, scabrous, 5.1-6.1 mm. long. This variety differs from the type in having the leaves broadest near or at the middle and especially in having the sepals more ovate or ovate-lanceolate with the dorsal [ 52 ] sepal slightly broader than the lamina of the lateral sepals. It occurs at a rather greater altitude than the type. Cuzco: Prov. of Convencién, Hda. Potrero, Sapan Sachayoce, at 2200 meters altitude, epiphytic in dense forest, March 5, 1942, C. Vargas 2561 (Tyrer in Herb. Ames No. 65371); “‘Pillahuata,’’ Cerro de Cusilluyoc, above Rio Pillahuata, at 2300-2400 meters altitude, epiphyte in forest, May 38-6, 1925, F. W. Pennell 14005. Pleurothallis Lindenii Lind/ey var. longiracema C. Schweinfurth var. nov. Herba mediocris, inflorescentia stricta quam folium longiore et petalis oblongo-linearibus et labello trilobu- lato a specie differt. Plant medium-sized, up to 89 cm. tall. Stems often proliferating, the lower stems about 9-21 cm. long from a decumbent base, unifoliate, with 8 or 4 close, tubular, evanescent sheaths. Leaves sessile or very short-petioled, ovate-oblong to elliptic-oblong, acute or acuminate, cu- neate at the base, up to 11.5 cm. long and 2.8 cm. wide. Inflorescences solitary, strict or nearly so, loosely few- to several-flowered, up to 18 cm. long, rising from a more or less conspicuous, conduplicate spathe up to 1.9 em. long. Flowers bilabiate, rather similar to those of the type, apparently densely spotted. Dorsal sepal ob- long-ovate, acuminate, concave, about 10 mm. long and 5.2 mm. wide when expanded. Lateral sepals entirely connate into a deeply concave suborbicular-ovate lamina which is slightly longer but much broader than the dor- sal sepal. Petals oblong-linear, acute, suboblique, about 7.5 mm. long. Lip much shorter than the other seg- ments, suborbicular-ovate in outline, lightly but dis- tinctly 3-lobed, cordate at the base, acute at the apex, about 4 mm. long and 5.2 mm. wide; lateral lobes ob- liquely ovate-rounded, apparently contracted by a me- dian transverse keel; mid-lobe larger, triangular-ovate. Column minute, stout. This variety differs from the type in having strict ra- cemes which are more or less longer than the leaf, much narrower petals and a trilobulate lip. Cuzco: Prov. of Urubamba, Tuncapata, Sta. Rita, at 2800 meters altitude, epiphytic, flower yellow-brown, March 28, 1942, C. Vargas 2663 (Tyrer in Herb. Ames No. 65369); Near Tuncapata, at 2600 meters altitude, “‘laderas junto al camino entre herbaceas,’’ March 10, 1944, C. Vargas 4098. Pleurothallis pedunculata (K/.) Reichb.f. var. peruviana C. Schweinfurth var. nov. Herba inflorescentiis quam folium multo longioribus et floribus minoribus et segmentis angustioribus et labello magis ovato a specie differt. Plant large, 80 to over 40 cm. high. Stems approxi- mate, rather slender, unifoliate, 8-18 cm. long, with two or three close, tubular, evanescent sheaths. Leaf shortly but distinetly petioled; lamina ovate-oblong to oblong- elliptic, acute with a minutely tridenticulate apex, broad- ly cuneate at the base, up to 18.5 cm. long and 4.3 cm. wide. Inflorescences one to three, suberect to flexuous, about 23 cm. or less long, loosely few- to several-flowered, subtended by a conspicuous conduplicate spathe up to 5.7 em. long. Flowers smaller than in the species, ‘‘red- dish-greenish-yellow.’” Dorsal sepal linear-lanceolate, gradually narrowed to an obtuse tip, about 1.6 cm. long and 2.8 mm. wide at the concave base. Lateral sepals connate into an oblong-lanceolate, ]long-acuminate lam- ina which is slightly shorter but much broader than the dorsal sepal. Petals lanceolate-linear, caudate-acuminate and fleshy above, almost as long as the dorsal sepal. Lip much shorter than the other segments, tubular-involute in natural position, oblong-ovate when expanded, rounded at the apex, about 5.9 mm. long and 4 mm. wide near the base, with the anterior half rather abruptly con- tracted. Column minute. [ 54 | This variety diverges from the type in having the in- florescences much longer than the leaf, smaller flowers with narrower segments and a more ovate lip. Hvuanuco: Yanano, at about 1800 meters altitude, in patches on humus, May 18-16, 1923, J. Francis Macbride 3849 (Tyre in Herb. Ames No. 60876). Pleurothallis tenuifolia C. Schweinf. var. longi- sepala C. Schweinfurth var. nov. Herba mediocris, caespitosa, caulibus gracilioribus et foliis angustioribus et sepalis longioribus a specie differt. Plant medium-sized, caespitose, epiphytic, with an abbreviated rhizome. Stems approximate, numerous, unifoliate, very slender, up to 11 cm. long, provided with two or three close, tubular, evanescent sheaths. Leaf erect or spreading, narrowly linear, abruptly obtuse to subacute with a minute apicule, gradually narrowed be- low to a subpetioled base, the margins being conspicu- ously revolute, up to 18.5 cm. long and 5 mm. wide in the dried specimen. Inflorescences commonly solitary (sometimes the remnants of four peduncles are present), lax or nodding, loosely several- to many-flowered almost to the base, markedly shorter than the leaf, up to about 9cm. long. Flowers yellowish white. Sepals free, spread- ing above, finely papillose-pubescent within, 3-nerved. Dorsal sepal lanceolate-linear, concave below, about 20 mm. long and 2 mm. wide near the base. Lateral sepals very similar, lightly oblique. Petals much smaller than the sepals, cuneate-oblong, subacute, lightly oblique, about 4mm. long and 1.8 mm. wide above. Lip lightly recurved in natural position, oblong-oblanceolate, obtuse at the apex, gradually narrowed toward the base, about 5 mm. long and 1.8 mm. wide above, lightly bicarinate near the middle. Column very small, shorter than the petals, wing-dilated above, denticulate at the apex, about 3 mm. high at the back. This plant differs from P. tenuifolia in being more slender throughout, with narrower leaves, shorter inflo- rescences and much longer sepals. Hvuanuco: Prov. of Pachitea, above La Molina near Panao, Sep- tember 12, 1940, Erik Asplund 13697. Pleurothallis tunguraguae Lehmann & Kriinzlin in Engler Bot. Jahrb. 26 (1899) 439. A collection of this Ecuadorian species has recently been made in Peru. It differs from an isotype specimen in the Ames Herbarium in having markedly broader leaves (8.1-3.6 cm., as contrasted with 2.5 cm.), only two inflorescences (as compared with five in the isotype) and the flowers somewhat larger than those described, although but little larger than those in the isotype. Also the mid-lobe of the lip in the type is relatively larger than the lateral lobes. Amazonas: Prov. of Bongara, on ridges above Quebrada Santa Rosa, across northeast end of Lake Pomacochas, at 2430 meters alti- tude, epiphytic in mountain rain forest, flowers white, April 4, 1944, W. H. Hodge 6154. BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CampripGak, Massacnusetts, June 19, 1953 VoL. 16, No. 4 PLANTAE AUSTRO-AMERICANAE VIII DE PLANTIS PRINCIPALITER VALLIS AMAZONICIS NOVIS VEL CRITICIS NOTAE DIVERSAE BY Ricuarpb Evans Scuutres' StrupiEs which have been carried on during the past two or three years have disclosed in recent plant collections from South America a number of species which are new to science or otherwise noteworthy. This article is a con- tinuation of a series dedicated to the widening of our understanding of the flora of northern South America, with special reference to the Amazon basin. [t is with pleasure that I acknowledge the collabora- tionof Mr. Jason R. Swallen (Gramineae); Dr. Harold N. Moldenke (Hriocaulaceae); and Mr. Emery C. Leonard (Acanthaceae). GRAMINEAE Panicum molliculmum Swadlen sp. nov. Annuum (/); culmi tenuissimi, decumbentes, 5-15 em. longi, glabri; vaginae internodiis 4} breviores, pilo- sae; ligula brevissima, ciliata; laminae 5-10 mm. longae, 2-4 mm. latae, obtusae, basi subtruncatae, hirtellae; pan- iculae terminales et axillares, simplicissimae, pedunculis 1Botanist, Division of Rubber Plant Investigations, Bureau of Plant Industry, Soils, and Agricultural Engineering, Agricultural Research Administration, U. S. Department of Agriculture; Research Fellow, Botanical Museum of Harvard University. [ 57 | filiformibus, spiculis 1-5 breviter pedicellatis; spiculae 1.2-1.5 mm. longae; gluma prima minuta; gluma se- cunda et lemma sterile aequales, fructu paulo longiora, acuta, pilosa; fructus minute rugosus. Swallen states: ‘“This species is not closely related to any known species of the genus Panicum. It is un- usual in having few to several axillary inflorescences from the upper nodes, these long exserted on long filiform peduncles. ”’ Cotomsi1a: Comisaria del Caqueta, Cerro de El Castillo, Rio Apapo- ris. Damp sandstone ledge with mosses and Selaginellas. January 16, 1942, Gabriel Gutiérrez & Richard Evans Schultes 616 (Tyrr in U.S. Nat. Herb.). CYPERACEAE Cephalocarpus Dracaenula Nees in Martius FI. Bras. 2, pt. 1 (1842) 162, t. 18. The collection which is cited below establishes the oc- currence of this curious epiphytic cyperaceous plant far to the north of its only other known locality, the Cerro de La Pedrera on the Rio Caquetaé at the Colombian- Brazilian boundary (Schultes in Bot. Mus. Leafl. Har- vard Univ. 18 (1949) 298). Additional material and more exact knowledge of this genus may indicate that Schultes & Lopez 10066 repre- sents a variety of Cephalocarpus Dracaenula, but it seems to match this species as delineated by Gilly (in Bull. Torr. Bot. Club 69 (1942) 290). Cotoms1a: Comisaria del Vaupés, Rio Guainia basin, Rio Naquieni, vicinity of Cerro Monachi. ““Common epiphyte on rocks amidst mosses and lichens and on trunks. Flowers very small, white. On summit of mountain.’’? June 1948, Richard Evans Schultes & Francisco Lépez 10066. CYCLANTHACEAE Carludovica aurantiaca RP. 1’. Schultes sp. nov. Planta caulescens, terrestris, in rivulorum saxis colonice [ 58 ] erescens. Caulis comparate robustus, sublignosus, usque ad 10 cm. longus vel longior, 1.7 cm. in diametro, as- periter fibroso-squamosus. Petioli tenuissimi, usque ad 50 cm. longi, basi dilatati, vaginis linearibus inclusi. Foliorum laminae usque ad 35 cm. longae, 5/6 longitu- dinaliter bipartitae; segmenta anguste linearia (saepis- sime longitudinaliter lacerata), apice filiformia, basi plus- minusve 6 mm. lata. Pedunculi magnopere tenuissimi, usque ad 40 cm. longi, plerumque 2 mm. in diametro, fortiter fibrosi. Spathae hyalinae, vivo triangulares, apice acutae, usque ad 4 cm. longae, 2.5 cm. latae, superior lineari-triangularis, apice longissime attenuata, basi 2-3 em. lata, 6.5-10 cm. longa. Spadix cylindricus, apice rotundatus, 16 mm. longus, usque ad (sed saepissime mul- to brevior) 7-11 mm. in diametro. Flores pistillati parvi, 2 mm. in diametro, subconcrescentes, perianthio quadri- lobato; stigmatibus carnosis, in apicem ovarii confluen- tibus, cruciformibus, in circuitu oblongis, 1.3 mm. longis, 0.3 mm. latis. Flores staminiferi conspicul, 1-1.5 mm. in diametro; pedicellis valde complanatis, 1.5 mm. lon- gis, perianthii lobis liberis suborbicularibus, apice rotun- datis, 1.5 mm. longis, circiter 1 mm. latis; staminibus numerosissimis (usque ad sexaginta), filamentis brevibus, antheris in circuitu suborbicularibus, 0.4 mm. in diame- tro, bilocularibus ; staminodia conspicuissima, caduca, fili- formia, usque ad 7 vel 10 cm. longa, splendide aurantiaca. This new species of Car/udovica is a beautiful and grace- ful plant which grows in dense and extensive colonies. It can be distinguished from all other described members of the genus by its rich orange-colored staminodes. It ap- proaches in some respects Carludovica pygmaea Gleason, but can be separated from this species by its much longer and more narrowly linear leaves and its larger floral bracts. Cotomsia: Comisaria del Vaupés, Rio Guainia basin, Rio Naquieni, ee , Oey. . . 7 vicinity of Cerro Monachi. ~Epiphyte on rocks in rapids. Flowers [ 59 | white, excessively fragrant of narcissus. Staminodes bright orange. In caatinga forest at base of mountain.’’? June 1948, Richard Evans Schultes & Francisco Lépez 10085 (Tyrer in Herb. Gray). ARACEAE Anthurium atropurpureum Schu/tes & Maguire sp. nov. Herba terrestris, parva, in silva non densa arenosa hu- midacrescens. Caudiculus magnopere abbreviatus, robus- tus, internodiis brevissimis. Folia rigide erecta, cum peti- olis strictis, vivo aliquid complanatis, striatis, basi usque ad 5 mm. in diametro; lamina valde chartacea, clare viridis, glabra, lanceolato-elliptica, apice longe acuminata (acu- mine quam 2 cm. longiore), basi attenuata, leviter mar- ginata, nervis secundariis plusminusve novem, arcuatis, confluentibus, subtus prominenter elevatis, supra paulo elevatis. Inflorescentia erecta, pedunculo aliquid carno- sulo ut videtur, vivo probabiliter subcomplanato, plus- minusve 1 mm. in diametro, purpureo. Spatha membra- nacea, atropurpurea, lanceolata, apice acutissima, 2.5 cm. longa, 5 mm. lata, petiolum cingens. Spadix erectus, eylindricus, 2.5 cm. longus, 8 mm. in diametro, atro- purpureus. Flores valde regulares, quadrangulares. Anthurium atropurpureum would seem to be set apart from other small Amazonian species of the genus by the combination of its unusual leaf shape and the deep, rich purple color of the fertile parts. Although it was col- lected in the same locality as the following species, the two concepts are not closely related. Cotompra: Comisaria del Amazonas, trapecio amaz6nico, ‘‘Varial perto da Quebrada Agua Préta. Epiphyta no chao.’’ November 8, 1946, George A. Black & Richard Evans Schultes 46-375 (Tyrer in Herb. Gray). Anthurium fontoides PR. Ll’. Schultes sp. nov. Herba epiphytica, parva, gracillima. Caudiculus cras- [ 60 | sus internodiis abbreviatis. Folia coriacea, petiolis stric- tis, crassis teretibusque, sulcatis, basi latiuscule dilatatis, usque ad 2 cm. longis, 1.5—2 mm. in diametro; lamina pallide viridis, glabra, lineari-lanceolata, apice longissime acuminata, basi longe cuneata et decurrens, margine levi- ter revoluta, subtus minute tessellato-squamulosa, 18-26 em. longa, 1.2—-1.4 cm. lata, nervis crassis, utrinque pro- inentibus, secundariis duobus nervo centrali parallelis. Inflorescentia brevis, gracilis, pedunculo usque ad 6 cm. longo, leviter complanato. Spatha firme membranacea, lineari-lanceolata, apice acuta, usque ad 3 cm. longa, 2 mm. lata, vivo reflexa, ut videtur rubens vel purpurea. Spadix sessilis, cylindricus, apicem versus non attenua- tus, 6-7 cm. longus, plusminusve 8 mm. in diametro, probabiliter flavus. Anthurium fontoides is very closely related to A. gra- cule (Rudge) Engl., a widespread tropical American spe- cies of which several varieties have been recognized. Later studies and collections may indicate that the concept here described as a species is deserving only of varietal rank. Anthurium fontoides can be separated from A. gracile on first sight by its narrower leaves which are basally de- current, by the more closely packed basal bracts, by its much shorter petioles, and by its smaller spathe. Per- haps the greatest difference is to be found in its having an inflorescence which, instead of being subequal to the leaves, is less than half as long. The specific epithet was suggested by the graceful habit of the leaves. Cotomsia: Comisaria del Amazonas, trapecio amaz6énico, Loreto- yacu River. Alt. about 100m. ““Epiphytic.’’ September 1946, Richard Evans Schultes & George A. Black 8399 (Tyrer in Herb. Gray). Anthurium nemoricola Schultes & Maguire sp. nov. Herba terrestris, humillima, in silva non densa arenosa [ 61 | humida crescens. Caudiculus abbreviatus, erectus, inter- nodiis magnopere brevissimis. Folia rigide erecta, cum petiolis strictis usque ad 8 cm. longis, basi teretibus, apice aliquid canaliculatis; lamina valde coriacea, glabra, sub- tus tessellata, subsagittata (ad basim leviter lobata), 7-9 em. longa, basi 2.8-3.2 em. lata, apice acuminata, basi rotundata, margine integra, valde revoluta, nervis secun- dariis duobus nervo centrali subparallelibus, adscenden- tibus, tertianis decem vel duodecim, subtus nervis promi- nenter elevatis supra conspicue canaliculato-depressis. Inflorescentia erecta, pedunculo 11 cm. longo, cylindrico. Spatha vivo reflexa, anguste lanceolata, acuta, extus pulverulenta, 2.5 cm. longa, 6 mm. lata, sanguinea. Spadix erectus, albus, 2 cm. longus, cylindricus, tenuis (2mm. in diametro), ad apicem attenuatus. Flores quad- rangulares, regulares. The minute stature of Anthurium nemoricola, combined with the unusual shape, venation and texture of the leaves, serve to set this concept apart from other Amazonian species of the genus. It is highly adapted to the xero- phytic conditions which obtain, in spite of ample rainfall, because of high acidity. The habitat, as indicated by the specific epithet, is a light savanna-forest of small treelets and abundant light. Cotompi1a: Comisaria del Amazonas, trapecio amaz6nico, interior regions of trapecio between Amazon and Putumayo watersheds. Alti- tude above 100 m. “‘In sandy savanna-forests or varial.’’ November 1945, Richard Evans Schultes 6900A (Tyre in Herb. Gray). Anthurium pluviaticum RP. L. Schultes sp. nov. Herbaepiphytica, parva. Caudiculus abbreviatus, satis robustus, internodiis brevibus. Folia tenuiter papyracea, erecta, cum petiolis gracillimis vivo ut videtur teretibus, obscurissime striatis, basi usque ad 2.2 mm. in diametro, 4.5-6 cm. longis; lamina pallide viridis, glabra, regulari- ter hastata, apice breviter acuminata, basi valde cordata, [ 62 | omnino magnopere minutissime tessellato-squamulosa, 7.5-10 cm. longa, 6-7.5 cm. lata (parte centrali); nervo centrali robusto, nervis secundariis valde arcuatis, terti- anis angulatim 45° patentibus, nervis omnibus supra in- conspicuis sed subtus prominentioribus. Inflorescentia erecta, gracillima, pedunculo usque ad 16 cm. longo, eylindrico. Spatha viridis, vivo valde reflexa, membra- nacea et hyalina, oculo armato minutissime squamulosa, nervulis prominentibus, lanceolato-elliptica, apice acutis- sima, 4.5 cm. longa, 7.5-8 mm. lata. Spadix erectus, albidus, usque ad 5 cm. longus, cylindricus, comparate tenuis (3 mm. in diametro, ad apicem non attenuatus), stipite 1.5 cm. longo suffultus. Flores valde regulares, subquadrangulares. From other species with similar hastate leaves, 4n- thurium pluviaticum can be distinguished by its long- stipitate and unusually slender spadix, by its very abbre- viated caudicle and by the small size of all its parts. It would seem to resemble the type of Anthurium chloro- carpum Sodiro in some respects, but its leaves are smaller, more acuminate and more nearly membranaceous, and it has an inflorescence which, in addition to the stipitate spadix, shows important morphological differences. The specific epithet of Anthurium pluviaticum recog- nizes that the plant grows in an area which has one of the world’s heaviest rainfalls. Cotomsia: Departamento del Valle, Rio Calima, Quebrada de La Brea. Alt. 30-40 m. ‘‘Epiphyte. Spike white.’’ May 19, 1946, Richard Evans Schultes & Mardoqueo Villarreal 7348 (Typr in U.S. Nat. Herb. ). Anthurium tikunorum R. LE. Schultes sp. nov. Herba terrestris, usque ad 14 pedes alta, in silva non dense arenosa, humida crescens. Caudiculus multo ab- breviatus, erectus, internodiis brevissimis. Folia rigidis- sime erecta, cum petiolis strictis, 12-24 (plerumque plus- [ 63 ] minusve 20) cm. longis, basi teretibus, apice canalicula- tis, vivo valde complanatis; lamina valde coriacea, glabra, utrinque minutissime tessellata, hastata, 14-16 em. longa, parte centrali 8—5.5 em. (vel basi 6.5-10 em.) lata, apice acuta, basi truncata et abrupte triangulari-cuneata, lobis lateralibus rotundatis, margine integra, valde revoluta, nervis secundariis duobus nervo centrali subparallelis, adscendentibus, tertianis decem vel duodecim, subtus nervis prominenter elevatis supra conspicue canaliculato- depressis. Inflorescentia erecta, pedunculo robusto, cy- lindrico, conspicue striato-canaliculato, 80-35 cm. longo, 2 mm. in diametro sed basim versus aliquid dilatato. Spatha vivo reflexa, anguste lanceolata, acutissima, extus densius scrobiculata, 5.5 cm. longa, 6 mm. lata, probab- iliter atrosanguinea vel fusco-rubens. Spadix erectus, non stipitatus, flavus, 5.5 cm. longus, maturitate 8 mm. in diametro. Flores quadrangulares, regulares, tepalis extus dense squamulosis. Baccae carnosae, virides, glob- osae, densae. It is at once evident that Anthurium tikunorum is very intimately allied to A. nemoricola Schultes & Maguire from the same region, but there are differences which would seem to be of sufficient magnitude to warrant a specific, and not a varietal rank, for the concept described above. Anthurium tikunorum is very much larger in all its parts than A. nemoricola; the leaves of the former species are very definitely hastate with prominent lateral lobes basally whereas those of the latter are subsagittate, albeit with a slight suggestion of lobes which, however, do not develop. The specific epithet of Anthurium tikunorum refers to the Tikuna Indians who inhabit Colombia’s trapecio amazonico and adjacent areas of Brazil and Pert and who employ this plant to relieve a condition of pyorrhea which is not uncommon amongst them; the astringent [ 64 | juice of the fleshy fruits is applied to staunch the flow of pus from the gums. Cotompia: Comisaria del Amazonas, trapecio amazonico, interior regions of trapecio between Amazon and Putumayo watersheds. Alt. ee ° . Ps above 100 m. ‘‘Varial [caatinga] near headwaters of Pamaté. Septem- ber 1946, Richard Evans Schultes 8108 (Type in U.S. Nat. Herb.). ERIOCAULACEAE Paepalanthus fasciculatus (Rottb.) K6érnicke in Martius FI]. Bras. 3, pt. 1 (1868) 357. A rather abundant though highly localized little plant, Paepalanthus fasciculatus seems to occur on the sandy savannas or caatingas of both cretaceous quartzite and cambrian granite origin—one of the relatively few spe- cies common to the floras of both of these geological areas in the northwestern Amazon. It is a species of the northwestern Amazon Valley and the upper reaches of the Orinoco. Brazi_t: Estado do Amazonas, Rio Negro, Jucabi (at mouth of Rio Curicuriari. September 23-24, 1947, Richard Evans Schultes & Fran- cisco Lopez 8837.—Estado do Amazonas, Rio Negro basin, Rio Curi- curiari, February 1948, Richard Evans Schultes & Francisco Lopez 9715. Cotomsia: Comisaria del Amazonas, Rio Igaraparana, La Chorrera. ee . . - me * Witoto Indian name: hé-gé-ne.’’ June 6, 1942, Richard Evans Schultes 3943, Paepalanthus Schultesii Moldenke sp. nov. Herba gracilis, caulescens ; caulibus tenuibus, usque ad 12 cm. longis, foliosis ; foliis membranaceis, graminoideis, patentibus, 2-8.5 cm. longis, medio 1—2.5 mm. latis, mar- ginibus remote villoso-ciliatis, caeterum minutissime puberulis vel glabrescentibus; pedunculis paucis, 2-9 cm. longis, 4- vel 5-costatis, in Juventute pilosis, denique glabris, tortis; vaginis 1.5-2 cm. longis, parce adpresso- pilosis, glabrescentibus, ad apicem 8-lobatis; capitulis griseis, hemisphaericis, 5-8 mm. in diametro. Slender caulescent herb. Stems very slender, up to [ 65 ] about 12 em. long, long-pilose when young, glabrescent in age, rather densely leafy. Leaves membranuos, uni- formly dull green on both surfaces, grass-like, 2-8.5 cm. long, 1-2.5 mm. wide at the mid-point, attenuate-acute at the apex, 4- to 7-veined, but the veins not conspicuous, not fenestrate, rather remotely but very conspicuously villous-ciliate along the margins with stiff, divergent, white hairs about 3 mm. long, otherwise very minutely puberulous-pulverulent or glabrescent on both surfaces. Peduncles few, mostly 1 or 2 per season at the apex of the stem but several of previous seasons persistent far- ther down, erect, slender, 2-9 cm. long, 4- or 5-costate, twisted, antrorsely white-pilose when young, eventually glabrescent; sheaths slender, 1.5-2 cm. long, sparsely appressed-pilose or eventually glabrescent, conspicuously 3-lobed at the apex, the lobes subhyaline-scarious, lan- ceolate, about 4 mm. long, attenuate-acute at the apex, glabrous, erect. Heads grey, hemispheric, 5-8 mm. in diameter; involucral bractlets light brownish, ovate, about 3.5 mm. long and 1.9 mm. wide, slightly concave, rounded at the apex, densely short-ciliolate along the margins, otherwise subglabrous or sparsely pilose toward the apex on the back; receptacle long-villous; receptac- ular bractlets stramineous or light brownish, oblong- elliptic, about 2.8 mm. long and 0.9-1 mm. wide, blunt at the apex, ciliate-pilose toward the apex on the back and barbellate at the apex. Staminate florets: sepals 8, con- nate only at the base for about 1 mm., brownish, about 2.5 mm. long and 0.7 mm. wide, obtuse and barbate at the apex, otherwise glabrate; petals 8, whitish, connate into an infundibular tube about 2.5 mm. long, the upper free portions subhyaline and about 1 mm. long, glabrous, not glanduliferous; stamens 8, exserted, erect-spreading ; filaments about 1.5 mm. long, glabrous; anthers white. Pistillate florets: sepals 8, brown, separate to the base, [ 66 | hyaline, narrow-elliptic, about 2 mm. long and 1 mm. wide, acute at the apex, rather densely long-pilose with erect white hairs on both surfaces; pistil about 3 mm. long; ovary brown, 38-sulcate, 3-celled, 8-ovulate; style about 0.8mm. long, glabrous; the 8-stigmas and 8 style- appendages issuing from the apex of the style, erect, about 1.5 mm. long. Cotomsia: Departamento de Cundinamarca, Macizo de Bogota, Quebrada de Chapinero. Alt. about 9000 feet. September 24, 1941, Richard Evans Schultes 1024 (Tyre in Herb. Britton, N.Y. Bot. Gard. ). MENISPERMACEAE Abuta rufescens Audblet Hist. Pl. Guian. 1 (1775) 618, t. 250. This species is widespread from French Guiana south through the Amazon Valley to Minas Geraes. It has been identified as the species which Martius reported to be employed in the preparation of arrow-poison along the Rio Caqueta (Japura) (Krukoff & Moldenke in Brit- tonia 8 (1988) 67). Brazit: Estado do Amazonas, Rio Negro basin, Rio Curicuriari, near summit of Serra Curicuriari. April 26, 1948, Richard Evans Schultes & Francisco Lopez 9827 A. Chondodendron toxicoferum (Wedd.) Krukoff & Moldenke in Brittonia 8 (1939) 838. Chondodendron toaicoferum, known from the western ~ Amazonia of Pert and Brazil, has apparently not been hitherto reported from such a northwestern locality as the upper reaches of the Apaporis River where Schultes 5526 was collected. CotomsBia: Comisaria del Vaupés, Macaya River, vicinity of Cachi- vera del Diablo. “‘Vine. Fruits yellow, bitter.’? May 1943, Richard Evans Schultes 5526. [ 67 | LEGUMINOSAE Hymenaea oblongifolia Huber in Bol. Mus. Para. 5 (1909) 386. Cynometra Zamorana R. E. Schultes in Bot. Mus. Leafl. Harvard Univ. 18 (1949) 801, t. 84. The collection Philipson 1647 has very appreciably extended the known range of this beautiful tree which is known at present from the State of Para in Brazil (Ducke 9137, 14982, 16570) west to the eastern rim of the Macarena Mountains near the Andean Cordillera in Colombia. Philipson 1647 is apparently the fourth collection from Colombia. In 1912, Ducke collected it at the rapids of Cupati (now called La Pedrera) on the Rio Caqueta, at the Colombia- Brazilian boundary. The two other Colom- bian collections—Schultes 5424 and 5429—were made in the upper reaches of the Apaporis basin, half way be- tween the Macarena and the Cupati localities. Schultes 5424 and 5429 are in fruit and were described recently as Cynometra Zamorana R. EK. Schultes. Study of the flowering Philipson and Ducke collections has indicated the identity of the Schultes material with Huber’s Hymenaea oblongifolia. It is obvious that Hymenaea oblongifola in eastern Colombia is associated with the ancient cretaceous out- liers of the Venezuela-Guiana land mass. . 4s ee at foot of eastern slopes of Sierra Macarena; alt. 450m. — Large tree [ 68 ] 25 m., leaves coriaceous, shining, dark green above, sepals reddish brown, petals white.’’ November 29, 1949, W. R. Philipson 1647. Monopteryx Uaucu Spruce ex Bentham in Martius Fl. Bras. 15, pt. 1 (1862) 807. This enormous tree, especially abundant on the pro- terozoic granitic shield of the Rio Negro basin, has been cited by Ducke (‘‘As Leguminosas da Amazonia brasil- eira’’ (1939) 109) as occurring also along the lower Rios Ica (Putumayo) and Solimdées, as well as in Venezuela and Colombia. I had not seen any specimens from Co- lombian territory, but the species was certainly to be ex- pected in the Colombian Amazonia along the Brazilian boundary. ‘The collection cited below establishes the occurrence of Monopteryx Uaucu far into Colombian territory and near the very headwaters of the Rio Negro. Many collections have been made recently in the Brazil- ian part of the Rio Negro valley by collectors of the Instituto Agronomico do Norte which has been much interested in the wauwcw as an oil-tree. The seeds of Mon- opterye Uaucu are exceedingly rich in oil and, when roasted or boiled, are edible; the oil is employed as a food as well as a fuel for lamps (La Cointe: ‘‘A Ama- zonia brasileira. If. Arvores e plantas utéis’’ (1934) 452; Ducke l.c.). Amongst the Kuripaka Indians of the Rio Guainia, Monopteryx Uaucu is called ha-pét-o6-he. CotomsBiA: Comisaria del Vaupés, Rio Guainia basin, Rio Naquieni, a C6 Ty . ) at base of Cerro Monachi. ‘Enormous tree with buttress roots.’” June 1948, Richard Evans Schultes & Francisco Lépez 10126. IK. UPHORBIACEAE Nealchornia japurensis Huber in Bol. Mus. Goeldi 7 (1913) 298. The type of Nealchornia japurensis was collected in Colombia at Puerto Cérdova (above La Pedrera) on the [ 69 ] Rio Caqueta (Japura). Black & Schultes 46-257 repre- sents apparently the second record of this rare plant from Colombia. It has been found several times in the adja- cent parts of the Amazon in Brazil. A specimen of Ule 6292 was found amongst the un- identified Muphorbiaceae during my recent work at Kew. It also is referable to Nealchornia japurensis, a species which hitherto apparently has not been reported from Pert. It will be noted that the Ule collection was made at an earlier date than the type, but it has been overlooked. Cotomp1a: Comisaria del Amazonas, Rio Loretoyacu. “‘Arvore 15 m. Flor verde com cheiro de citrus.’’ October 31, 1946, G. A. Black & R. EF. Schultes 46-257. Peru: Departamento de Loreto, Yurimaguas. “‘Baum 5-10 m. Bl. gelb. in Walde bei Yurimaguas.’’ August 1902, E. Ule 6292. Sagotia racemosa Ballon in Adansonia 1 (1860— 1861) 54. Sagotia racemosa Baillon var. brachysepala Mueller- Argoviensis in Fl. Ratisb. 47 (1864) 516. Sagotia racemosa Baillon var. genuina Mueller-Argo- viensis ibid 516. Sagotia racemosa Baillon var. ligularis Mueller-Argo- viensis ibid 516. Sagotia racemosa Baillon var. macrocarpa Mueller- Argoviensis ibid 516, When Mueller published the descriptions of the vari- etal concepts of Sagotia racemosa, he had available a very limited number of collections; the differences which they exhibited seemed to be significant. A study of the material now available and a comparison of the newer with the older collections shows that all of the characters upon which Mueller had based his varietal concepts are either manifestations due to age or seasonal influences or else to individual idiosyncrasies. Therefore, I am reduc- ing the varietal epithets to synonymy under Sagotia racemosa. [ 70 | Brazit: Estado do Para, Lago Salgado, Rio Trombetas. Forest, August 23, 1910, dA. Ducke 11067.—FEstado do Para, Belém. March 12, 1901, 4. Ducke 2001.—FEstado do Para, Peixeboi, chemin de fer de Braganca. **Arataciti.’’ August 13, 1907, R. Siqueira 8277.—Es- tado do Parad. Igarapé-assti, chemin de fer de Braganea. “‘Aratacit.”’ February 6, 1903, R. Siquetra 3343.—Estado do Para, Serra de Parin- tins. September 15, 1907, 4A. Ducke 8724.—Estado do Para, Belém. December 11, 1914, 4. Ducke 15549.—Estado do Para. Distrito Ac- ara, Thomé Asst, up Rio Acard 6 km. on river bank in forest. ‘‘Ar- borescent 6 m. high; white flower, Murascaca.’’ August 5, 1931, Y. Mexia 6051.—Estado do Para, Belém. “‘Shrub 2 ft. Roots fra- grant, scraped in water and used to scent clothing.’’ December 18, 1942, W. A. Archer 7999.—FEstado do Pard, Belém. ““Tree. Root aromatic, sold in local market as perfume for bath.’’ January 11, 1943, W. A. Archer 8146.—¥Estado do Parad, Belém. Entroncamento. ** Arataciti.”’ September 26, 1944, 4. Ducke 1635.—Estado do Ama- zonas, Rio Uaupés. Ipanoré[Panuré]. ‘‘Small bushy tree, 15-20 feet. Flowers white. Caatinga.’’—Kstado do Amazonas, Municipality Sao Paulo de Olivenca, near Palmares. ‘‘Tree 50 feet high; trunk 4 inches diam. Terra firma; high lands.’’ September 11-—October 26, 1936, B. A. Krukoff’s 7th Exped. Brazil. Amazon. 8189.—Estado do Ama- zonas, Rio Negro, Santa Izabel. November 10, 1936, A. Ducke 341. Britis Guiana: River Quitaro. ““Tree. Flowers yellow.’’ 1838. Schomburgk 569.—Tributary of Habu Creek (New River). “‘Creek bank. Moist clay soil. A tree 40 feet high and 8 inches in diameter. (Wood preserved separately.) Small flowers in spray, both red and white on the same spray. (The white had turned to red by the time the specimens reached camp: 6 hours).’? November 10, 1937, Bed- dington 31.—Basin of Kuyuivini River (Essequibo tributary), about 150 miles from mouth; dense forest. ““T'ree 20 m. high.’? November 21-26, 1937, A.C. Smith 2539.—Northern slope of Akarai Moun- tains, in drainage of Shodikar Creek (Essequibo tributary); dense forest. Alt. 300-600 m. ““Tree 35 m. high.’? January 10-20, 1938, A.C. Smith 2905.—Moraballi Creek, River Essequibo. “*30/ under- growth tree in mixed forest, rather ropy on rock slope, bark grey, smooth; slash similar to Sandwithia guyanensis. % flowers paired on long stalks, calyx green, glabrous; corolla white, stamens °° white; ® flowers green perianth, leaf-like ovary trigonal; stigma 6-partite, rayed.’’ October 14, 1939, Field No. (Forest Dept. Brit. Guian.) F, 273, Record No. 3009. Durcu Gutana: No precise locality, no date. Hostmann 1156.— No precise locality. ‘‘Cacaoballi,’? May 1945, Zanderij 311. Frencn Gurana: No precise locality, no date. Martin 27.—No precise locality. 1840, H. F. Talbot s.n.—‘‘Environs de Godebert.’’ December 20, 1920, Wachenheim s.n. [71 ] Venezurta: Territorio del Amazonas. “‘In ripis fl. Pacimoni. Arbor subramosa, tenuis, 15-20 pedalis. Monoica. Flores % albi, odorati. Fls. 9 virides.’’ February 1854, R. Spruce 3342. Senefeldera contracta R. HE. Schultes sp. nov. Arbor usque ad sexaginta pedes alta, in sylvis crescens. Truncus robustus. Ramuli glabri ut videtur, cortice rufo obtecti. Folia valde coriacea, elliptica, petiolata (petio- lis robustis, 3.5-8 cm. longis, usque ad 0.5 em. in dia- metro), apice obtusa vel obscure acuta, basi rotundata, obscure biglandulosa, leviter marginata, glabra, sicco stramineo-fusca, nervo centrali robusto subtus valde ele- vato supra non conspicuo, nervis secundariis quattuor- vel quindecim, 28-81 cm. longa, 9.5-12.5 cm. lata. In- florescentiae stamineae erectae, subpaniculatim ramosae, 19-27 cm. longae, axibus centralibus omnino glabris, bracteis ovato-ellipticis, magnis, 10 mm. longis. Flores flavi, tres staminiferi in axilla bracteolae parvae, atroviri- dis, suborbicularis vel subtriangularis, irregularis cum margine eroso-hyalino, circiter 1.7 mm. longae, 2 mm. latae (sed bracteolae inter se multo variabiles), perianthio 0.5 mm. longo, obtuse trilobato, glabro, staminibus quin- decim minoribus, filamentis 0.1—0.2 mm. longis, liberis, antheris flavis, subglobosis, 0.25 mm. altis, 0.80 mm. latis. Fructus adhuc ignotus. Senefeldera contracta may be distinguished from other species by the very large size of the leaves, their coria- ceousness and high lustre on both surfaces; by the un- usually stout petioles; by the large number of stamens; and especially by the extremely contracted (not loose and much-branched) inflorescence. In the size of the leaves, Senefeldera contracta resembles 8. macrophylla Ducke, but the latter differs obviously from the former in the shape and dullness of the leaves, in the more slender pet- ioles and in having a long, lax and much-branched inflo- rescence. Like Senefeldera contracta, S. nitida Croizat is [ 72 ] very lustrous, but its leaves are very much smaller and its inflorescence is very dissimilar. The flowers of Sene- feldera nitida are unknown, so it is not possible to eval- uate the relationship of this species with S. contracta; it is not unlikely that these two are rather closely allied. Senefeldera contracta, with fifteen stamens, is approached by S. dodecandra with twelve; but a number of vegeta- tive and floral characters indicate that no relationship exists between the two. CotomsiA: Comisaria del Amazonas, entre Leticia y El Marco. ‘Virgin Forest. Tree 30 meters; trunk stout; flowers small.’? Au- gust 20, 1946, George A. Black & Richard Evans Schultes 46-36 (Typr in U.S. Nat. Herb. ). Senefeldera inclinata Mueller-Argoviensis ex Mar- tius Fl. Bras. 11, pt. 2 (1874) 530. The collection cited below appears vegetatively to match rather closely the type of Senefeldera inclinata which is a fruiting specimen from the region of the Casi- quiare. This species has apparently not hitherto been reported as an element of the flora of Colombia. Cotomsta : Comisaria del Amazonas, Rio Amazonas, Leticia. ‘‘Bush, 34 m.; Flowers greenish-white.’? August 16, 1946, George A. Black & Richard Evans Schultes 46-4. SAPINDACEAE Paullinia Yoco R. Ll. Schultes & Killip ex Schultes in Bot. Mus. Leafl. Harvard Univ. 10 (1942) 302, t. 27. In 1942, world conditions prevented my consulting the specimens of yoco which were collected in Colombia in 1923 by the Belgian botanist Florent Claes. Mention was made (Schultes |.c. 309) of this material and the fact that De Wildeman (in Compt. Rend. 188 (1926) 1850) had identified yoco as Paullinia searlatina Radl. from this material. When I visited Belgium in 1950, I was fortunate in being able to examine Claes’ material, preserved in the [ 73 | herbarium in Brussels. Claes made several collections of the plant from which the Indians prepare yoco. Of these, one (Claes 30) is in excellent flower, one (Claes 24) has partly disintegrated remains of flowers, and two are ster- ile. There is an appreciable variation in leaf size, but all of the collections may be referred without hesitation to Paullinia Yoco. The flowering specimen (Claes 30) has the smallest leaves and is the specimen which De Wilde- man has annotated as representing Pau/linia scarlatina. It has the same characteristic short, stiff inflorescence as the type. Claes 24 has much longer inflorescences, but the same woody tendrils as the type. The other speci- mens had no determinations, but I have annotated all the material as Paullinia Yoco. There apparently is no information on the Claes speci- mens relative to the different ‘‘kinds of yoco’’ to which Zerda Bayon, Klug and Schultes have referred (cf. Schultes l.c. 811). Through the kindness of Professor W. Robyns, direc- tor of the Brussels Herbarium, I was able to obtain an excellent biographical article on the life and work of the late Florent Claes. Claes, it would seem, carried out rather extensive explorations for plant introduction. The vast amount of work which he did _ has certainly not re- ceived merited notice. For this reason, I am calling at- tention to the article which appeared in a publication not widely circulated, at least in the New World: L. Pyn- aert: ‘‘Florent Claes: Botanist belge, explorateur,’”’ Publication de botanique d’ Agrément, Bulletin mensual de documentation de botanique horticole et d’horticul- ture pratique (Régions tempérées et tropicales) (1987) Bruxelles. Pynaert’s article reviews in some detail much of the work which Claes accomplished during his several trips to Colombia and considers the discovery of yoco at length (lc. 44-46). [ 74 ] Cotomara: [Comisaria del Caqueté, Rio Orteguaza], 1923, Claes 12, 23, 24, 30. BOMBACACEAE Bombax sordidum PR. FE. Schultes sp. nov. Arbuscula parva, usque ad tres pedes alta, a Bombace coriaceo foliis coriaceoribus, majoribus, margine magno- pere revolutis, infra densissime et sordide pulveraceis cum pilis fulvis et fructibus aliquid majoribus differt. Flores adhue ignoti. The collection Schultes & Lopez 9342, first identified as Bombax coriaceum Mart. & Zucc., but shown by later studies to be distinct, represents one of the most curious species of the genus. Most closely allied to Bombawz cori- aceum, Which it resembles in its diminutive bushy habit, B. sordidum can be distinguished by its larger and much more coriaceous and more marginate leaves, which are densely clothed beneath with a very dark brown, dust- like indumentum, and by its larger capsules. The distribution of Bombax coriaceum as given by Schultes (in Bot. Mus. Leafl. Harvard Univ. 13 (1949) 303) must be slightly modified to accord with this new disposition of the collection cited below. Bombaz coria- ceum is, so far as we now know, confined to cretaceous quartzitic areas in Amazonian Colombia and apparently does not occur, as does B. sordidum, on the proterozoic granitic shield. The two species are, however, very closely allied. Cotomspia: Comisaria del Vaupés, Rio Negro, El Castillo or San Felipe, near confluence of Rios Guainia and Casiquiare., Caatinga. ““Low shrub, 3 feet tall. Leaves very coriaceous.’? December 12, 1947, Richard Evans Schultes & Francisco Lépez 9342. STERCULIACEAE Herrania albiflora Gowdot in Ann. Sci. Nat., sér. 3, 2 (1844) 230, t. 5, figs. 1-10. [75 | In June 1950, when I visited the Jardin Botdnico de Madrid, it was my exceedingly great pleasure to see the famous Mutis water-colors of Colombian plants, made by the artists of the Expedicién Botanica which investi- gated the flora of Colombia under the direction of Celes- tino Mutis from 1788 to 1808. Amongst the six thousand plates, there are several extraordinarily artistic and accurate illustrations repre- senting species of Herrania. These plates, labelled sim- ply ‘‘“Theobroma’’ by the artists, are all included in Vol- ume No. 28 under number 5333. They were determined as the several concepts of Herrania by Triana who an- notated each plate in his own hand in pencil and signed his annotations. It is of great interest to recall that these plates were made about half a century before the genus Herrania, and the three species so beautifully repre- sented by the Mutis plates, were described by Goudot in 1844. One of the plates represents a fruiting and flowering branch of Herrania albiflora. Not only are ripe and un- ripe fruits shown in excellent detail, but a large number of flowers are depicted so painstakingly that it is clear that Mutis was able, so long ago, to differentiate between the patelliform calyx of Herrania albiflora and the sub- cymbiform calyx of the other species illustrated. No foli- age is drawn on the plate of Herrania albiflora. Triana correctly annotated this plate as ‘‘Herrania albiflora Goudot.”’ In the Mutis collection of plants in Madrid, there is asterile collection of leaves, misidentified as ‘‘ Theobroma Mariae,”’’ which also represent Herrania albiflora. They undoubtedly belong to the plant the flowers and fruits of which are portrayed on the Mutis plate of Herrania albiflora. Cotomsia: Exped. Bot. Mutisii Novae-Granat. (1788-1808) 3759. [ 76 ] We have always presumed that the type of Herrania albiflora, the type-species of the genus, was preserved in the herbarium at Paris. In June 1950, however, I found Goudot material at Geneva which may very well be the type of Herrania albiflora. In Geneva, there are three sheets representing the Goudot collection of this species, all labelled, in Goudot’s handwriting, ‘‘C. N. 1 Herrania albiflora mihi. Annales Sc. nat. 1844. Muzo.’’ One sheet has several very young and membranaceous leaves and an envelope in which there are fragments of a fruit belonging possibly to an annonaceous plant and which, by some error, have been associated with the Herrania collection. Another sheet has acomplete and mature leaf. he third sheet has three envelopes: one contains several seeds of Herrania alli- Jlora; another has a few flowers and a very young cap- sule; the third has a flower completely dissected with the parts glued flat to the envelope. An examination of these floral parts and of Goudot’s description and drawing of Herrania albiflora lead me to the conclusion that, at least for the flowers and fruit, the Geneva material is the type of the species and genus. How has it been possible for Goudot type material to find its way to Geneva? There are, of course, many Goudot collections in the Delessert Herbarium (cf. A. Laségne: ‘‘Musée Botanique de M. Benjamin Deles- sert”’ (1845) 471). It may be of interest to note that acomparison of the Goudot floral dissection with the description of Herrania albiflora has uncovered several minor discrepancies or omissions. The sepals, described as glabrous within, have a very minute and sparse puberulence on the lower portion of the inner surface; and the petals are extremely muri- cate-granulose externally, as are also the very short ligules in the basal portion near their junction with the petal. eve Cotoms1a: Departamento de Cundinamarca, Muzo, 1825, J. Goudot sine num. Herrania breviligulata R. EH. Schultes in Caldasia (1942) 19. Herrania breviligulata was described from flowering material collected in the Putumayo of Colombia in 1941. Hitherto, no additional material has been available for study. In July 1950, I found the collection Mewia 7328 in the Riksmuseet in Stockholm. Although, unfortu- nately, it is nearly sterile (having only a few loose seeds), the collection can be referred, on the basis of very speci- fic vegetative characters, to Herrania breviligulata. The Provincia de Napo-Pastaza is adjacent to the Colombian Comisaria del Putumayo, but the known distribution of the species is extended to include an additional drainage- area. Ecuapor: Provincia de Napo-Pastaza, near Archidona. Alt. 650 m. “*Dense forest. Erect shrub, 2-5 m. high. Fruit green, deeply ribbed. ”’ No date. Ynes Mexia 7328. Herrania kanukuensis P. F. Schultes in Caldasia 2 (1943) 11; in Bot. Mus. Leafl. Harvard Univ. 18 (1949) 227; ibid 14 (1950) 126, t. XX XIII. Theobroma Mariae (Mart.) Decaisne ex Goudot var. lobata Pulle in Rec. Trav. Bot. Neérl. 9 (1912) 151. In 1948, I described Herrania kanukuensis—extraor- dinarily distinct in the form of its fruit—from a collec- tion from southern British Guiana. Recently, a sterile collection referable to this species was reported from an adjacent part of the Territorio do Rio Branco in Ama- zonian Brazil (in Bot. Mus. Leafl. Harvard Univ. 14 (1950) 126). In May 1950, while engaged in taxonomic studies in the Botanisch Museum en Laboratorium at the University of Utrecht, Holland, a large series of collec- tions representing this species was kindly made available to me for examination. [ 78 ] In 1932, Uittien (in Pulle: FI. Surin. 8 (1982) 44) re- duced Theobroma Marae var. lobata to synonymy under Th. Mariae, identifying all of the then available material from Dutch Guiana as representing this Amazonian con- cept. In 1943, it appeared to me that the Surinam 7Theo- broma Mariae var. lobata represented the concept de- scribed by Schomburgk from nearby British Guiana as Lightia lemniscata, and I placed it in synonymy under Herrania lemniscata (Schultes in Caldasia 2 (1948) 18), a species with remarkably lobate leaflets. During the war, the Utrecht material was unavailable, but my recent studies have convinced me that Theobroma Mariae var. lobata and Herrania kanukuensis represent the same concept. From the numerous collections, for the most part from Surinam, it is now obvious that Herrania kanukuensis is both a widespread and, at least locally, an abundant element of the flora of Surinam, eastern British Guiana and the adjacent rim of northern Brazil. Dutcn Gutana: Am oberen Corantijne-fluss. December 27, 1910, J. F. Hulk 26 (Tyee of Theobroma Mariae var. lobata in Herb. Utrecht); River Coppename, ‘‘Maripoele kakaoeleo. Herrania. Vrucht eetbaar. Eerste doorsteek,’’ November 6, 1916, J. W. Gonggrijp 2565; River Corantijn, Kaurikreek. ‘““Onvertaktboompje 3-5 m. hoog met arm- dikken stam en cauliflore vruchten,’’ June 20, 1916, J. W. Gonggrijp 2111; River Corantijn, Kaurikreek, ““Maipoelie doron doron (Kar.). Boompje 5 m. hoog onvertakt Blm. donker purperrood.’’ November 22,1916, Stahel & Gonggrijp 3015; Placer L’Aiva, October 31, 1918, J. W. Gonggrijp 4126; Flum. Marowijne, November 20, 1918, J. W. Gonggryp 4101; River Tapanahonie, Jaikreek. November 17, 1918, J. W. Gonggriyp 4117; River Marowijne, hill forest near Amerikan Kondre. “‘Boompje; groen op stam. Boscacao.’’ February 21, 1949, J. Lanjouw & J. C. Lindeman 2304. Herrania laciniifolia Goudot ex Triana & Planchon Fl. Novo-Granat. (1862) 209, nomen subnudum; Garcia- Barriga in Caldasia 1, No. 2 (1941) 55, t. 1, 4. In the collection of Mutis plates preserved in Madrid, [ 79 | there are several illustrations representing Herrania laciniufolia. These are all in black and white, not in color. One plate has a leaf with one complete leaflet and a length of stem with several flowers and buds; another plate has analytical drawings of the flowers and fruits; a third has analyses only of the floral parts. The Mutis specimen in Madrid (cited below) is sterile, but it is undoubtedly from the tree from which the plate was made. In view of the scarcity of collections of Her- rania lacinifolia, it is unfortunate that a definite locality for the Mutis collection is not available. Both the speci- men and the plates agree perfectly with the type and later material of this remarkable species. Cotompia: Exped. Bot. Mutisii Novae-Granat. (1783-1808) 937. Herrania nitida (Poepp.) R. EF. Schultes in Caldasia 2 (1943) 16, t. p. 17. Herrania atrorubens Huber in Bull. Soc. Bot. Genéve, sér. 2, 6 (1914) 187. Early in my study of the genus Herrania, I published (in Caldasia 2 (1944) 829) a note pointing out that with- out an examination of the type of Huber’s HZ. atroru- bens, it was not possible to evaluate the validity of the concept and that ‘‘the colour character alone would hardly suffice for the creation of a new specific concept. ”’ Now, having completed an extensive study of the clas- sical material of the genus, I have been unable to locate the type of Herrania atrorubens. One would expect it to be preserved in the Museu Goeldi in Belém do Para or in the Herbier Boissier in Geneva, but a search in these two institutions, as well as in other Brazilian and Kuropean herbaria, has not uncovered Huber’s material. From an evaluation of the meager characters given by Huber and from the geographical data given for the type collection, Herrania atrorubens may, it would seem, in [ 80 ] the light of experience gained during the study of a wide range of material, safely be referred to H. nitida. Herrania pulcherrima Goudot in Ann. Sci. Nat. sér. 8, 2 (1844) 282, t. 5, figs. 11-12. In the collection of Mutis plates, there is a most strik- ingly beautiful and accurate water-color of a section of a stem of Herrania pulcherrima in full flower. A number of diagnostic characters of Herrania pulcherrima are most clearly shown: the congested, many-flowered inflores- cences; the very abbreviated pedicels; and the long and membranaceous ligules with alternate scarlet and whitish bands. Of this colored plate, there are two copies in black and white. No foliage seems to have been drawn. A search in the Mutis collection of plants in Madrid has failed to produce aspecimen of Herrania pulcherrima. Herrania purpurea (Pitt.) R. 1. Schultes in Cald- asia 2 (1944) 3338. Further collections of Herrania purpurea in the Uraba area of Antioquia emphasize the great abundance of this Middle American species in the northwesternmost corner of the South American continent. CotomsiaA: Departamento de Antioquia, Municipio de Pavarando- cito, Villa Arteaga. *‘Small tree 8 feet tall.’’ December 1948, Richard Evans Schultes & Francisco Lopez 10464.—Departamento de Antioquia, Municipio de Pavarandocito, Pavarandocito and vicinity. October 1950, Gabriel Gutiérrez V. 2000. DILLENIACEAE Saurauia pruinosa R. FE. Schultes sp. nov. Arbor frondosa, usque ad viginti quinque vel triginta pedes alta. Ramuli teretes, brunneo-cinerei, maxime densissime et grossiuscule echinato-setosi (setis ipsis albido-scrobiculatis) atque inter setas densissime pilis minutis roseo-niveis aloeformibus vestiti. Folia valde coriacea, siccitate fragilia, 22-31 cm. longa, 6.5-12 cm. [ 81 ] lata, elliptica, apice longe acuminata, basi rotundata, margine minute et regulariter denticulata, supra atro- viridia et densiore praecipue venas versus aspero-setosa, subtus densissime et molliter praecipue prope venas cinereo-echinato-setulosa et maxime densissime lanato- tomentosa cum pilis roseo-niveis aloeformibus vel dendri- formibus. Petioli robusti, teretes, 2-8.5 cm. longi, 4mm. in diametro, ramulorum indumento vestiti. Inflorescen- tiae magnae, foliis subaequales, usque ad 25 cm. longae, 15 cm. latae, ramulorum indumento sed colore cinereo- roseo omnino obtectae; rhachide centrali robustissima, 5-6 mm. in diametro. Bracteae subulatae, usque ad 1 em. longae, intus brunneo-setulosae, extus rhachidis in- dumento vestitae. Alabastra globosa, usque ad 1 cm. in diametro, roseo-tomentulosa. Flores numerosiores, max- imi (usque ad 3.8 cm. in diametro) et pulcherrimi, valde aromatici, fragrantes. Sepala quinque, rosea et alba, in- aequalia, chartacea, ovato-elliptica, apice obtusiuscula; tria majora 10-11 mm. longa, 8-9 mm. lata, extus roseo- albida minute lanato-tomentulosa cum pilis aloeformibus et cum carina centrali setarum roseo-brunnearum armata; duo minora usque ad 10 mm. longa, 7-8 mm. lata, extus omnino setis roseo-brunneis (setis ipsis cum appendicibus setuliformibus dense vestitis); omnia intus magnopere dense tomentulosa pilis albis lanato-aloeformibus, basim versus vulgo glabra sed prope insertionem cum pilis aureis vestita. Petala quinque, alba, aequalia, membranacea, rotundato-obovata, integra vel saepe apice aliquid incisa, utrinque glabra. Stamina plusminusve centum sexaginta, filamentis debilibus, 4 mm. longis, antheris flavis, 1.5 mm. longis. Ovarium perfecte globosum, 5-6 mm. in diametro, glabrum, quinque (rarenter sex) cum stylibus carnosis, usque ad 6 mm. longis, stigmate capitato. Although Sawrauia pruinosa resembles S. brachybotrys Turcz. and S. Spraguiana Busc. in some respects, it can [ 82 ] be distinguished at once from these and all other species by its extraordinary number of stamens. Saurauia brachybotrys has about sixty and S. Spraguiana from sixty to eighty stamens; Saurauia pruinosa, with one hundred and sixty stamens, has the highest number among the American species of the genus. Saurauia pruinosa may also be distinguished from most other American species by the peculiar hoary indumentum which has a beautiful rose-pink bloom. Saurauia pruinosa appears to be most closely allied to S. roseotincta R. E. Schultes of Perti. The latter species, like the former, has a very high number (130) of stamens and has a beautiful pink-pruinose indumentum on the under surface of the leaves. It differs in number of sta- mens; in having an inflorescence subequal to (not shorter than) the leaves; in having flowers only half as large; and in several floral characters. Cotomsia: Comisaria del Putumayo, Valley of Sibundoy, Sibundoy. Altitude about 2225-2300 m. Moquillo. Flowers excessively narcissus- fragrant. Petals white; hairs on sepals pink. Leaf hairs pink, espe- cially on young leaves. Leaf backs and young branches a tawny ash- brown, giving peculiar characteristic appearance at a distance — the two colours of hair. Fruit ripens red. Flower buds large.’’ May 29, 1946, Richard Evans Schultes & Mardoqueo Villarreal 7651 (Type in Herb. Gray). Saurauia roseotincta 2. l. Schultes sp. nov. Arbor gracilis, quindecim ad viginti pedes alta. Folia subcoriacea, elliptica, margine minutissime denticulata, apice subacuta, basi cuneata, 21-34 cm. longa, 7-11 cm. lata, supra bulbata, nervos versus et sparsiore in lamina aspero-setosa, subtus tactu mollia, omnino minutissime albo-stellato-pilosa atque in nervis omnibus densius roseo- stellato-pilosa; petiolus robustus, usque ad 8.5 cm. lon- gus, 4-5 mm. in diametro, grossiuscule rufo-ferrugineo- setosus. Inflorescentiae quam folia breviores, usque ad 20-22 cm. longae, plusminusve viginti-florae, partibus [ 83 ] omnibus rufo-ferrugineo-setulosae. Bracteae setoso-pilo- sae, subulatae, usque ad 5 mm. longae. Flores non nu- merosi, 13-14 mm. in diametro. Sepala quinque; interi- ora rhomboideo-ovata, apice obtusa, extus setarum carina mediana armata, intus prope apicem dense pulverulenta, basi glabra, 9-10 mm. longa, 9-10 mm. lata; exteriora crassiora, elliptico-ovata, apice subacuta, extus densissime et grossiuscule aureo-setosa, intus pulverulenta sed basi glabra, 9-10 mm. longa, 5-6 mm. lata. Petala mem- branacea, alba, rotundata, margine subintegra (leviter subundulata ut videtur), 6-7 mm. longa, 5-6 mm. lata. Stamina plusminusve centum triginta, antheris parvis, 0.7 mm. longis, filamentis 23 mm. longis, basi rufo-setoso- barbatis. Ovarium globosum, 4 mm. in diametro, glab- rum, quinque cum stylibus usque ad 1.5 mm. longis. Saurauia roseotincta very closely resembles S. pseudo- ruiziana Buse., 8. Ruiziana Steud., and S. tomentosa Spreng., from which concepts it may be immediately dis- tinguished by having 130-140 stamens instead of 80, 30-40 and 40-70 respectively. There are also differences to be noted in the pilosity of the leaves and in the size of the floral parts. There are a number of notes appended to the type specimen. These were made by Dr. E. P. Killip when he compared the specimen with material at Kew and in Berlin. Killip wrote: ‘‘ Aff. tomentosa, but hairs yellow, not white. Aff. Ruizana [sic], but that has long soft, appressed hairs above, and on nerves beneath; flowers very similar. [Notes at Berlin. ] Aff. pseudruizana [sic] but leaves here are more bullate and without long ap- pressed hairs of pseuoruizana. [Notes at Kew. ] This agrees well with description of Ruizana var. Weberbauert Buse. in Monagr.; perhaps, though, it is a distinct spe- cies. (Oct. 1925).”’ According to the collector’s data, the pubescence of [ 84 ] the under surface of the leaves was pink-red in life. In the dried specimen, it has changed to a rich golden rust- color. It is probable that the finest of the stellate hairs (that is, those on the blade itself) were, in life, whitish as they are in the dried material and that the coarser stellate hairs along the primary, secondary and tertiary nerves (now rust-colored, but in places still pinkish) were responsible for the general hue of redness. The specific epithet refers to this beautiful character of the under surface of the leaf. Peru: Chaglla. Alt. about 9,000 ft. ‘‘Slender, open, 15-20 ft. tree; leafy only above. Pubescence lower sides leaves pink-red; flow- ers white.’’ May 12, 1923, J. Francis Macbride 3652 (Tyrer in U.S. Nat. Herb. ). OcHNACEAE Leitgebia colombiana R. L. Schultes sp. nov. Fruticulus usque ad 4-pedalis, parce fastigiato-ramosus. Caules nigri, basi 8-10 mm. in diametro, praeter apicem denudati, foliorum delapsorum cicatricibus 1.8-2.2 mm. distantibus asperati, superne stipulis elongato-triangu- Jaribus, 3 mm. longis, 1 mm. latis, pectinato-fimbriatis, extus striatis et subcarinatis, sublignosis, persistentibus, nigrescentibus et irregulariter ferrugineo-setulosis, erec- to-amplectentibus. Folia caulem prorsus velantia, cori- acea, lanceolato-elliptica, plerumque 10 mm. longa, 3-3.5 mm. lata, apice obtusa sine mucrone glanduloso, basi attenuato-cuneata, margine callosa et valde revoluta, re- motissime et obscure subundulato-denticulata (utroque margine cum sex vel septem denticulis glandulosis), supra lucida, nervo medio et sex ad novem nervis secundariis densis obliquis prominentioribus, subtus pallidiora min- usque lucida, praeter nervum medium prominulum ap- parenter avenia. Flores solitarii, inter folia summa sub- immersi, 5-6 mm. in diametro, bibracteolati. Pedicellus 1 mm. longus. Bracteolae oppositae, subulatae, margine [ 85 | omnino irregulariter fimbriolatae, circiter 8 mm. longae sed saepe leviter inaequales. Sepala quinque hyalino- membranacea, omnino glabra, ovato-lanceolata, 8 mm. longa, 1 mm. lata, apice longe acutissima, margine in- tegra, valde concava. Petala quinque, tenuiter membra- nacea, ovata, circiter 8 mm. longa, basi 1.5 mm. et apice 1 mm. lata, apice rotundata, margine integra, nervis numerosis percursa, basi alba, apicem versus violacea. Staminoidea quinque, petaloidea, membranacea, concavo- oblonga (non spathulata), apice rotundata, circiter 8 mm. longa, 1.5 mm. lata, rosea, nervo medio prominentiore. Stamina quinque, antheris linearibus, flavis, usque ad 1.8 mm. longis, ad basim staminodei filamento brevissimo coalita. Ovarium subtriangulari-ovoideum, 0.8 mm. lon- gum, stylo filiformi 8 mm. longo, stigmate obtuso coro- natum. Fructus adhuc ignotus. Leitgebia, a genus hitherto unknown from Colombia, is characteristic of the isolated mountains of the Vene- zuelan-Guianan land-mass. The type and, until very recently, the only known species was Leitgebia guianensis Eichler, from Mount Roraima and from Mount Duida. Oliver described a concept from British Guiana as Leitgebia Imthurniana. Later, Gleason made this the type of anew genus: PRo- raimanthus. Recently it has been placed in the genus Sauvagesia by Dwyer. In 1945, Lasser (in Bol. Acad. Ciéne. Fis. Matem. Nat. Caracas (1945) 246, fig. p. 247), described Leitgebia Gleasoniana from Mount Duida and Mount Paraque in southern Venezuela; this species is at once distinguished from Leitgebia guianensis by its leaves which, measuring 18-20 mm. by 2 mm., are twice as long as those of the earlier species. Leitgebia colombiana appears to approach most closely to L. guianensis, from which it can be distinguished at [ 86 ] once by having the leaves much less dense and much less closely appressed around the stems; where the leaves have fallen, the scars are less conspicuous and are more distinctly placed on the stem. The former has the stip- ules much less conspicuous and spreading than the latter ; and it has flowers which are only half as large; slightly larger leaves which are more nearly lanceolate-elliptic (not oblanceolate) and remotely and obscurely subundu- late-denticulate along the entire margin (instead of closely and definitely sharp-dentate along the upper half of the margin only); petals which are apparently non- deciduous, ovate or broader at the base than at the apex (instead of conspicuously obovate); and sepals which are entire (not serrulate) near the apex. The fruit of none of the three species of Leitgebia is known. Leitgebiacolombiana flowers on Mount Chiribi- quete in May and sets fruit probably from October to December. I collected at this locality in May, July, and January and was unable to find fruiting material of this curious shrub. Cotomsria: Comisaria del Vaupés, Macaya-Ajaju River confluence, Mount Chiribiquete. Quartzite base. Summit 800-1200 feet above forest floor; 1700-2100 feet above sea-level. “‘Small shrub. Flowers pinkish.’? May 15-16, 1943, Richard Evans Schultes 5479 (Tyee in Herb. Gray). CoMBRETACEAE Combretum laxum Jacquin Knum. PI]. Carib. 19 (1760) 115. Black & Schultes 46-293 represents the same variant of the widespread and variable Combretum lawum which occurs in adjacent parts of Amazonian Pert and which was recently reported from the Rio Igaraparana in Am- azonian Colombia (Schultes in Bot. Mus. Leafl. Harvard Tniv. 14 (1949) 40). . , , 6 bry . Cotomsria: Comisaria del Amazonas, Rio Loretoyacu. ‘Trepadeira. [ 87 | Flores brancas, cheirosas,’” November 2, 1946, G. A. Black & R. E. Schultes 46-293. Combretum rotundifolium ZL. C. Richard in Act. Soc. Hist. Nat. Paris 1 (1792) 108. One of the most common riverine lianas in Amazonian Colombia, Combretum rotundifolium has not been fre- quently collected there; Schultes 3984 is apparently but the second collection reported in the literature. The first (Schultes in Bot. Mus. Leafl. Harvard Univ. 14 (1950) 135) was from the distant Apaporis River basin. Cotoms1A: Comisaria del Amazonas, Rio Igaraparana, between Que- brada Menaje and Rio Putumayo. Alt. 150-170 m. ‘‘Fruit fiery red. Bush.’’ June 15-17, 1942, Richard Evans Schultes 3984. Combretum Wandurraganum P. I. Schultes sp. nov. Frutex scandens, extensus, robusto cum trunco et ramis glabris laevibus teretibusque et ramulis fusco- lepidotis. Folia chartacea, elliptica vel subovata, apice abrupte acuminata, basi rotundata, integra, statu adulto 7-17 cm. longa, 4-8 cm. lata, supra nitidula, subtus vivo aureo-lepidota, venis secundariis plerumque novem sub- tus aliquid conspicuis. Paniculae axillares, apparenter densiflorae, plerumque usque ad 24 cm. longae, rhachidi- bus robustioribus 4 mm. in diametro, dense fusco-lepi- dotis. Flores adhuc ignoti. Samarae pulchrae, atrosan- guineae, quadrialatae, orbiculares, plerumque 2.2-—2.4 em. latae (alis computatis), plusminusve 2—2.2 cm. altae, longe et graciliter stipitatae (stipitibus 10-12 mm. longis, dense lepidotis), omnino sparsissime lepidotae. Combretum Wandurraganum resembles most closely perhaps C. rotundifolium Rich., from which it can be distinguished by differences in the size and shape of the leaves; in the color and density of the lepidote indu- mentum on the under surface of the leaves; and, most [ 88 ] strikingly, by the peculiar round shape of the long-stipi- tate samara. Cotompra: Comisaria del Amazonas, trapecio amazénico, Loretoyacu River. Alt. about 100 m. ‘*Extensive vine. Fruit red.’’ September— November 1944, Richard Evans Schultes 6070 (Tyrer in Herb. Gray). STYRACACEAE Styrax Tessmannii Perkins in Notizbl. 10 (1928) 459, This species, described from material collected along the Rio Ucuyali in the Departamento de Loreto, Pert, has hitherto not been reported from Colombia. Schultes 7144 has larger leaves than those described from the type, but all other reported characters appear to agree. Cotompra: Comisaria del Amazonas, trapecio amazénico, Amazon River watershed, Loretoyacu River. Alt. about 100 m. ‘“Small tree on highland,’’? March 1946, Richard Evans Schultes 7144. APOCYNACEAE Couma utilis (Wart.) Mueller-Argoviensis in Mar- tius Fl. Bras. 6, pt. 1 (1860) 19, t. 5. The collections cited below are apparently the second and third known from the Comisaria del Vaupés in Co- lombia (J. Monachino in Lloydia 6 (1948) 237, ibid 9 (1946) 801). It would seem that this species is associated only with the granitic ‘‘Brazilian shield’’ portion of east- ern Colombia. It is known also from the Amazon of Brazil (especially the Rio Negro and Rio Madeira basins) and Venezuela. Cotomsra: Comisaria del Vaupés, Rio Negro, Cocui. ‘‘Sorva. Col- umnar tree, 24 feet in diameter, 75 feet tall. In flood-bank. Bark very thick, smooth but checked outside, grey and black, soft, reddish inside. Latex extremely thick, abundant, white, very sticky. Leaves dark green above, pale beneath. Flowers fragrant, pinkish to white.”’ December 26, 1947, Richard Evans Schultes & Francisco Lopez 9467 .— Same locality and date. ““Small columnar tree, 15 ft. diameter, 50 feet tall. Crown wide. Flowers pink, fragrant. Latex very abundant, thick, white. Bark smooth, black-grey.’’ Schultes & Lopez 9485. [ 89 | Malouetia Tamaquarina (4ubl.) A. de Candolle Prodr. 8 (1844) 878. Malouetia Tamaquarina, widespread in Amazonian Brazil and in the Guianas, occurs in heavy densities in the trapecio amazonico of Colombia. It inhabits the low flood-land or rebalsa accompanying Hevea brasiliensis (Willd. ex Juss.) Muell.-Arg. It has an extremely abun- dant white latex which, according to persistent reports, was formerly used as an adulterant of Hevea rubber. The Peruvian name euchara-caspi (‘‘spoon-tree’’) is used by the inhabitants of the trapecio; the name refers to the custom of making spoons and other utensils from the soft, white wood of the treelet. The ripe fruit of Malouetia Tamaquarina forms, in season, a major part of the diet of the pajuz/, a bird na- tive to the region. There is a widely accepted belief in the upper Amazon that the bones of the pajui/, at the time of abundant fruiting of the euchara-caspi, are highly poisonous to dogs who may obtain and eat them. The poisoning agent is held to be the fruit. Such a generally accepted belief is, indeed, worthy of phytochemical in- vestigation. Cotompia : Comisaria del Amazonas, trapecio amazénico, Loretoyacu River. Alt. about 100m. “Small tree. Flowers white. Cuchara-caspi.’’ September—November 1944, Richard Evans Schultes 6083. VERBENACEAE Citharexylum Ulei Moldenke in Fedde Repert. Sp. Nov. 37 (1984) 237. This collection provides the first record of the Brazilian Citharevylum Ulei from Colombia. It is shown, through the collection cited below, to occur right up to the base of the Andes in Colombia. CoLtomBiA: Comisaria del Putumayo, Rio Putumayo, Puerto Asis. March 9-10, 1942, Richard Evans Schultes 3388. [ 90 | SOLANACEAE Datura suaveolens Humboldt & Bonpland in Willdenow Hort. Berol. (1809) 227. This species, native apparently in southeastern Brazil, is widely cultivated in the New World tropics. Amongst the Kofan Indians on the Colombian- Ecuadorian border, an infusion of the leaves is said to be taken occasionally as a narcotic. It is known to the Kofanes as /u-a-vd-u, which, translated, signifies ‘‘pink Datura.’’ The Datura arborea L., which has large white flowers but which the Kofanes apparently do not like to use as a narcotic be- >is called tu-to-a-vad-u, liter- ee cause “‘it is too poisonous, ’ ally ‘‘white Datura. ’’ Ecuapvor: Rio San Miguel o Sucumbios, entre el Rio Putumayo y la Quebrada Teteyé. Alt. 260 m. “‘From Inga Indian garden. A narcotic. Kofin name: ku-a-vd-u,’’ March 29, 1942, Richard Evans Schultes 3472. BIGNONIACEAE Arrabidaea Fanshawei Sandwith in Bull. Torr. Bot. Club 75 (1948) 662. Schultes 5394 represents, according to Sandwith, a form of the British Guianan Arrabidaea Fanshawei ‘with leaflets persistently tomentellous beneath.’’ Its presence in eastern Colombia once more emphasizes the relation- ship of the flora of the upper Apaporis basin with that of the Venezuelan-Guianan area. Cotomsr1A: Comisaria del Vaupés, Rio Macaya, between Rio Apa- poris and Cachivera del Diablo. “‘Extensive vine. Flowers purple, covered with an ashy puberulence (?). Basal diameter of vine 4 inches. Bark rough, brown at base. Leaves pale ashy green beneath, glossy dark green above.’’ May 15-16, 1943, Richard Evans Schultes 5394. Arrabidaea xanthophylla Burret & K. Schumann in Martius F]. Bras. 8, pt. 2 (1896) 70. The yellowish leaves of this vine, which is rather com- mon in the trapecio amazénico of Colombia, are made [ 91 | into an infusion and used by the Tikuna Indians as an eye-wash in the treatment of the severe conjunctivitis which often spreads throughout the area in epidemic form. The Tikunas call the plant Adh-pe-ree. This spe- cies is especially well represented in herbaria from east- ern Pert. Cotoma1a : Comisaria del Amazonas, Rio Loretoyacu (Lago de Soc6). ““Vine. Flowers yellow.’’ November 1945, Richard Evans Schultes 6798. Paragonia pyramidata (Rich.) Burret in Vidensk. Meddel. 1898 (1898) 104. Paragonia pyramidata, rather widespread in tropical America, is a conspicuous element of the flora of the upper Apaporis basin. Cotompia: Comisaria del Vaupés, Rio Macaya, vicinity of Cachivera del Diablo and mouth of river. ‘‘Vine. Flowers purple.’’ May 1943, Richard Evans Schultes 5498. ACANTHACEAE Mendoncia lasiophyta Leonard sp. nov. Suffrutex volubilis, grandis, caulibus subquadrangu- laribus, sursum dense pilosis, pilis fulvis adscendentibus usque ad 4 mm. longis, deorsum glabratis vel strigosis, sursum incurvo-adpressis; lamina foliorum orbicularis vel ovata, usque ad 24 cm. longa et 13 cm. lata, abrupte subacuta, basi rotundata, chartacea, integra vel undulata, recurvata, supra rugosa, glabra vel strigosa, pilis in costa et venis positis, venis (circa 5-paribus) et venulis valde et crasse reticulatis, profunde impressis, subtus pilosa, pilis cinerels, circa 2 mm. longis, patulis, curvatis, costa et venis prominentibus; folia novella utrinque dense fulvo- pilosa; petioli crassi, usque ad 8 cm. longi, plusminusve strigosi, saepe curvati; flores solitarii vel pauci, in cal- ‘aribus axillaribus compressis dispositi; pedicelli crassi, 3.5cm. longi, dense tomentosi, pilis fulvis usque ad 5 mm. [ 92 ] longis; bracteae lanceolatae, 3.5—4 em. longae, 1.5 em. latae, leviter faleatae, dense tomentosae, pilis fulvis usque ad 5 mm. longis; calyx glaber vel parce pilosus; corolla 4.5 cm. longa, rubra, glabra, minute papillosa, tubo cylindrico, basi’7 mm. lato, usque ad 15 mm. supra basim 3.5 mm. lato, prope oram 6 mm. lato, lobis suborbicu- laribus, 4 mm. longis, 3 mm. latis, emarginatis; stamina inclusa, ad medium tubi corallae affixa, antheris sagitta- tis, 15 mm. longis, basi 2.5 mm. latis, apice acutis, lobis apice barbatis; stylus 8.5 em. longus, glaber; stigma bilobatum, lobis aequalibus, cupuliformibus; ovarium glabrum. In connection with his description of this new species, Leonard writes: The leaf blades of Ducke’s material are more ovate and less rounded at the base than are those of the type. The type plant, Schultes informs us, is a rampant liana clambering over grasses or low shrubs. Only one plant was observed. The specific epithet is derived from Adsios, shaggy, and #vrov, plant, and alludes to the densely fulvous-tomentose inflorescence, young leaves, and stem tips. This new species resembles and is probably closely related to Men- doncia gigas Lindau. The size and shape of the bracts and the nature of the pubescence covering them and the peduncles are much the same for the two species. The leaf blades, however, of Mendoncia gigas are generally smaller, not exceeding 13 cm. in length and 7 em. in width, and no mention is made of the suborbicular type of leaf sometimes subtending the inflorescence. Again the calyx of Lindau’s species is described as puberulous, whereas in Mendoncia lasiophyta it varies from glabrous to sparingly long-pilose. The corolla lobes of Mendoncia gigas are larger, 6 to 7 mm. in diameter instead of 4 mm. long and 3 mm, wide, and their tips are rounded instead of emarginate. Finally, the ovary, pronouncedly puberulous in Mendoncia gigas, is definitely gla- brous in M. lasiophyta. A photograph of Ule 9800, the type of Mendoncia gigas, is in the U.S. National Herbarium, The type was collected at Seringal Auris- tella, along the Rio Acre in Perit. Brazit: Estado do Amazonas, Esperane¢a, at mouth of Rio Javari, October 18, 1945, A. Ducke 1851. Cotomsia: Comisaria del Amazonas, vicinity of Leticia, Rio Ama- [ 93 ] zonas. November 1948, Richard Evans Schultes & Francisco Lopez 10400K (Tyre in U.S. Nat. Herb.). Sanchezia thinophila Leonard sp. nov. Frutex, caulibus subquadrangularibus, hirtellis, pilis usque ad 1 mm. longis, patulis vel adscendentibus; lam- ina foliorum elongato-lanceolata, acuminata (apice ipso obtuso), basi angustata, in petiolum decurrens, subcoria- cea, marginibus leviter crenatis, supra glabra, minute aspera, cystolithis obscuris, costa et venis lateralibus (15- 20 paribus) utrinque prominentibus, subtus hirsuta, cum pilis brunneo-gilvis, patulis vel adscendentibus, rectis vel leviter curvatis, usque ad 1.5 mm. longis; petioli circa 2 em. longi, canales glabri, subtus hirtelli; paniculae ter- minales, parce ramosae, 22 cm. longae, rhachi et pedun- culo (5.5 em. longo) hirtello; bracteae rubrae, ovatae vel lanceolatae (apice ipso obtuso) pari infimo 7 em. longo, 4 cm. lato, aliis deinceps minoribus, glabris vel basi hir- tellis, ciliatis, fasciculi multiflori (floribus 10 vel pluribus) ; bracteolae ovatae, maximae et infimae 8 cm. longae, 1.8 em. latae, apice rotundatae, glabrae vel sursum hirtellae; calycis segmenta subligulata, usque ad 2.5 cm. longa, sursum 4-7 mm. lata, obtusa vel rotundata, basi angus- tata, glabra vel apice pubescentia, pilis circa 0.25 mm. longis; corolla 4 cm. longa, flava, deorsum glabra, sur- sum hirsuta, pilis retrorse adpressis, albis, circa 0.5 mm. longis, tubo cylindrico, basi 8 mm. lato, prope oram 7 mm. lato, lobis ovalibus, circa 8 mm. longis et latis, emarginatis ; stamina cirea 4cm. longa, filamentis planis, parce pilosis cum pilis patulis, usque ad 2 mm. longis, antheris 6 mm. longis, 2.5 mm. latis, dorso glabro, ventre puberulo, calcaribus basalibus 0.75 mm. longis; ovarium glabrum; stigma bilobatum, lobis inaequalibus. Leonard states further: Schultes 6937, collected in the vicinity from whence the type of Sanchezia thinophila was taken, has the same general appearance, but [ 94 ] differs in being essentially glabrous. The specimen is very immature, the corollas being barely formed. Until mature collections are avail- able, it cannot be determined definitely whether Schultes 6937 is only a glabrous form of Sanchezia thinophila or a new species. The type plant, Schultes states, was growing on a sandy beach at the mouth of the Rio Loretoyacu, hence the specific epithet thinophila (Gs [@mo-], beach, and $Aéw, loving). Cotoms1A: Comisaria del Amazonas, bank of Loretoyacu River, 100 meters altitude. October 20-30, 1945, Richard Evans Schultes 6607 (Type in U.S. Nat. Herb.). CoMPOSITAE Gongylolepis maroana Badillo in Bot. Soc. Venez. Ciéne. Nat. 8 (1943) 237. Recently, a number of collections of this curious composite have been reported from eastern Colombia (Schultes in Bot. Mus. Leafl. Harvard Univ. 18 (1949) 310, 14 (1949) 47). The two additional collections from the Comisaria del Vaupés indicate that Gongylolepis maroana is, indeed, one of the most widespread as well as dominant shrubs on the isolated quartzite areas in Amazonian Colombia. Cotomspia: Comisaria del Vaupés, Macaya River, Mount Chiribi- quete. ““Bush’’. July 24, 1943, Richard Evans Schultes 5742.—Comis- aria del Vaupés, Rio Vaupés, Cafio Pact. Sandy savanna, ‘‘Bush; 15 feet.”’ March 6, 1944, Richard Evans Schultes 5816. [ 95 ] EXPLANATION OF THE ILLUSTRATION Pirate VII. Cartupovica aurantiaca R. EF. Schultes. 1, habit of plant, about one tenth natural size. 2, early stage of inflorescence, showing spadix with staminodes still wrapped in translucent spathes. One spathe is detached at base to show spadix and coiled mass of staminodes within. About one half natural size. 3, female flower from above; areas shaded with diagonal lines indicate scars left by the caducous staminodes. About ten times natural size. 4, male flower, lateral aspect. About ten times natural size. 5, spadix with young fruits. About three times natural size. 6, leaf, about one third natural size. Drawn by Ermer W. Situ Rf /| i/ l/ | H/ , Wa Vat CARLUDOVICA aurantiaca RL Schultes i EXPLANATION OF THE ILLUSTRATION Prare VITE. Cartupovica AuRANTIACA. A photo- graph of the colony from which the type collection was taken. Photograph by Ricuary Evans ScuuLres Piatre Vill EXPLANATION OF THE ILLUSTRATION Prare IX. 1, ANruurtum ronrorpes I. Eb, Schultes. 2, AnruurtumM pLtuviaticum PR. E. Sehultes. Habit drawing of plants, three eighths natural size. Drawn by Dororuy H. Marsu PratEe LX ANTHURIUM Wi) pluviaticum VAN THURIUM fonto ides R.G. Schultes EXPLANATION OF THE ILLUSTRATION Puare X. Anruurium rikunorem I. bh. Sehultes. Habit drawing of plant, three eighths natural size. Drawn by Dororuy H. Marsu PLATE N AN THURIUM tikunorum RE. Schultes EXPLANATION OF THE ILLUSTRATION Puare XI. Parpatanruus rascrcutatus ( Rolth.) Korn. in sandy spot at La Chorrera, Rio [garaparana, Co- lombia. Photograph by Racuarp Evans Scuurres @ | PLATE EXPLANATION OF THE ILLUSTRATION Prare XIT. Herrania previniotiara R. Ee. Schultes. 1, leaf, about one third natural size. 2, flower, about one half natural size. 8, petal, twice natu- ral size, 4, staminode and anthers, twice natural size. 5, ovary and style, four times natural size. Draven by Fimer W. Suivi Prare XII HERRANIA / brevi ligulata es R.E.Schuttes pas | EXPLANATION OF THE ILLUSTRATION Prare XID. Herranta purpurea (Pitter) Re E. Nchultes. 1, leaf, about one third natural size. 2, flower, about one half natural size. 8, petal, twice natural size. 4, staminode and anthers, twice nat- ural size. 5, fruit, one half natural size. Drawn by Kimer W. Suivi rrare NIT HERRANIA purpurea (Pittier) RE Schultes EXPLANATION OF THE ILLUSTRATION Prare XIV. Comprerem Wanpurracanum R. EF, Schultes, Fruiting branch of the type plant. Photograph by Ricuarp Evans Scuurres XTV PLATE EXPLANATION OF THE ILLUSTRATION Prare XV. Menponcia LastopuyTa Leonard. a, leat blade, one half natural size. b, portion of leaf blade (upper surface), one half natural size. c¢, portion of leaf blade (upper surface enlarged to show stelli- form cystoliths), three and one half times natural size. d, part of inflorescence, one half natural size. e, hair from bract, six times natural size. f, sta- men, twice natural size. g, corolla lobe, twice nat- ural size. h, calyx, one and one half times natural size. i, stigma, three times natural size. J, a cys- tolith shown in detail, about eleven times natural size. Drawn from the type specimen Schulles & Lopez 10400Kk. be PLATE EXPLANATION OF THE ILLUSTRATION Phare NVI. Sancuezia ruinopuita Leonard, a, basal portion of inflorescence, one half natural size. b, bract, one half natural size. c, leaf, one half natural size. d, calyx, natural size, e, stigma, about three times natural size. f, corolla, natural size. g, anther (dor- sal view), about twice natural size. h, anther, (ven- tral view), about twice natural size. Drawn from the type specimen Schultes 6607, AVI PLATE BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 16, No. 5 CAMBRIDGE, MassaAcH USETTs, JULY 6, 1953 NOTES ON THE CULTIVATED LULO BY Ricuarp Evans Scuuures! AND JOSE CUATRECASAS” ONE of the most delicious of those cultivated fruits pe- culiar to the northern Andes is the /u/o or naranyillo. This fruit is very common from Pert and Ecuador to northern Colombia and Venezuela. Its area of greatest production centers probably in Ecuador and southern Colombia. In spite of the fact that a number of articles on this economic fruit have recently appeared (Pérez-Arbelaez, ‘*Plantas medicinales y venenosas de Colombia’’ (1937) 246; Chalons in Agric. Amer. 4 (1944) 110-112; McCann, ibid. 7 (1947) 146-149; Hodge in Rev. Fac. Nac. Agron. 7 (1947) 147-154; Hodge in Journ. N.Y. Bot. Gard. 48 (1947) 155-159; Pérez-Arbeléez, ‘‘Plantas litiles de Colombia”’ (1947) 451), little of a detailed na- ture has been known about the taxonomy of the source- plant. Although it has been generally accepted that the lulo represented Solanum quitoense, there is sufficient variation between the lulo plants from different parts of Colombia to raise some doubt that only one species is 1 Botanist, Division of Rubber Plant Investigations, Bureau of Plant Industry, Soils, and Agricultural Engineering, Agricultural Research Administration, U.S. Department of Agriculture; Research Fellow, Botanical Museum of Harvard University. ? Curator of Colombian Botany, Chicago Natural History Museum. [ 97 | involved (Schultes in Bot. Mus. Leafl. Harvard Univ. 14 (1949) 45, t. 10). We have recently attempted to study the numerous specimens of lulo now available. Although many more field observations and much more material from a wider area are greatly to be desired, we believe that our studies have led to at least a preliminary clarification of the prob- lem. It is merely as a preliminary contribution that we offer the following notes. Both of us have seen the lulo under cultivation in the field. Our experience and field notes are in complete agreement with the results of our study of the available herbarium material. We wish to thank the Directors of the following botan- ical institutions for their kindness in allowing us to con- sult the material entrusted to their care: Gray Herbar- ium, Arnold Arboretum, Economic Herbarium of Oakes Ames (Harvard University); Chicago Natural History Museum; National Arboretum Herbarium (Beltsville, Maryland); U.S. National Herbarium (Washington, D.C.); Royal Botanic Gardens (Kew); British Museum (Natural History) (London); Musée d’ Histoire Natu- relle (Paris); Jardin Botanique de |’Etat (Brussels) ; Jardin Botanique (Geneva); and the Jardin Botanico (Madrid). The two concepts which are involved may be distin- guished by the character in this key. I. Planta perfecte inermis. Peruvia, Kcuadore, Colombia meridi- onali. Solanum quitoense II. Planta ramulis, petiolis, nervisque variabiliter armata, Colombia centrali et septentrionali. Solanum quitoense var. septentrionale [ 98 ] Solanum quitoense Lamarck Illustr. 2 (1797) 16. Solanum angulatum Ruiz & Pavon FI. Peruv. 2 (1799) 36, t. 170, fig. a. Solanum quitense Humboldt, Bonpland & Kunth Nov. Gen. et Sp. 8 (1818) 25. Cotomsia: No definite locality. 1842, Sinclair s.n.—Comisaria del Putumayo. Valley of Sibundoy, Sibundoy. Alt. about 2225-2300 m. February 16, 1942, R. E. Schultes 3271.—Same locality. ‘‘Large shrub. Stem and under surface of leaf purplish, hairy. Flowers white, anthers yellow. Fruit orange coloured and size of oranges. Naranjillo. Kamsé Indian name: ma-sha-kve.’’ May 29, 1946, R. E. Schultes & M. Villarreal 7616. Ecuapor: No definite locality. “‘Nuqui.’? January 1848, Seemann s.n.—Guayaquil. [Cultivated from seed collected in Chimbo. ] Novem- ber 27, 1897, J. V. S. Muller s.n.—Vicinity of Huigra, ‘‘mostly on the Hacienda de Licay.’’ August 1918, J. N. Rose & G. Rose 22398. —Garden in Otavalo. Alt. 8500 feet. February 20, 1921, W. Popenoe 1266.—Bafios, Tungurahua. Alt. 6500 feet. March 6, 1921, W. Popenoe 1267.—Provincia de Tungurahua, Valley of Pastaza River. Between Bafios and Cashurco. Alt. 1300-1800 m. September 25, 1923, A. S. Hitchcock 21787.—Provincia de Tungurahua, between Hacienda San Francisco and Rio Margarjitas, alt. 1225 m. March 20, 1939, C. W. Penland & R. H. Summers 168.—Provincia de Pichincha, Valley of Rio Saloya, Los Paysanes. Alt. 1400 m. June 28, 1939, E, Asplund 7298.—Provincia de Imbabura, above Garcia Morena. **Common at edges of clearings. Alt. 4200 feet. Suffructescent herb, to 2 meters. Corollas whitish to pale magenta. Naranjilla.’’ August 9, 1944, W. B. Drew E-531. Peru: ‘38. 786—Solanum. Planta culta in Hort. Oppido la Mag- dalena. Solanum.’’ No collector, no date. [Possibly same collection as following. |—‘‘Ex oppida La Magdalena in Peruvia. Habitat in Limae hortic. Floret. Jun. Julio, Anoto et Sept. Fructus valde odorus. Vulgo Naranjas de Quito. Naturales Limae exprimunt aliquot gutas huju fructu in potum (vulgo maté) protectorem sauvem. Vulgo Naranjitas de Quito.’? August 1782, H. Ruiz & Pavén s.n.—‘*Lima Hort.’’? 1778, H. Ruiz & Pavén s.n.—‘‘Maynas,’’ 1831. Poeppig 2222,.—‘‘Chacapoyas, Pert.’’ Nodate. Matthews s.n. Country Unknown: ‘Ex hb. de Jussieu No. 6458.’’—‘‘Ex hb. Lamark.’’—‘‘Ex hb. De Candolle.’’ [Apparently a Pavén specimen. ] [ 99 ] Solanum quitoense Lamarck var. septentrionale R. E. Schultes & J. Cuatrecasas var. nov. Haec varietas a Solano quitoensi principaliter ramis, petiolis foliorum superficiebus superioribus et inferioribus in nervis spinosis spinulosisque differt. Cotompia: Departamento de Antioquia, no date, Jervisse sn.—De- partamento del Tolima, Ibagué. 1844. J. Goudot s.n.—Departamento del Valle de Cauca, La Paila, April 1853, Holton 23.—Departamento Norte de Santander, near Ocafia. “‘Bush, open spaces. Alt. 8,500’. Shrub or small tree, 10-19’, slender growth stems with thorns; flow- ers whitish, calyx and leaf with long, dark violet velvet, latter 14—2/ broad and 24-3/ long; fruit yellow with prickles, plum-shaped, edible, sour.’ January 1878, Kalbreyer 543.— Departamento de Santa Marta, [Santa Marta mountain], alt. 6,000 ft. January 1903, H. H. Smith 1853.—Departamento de Cundinamarca, cercanias de San Bernardo hacia Sasaima, alt. 1600-1800 m. “‘Matotales y cafetales. Gran hierba de tallo grueso y aquifolioso; cdlices algo violdceos, corola blanca o blanco-violacea.’’ June 23, 1940, J. Cuatrecasas 9604.—Departamen- to de Antioquia, Itaqui, July 1944, Br. Daniel 3352.—Departamento de Antioquia, alrededores de Medellin, Rio Negro, alt. 1,560 m. “*Lulo. Cultivated,’’ February 11, 1946, W. H. Hodge 6712.—De- partamento del Valle, Cordillera Occidental, vertiente oriental, Que- brada del Tigre, Quebradita de Pultabrava, alt. 1,440 m. ““‘Hierba muy robusta, 1 m. alt. Hoja blanda rigida, verde oscura haz, violeta envés. Hoja tierna violeta, ramos id. CAaliz violeta. Corola blanca o blanco violacea. Anteras amarillas. Fruto 5 cm. didmetro.’’ October 28, 1946, J. Cuatrecasas 22694.—Departamento del Valle, Cordillera Occidental, vertiente occidental, hoya del Rio Digua, Rio San Juan, abajo de Queremal a la derecha del rio entre km. 52 y 58, alt. 1,300- 1,500 m. ““Hierba robusta. Hoja muy blanda, envés y peciolo vio- leta. Haz verde, oscuro. Caliz violeta claro. Corola exteriormente violeta claro, interiormente blanca. Anteras amarillo claras. Fruto maduro amarillento, sabroso—5 ecm. didim., hirsuto. Lulo morado.’’ March 19, 1947, J. Cuatrecasas 23853.—Departamento del Valle, Cordillera Occidental, vertiente occidental, hoya del Rio Digua, lado derecho, entre Queremal y La Elsa, alt. 1,200-1,160 m. “‘Hierba robusta, 1 m,o mas. Hoja verde oscura haz, verde clara envés. Peci- olo verde palido (verdoso blanquecino). Céliz verdoso blanquecino. Corola blanca con ligero tono lila. Anteras amarillas. Frutos amarillo- anaranjados (maduros), 4.5 cm. diém., cubiertos de pelusa patente y punzante. Pulpa acidula, muy sabrosa. Lulo.’’ March 27, 1947, J. Cuatrecasas 23992.—Departamento del Valle, Cordillera Occiden- tal, vertiente occidental, hoya del Rio Digua, lado izquierdo, Piedra [ 100 ] de Moler. Bosques, 1050 m. alt. “‘Hierba muy robusta y ramificada. Hojas blanquecinas. Corola blanea. Anteras amarillas. Bayas 4 cm. diim., amarillo-anaranjadas, erizadas. Lulode perro.’? August 19-28, 1943, J. Cuatrecasas 15031. Ecuapor: “‘Ad radices M. Chimborazo, alt. 2,300’. Suffrutex 5- pedalis.’’ June 1860, R. Spruce s.n. VENEZUELA: Colonia Tovar, 1854-55, A. Fendler 1001.—Same lo- cality and date, Ad. Fendler 1002.—Same locality and date, W. Son- dar s.n.—Caracas, neighborhood of Guaranas. Alt. 3,000 ft. ‘‘Flow- ers blue, fruit without prickles, small, veined, tall shrubby plants.’’ December 1854, Birschel s.n. (K).—‘‘Habitat ad Orinocum, prope Carichana.’’ No collector, no date [ex ‘““hb. Bonpland’’ ]. Our studies lead us to the conclusion that typical Sol- anum quitoense is confined to the southern part of the range of the species complex and occurs in Pert, Ecua- dor (where it appears to be most abundant) and southern Colombia. This concept is easily recognized because it is completely devoid of spines along the branches, peti- oles and veins. It is deeply significant, we feel, that all of the collections from this southern periphery, with one exception, have not the slightest trace of spines. In other respects (such as color and density of indumentum on the leaves) they are also rather homogeneous. The collection Spruce s.n., from the base of Mt. Chimborazo, is the single exception; it has slight and very remote spines, and the Kew specimen is annotated as representing one of the types of Solanum, the fruit of which is gathered for food in Ecuador. It may well represent a distinct variant, but paucity of material precludes a more pre- cise disposition. Herbarium material from the northern periphery— most of Colombia and part of Venezuela—is, without a single exception, spiny. There is tremendous variation in the abundance and size of the spines which are borne along the branches, on the petioles and along the nerves of the upper and lower surface of the leaves. Some speci- [ 101 ] mens have very small and remotely placed spinules; the other extreme has stout spines up to 10 or 12 mm. in length. Cuatrecasas 23992, from the Rio Digua, is al- most devoid of spines, whereas Cuatrecasas 15031, from the same area, is well armed on the petioles and leaves with spines up to more than 1 em. in length. Hodge 6712, from Antioquia, has remote but strong spines along the petioles and the midrib and, occasionally, even on the tertiary veins. The collection Cuwatrecasas 9604, from Cundinamarca, has stouter spines, even on the upper surface of the leaves. If we can judge from the admit- tedly limited material at hand, we may suggest that there is evident a perceptible increase in density and size of spines as one proceeds northwards. Some of the material from Cauca is only weakly armed; the stoutest armature is found on specimens from the central and northern Andes of Colombia and Venezuela. Collections from in- termediate regions, such as Cuatrecasas 22694, from the Departamento del Valle, would seem to be links between Solanum quitoense and its var. septentrionale. The recognized fact that the density and size of the spines vary so much does not, we feel, argue against the separation as a distinct variety of the spiny from the unarmed variants. The genus Solanum tends to be ex- tremely variable in respect to spines where they occur. There is a very distinct possibility that, when ample material is available for study and when adequate field studies have been carried out, the variation in color and density of the soft indumentum of the leaves may also be found to be sufficiently important to use in the recog- nition of additional varieties. However, we cannot, at the present state of our knowledge, evaluate the charac- ters which may reside in the differences of leaf pilosity in the Solanum quitoense complex. We have chosen the varietal epithet septentrionale to [ 102 ] indicate our belief that the spiny material represents a northern variant of Solanum quitoense. In the Mutis collection of water-colors of Colombian plants, executed between 1783 and 1808 and preserved in the Jardin Botanico in Madrid, the two concepts Sola- num quitoense and S. quitoense var. septentrionale are clearly distinguishable. Plate 88 in volume 19 consists of two double-sized black and white sheets, one depicting a flowering branch with a floral dissection, one with a fruiting stem and a dissected fruit; this is the spineless Solanum quitoense. Plate 39, representing Solanum quito- ense var. septentrionale, has one sheet showing, in colors, a leafy branch in flower and one piece of stem in fruit; the stems, petioles and veins on the under surface of the leaf are armed with heavy spines and the flowers are larger than those shown in plate 88. There is a large colored plate of Solanum quitoense (C. M.-Curtis del.) in the De Candolle herbarium in Geneva. It is annotated as follows: ‘‘Solanum angula- tum. Imported from Peru in 1824, by Robert Barclay. Bury Hill’’; and in De Candolle’s hand: ‘*‘Gravure don- née par M. Barclay. Ne fait partie d’aucun ouvrage. A.DC, 1839.”’ The most extensive field notes found on an herbarium collection are preserved on the specimen of J. VY. Sigvald Muller s.n, at Kew. Because of their completeness, we hereby publish them almost in full: ... the plant is, as you no doubt know, a climber, the fruit bright orange, nearly round or spheroid, about 1} inch to 2 inches in diam- eter. The pulp is bright green, very juicy and very aromatic. The seeds are mixed with the pulp when ripe... . The seeds are fixed to a softer body, than what I describe as solid white, but this solid part gets broken up in the pulp (and is eaten as well) when the pulp is squeezed out. The pulp looks like the pulp from green gooseberries ; it is eaten with a little sugar. Is exceedingly pleasant and cooling. With cream it must be a delicious dish. The sugar is mixed with the pulp to taste. The pulp alone is not more acid than to make it pleas- { 103 | EXPLANATION OF THE ILLUSTRATION PLate XVII. Flower and young fruit of the type plant of Solanum quitoense var. septentrionale. Photograph by J. Cuarrecasas [ 104 | > XVO PLATE ant, even without sugar, which is a costly luxury in the Andean Val- ley. Large quantities come down to Guayaquil where they are rather expensive, as the mule transport over the Cordilleras is long and tends towards ruining the delicate fruit. Of late the pulp, strained from the seeds and added to water, is used for ice-making. The flavour is ex- cellent and was quite new to me. Hence, I went into the market, and went into the question what fruit it was. I was told by the English people, that it was a guava (there are many varieties here), but that could not be. Ruiz, who encountered Solanum quitoense in Pert in 1777, wrote of it (under the synonym S. angulosum) “©... Narangitas de Quito, por haber sido transplan- tado de esta Provincia, y tener sus frutos la figura y color de una Naranja pequefia; las mugeres estiman estos frutos por su olorcillo y por el gusto particular que da 4 la beb- ida del Mate, en la que acostumbran echar algunas gotas de su xugo; tambien los ponen entre las mixturas de flores para que hermosée y contribuya con su olor 4 hacer mas grata la mixtura’’ (Ruiz, H.: ‘‘Relacién historica del viage a los Reynos del Perti y Chile’’ Jaramillo- Aranjo ed. (1952) 80). [ 105 ] PLANTAE COLOMBIANAE XIV RHYTIDANTHERAE MONTIS MACARENAE NOVA SPECIES BY Ricuarp Evans Scuutres’ REcENT collections of plants from the northern part of the Cordillera de La Macarena in the Intendencia del Meta, Colombia, have been replete with extraordinary novelties and endemics. One of the most outstanding of these plants is a hitherto undescribed species of the ochnaceous genus Rhytidanthera. Because of its unusual size and beauty, it may be named Rhytidanthera regalis Rk. HL. Schultes sp. nov. Arbor magna usque ad septuaginta pedes alta, fron- dosissima, aspectu nobilis. ‘Truncus columnaris, basi saepissime aliquid arcuatus, usque ad duo pedes in dia- metro, rufo-brunneo cum cortice subruguloso et minute lenticellato; ligno duro, albido. Folia imparipinnata, conspicue pendula; petioli 20-30 cm. longi, basi usque ad 2 mm. in diametro; foliola plerumque tredecim (cum petiolis 4-6 mm. longis), lateralia alterna, 8-4.5 cm. dis- tantia, lanceolata, apice acuto-acuminata, basi late cu- neata vel subrotundata, 15-22 cm. longa, 8—5.5 cm. lata; foliolum terminale liberum, perfecte lanceolatum, apice 'Botanist, Division of Rubber Plant Investigations, Bureau of Plant Industry, Soils, and Agricultural Engineering, Agricultural Research Administration, U. 8. Department of Agriculture; Research Fellow, Botanical Museum of Harvard University. [ 106 ] sensim acuminatum, basi cuneatum, usque ad 15.5 cm. longum, 8-8.5 cm. latum; alia marginem versus vix retrocurvata, margine grossiuscule serrata, dentibus apice paulo incurvis et plerumque 5-8 mm. distantibus, supra nitida venis non elevatis, subtus pallidiora, venis promi- nenter elevatis, secundariis plusminusve viginti quinque ad triginta; stipulis caducis. Inflorescentiae terminales, longae, quam folia paulo breviores, usque ad 20 cm. longae, rhachide 2-3 mm. in diametro, pauciflorae (flori- bus usque ad viginti); rhachide rami usque ad 10 cm. longi; pedicelli ad rhachidem articulati, robustiores lig- nosique, usque ad 10-12 mm. longi, apice subclavati, basi 1.5 et apice 2.5 mm. in diametro. Flores solitarii, aromatico-fragrantes, mucilaginosi, alabastris comparate parvis, usque ad 15 mm. longis, 6 mm. in diametro. Sepala quinque, imbricata, concava; duo exteriora sub- coriacea sed marginem versus aliquid membranacea, late rotundata, margine subintegra, apice retuso-incisa, supra minutissime pulverulento-substriolata (non strigillosa), plusminusve 7 mm. longa, 12 mm. lata, basi cum plus- minusve quattuor ad sex glandulis minutis digitaliformi- bus, usque ad 1.5 mm. longis, 0.25 mm. in diametro; in- teriora majora, membranacea vel apicem versus aliquid papyracea, concava, rotundata vel late oblonga, apice rotundata, margine irregulariter sublacerata, 11 mm. longa, 15 mm. lata. Petala quinque, alba, membranacea, leviter inaequalia, oblonga (80 mm. X15 mm.) vel ob- longo-spathulata (80 mm. Xapice 10 mm.) apice leviter fissa, margine integra. Stamina viginti octo ad triginti duo, brunneo-flava, subaequalia, usque ad 12 mm. longa. Antherae 10 mm. longae, 0.8 mm. latae, longitudinaliter rugulosae, filamentis 2 mm. longis, basi 0.7 mm. latis. Ovarium crassum, nigrum, inverse claviforme, apicem versus in stylum indeterminatum, erectum, luteum, paulatim attenuatum, usque ad 15 mm. longum, 2.5 { 107 | EXPLANATION OF THE ILLUSTRATION Pirate XVIII, Ruyripanruera reeAtis R. E. Schultes. Characteristic habitat on steep slopes or edges of cliffs. Macizo Renjifo, northwestern part of the Cordillera de La Macarena. Photograph by R. E. Scuuttes [ 108 ] Piavre SV Ll mm. in diametro, quinque cum stigmatibus sessilibus. Capsulae usque ad 3.5—4 cm. longae, 0.5 cm. in diametro, apicem versus anguste falcatae. Largest of the five known species of this endemic Colombian genus, Rhytidanthera regalis appears to be somewhat intermediate between FP. splendida and R. magnifica. In addition to marked differences in the leaves, Rhytidanthera regalis can be distinguished from all other species in the important character of the num- ber of stamens. Rhytidanthera splendida has 18—20 sta- mens; R. magnifica 40-50; R. sulcata, about 50; R. mellifera, 64; whereas R. regalis has 28-32. The fruit of Rhytidanthera regalis resembles in its size that of RP. sulcata, but the latter isa small shrub; the margin of its leaflets is doubly serrate and its inflorescences are longer than the leaves. From Rhytidanthera mellifera, the new species differs strikingly in shape, size, and consistency of the leaves and in several important floral characters. Worthy of note is the presence in Rhytidanthera regalis (sometimes on the same flower) of both obovate-spatulate and oblong-rotund petals, although the great majority are oblong-rotund; this is, apparently, an unusual con- dition, for Dwyer (in Lloydia 9 (1946) 51) employs the difference in petal shape as the major character in his key to the species. Rhytidanthera regalis is the second surprise which this genus—the only compound-leaved one in the Ochnaceae —has given us in the last decade. For one hundred years, Rhytidanthera has been known, although until 1904 it was not recognized as distinct from Godoya. In this century, only three species, all from Colombia (Santan- der, Magdalena and Boyaca), were collected (ef. Dwyer loc. cit. 50-54). This genus, native to the northern parts of the eastern Cordillera of Colombia, constituted then an extremely restricted endemic. [ 109 ] In 1943, Rhytidanthera mellifera was discovered on ancient (Cretaceous) sandstone hills in the Amazonian drainage of Colombia —a significant range extension. Although still confined within the boundaries of Colom- bia, Rhytidanthera had been located in the great Ama- zonian area, astonishingly distant from the eastern Cordil- lera. Furthermore, since the isolated sandstone hills of the Vaupés and Caqueta (where Rhytidanthera mellifera was collected) are remnant outliers of a once more or less continuous land-mass the core of which lies in southern Venezuela and British Guiana, the possibility of the dis- covery in the future of Rhytidanthera far to the east was open to suggestion. This curious distribution of Rhyti- danthera was accepted (Schultes in Bot. Mus. Leafl. Harvard Univ. 14 (1949) 84) as evidence for believing in an ancient migration or flow of Andean elements east- ward over the old Venezuelan-Guianan land-mass. The collection in the Macarena Mountains of Rhyti- danthera regalis stands out as one of the most significant phytogeographical discoveries of the last two decades. Both morphologically and geographically intermediate between the western species and the Amazonian Rhyti- danthera mellifera, the Macarena species provides us with an unexcelled ‘‘missing link.” Cotompra: Intendencia del Meta, Cordillera La Macarena (extremo nordeste), Macizo Renjifo, faldas orientales. Alt. 600-1300 m. ‘‘Tree 60 ft. tall. Crown heavy. Flowers white, anthers yellow.’’ Decem- ber 30, 1950-January 5, 1951. Jesus M. Idrobo & Richard Evans Schultes 871 (Tyrer in Herb. Gray; Dupticarr tyre in Herb. Nac. Colomb. ).—Intendencia del Meta, Sierra de La Macarena, Central Mountains, North Ridge. Alt. 1500 m. “‘Dense forest. Tree up to 15 m. high. Petals white ephemeral, stamens yellow, inclined to lower side of flower.’’? December 30, 1949, W. R. Philipson & J. M. Idrobo 2011.—Intendencia del Meta, Cordillera La Macarena (ex- tremo nordeste), Macizo Renjifo, faldas orientales. Alt. 600-1300 m. ‘Tree growing on steep, rocky slopes. Height 60-70 feet. Diameter at base 20 inches. Bark dark reddish brown, slightly rough. Wood very hard. Usually somewhat gnarled. Fruit green.’’ December 30, [ 110 ] 1950—January 5, 1951, Jesus M. Idrobo & Richard Evans Schultes 852. —Same locality and date. ‘‘Tree 50 ft. tall, very extensive crown. Branches gnarled. Fruits green.’’ Jests M. Idrobo & Richard Evans Schultes 860.—Same locality and date. ““Flowers white. Tree 65 feet tall.’’ Jesus M. Idrobo & Richard Evans Schultes 869.—Same locality. **Al borde de un pefiasco. Arbol de 20 m., tortuoso. Tronco de 2 pies de didmetro. Flores blancas.’’ January 22, 1951, Jesus M. Idrobo & Richard Evans Schultes 1188.—Intendencia del Meta, Cordillera La Macarena, Mesa del Rio Sansa. Alt. ca. 1000-1300 m. ‘‘Tree about 50 ft. tall, inclined over cliff.’’ January 23, 1951, Jestis M. Idrobo & Richard Evans Schultes 1300. [111 ] A NEW SAURAUIA FROM MEXICO BY RicHarp Evans SCHULTES Saurauia Comitis-Rossei PR. LH. Schultes sp. nov. Arbor robusta ut videtur. Rami secabrido-hirsuti sed denique subglabrescentes, fulvo cum cortice. Foliorum lamina petiolata, valde coriacea, obovata, apice acuta, basi rotundata, minutissime serrulata, supra atroviridis et muriculata et sparsissime setoso-pilosa, infra velutina et densissime pilosa albo-canis cum pilis atque cum pilo- rum floccis in nervorum axillis, nervis plusminusve tres et viginti parallelis, 15-18 cm. longa, 5.5—-6 em. lata. Inflorescentia comparate pauciflora, foliis brevior, cum pedunculo 10-15 cm. longa; pedunculus dense adpresso- hirsutus vel setoso-pilosus; pedicelli robusti, hirsuto- pilosiusculi, 4 mm. longi; bracteae conspicuae, lineares, setoso-pilosae, usque ad 1 cm. longae. Flores 1.8 em. in diametro, albi. Sepala quinque, in maturitate sicca atque persistentia, obovata vel subrotundata, 6 mm. longa, 4 mm. lata, extus scabridule et minute setoso-pilosiuscula, margine minutissime ciliata. Petala quinque, fere usque ad basim libera, glabra, membranacea, late et subquad- rangulateque rotundato-ovata, 8 mm. longa, 5 mm. lata. Stamina plusminusve viginti, corollae basi adhaerentia, basi setoso-barbata. Staminum filamenta 8-4 mm. longa; antherae versatiles. Ovarium globosum, quinque-parti- tum, glabrum, quinque cum stylis carnosis minutisque. C112] Fructus niger, 3 mm. longus, baccatus, quinque-carpel- latus. Saurauia Comitis- Rosset appears to have no close allies among the known species of Mexico and adjacent parts of Middle America. It is easily distinguished by its ex- tremely thick and coriaceous leaves, as well as by a num- ber of less conspicuous vegetative and floral characters. The common name of Saurauia Comitis-Rossei is re- ported by Reko to be “‘mameyito.’’ In southern Mexico, this name refers to a number of species of Saurauia and other genera with small, edible fruits. At the request of the collector, I have named Saurauia Comitis- Rosset in honor of the Earl of Rosse in recogni- tion of his active and enthusiastic interest in Mexican horticulture. Mexico: Estado de Oaxaca, Distrito de Juquila, Pochutla, Cafetal de San Antonio, 1700 meters altitude, February 1941, B. P. Reko 6183 (Typr in Herb. Gray). [113 ] EXPLANATION OF THE ILLUSTRATION Prare XIX. Savuravura Comitis-Rosser R, EF, Schultes. 1, plant, one half natural size. 2, flower, twice natu- ral size. 3, petal, three times natural size. 4, sepal, three times natural size. 5, sepal, inside view, three times natural size. 6, sepal, outside view, three times natural size. Drawn by Gorvon W, DiLion [ 114 ] PLATE XIX SAURAVIA Comitis~Rossei ARE. Schultes vy & as oO SS Sorel e wee TNE Nat ee = ee nz INS ft} x ez Se SS NS Ex 4 WV? 5 \ Py = <9 SE CRIS ROE IRIE SER rs REA RA Se SS Se EFS aS Nx? Cae RO IS SASS 3 : ASE DAN RSRSS Ae re ic lem. cee eS hy Santas Se \ on tr SG iret RS $ st) =a" see fA Tos: + eal Fath ie ¥e9) i, ONENESS Sk ait re eS SEs e" BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 16, No. 6 CamsripGr, Massacuusetts, DecemBer 4, 1953 THE PRE-COLUMBIAN CULTIVATED PLANTS OF MEXICO BY Rosert L. DrEssLter INTRODUCTION IN recent years many important papers have been pub- lished dealing with cultivated plants, their origins and their relationships to human cultures. ‘These studies have served to increase greatly our knowledge and under- standing both of the plants and of the people by whom they were used, as well as to point out some of the areas where further research is most urgently needed. Middle America, especially from central Mexico to Guatemala, was one of the two great centers of agriculture in the New World, the other being the Andean area, and a very considerable variety of cultivated plants were known to the peoples of this region. It seems, therefore, worth- while to attempt an enumeration of the plants which were cultivated in this area, with a discussion of the per- tinent botanical literature as to their origins, distributions and importance. At this point the author wishes to acknowledge his indebtedness to the persons who have aided in this study by their discussion and suggestions concerning many phases of the problem. Special thanks are due Drs. Edgar Anderson, Howard S. Gentry, Gordon W. [115 J Hewes, Albert F. Hill and Paul C. Mangelsdorf and Mrs. Margaret A. Towle. In the understanding of any biological entity, it is first necessary to recognize and to characterize the ele- ments of the group in question. When this is accom- plished, one is able to investigate profitably the distri- bution, history and relationships of the form or group of forms. Unfortunately, the cultivated plants present the botanist with problems the complexity of which is rarely equaled among other organisms. By becoming associated with man the plants are partially freed from the restrictions of natural selection and carried to new areas where they may hybridize with related types from which they would otherwise be isolated. Man, in addi- tion, aids in the development of new types by conscious and unconscious selection. Al] this leads to an inordinate degree of variability in such populations, an understand- ing of which can rarely be achieved by any one limited field of approach. However, with careful and discerning morphological study and the application of cytology and genetics, the newer tools of taxonomy, and with the cooperation of the ethnologist and the archaeologist it is possible to obtain results of very great value to all concerned. The plants and animals domesticated by man have certain unique qualities which cause them to be of in- terest to the anthropologist. These center about the fact that they, themselves, are organisms, biological entities which may be studied as such, as well as in their rela- tionships to man. In other words, though shaped by their association with man, they do not so nearly repre- sent mere products of the human mind as do many other phases of human culture which we study. Aside from the very basic importance of cultivated plants to all advanced cultures, there are more practical [ 116 ] reasons for studying them. Since these plants are basic to our present cultures as well, a more complete under- standing of them allows us knowingly to reshape and im- prove our sources of food and raw materials in a way which was not possible for our ancestors. In the present paper particular attention will be given to the geographic origins of the plants of this area. A number of different criteria have been used in attempts to determine the centers of origin of the plants which are more widely distributed in cultivation. These criteria have largely been developed and elaborated by de Can- dolle (85) and Vavilov (205, 206, 207); among the more important ones may be listed the following: 1. Distribution of the same species or its apparent an- cestor ina wild state. This is a point of great importance, if one can be certain that the ‘‘wild’’ plants are not nat- uralized from cultivation. his, however, is not easy, particularly considering the grave doubts which exist as to the occurrence of ‘primeval forest’’ in much of trop- ical America. If, of course, a useful wild plant occurs over a wide area, then it is not only possible but quite probable that it will have been cultivated independently or simultaneously in two or more areas. The avocado (Persea americana) may be an example of this. 2. Distribution of primitive forms within the culti- vated species. ‘This, too, is of great value, if properly in- terpreted. The primitive forms of a group may, however, occur either in the center of origin or on the periphery of the plant’s range. Each case must be decided on its own merits. 3. Center of diversity. This valuable concept, devel- oped principally by Vavilov (205, 206, 207), is based on the idea that the greatest diversity within a species will occur in the area where it has been in cultivation for the longest time. Also, the area where varieties of the wild [ 117 ] parent species occur will be the area in which the great- est hybridization might be expected. While this concept is of great interest, both to the botanist and to the plant breeder, it must be realized that secondary centers of diversity may occur, through other factors such as hy- bridization or topography. 4. Distribution of closely related species. Within broad limits, this criterion is quite useful. Some of the squashes (Cucurbita), for example, have been thought to be natives of Asia, but there are no wild species of the genus in Asia, whereas there are many in America. In studying the relationships of species, cytogenetics may at times give evidence which is of great value, as will be shown to be the case with tobacco (Nicotiana). 5. Archaeological evidence. If sufficiently extensive, the record of archaeology may be of very great value. With the interest in carbon 14 dating, which is possible wherever there remains much organic matter, and in- creased attention to plant materials, this approach will become of increasing importance. Paleobotanical evi- dence, which is scanty for the cultivated plants, may conveniently be included in this category. 6. Historical data. Written accounts may often be of value in interpreting recent dispersals of cultivated plants, and there remains to be done a good deal of care- ful work along these lines. 7. Linguistic. The interpretation of routes of dispersal through the study of plant names is of interest and of some value, but this is probably the least reliable type of evidence and must be used with great caution. As in other such cases, the most dependable studies are those that use all available data, from every source or aspect of the problem. One of the untouched fields in ethnobotany is the study of the development and diffusion of geographical races [ 118 ] within a crop plant and of their relations with culture complexes. Such races have been studied in some degree for two Middle American crop plants, maize and cotton (104, 213). In each case it has been pointed out that the development of these races must have occurred in partial isolation and must have been related to cultural centers. It is for this reason especially desirable that extensive col- lections, if not studies, of truly native crop plants should be made before they are further decimated by the impact of modern cultural changes. For the present paper, material has been drawn from the literatures of botany, ethnology and archaeology, but no pretense is made of a complete coverage of any one. A better representation of the historical writings of the earlier Spanish explorers would be especially desirable, but it is a subject worthy of study in itself. Several of the important references in this field are cited, but none has received the time and attention which it deserves. Some cultivated species may have been overlooked. It is hoped, however, that most of the important cultivated plants have been included, particularly those which have changed in their biological nature and their distribution through their association with man. One cannot, in a paper of this scope, list all the useful wild plants of Mexico, though the more important of these were doubt- less planted at times by the early Mexicans; information is meager or lacking for those which have been intention- ally omitted. ‘The evaluation of the published informa- tion has in some cases been aided by frequent reference to herbarium material and by some field experience. It should be borne in mind that the peoples of this region had developed both agriculture and horticulture to a rather high degree and actually possessed botanical gar- dens which were, at the time, unrivaled among the Euro- pean peoples. | 28-4 List oF SPECIES The species are listed alphabetically, using the correct scientific names. Other names which have achieved a wide usage in the literature are in some cases cited in parentheses; these may be synonyms or names properly belonging to other species. No attempt has been made to list all of the common names; these may be sought in various works listed in the bibliography, particularly those by Standley and by some of the Mexican authors. For the sake of brevity, the number system of citation is used; and, for the convenience of the reader, the more recent or more comprehensive studies are indicated by an asterisk in the citations which follow the discussion of each species or group. Achras: see Manilkara. Agave: Metl, maguey, henequén. The species of Agave are, to the present day, of great importance in Mexico as sources of food, drink and fiber. Among the Aztecs they entered directly and indirectly into religion, and it has been suggested (182) that the name, Mexico, may be derived from the Nahuatl word metl, which is a generic term for agaves and other plants of similar appearance. Among the species anciently grown in central Mexico for aguamiel and pulque are A. atrovirens Karw., which is the most important spe- cies, with many variations recognized by the growers, A. latissima Jacobi and A. mapisaga 'Trel. The types now used as a source of distilled beverages, such as 4. tequlana Weber, were no doubt used in pre-Columbian times at least for food, but may not have been cultivated. Since both the cultivation and the uses of the pulque agaves center in central Mexico, where the species used seem to be native, these cultivated plants may be con- sidered as having originated in that region. [ 120 | A. sisalana Perrine and A. fourcroydes Lem. are both extensively cultivated as fiber plants and are apparently natives of the Yucatan Peninsula, where they were prob- ably cultivated on a smaller scale in ancient times. Many wild species of Agave, and of Mourcroya, Yucca and Bromela as well, may be utilized as fiber sources. Even in the wild state the agaves are not too well understood, because, like the cacti, their habit is so poorly suited to the preparation of specimens. Extensive field work is necessary to study such a group properly. (48, 182%, 156, 164*, 191*, 192) Amaranthus leucocarpus S. Wats., A. cruentus L.: Huauhth, bledos, alegria, amaranth. The great importance of these plants as ‘‘cereals’” in ancient Mexico is clearly indicated by the tribute lists for the empire of Moctezuma (178), which indicate an annual levy of about 200,000 bushels of ‘‘huauhtli’’ or amaranth seed, as compared to about 280,000 bushels of maize and 230,000 bushels of beans. The grain amaranths have been poorly understood, but a recent careful ethno- botanical study by Sauer (178) has gone far to improve the situation. 4. leucocarpus is still grown over a wide area in Mexico and Guatemala. A. cruentus occurs in the same area, but less extensively, and appears to be more important in Guatemala than A. lewcocarpus. The great decline in cultivation of this useful plant in post- Columbian times is due, at least in part, to its suppres- sion by the Spaniards because of its important role in Mexican religious ceremonies. The grain amaranths are, by their botanical relation- ships, undoubtedly of American origin. Sauer suggests that A. leucocarpus may be most closely allied to 4. hybridus LL. and A. Powell S. Wats., both of Mexico and Central America, while A. cruentus seems most F121 | closely allied to the Central American 4. dubius Mart. The exact origins are not certain, but Sauer and some other writers, have suggested that the cultivation of the grain amaranths may have preceded maize culture. There is not yet adequate archaeological evidence on this point. (5*, 17, 18, 102*, 132*, 148, 171, 178*) Anacardium occidentale L.: Maranon, cashew. The cashew is thought by some (192) to occur natu- rally in southern Yucatan and may possibly have been cultivated there. It has the appearance of being native from Brazil to the Antilles, especially as a strand plant. (153, 191, 192) Ananas comosus (L.) Merril/: Pina, pineapple. The pineapple is known to have been cultivated in Mexico before European contact (52). Since it is usu- ally propagated by vegetative means, it must have been easy for the early Americans to select and grow seedless forms. The wild species of Ananas are all native to the Brazil-Paraguay region, though 4. comosus sometimes occurs as an escape from cultivation in many parts of the world (15). Though the exact ancestry of the cultivated pineapple is not known, it is almost certainly a native of South America, probably of the south Brazil-Paraguay region. (15*, 52*, 58, 54°*, 149, 186*, 196, 208) Species of the related genus Bromelia are sometimes planted as hedges (149) and may have been so used in pre-Columbian times. These spiny plants produce an edible, acid fruit. Annona The species of this genus are widely cultivated, but their histories are quite imperfectly known. The first accounts (98) indicate that several kinds were known at an early date in Mexico. There is little agreement [ 122 | among authors as to their origins. Among the species which have been cited as cultivated in Mexico are the following : A. Cherimolia Mill. : Quauhtzdpotl, chirimoya, cheri- moya. The cherimoya is probably the best known and surely one of the best liked of the annonas. The evidence as to its origin and early distribution appears to be conflicting, doubtless due, in part, to confusion between the different species of Annona. Popenoe describes what he believes to be groves of wild cherimoyas in the mountains of Keuador (151). Safford (170) and Costantin and Bois (55) present archaeological evidence of its early occurrence in Peru. Cobo (50) writes of introducing the cherimoya from Guatemala into Peru about 1680 and implies that it was unknown in Peru previous to that date. The name cherimoya is said to be of Quechua origin. Fries, in his monograph of Annona (80), does not comment as to the origin of A. Cherimolia. A. diversifolia Safford: Ilama. This species, though said to be of good quality, is not well known. It ranges from Colima and Guerrero into Central America. Safford (168) thinks it to be one of the types described by Hernandez. A. glabra L. Reports as to the quality and cultivation of this spe- cies do not agree. It appears to have a very wide natu- ral distribution as a strand plant. A. muricata L.: Guandbana. The origin of this widely distributed species is quite uncertain. A. purpurea Moc. & Sessé This species now ranges from Mexico to northern South America. Daies | A. reticulata L.: Anona. While Popenoe (158) and Standley (191) think that this is probably native in parts of Mexico and Central America, Fries (80) considers it to be West Indian. A. squamosa L.: Ahate. Fries (80) considers this species to be probably West Indian, because this section of the genus (Atta) has the greatest number of its species there. The section is not, however, limited to that area. This appears to be the tree which Hernandez describes as native to the Panuco region (northern Veracruz) and then introduced into Cuernavaca (98, 199). (55, 78*, 80*, 98, 150, 151, 153, 191, 196, 199) Arachis hypogaea L.: Tlaleacduatl, cacahuate, mant, peanut. The peanut was apparently not of great importance in early Mexico, and it may actually have been introduced from the West Indies by the Spaniards as implied by Hernandez (93). The compound name, ‘‘tlalcacduatl,”’ or earth cacao, has been cited as evidence of its late arrival in Mexico (238, 129). In any case, all evidences point to a South American origin for this plant. Closely related wild forms are known from the south Brazil- Paraguay region. (5, 23, 32*, 47, 103*, 209) Bixa Orellana L.: Achiotl, achiote. This variable tree is grown for the pulp surrounding the seeds, which is used as a dye, food coloring, flavor- ing material and cosmetic. It isnow known throughout tropical America. No careful study of Biva is available, but it may possibly be of Amazonian origin. Other spe- cies which may be distinct have been described from that area. (49, 71, 98*, 183, 191, 207) [ 124 | Bomarea edulis (7'uss.) Herb.: Coyolxéchitl. This is probably the species of Bomarea which the Mexicans cultivated for the edible, tuberous roots and as an ornamental (93, 201). B. edulis is apparently a variable species with a wide natural range in Mexico and Central and South America. The Mexican members of this largely South American genus are not well under- stood. (14, 98, 162, 201*) Bromelia: see under Ananas. Brosimum Alicastrum Swartz: Ox, ramon. Lundell (118) believes this to have been an important tree cultivated by the Mayas. The pulp of the fruit is sweet and edible, and the seed is eaten boiled or roasted or may be dried and ground into a flour. It is said to be quite palatable and nutritious. The ram6én, now impor- tant as a source of forage, is abundant about ruins in Yucatan, implying former cultivation. The species is widespread and probably native from northwestern Mex- ico into Central America. (113*, 191*, 192) Byrsonima crassifolia (L.) DC.: Nantzinaécotl, nance, nanche. This shrub or small tree is of great importance in some areas. It appears to be the most important fruit of south- ern Veracruz and the Yucatan Peninsula during the summer months. The fruit, about the size of a large cherry, has a strange and at first unpleasant flavor, but is universally popular. The species appears to be native in savanna areas in Veracruz and Campeche and_ probably elsewhere. When natural groves occur near homes or villages, they are preserved and the fruit gathered for market and home consumption. (17, 1383, 177, 191, 194) Calocarpum mammosum (L.) Pierre (Pouteria mammosa (I..) Cronquist): Tezonzdpotl, sapote, mamey sapote, mamey colorado. [ 125 ] The mamey sapote is a widely cultivated fruit tree which occurs from Mexico to northern South America and in the West Indies. Standley (191) and Popenoe (158) consider it as probably native to southern Mexico and Central America. (64*, 183, 158, 191*, 199) Calocarpum viride Pitt. (Pouteria viridis (Pitt.) Cronquist): Jnyerto, green sapote. This species, variable, though generally smaller-fruited than the last, ranges from southern Mexico to Costa Rica, but is apparently most frequently cultivated in Central America. (64*, 153*) Canavalia ensiformis (.) DC.: Jack bean. This bean is now of relatively little importance. It is known from ancient archaeological levels in Peru (19, 55, perhaps not this species’), and is reported from archae- ological evidences in North America (116). It may be that this large-seeded species was cultivated very early, but has since declined in popularity and usage due to the development of the better types of Phaseolus beans (177). Piper (147) considers it “‘practically certain that the plant is native to America.’’ Its nearest relatives appear to be Mexican, Central American and West Indian in distri- bution. Vavilov (207) assigns it to the Mexican-Central American center of origin (diversity) with a query. (55, 115, 117, 147*, 207) Capsicum annuum L., C. frutescens L.: Chile Pepper. The importance of the chile in Mexican diet is well known and doubtless of great antiquity. The common species, C. annuum and C. frutescens, have long been confused, but Smith and Heiser (187) have recently found that they appear to be truly distinct with strong sterility barriers preventing hybridization. Both species include [ 126 ] perennial forms in the tropics. The slight, but fairly constant, morphological differences listed by Smith and Heiser include characters that sometimes serve to dis- tinguish the seeds of the two species, a fact which should be of interest to archaeologists. Centers of diversity for the peppers occur both in Mexico and in Brazil (206). The small-fruited peppers are weedy and now occur spon- taneously throughout the tropics. C. frutescens is found under seemingly natural conditions in the canyons of northwestern Mexico (81) and even as far north as the Baboquivari Mountains in Arizona(41). Too little is now known to determine with any certainty the origin of either species. They may have spread as useful weeds at avery early date and then been cultivated independently in two or more areas. (25*, 31, 41, 81, 92, 92a*, 93, 106, 187*, 196, 206) C. pubescens R. & P., distinguished by purple flowers and purplish-black seeds, is known from both Central and South America (92). Carica Papaya L.: Papaya. The papaya, an herb of tree dimensions with melon- like fruit, is believed to have been known to the Mayan and perhaps to the Aztec cultures (118, 149). Several wild and cultivated species of this genus are found in South America, but Solms-Laubach (188) considered C. Papaya to be most closely allied to wild forms occur- ring in Mexico and the West Indies. Hybridization may have played a part in the origin of the cultivated form. Sauer (177) gives reasons for believing it to be Central American. (113, 138, 149, 177*, 188*, 191*, 192) The smaller-fruited C. cauliflora Jacq., which ranges south to Colombia and Venezuela, is listed by Standley as cultivated and perhaps native in Veracruz and Chia- pas (191). Casimiroa edulis La Llave & Lew.: Iztaczipotl, cochitzdpotl, sapote blanco, white sapote; C. Sapota Oerst.: matasano. These trees are grown in Mexico and Central America for their sweet fruits, which apparently vary in quality. A good account is given by Martinez (186). C. edulis, the sapote blanco, is Mexican in its present occurrence, while C. Sapota, the matasano, is largely Central Ameri- can. These distributions probably reflect their origins. (133, 186*, 158, 191, 199) Chamaedorea Tepejilote Lichm., C. Wendland- iana (Oerst.) Hemsl.: Tepejilote, pacaya. At least one, and probably several, species of the small, attractive palm, Chamaedorea, are cultivated in southern Mexico and Central America for the young staminate flower clusters, which are used as a vegetable. ‘The un- opened inflorescence is said to resemble an ear of maize in appearance and, at times, in size. Standley (198) re- ports them to be quite palatable. Chamaedorea, like most palm genera, is in need of study, and the names given here are, at best, tentative. (81, 191, 192, 198*) Chenopodium Nuttalliae Safford (C. pueblense Reed?): Cuauhzonth, huauhtzonth, apazote. Though less well known than the species cultivated in South America, this Mexican Chenopodium was of some importance, as indicated by its present relict occurrence in cultivation in many parts of Mexico (159, 172). It now is used principally as a green vegetable (the unripe fruit clusters), though it may have been used to some ex- tent as a cereal, as are the South American species. C. pueblense Reed (58) was not distinguished from C. Nut- talliae by its author and may be the same species. C. Nuttalliae is closely related to C. Quinoa Willd. of the [ 128 ] Andean area, and Aellen (8) and others have considered it to be that species. The supposed diffferences in seed color are of no value. Hunziker (102), however, points out floral differences between the two plants. While C. Quinoa has close allies occurring wild in the Andean region, the situation as regards the Mexican Chenopodium is not clear. It may have been derived from the South American species in cultivation, though Aellen (1) im- plies a relationship to the North American C. Berlandieri Mogq. A careful study of these plants should be reward- ing, (1, 2; 8, 101, 102*, 148, 158, 159*, 172*,. 178) C. ambrosioides L. (Apazote or wormseed) is widely distributed and has medicinal uses. It is said to be cul- tivated at times (17, 194). Cnidosculus Chayamansa MecVaugh (Jatropha aconitifolia Mill.): Chaya, chay. The chaya is a shrub cultivated in the Yucatan area for its young shoots and leaves, which are eaten as a pot herb. It is related to C. aconitifolia (Mill.) I. M. Johnst. and C. Chaya Lundell, both of which are more abundantly supplied with the stinging hairs which have earned Cnid- osculus the generic name of ‘*Mala mujer’? in Mexico. The less objectionable forms of C. Chayamansa have doubtless been selected under cultivation. (113, 114*, 118*, 166, 192, 198, 196) Cocos nucifera 1.; Coco, coconut. There has been a good deal of controversy concerning the pre-Columbian distribution of the coconut. There is now little doubt of its Old World origin. Bruman (29), after a review of the historical data, has concluded that the coconut did occur in Colima and probably elsewhere on the west coast when the first Europeans arrived. It was no doubt used by the natives, but may or may not have been cultivated. (29*, 177a, 191) [ 129 ] Crataegus pubescens (HBK.) Steud. (C. mexi- cana Moc. & Sessé?, C. stipulosa (HBK.) Steud.): Teax- dcotl, tejocote, manzanilla. The tejocote is still widely cultivated in Mexico and Guatemala for the apple-like fruit which is eaten raw or variously cooked. ‘These are probably the apples referred to by Pomar (158) as being equal in size and flavor to the Spanish ‘‘San Juan’’ apples. Standley and Steyer- mark (194) consider it to be a native of Mexico intro- duced into Guatemala. (9, 18, 138, 148, 158, 158, 191, 194°", 206, 207) Crescentia Cujete L.: Jicara, tecomate, calabash. The calabash, a tree quite unrelated to the bottle gourd Lagenaria, produces a large fruit, the shell of which is used for utensils, as is that of the gourd. It ranges from Mexico to northern South America and oc- curs also in the West Indies. It is probably native in southern Mexico. C. alata HBK. is a smaller-fruited species of western Mexico and Central America. (191*, 192, 196, 200) Crotalaria longirostrata Hook. and Arn.: Chip- ilin. This large herb of the legume family is grown in Guatemala and probably southern Mexico as a pot herb. It is apparently native to much of Mexico and Central America. (115, 191, 194%) Cucurbita: Ayotli, calabaza, squash, pumpkin. The pumpkins and squashes occupied a place of impor- tance in the agriculture of both North and South Amer- ica. In Mexico, they are a valuable source of edible oil seeds and the flowers and young foliage are used as vege- tables. It has been suggested that the squashes were [ 180 ] utilized for their edible seeds long before the flesh was eaten (7, 36), because the flesh of all the known wild species is scant, bitter and unpalatable. Whitaker and Bohn (216) have summarized the available information on all the cultivated species. C. ficifolia Bouché is a perennial species known from Mexico to Chile at higher altitudes. It is believed to have occurred in Peru at a very early time, but its origin is not known. Like Canavalia, this may be avery ancient cultigen which has since been largely replaced by the other and superior species. C. maxima is a South American species which did not reach Mexico in pre-Columbian times. C. mixta Pang. is believed by Whitaker and Bohn to be the same as the Mexican form of C. moschata, though the Russian workers (31, 146) believed them to be dis- tinct and list two varieties of C. mivta, one from Mexico and the other from Guatemala. C. moschata Duch. occurs from the southwestern United States to Colombia, though the South American forms seem to constitute a group distinct from those of Mexico and Central America. The origin of this species has not yet been determined, but it may be Central American. C. Pepo L. is believed to have arisen from C. texana Gray of the southern United States, or a similar wild plant. It has been suggested (86) that this species was independently domesticated in the southwestern and in the southeastern United States; while this may be so, the endemic forms known to occur in Mexico and Cen- tral America (31, 86) should be considered in any study [ 131 | of the species. (9, 12, 13, 81, 36, 118, 115, 146, 177a, 200, 206, 207, 214, 216*, 217*, 220) Dahlia coccinea Cav. (D. rosea Cav.); D. pinnata ‘av. (D. variabilis Desf.); D. Lehmannii Hieron. (D. Mawonii Saff.): Acocowdchitl, acocoth, dahlia. That dahlias had long been cultivated before their dis- covery by Europeans is indicated by the great degree of variability which these plants showed when first intro- duced into Europe, and by the great variety observed by Hernindez (98, 173, 188). The tuberous roots of the dahlias are edible, and Camp (84) believes that they were first cultivated as food plants. The huge ‘‘tree’’ dahlias, D. Lehmannii (D. Maxonii) and perhaps D. excelsa Benth., are much cultivated in southern Mexico and Guatemala, where they serve as living fences. There is no comprehensive study of these plants available, and their relationships are but poorly understood. It is al- most certain that hybridization has played a considerable part in the origin of the cultivated forms, some of which are polyploids. (84, 62, 178*, 183*, 184) Diospyros Ebenaster Retz.: Tli/zapotl, zapote prieto, black sapote. Though a popular fruit in parts of Mexico, the black sapote is found unattractive by some because of its dark colored pulp. It is related to the better known persim- mon and produces a fruit of good size. Some have thought it a native of the East Indies, but the evidence seems to indicate a Mexican origin (138). There is no recent study of this genus. (31, 95, 133, 188, 158*, 164%, 191, 199) Gossypium hirsutum L.: [zcat/, algodén, cotton. Cotton was an important fiber plant in much of Amer- [ 182 ] ica and is of special interest in that it has recently been the subject of detailed cytogenetic investigation (105). The cytological evidence shows that the New World cottons are allopolyploids (amphidiploids); that is, a type of stable hybrid which arises through the doubling of chromosomes in the progeny of an interspecific cross. One of the parents of the American allopolyploid cottons was G. Raimondu Ulbrich of Peru, or a similar (perhaps ancestral) type, while the other was an Old World type similar to G. arboreum L. While the explanation offered by Hutchinson, Silow and Stephens (105), that the Old World cotton was carried across the Pacific by man, has been the subject of some controversy, we are here pri- marily concerned with the evidence that the American cultivated cottons arose in the Andean region. The Mexican cotton, G. hirsutum, appears to have arisen, as a species, in the south Mexican-Guatemalan region. Three varieties of this species were recognized at the time of the comprehensive work cited above, but Hutchinson (104) has since been able, with more ade- quate material, to recognize seven geographic races, five of which are cultivated in Mexico, asixth occurring only in the naturalized state in coastal Yucatan. The seventh race, ‘‘Marie-Galante,’* the most primitive of the group, occurs extensively in Central America, the West Indies and northern South America. The evidence that the distinct species, G. barbadense L., of South America, and G. hirsutum, have differen- tiated from a common ancestor while under cultivation is of very great interest. This implies a considerable an- tiquity for agriculture in both hemispheres and shows that definite conclusions concerning the relationships and origins of cultivated plants can be reached only after the most careful study. (104*, 105*, 188, 164, 191, 194, 203, 206, 207, 210) [ 133 ] Helianthus annuus L.: Acahualli, chimdlatl, chim- alacatl, maiz de teja, girasol, sunflower. The sunflower is to be counted among the cultivated plants of early Mexico, as shown by the endemic varie- ties occurring there and by the descriptions of the early writers (90, 93). The cultivated sunflower, HZ. annuus var. macrocarpus (DC.) Ckll., which was also grown in the eastern, central and southwestern United States, is thought to have been derived from the wild HZ. annwus var. lenticularis (Dougl.) Ckll., which is widespread in the western United States and is limited in Mexico to the northern states. A form from Jalisco is found to resem- ble the Hopi sunflower as well as archaeological material from the eastern United States. While the history of the cultivated sunflower cannot yet be given in detail, it would appear to have arisen to the north of Mexico, where the wild form occurs and archaeological remains indicate long usage by man (185). (5, 89, 90*, 91*, 1838, 185). Hylocereus undatus (Haw.) Brit. & Rose: Pita- haya. The cacti, which are so prominent in the landscape of the more arid regions, were of great importance to the early inhabitants, and a number were cultivated (see also Opuntia). The present species, a climbing vine, is widely grown on walls and fences as an ornamental and as a source of large edible fruits. (17, 24, 26*, 28, 68, 191) Hyptis suaveolens Poit.: Chia grande, chia de Co- lima, chan. The seeds of this labiate are used in the same manner as those of Salvia hispanica and it frequently goes under the same common name, chia. Like that species, it is a somewhat variable and weedy plant and now occurs in [ 134 ] many parts of the tropics. It is known in cultivation prin- cipally in western Mexico from Sonora and Chihuahua to Oaxaca, and perhaps in San Luis Potosi. (75*, 1838, 145*) Indigofera suffruticosa Mill. (I. Anil L.): Xiu- quilitl, jiquelite, anil, indigo. This American species of indigo was apparently culti- vated in Mexico and Guatemala as a source of a blue pig- ment used extensively to dye clothing; it was also used by the Mexican women to tint their hair (98, 172). The cultivation of indigo for export was promoted by the Europeans, but has declined because of the development of synthetic dyes. The plant is now widespread as a weed and may have had a rather extensive natural distribution (33). Its cultivation probably originated in the Guate- mala-southern Mexico area. (33*, 115, 183, 155, 162, 166, 191, 192, 194*, 201) Ipomoea Batatas (L.) Poir.: Camotli, camote, ba- tata, sweet potato. The sweet potato was and is extensively grown in Mexico, the region of Querétaro being long noted for the excellence of this crop. It was not only widely grown in tropical America, but in Polynesia as well, a fact which has led to much discussion (58, 69, 99). Vavilov (207) assigns it to the south Mexican-Central American cen- ter of origin (diversity), though some other writers favor a South American origin. J. tilacea (Willd.) Choisy UI. fastigiata (Roxb.) Sweet) is thought to be the ances- tral form (58, 100), but its present distribution is too broad to pin-point the area in which it was first culti- vated. It ranges from Florida and Mexico to South America. The information presently available seems in- adequate to reach a conclusion as to the exact geographic origin of J. Batatas. (49, 58*, 69, 100*, 188, 197, 202, 207) [ 185 | Jatropha Curcas L.: Pinoncillo, physic-nut. (see also Cnidosculus. ) McVaugh (119) says of this species: ‘“The original range of J. Curcas doubtless included the tierra caliente of southern Mexico and Central America, but as it is widely planted and has been so since before the advent of Kuropeans... ”°’ Now of wide distribution in the tropics, this shrub or small tree is planted as a hedge and has medicinal uses. The seeds are said to be edible if thor- oughly roasted, but are strongly purgative when fresh. Jatropha and Spondias were used as host plants for a coc- cid insect, the aw or avin, which was cultivated in Vera- cruz for a yellowish wax which it produced. This is a ‘‘domesticated animal’’ of the Mexicans which is not well known. The wax was and is used as a varnish and also had medicinal uses. (119*, 183, 191*, 194, 200) Lagenaria siceraria (Mol.) Standl. (L. vulgaris Sér.): Tecomate, bule, bottle gourd. The gourd is still of some importance in primitive cul- tures and must have been much more so to non-ceramic groups. It is thought to be a native of the Old World, perhaps of Africa, but is known from archaeological evi- dence to have had a very wide distribution in America at an early time. J. siceraria is the only species in the genus, but the closely related genera are Old World plants. Kobiakova (109) has attempted a preliminary study of this species and believes that the American gourds are derived from Africa. However, this study is admittedly based on insufficient material and American students (72, 141) have found it to be inadequate. The dispersal of the bottle gourd in America may parallel that of cotton. (51, 72*, 108, 109, 141, 177a, 198, 215*) Lemaireocereus: see under Pachycereus. Lucuma: see Pouteria. [ 136 ] Lycopersicon esculentum Mill.: Xitématl, jito- mate, pak, tomato. The wild species of Lycopersicon are native to western South America, and it has long been thought that the cultivated tomato was brought into Europe from Peru. Jenkins (107), however, has given good reasons for be- heving that the tomato was introduced, with its Mexican name, from Mexico to Kurope. It appears that the to- mato was cultivated in southern Mexico and Veracruz, but not in the central Mexican highlands, where Physalis (q. v.) was, and still is, more important. The cherry to- mato, L. esculentum var. cerasiforme (Dun.) A. Gray, the ancestral form, is now a pan-tropic weed. It is thought by Jenkins that it may have spread as a weed from South America to Mexico where it was brought into cultivation. It may be that the development of large- fruited forms was facilitated outside of the native home, because the absence of the normal pollinating agents forced self-pollination (161). (107*, 111, 112, 142, 161) Manihot esculenta Crantz (M. utilissima Pohl), M. dulcis (J. ’. Gmel.) Paw (M. Aipi Pohl): Quauh- cdmotl, guacamote, yuca, Manioc. Manioc, a starchy root crop, was cultivated as a vege- table in Mexico, though not of such importance there as in some other areas, where it is a staple food. ‘The genus is deserving of careful study, and it is uncertain, from the available literature, whether MW. dulcis is not one of the ‘*sweet”’ varieties of M. esculenta. The plants of Mexico appear to have been largely of these ‘‘sweet’’ or less poisonous forms (4, 177). Manioc is generally thought to be of Brazilian origin, but, here too, a careful study is much to be desired. (4, 35, 71, 98, 118, 138, 177a, 179, 191, 194*, 196, 202) [ 187 | Manilkara Zapotilla (Jacqg.) Gilly (Achras Zapota L.): Ttcozdpotl, chicozapote, sapodilla. The sapodilla is widely cultivated in tropical America as a fruit tree, and the wild trees are of importance as the source of chicle, a substance known and used by the an- cient Mexicans. Its wood is extremely durable and is said to have been used by the Mayas in temple construc- tion. The name ‘‘Achras Zapota’’ is well established in the literature, though there may be good botanical and nomenclatural reasons for abandoning it (82). Further study or action of the International Botanical Congress may conserve the older and better known name. Manil- kara Zapotilla is a highly variable species, thought to be native from southern Mexico to Costa Rica. (63*, 82%, 133, 158, 191) Nicotiana Tabacum L.; N. rustica L.: Yetl, picietl, tobacco. Various species of tobacco were cultivated and used almost throughout the Americas. There is no agreement as to which species was the most important in_pre- Columbian Mexico. Setchell (182) considered N. Taba- cum to be the principal tobacco of Mexico, while Spinden (190) considers N. rustica to have been the only species cultivated there, at least in the central highlands. It is probable that N. T'abacum was known and cultivated at least in southern Mexico. The plant figured by Hernan- dez (Plate XX) is clearly NM. rustica, but the ‘‘quauhyetl”’ which he mentions may be N. T'abacum. Nicotiana has been the subject of intensive cytogenetic study with results of value both to botany and to anthro- pology. Like the New World cultivated Gossypium, both of these species of tobacco are allopolyploids. N. rustica appears to have arisen as a hybrid between the progenitors of two modern species which occur in the [ 188 ] Peruvian Andes (83, 84), while N. Tabacum probably arose in a similar manner on the eastern slopes of the Andes, perhaps in the region of Northern Argentina (84, 88). The latter, at least, is not known from the wild state (85), and it or both may have arisen in cultivation. From their distribution at the time of European discov- ery, it would seem that N. rustica, which was then the cultivated tobacco of the eastern United States, was the first to be cultivated, or at least the first to be widely dispersed. N. Tabacum, arising or entering cultivation at a later date, was superior to the earlier species, especially at lower altitudes, and largely replaced it in South Amer- ica and parts of Middle America. The European cultures served to hasten the replacement of NM. rustica, and it is still peripheral in its distribution, being grown principally as a garden tobacco in the Old World. It may be that indigenous species of tobacco were early cultivated in Mexico, as they were in much of the western United States (182), and later replaced by the South American species. A monograph of Nicotiana, by Goodspeed, is in press and should be of interest. (88, 84*, 85-88, 98, 177a, 182*, 190, 196a) Nopalea cochenillifera (L.) Salm-Dyck: Nopal nochezth, A cactus similar to Opuntia in aspect, this species is well known as the plant on which the early Mexicans cultivated the cochineal insect, the source of a highly valued red dye (26). This cactus is spineless, apparently the result of selection under cultivation. ‘This species of Nopalea is probably native to some part of southern Mexico. The present cultivation of cochineal in Oaxaca is said to utilize varieties of Opuntia ficus-indica. (24, 26*, 49, 68, 93, 158, 191*) [ 1389 ] Opuntia ficus-indica (L.) Miller; O.megacantha Salm-Dyck; O. streptacantha Lemaire; O. amyclaea Tenore: Nochth, nopalli, tuna (fruit), nopal (plant), prickly pear. Opuntia, which figures in the ancient Mexican symbol of the eagle and the nopai, is and has long been of great economic importance in parts of Mexico. The early writers were impressed by the great variety of tunas or nochtli in the Mexican gardens and markets (98, 158, 174). Several species are extensively grown for their edible fruits, which are variously prepared, and the young stems, or ‘‘joints,’’ are used as a vegetable. The plants are easily multiplied by cuttings, so that fine hybrids and varieties can be selected and propagated true to type. Most of the cultivated Opuntias are doubtless native to central Mexico. (17, 24, 26*, 28, 49, 68, 98, 158, 164, 174, 191*) Pachycereus emarginatus (DC.) Brit. & Rose: Organo, pitayo. This cactus is used for picturesque living fences in many parts of Mexico. Other columnar types, such as species of Lemaireocereus, were probably also cultivated for fruit, protection and ornament. (24, 26*, 28, 68, 191) Pachyrrhizus erosus (L.) Urban: Xicama, jicama, yam bean. The yam bean is extensively cultivated in Mexico for its turnip-like roots, which are usually eaten raw and are said to be very palatable. It appears to be native to lower elevations in central and southern Mexico and northern Central America, and is now widely cultivated and nat- uralized in many parts of the world. Other species were cultivated in South America. (17, 48*, 118, 138, 192, 202) [ 140 | Panicum sonorum Beal: Sauwi, panic grass. This is a little-known grass cultivated by the Warihio of Sonora and Chihuahua as a cereal (81). It is said to have been cultivated also by the Cocopa (41, 96). It is probably an indigenous cultigen of this area, though the species also occurs in southern Mexico. It is of particular interest since very few true cereals, other than maize, were cultivated in the New World. Two South Ameri- can grasses of similar status are now believed to be ex- tinct (177). (41, 81*, 96). Parmentiera edulis DC.: Quauvilotl, cuajilote. This tree, related to the calabash, is cultivated in many parts of Mexico for its sweet fruits which are eaten either raw or variously cooked. The tree now ranges from 'T'am- aulipas and Sinaloa to Central America and is probably native at least to southern Mexico. (17, 133, 191*, 200) Persea americana Mill.: Ahudcatl, pahua, agua- cate, avocado; P. Schiedeana Nees: Chinini, coy6. The nutritious avocado, now becoming more popular in the North, has long been an important food in Middle America. There are three groups recognized within P. americana: the Mexican race, a thin-skinned, small- fruited type centering in the Mexican highlands and sometimes designated as P. americana var. drymifolia (Schlecht. & Cham.) Blake; the Guatemalan race, prin- cipally Central American; and the ‘‘West Indian”’ race which occurs in the lowlands of Central America and northern South America. Williams (219) states that the complex of wild forms, to which P. americana is most closely related, ranges from Mexico to Honduras and probably to Costa Rica. Popenoe (152) has found what he believes to be wild trees of the Mexican race on the slopes of Orizaba and supposed wild trees of the Guatemalan [ 141 | race in central Guatemala. These probably indicate the centers of origin of these two cultivated races from varie- ties of the same species or from closely related species. The West Indian race, which may not have occurred in Mexico until recent times, seems to be more closely re- lated to the Guatemalan race than to the Mexican and probably arose somewhere in lowland Central America. Hybridization may have played some part in the develop- ment of these races and certainly has been important in the formation of the modern commercial varieties (8). (8, 21, 97*, 188, 152, 191, 194*, 219) P. Schiedeana is a distinct species ranging from south- ern Mexico to Panama, but it is rarely cultivated except near Orizaba, Mexico, where it is of considerable impor- tance (153). Phaseolus: Ayecote, bul, frijol, bean. The bean is one of the ancient American trinity, maize- bean-squash, and is nowhere more important than in Mexico. The genus is not as well known as its economic value merits. A recent Russian paper, which the present author has not seen, is cited by Carter (86). Several spe- cies of beans are cultivated in Mexico: P. acutifolius 4. Gray: Tepary. This species is less important in Mexico than it is in the southwestern United States, where its resistance to drought and heat give it a great advantage over P. vul- garis and where it has its center of diversity. The wild forms of the tepary bean occur from western Texas to Arizona and southward to Jalisco. The most probable center of origin for the cultivated plant is northwestern Mexico (36). It is now found in cultivation in Chiapas and Guatemala, but its antiquity in these regions is not known (36, 79*). [ 142 ] P. coccineus L. (P. multiflorus Willd.): Ayécotl, cimatl, scarlet runner bean. This species appears to be known in the wild state in Mexico and Guatemala (148), and is thought by the Russian workers (31) to have its center of diversity in Guatemala. It appears to be of some importance both in Mexico and in Guatemala. The fleshy root may also have been eaten (202). P. lunatus Z.: Lima bean. This species, too, is known as a wild plant in Mexico and Guatemala. Mackie (117) places its center of diver- sity and origin in Guatemala and traces three routes of diffusion from this center, each with a different type. One group of the lima bean, the northern or ‘‘Hopi”’ branch, was dispersed northward through Mexico and into the United States. A second group, the ‘‘Inca”’ branch of the species, extends into Andean South Amer- ica and includes the varieties with the largest seeds. The third group, the ‘‘Carib’’ branch, occurs in the West Indies and lowland South America. While the routes followed by these three groups may not be quite as traced by Mackie (86), the groups and trends seem to be valid. For a discussion of the synonymy of this species and its subdivisions see Van Eseltine (204). P. vulgaris L.: Bul, frijol, common bean. This is the most important bean of most of America. The center of diversity of this species is in Mexico and Guatemala (31), and wild plants of P. vulgaris have been found in this region in recent years (115, 177). The cen- ter of origin for this important species was probably some- where in the Mexican-Guatemalan area, though Burkart (38a; pp. 429, 545) reports what he believes to be wild forms of this and of the preceeding species from Argen- tina. (31, 36*, 37, 79, 115, 117*, 133, 158, 194*, 204) [ 1438 | Physalis ixocarpa Brot.: Tématl, miltématl, tomate, tomatillo, husk tomato. This is an important plant in the central Mexican high- lands where the fruits are used in stews and sauces. The fruits are quite large and the plants highly variable. In some areas two varieties are recognized, at least in the markets; one with the husk close-fitting about the fruit and the other with a larger husk. ‘These seem to be the two forms illustrated by Rose (164). The species is prob- ably native to central Mexico. The cultivated Physalis of the Guatemalan highlands is usually referred to P. pubescens 1.., but may be P. ixocarpa or some other spe- cies. P. peruviana L., of South America, is not closely related. (115, 137*, 149, 164, 167, 193) Polianthes tuberosa L.: Nardo, tuberose. The tuberose, a widely cultivated ornamental, is not definitely known as a wild plant, but it is almost certainly a native of Mexico, as are the other species of the genus. Double forms are frequent, and the plant probably has a long history of cultivation. (14, 165*, 194*) Pouteria campechiana (H BAK.) Baehni (Lucuma saleifola HBK.): Costiczdpotl, yellow sapote. This fruit tree ranges from southern Mexico to Pan- ama, but is cultivated principally in Mexico according to Popenoe. P. hypoglauca (Standl.) Baehni, a similar species, is cultivated and perhaps native from San Luis Potosi and Veracruz to northern Central America. (64*, 1538*, 191, 199) Protium Copal (Schlecht. & Cham.) Engler: Copal, pom. This tree was important among the ancient Maya for [ 144 | its resin, which was used principally as an incense in re- ligious ceremonies and which is still of considerable im- portance in highland Guatemala (194). Although the tree is reported as having been cultivated at the time of the conquest (196), the resin is now obtained from wild sources. P. Copal is restricted to southern Mexico and northern Central America and is doubtless native there. (192, 194*, 196*) Prunus serotina hr). subsp. Capuli (Cav.) McVaugh (P. Capuli Cav., P. salicifolia HBK.): Capu- lin, cereza, capulin cherry. The Mexican cherry was noted by the early European observers as equal in size and quality to the European cherries but of a different flavor (49, 98, 191). It is ap- parently native to the Mexican highlands, though early introduced into South America, where it is now much grown, especially in Ecuador (154, 194). Cobo (50) tells of its introduction into Peru. McVaugh (120) has re- emphasized the close relationship of the capulin to P. serotina subsp. serotina, a wild cherry of the United States and Mexico. (49, 50, 120*, 188, 158, 154*, 158, 191, 194*) Psidium Guajava L.: Xalxécotl, jalocote, guayaba, guava; P. Sartorianum (Berg.) Niedenzu; Arrayan, guayabilla. The guava is a shrub or small tree which is widely dis- tributed in tropical America and, though valued for its edible fruit, may occur in such abundance as to be a weed and a nuisance. P. Guajava, the best known species, ranges from Mexico to Peru. Whether or not man played a part in bringing about this range, it is difficult to deter- mine. P. Sartorianum, of Mexico and Central America, is also cultivated to some extent. [ 145 ] P. Friedrichsthalianum (Berg.) Niedenzu is largely Central American and may never have been cultivated in Mexico. The wide ranging P. guineense Sw. (P. molle Bertol.) is apparently inferior and little grown. (116, 133, 149, 150, 151*, 191*) Salvia hispanica L. (S. Chian La Llave, S. poly- stachya Ort.): Chia, chiantzozolli (see also Hyptis). The seeds of chia are widely used in Mexico to prepare a nourishing and refreshing drink which is highly es- teemed by many. They may be toasted and ground or merely stirred into water, and produce a copious muci- laginous jelly. This drink is frequently sweetened and variously flavored. The seed has also long been the source of an excellent drying oil used in painting. This species is evidently a native of central Mexico. Bukasov (81) states that it is also cultivated in Guatemala (as 8. Chia Fern.). (31, 49, 74*, 110, 183, 145*, 162) Sambucus mexicana Pres/.; Sauco, elderberry. This shrub or small tree is occasionally seen in gardens or hedgerows (personal observation) and is said to be cultivated for its small fruits (191). The available infor- mation concerning this plant is inconclusive. (17, 191) Sechium edule Sw.: Chayotli, chayote, giiisquil. The chayote has long been an important cultivated plant in Mexico. Not only are its somewhat squash-like fruits produced in abundance, but the young leaves and shoots are useful as greens, and the large, starchy roots are also eaten, only a part of the root cluster being har- vested at any one time to avoid killing the vine. The greatest diversity of this species occurs in Guatemala, where a wild form is said to occur (115, 206). (9, 56, 98*, 115, 133*, 198, 202, 206) [ 146 ] Solanum tuberosum L.: Papa, potato. While the potato seems scarcely to be considered as a pre-Columbian cultivated plant of Mexico, McBryde’s observation of a small, semi-cultivated form (‘‘S. andi- genum Juz. & Buk. forma guatemalense Buk.’’) in Guate- mala, which he considers to be a pre-Columbian intro- duction from South America, deserves attention (115). If this potato was actually pre-Columbian, it would ap- pear that cultivated JS. tuberosum was present in Middle America, but that it was not sufficiently attractive to the people of this region to spread widely. Close relatives of S. tuberosum occur in Mexico and the tubers of wild plants were utilized to some extent. Correll (60) inter- prets S. tuberosum and S. andigenum as ecological forms of one species. (60, 115) Spondias purpurea L.: XYocotl, jocote, jobo, ciruela; S. Mombin (S. lutea L.): Jobo, cozticwécotl, ciruela amarilla. The jocote is a widespread and important fruit tree in Mexico and Central America. It is easily propagated by cuttings and is therefore often grown as a hedge or fence- row plant. Spondias, like Jatropha (q. v.), was used as a host for the wax-producing coccid insect, awin. iS. pur- purea, the better known of the two species, is widespread in tropical America and probably native in parts of Mid- dle America, and perhaps elsewhere. S. Mombin is gen- erally stated to be inferior and less cultivated ; it is prob- ably a native of Central America and perhaps also of southern Mexico. Both species are highly variable and a careful study of Spondias would be most welcome. (1338, 153, 164*, 191, 192, 194*) Tagetes patula L.; T. erecta L.: Cempoalxochitl, flor de los muertos, marigold. [ 147 ] Species of Tagetes are widely used in Mexico as me- dicinal plants and at times seem to carry some ceremonial significance (17, 164). The showy and variable 7°. erecta and 7". patula must, like the dahlia, have had a long his- tory of cultivation as ornamentals in Mexico. (17, 49, 50, 98, 138%, 143, 164, 192) Taxodium mucronatum 7Ten.: Ahuéhuetl, ahue- huete, Mexican bald cypress. Though perhaps not properly listed as a cultivated plant, the noble ahuehuete surely deserves mention. It is known to have been planted by the ancient Mexican rulers in their parks and gardens, and a number of the trees thought to have been planted by Netzahualcéyotl and Moctezuma II are still living. A good account is given by Martinez. (135*, 191) Theobroma Cacao L.; T. angustifolium DC:: Cacao; T. bicolor Humb. & Bonpl.: Patachtli, pataxte. Cacao was highly valued in Mexico as the source of wocdatl, a drink somewhat different from the modern chocolate. The seeds were often used as a medium of exchange. 7 Cacao is a rather variable population and some authors have recognized several species on the basis of fruit shape and other characters. Cheesman (45) recog- nizes two main groups. The ‘‘Criollo’’ varieties, with plump seeds and pale or unpigmented cotyledons, pro- duce the highest quality of seeds, and are thought to be the original cacao varieties of Mexico and Central Amer- ica. ‘These are known to occur also in Colombia and Venezuela, but were apparently not cultivated there in pre-Columbian times. The ‘‘Trinitario’’ varieties (7°. leiocarpa Bernoulli), with flattened seeds and purple coty- ledons, have been brought into cultivation in relatively recent times from wild Amazonian trees. The early de- [ 148 ] scriptions by Sahagun and Cobo (50, 174), however, seem to refer to ‘“Trinitario”’ varieties in Mexico rather than to ‘‘Criollos’’; perhaps the seed of 7. angustifolium re- sembles that of the ‘‘Trinitario’’ forms. The present commercial cacaos of America are mostly hybrids involv- ing both groups. Cheesman places the ‘‘center of origin”’ of 7. Cacao in the upper Amazon. This may have been a center of dispersal of wild 7. Cacao, but it was clearly brought into cultivation somewhere in Central America or southern Mexico. For the purpose of the anthropolo- gist, a distinction must be drawn between the origin as a cultivated plant and the history of the plant before cultivation or association with man. The latter subject is, however, of great interest to botanists. T. angustifolum, of Central America and southern Mexico, is said to be cultivated to a considerable extent in Mexico, particularly about Soconusco, Chiapas, and apparently produces a good grade of cacao (191, 194). 7" bicolor ranges from southern Mexico to northern South America and is generally considered inferior in quality to 7’. Cacao, but is cultivated and used in some areas. (44, 45%, 46, 49, 50, 70, 71, 133, 191*, 194*) Tigridia pavonia (L.f.) Kerr.: Ocelowéchitl, cacém- itl, cacomite, tiger-flower. Now known principally for its beautiful flowers, the tiger-flower was cultivated by the Mexicans for its edible corms as well as for ornament. It is evidently a native of central Mexico. (50, 98, 133*, 143) Vanilla planifolia Andr. (VY. fragrans (Salisb.) Ames): Tlhilwxéchitl, vainilla, vanilla. Vanilla was known to the Mexicans and used especially for flavoring chocolate. This vine occurs from tropical Mexico to northern South America, but its pre-Colum- [ 149 | bian use extended only south to Costa Rica (30). In Mexico its cultivation has long centered in southern Veracruz, and it was probably little cultivated in other areas. (30*, 59*, 60a*, 133) Yucca elephantipes Regel: Iczotli, izote, yucca. This large tree is much planted for hedges, especially in Central America, where it is apparently introduced. The flowers are valued as a vegetable. Standley (192) considers it to be a native of Veracruz. Y. aloifolia L. is also cultivated at times. (115, 133, 191*, 192, 198, 194*) Zea Mays L.: Tlaolli, centli, maiz, maize, Indian corn. Maize has long been the most important crop plant for most of the Americas and certainly retains that title in Mexico. The interest in this plant has been such that an overwhelmingly voluminous literature has developed concerning the genetics, cytology, morphology, relation- ships, importance and origin of this cereal. There remain, nonetheless, many unanswered questions about maize and it will doubtless provide a fertile field for investiga- tors for many years to come. The nearest ally of maize is teosinte, Muchlaena (Zea) mexicana, Which occurs apparently as a wild plant in Guatemala and Chiapas and as a weed of cultivated areas in many parts of Mexico. Teosinte has at times been thought to be the wild ancestor of maize, but this idea now has very few adherents. Mangelsdorf and Reeves (129) have suggested that teosinte is actually of hybrid origin, maize and a species of T'ripsacum, a more dis- tantly related grass, being the parent species. Stebbins (195, p. 277) suggests that a cross might have been more readily effected between primitive maize and some ex- tinct species of T'ripsacum, with a lower chromosome [ 150 | number, than between the modern representatives of these genera, which cross only with difficulty. Though there is not total agreement as to the origin of teosinte, there can be little doubt that it has played a major role in the evolution of modern maize, through hybridization and introgression. Mangelsdorf and his collaborators (123-131) have supported the hypothesis that the most primitive maize is both tunicate (a pod corn) and a pop- corn. This seems to have been supported by archaeologi- cal and other studies (181, 213). Mangelsdorf and Reeves further hypothesized that the ancestral region for maize should be sought in the South American lowlands, where some primitive forms are still found. More recent evi- dences seem to indicate a peripheral nature for these South American types. There is paleontological and archaeological evidence that maize is of greater antiquity in North America than in South America (16, 77, 181). Maize pollen was recovered at a depth of seventy meters in the Valley of Mexico; this is interpreted as represent- ing wild maize growing in that region in the Pleistocene (16, 67,77,181). Archaeological material from Bat Cave, New Mexico is of particular interest (181). The maize from the lowest level of these deposits (about 2500 to 2000 B.C.) is a very small-eared form and appears to be both a pod corn and a popcorn. Cobs in later levels are larger and show evidences of introgression from teosinte. Wellhausen et al (213) have recently published a val- uable book on the races of maize in Mexico. They have recognized and characterized at least twenty-five dis- tinct races which are grouped into four classifications: ‘Ancient Indigenous,’” primitive popcorns of relictual distribution, two of which are weakly tunicate; ‘*‘ Pre- Columbian Exotic,’ races believed to have been intro- duced into Mexico from the south in prehistoric times: **Prehistoric Mestizos,’* derived from races of the first Leet} two categories through hybridization; and ‘* Modern Incipient,’’ less stable races, apparently of relatively re- cent origin and dispersal. With the basis given by these authors, it should eventually be possible to make an ex- tensive study of archaeological remains of maize from Mexico and adjacent areas and to correlate in time and space these races with their phylogeny and with other culture traits. Indeed, a better framework for American prehistory could scarcely be desired. The exact origin of maize is yet unknown and will probably continue to be a source of speculation and con- troversy for some time to come. At present, the evi- dence seems to favor the Mexican-Central American area as the home of this cereal. The Chiapas-Guatemala area may have been the center of origin for races affected by introgression from teosinte (127). The Andean area has also been a center of dispersal, especially, it would seem, for large-kerneled races, including flour and sweet forms. (6*, 11, 16, 27, 88, 65, 66, 67, 76, 98, 94, 128, 124, 125%, 127*, 128, 129*, 180, 181, 157, 160, 195, 211, 212*, 213%, 218) THE GEOGRAPHIC ORIGINS oF MEXICAN CULTIVATED PLANTS On the basis of the data which have been discussed, it is possible to arrange the cultivated plants tentatively according to geographic origins. The divisions which are recognized within the Mexican-Central American region do not represent well defined centers or culture com- plexes, but rather geographic regions with some ecologi- cal character. Some plants probably overlap two of these subdivisions in their origins and a few are doubtless as- signed to the wrong subdivision. The divisions and assign- ments are, of course, distinctly tentative and subject to revision. [ 152 ] 1. Mexican-Central American Area 71 species ' a. Plants whose origins within this area are as yet diffi- cult to localize Cucurbita mixta C. moschata Hylocereus undatus Phaseolus vulgaris _ 8 species Theobroma Cacao T. angustifolium T.. bicolor Zea Mays b. Northern Mexico and the adjacent United States Cucurbita Pepo Helianthus annuus ce. Central Mexico (w=western Mexico) Agave atrovirens A, latissima A. mapisaga Amaranthus leucocarpus Annona diversifolia (w?) Casimiroa edulis Crataegus pubescens Dahlia coccinea D. pinnata Diospyros Ebenaster Hyptis suaveolens (w) Opuntia amyclaea 4 species Panicum sonorum Phaseolus acutjfolius 24 species O. ficus-indica O. megacantha O. streptacantha Pachycereus emarginatus Persea americana (Mexican race) Physalts ixocarpa Polianthes tuberosa Prunus serotina subsp. Capult Salvia hispanica Tagetes erecta T. patula Tigridia pavonia d. Southern Mexico and northern Central America (except for Guatemalan highlands). This is doubt- less the least well defined and most inclusive of the subregions delineated here. (y=Yucatan, m= southern Mexico, ca=Central America) 27 species Agave fourcroydes (y) A. sisalana (y) Annona purpurea ? Brosimum Alicastrum Byrsonima crassifolia Calocarpum mammosum C. viridis (ca) Carica Papaya Chamaedorea Tepejilote C. Wendlandiana Cnidosculus Chayamansa (y) Crescentia Cujete Indigafera suffruticosa Jatropha Curcas (m) Manilkara (Achras) Zapotilla Nopalea cochenillifera (m) Pachyrrhizus erosus (m) Parmentiera edulis Persea Schiedeana Pouteria campechiana P. hypoglauca Protium Copal Psidium Sartorianum Spondias Mombin (ca) S. purpurea Vanilla planjfolia (m) Yucca elephantipes (m) 1 Note that races of the avocado, Persea americana, are listed under two subregions. [ 158 ] e. Guatemalan Highlands. . . . . . . . 8 species Amaranthus cruentus Persea americana (Guatemalan race) Casimiroa Sapota Phaseolus coccineus Crotalaria longirostrata P. lunatus Dahlia Lehmannii Sechium edule 2. Andean Area. .......... . . 6 Species Annona Cherimolia Lycopersicon esculentum (or cultivated independently in Mexico) Chenopodium Nuttalliae Nicotiana rustica Gossypium hirsutum N. Tabacum 3. Lowland South American (Brazil-Paraguay) Region 3 species Ananas comosus Arachis hypogaea (post-Columbian in Mexico?) Manihot esculenta 4, Plants which are at present difficult to assign to any of the above areas, with the author’s guesses in paren- theses .......... =... . . . 8 Species Bixa Orellana Ipomoea Batatas Canavalia ensiformis (Mex.) Lagenaria siceraria (via Andean Capsicum annuum region?) C. frutescens (Mex.) Psidium Guajava (Andean?) Cucurbita ficifolia From the above lists it will be seen that over eighty species of plants are considered as having been cultivated in Mexico before Kuropean contact. Some may think this number to be excessive; it must be recalled, how- ever, that agriculture was developed to a high degree in parts of Middle America and that the ecology of the area is conducive to a great variety of crop plants. In almost all instances, I believe, a very good case can be made for considering these plants to have been cultivated in pre- Columbian Mexico. The greater number are indigenous either to Mexico or to adjacent areas. Some of the in- digenous plants are of secondary importance, but the list also includes plants of such prominence as maize, beans, agave, avocado, and species of amaranth and squash. Six species, including cotton and tobacco, are believed to [ 154 ] be of Andean origin. Though the species of cotton in Mexico is not the same as that of the Andean area, it is believed that they both diverged from a common ancestor under cultivation and that cotton culture in the New World stemmed from the Andean area. Two or three plants may be considered as having diffused from lowland South America under cultivation. There remain eight species which the author hesitates to assign to any of these areas. All, of course, are now very widespread, and this portion is not completely comparable to the rest of the list, which includes many local types. It may be ex- pected that several, if not most, will prove to be natives of the Mexican region. Mangelsdorf and Reeves (129), after a survey of the American cultivated plants, conclude that there had been no direct interchange of crop plants between the Andean and the Middle American areas, but that both regions had received some plants from lowland South America, thus accounting for the relatively few species that occur in both areas. Hutchinson, Silow and Stephens (105), on the other hand, hypothesize an early exchange of plants between these two centers, with little or no sub- sequent exchange. It is doubtful that ‘‘direct’’ transfer of plants between Mexico and the Andes occurred at any time before European contact. Plants traveled by grad- ual diffusion and those which occurred in both regions were mostly the ones which could and would be grown in the intervening areas. The indigenous tuber crops of the Andes would do poorly at lower elevations, and there would seem to be little incentive for their cultivation where Manihot and Ipomoea were both available and better adapted. The different Andean plants in Mexico were probably not of contemporaneous introduction. Cotton was clearly early, as was probably Chenopodium, if it is Andean. Nicotiana rustica was much earlier than [ 155 ] N. Tabacum, and Lycopersicon may have been fairly late. If Annona cherimola, Chenopodium Nuttalliae and Sola- num tuberosum are all truly Andean in origin and _ pre- Columbian in Mexico and Guatemala, a good case might be made for direct transport over a considerable distance ; all are plants that thrive only in the temperate highlands. The plants which are definitely of lowland South American origin are few and of relatively little impor- tance. Ananas is almost certainly South American and pre-Columbian in Mexico, Manihot is of less certain ori- gin but also pre-Columbian in Mexico, although both may be relatively late. Arachis is certainly late and per- haps post-Columbian. While some of the doubtful domesticates may ulti- mately prove to be exotic, the bulk of the plants culti- vated in the Mexican area are indigenous, and the native agriculture has retained its character to a remarkable de- gree under the impact of European domination (81). It must be concluded that the complex of domesticated plants which developed in Middle America constituted a well balanced and stable system of agriculture, which is thereby resistant to extensive establishment of new cul- tivated plants (with domestic animals, it was quite other- wise). The inertia of cultural patterns is no doubt also involved, but many other areas have proven to possess far less stable systems. A really comparable list of the pre-Columbian culti- vated plants of the Andean region is not available, but would be of great interest. A nearly or quite equal num- ber of cultivated species could probably be enumerated for this area which is noted for its endemic crops. It has been argued, on competent grounds, that agri- culture has had several independent centers of origin in the New World (86, 175). While the origins and present distributions of the cultivated plants cannot alone throw [ 156 | much light on this, the data presented here in no way conflict with multiple and independent origins for Amer- ican agriculture. Archaeological evidence seems at pres- ent to support this hypothesis and may be expected to add important information in the future (19, 20, 86, 181). Some writers (9) have upheld a very great antiquity for New World agriculture. The information on this is yet fragmentary, but here too, the joint efforts of archaeology and botany are doing much to clarify the picture and indicate a quite respectable age for agriculture in this hemisphere (19, 131). With attention being focused on the ultimate origins of agriculture, more thought is being given to the man- ner in which plants first entered cultivation. However, less is being heard of the ingenious savage who decided to return seed or roots to the soil and thereby revolution- ized culture overnight (I suspect that early man had an adequate understanding of the seed long before he used this knowledge in agriculture). It seems more probable that the development of agriculture, whatever its pat- tern, was a gradual process. Some authors (9, 10, 176, 178, 205, 206) have drawn attention to the weedy ‘‘camp- follower’’ element among our cultivated plants, those plants which might be expected to invade camp sites and trash heaps and to be encouraged by man. The hypo- thesis of a gradual development from such a nearly com- mensal relationship has much to recommend it. The al- liance was probably never purely commensal, but to some degree mutualistic or symbiotic from the beginning, with both members profiting from the association. Instances of semi-cultivation in peripheral areas of North America are described elsewhere (42), and several are known for tropical America. A considerable number of New World cultivated plants are such ‘‘weedy’’ types and many are still to be found on the trash heap; such plants include: [ 157 | Amaranthus, Capsicum, Chenopodium, Cucurbita, Heli- anthus, Lycopersicon, Psidium, Salvia and Solanum. Many other plants, such as the cacti and fruit trees, must have been selectively encouraged before agriculture be- gan, just as they are at the present day. Anderson, in particular, has emphasized the role played by man in the dispersal of useful plants into new areas. This allows hybridization between varieties, and hybrid- ization and introgression between related species. Both of these processes serve at times to promote ‘‘weediness’’ and always to increase the variability of the populations involved, a very important factor in the origin and de- velopment of cultivated plants (7, 10, 126). This rela- tionship between man and useful plants doubtless began before agriculture, when man first carried a fruit with him for a while before eating it, thus discarding the seed in a new area. One cannot discuss the origins of New World agri- culture without mentioning the question of trans- Pacific cultural contact. Evidence has been presented concern- ing several cultivated plants, notably Gossypium, Ama- ranthus, Ipomoea, Lagenaria and Cocos, which has been interpreted as indicating early cultural contact between southeast Asia and tropical America (69, 105, 178). This evidence has been discussed at length and opinion is sharply divided (89, 40, 108, 128, 188, 189). This con- troversy has brought forth, on one extreme, vituperative and dogmatic insistence that no such contact has or could have occurred, and, on the other hand, the wildest flights of faney, in which all high cultures are derived directly and apparently recently from a single source. I have not cited the extremes on the more fanciful side as they are based largely on non-botanical arguments, if any. They are, however, often associated with other less imaginative views, much to the detriment of the latter. Though the [ 158 |] evidence now available may not prove early cultural con- tact across the Pacific, it is highly suggestive of such contacts. Our knowledge of cultivated plants and of the earlier cultures is not yet so perfect that we may wrap our minds in the comforting cloak of dogma (nor should it ever be thought so). The evidences should be consid- ered on their own merits; further unbiased studies along these lines cannot fail to be of great value, regardless of the light they may or may not throw on the question of early contacts between the New World and the Old. Whatever the relationship between these two areas, the agriculture of the Americas is, in its broad outlines, dis- tinct both in crops and in techniques from that of the Old World, and particularly so from that of temperate Eurasia. In conclusion, the geographic origins of most culti- vated plants can be stated only in regional terms and can never be very narrowly localized. Of the chronological origins, we can scarcely speak yet even in general terms. The cultivated plants are not easy subjects for compre- hensive study, and both botanists and anthropologists have too often neglected them. The outlook, however, is very good. With the use of newer tools (carbon 14 and cytogenetics, for example) and the careful reapplication of the older tools of the many phases of both anthropol- ogy and botany, there is every reason to believe that a good culture history can be developed for the most basic and important material culture traits of man, his culti- vated plants. SUMMARY The problems and importance of studying cultivated plants are considered, and the criteria used for determin- ing the center of origin of a cultivated plant are reviewed. The cultivated plants of pre-conquest Mexico are enu- [ 159 | merated and discussed, with special attention given to the botanical literature and the question of geographic origins. The high development of agriculture and the great diversity of habitat in Mexico are favorable for the development of a great variety of cultivated plants. Over eighty species are listed, which may be summarized as to origins as follows: 1. Mexican—Central American region . . . 71 species la. Origin not further localized (including maize, beans and cacao) 8 species 1b. Northern Mexico and adjacent areas (sunflower and a pump- kin species) 4 species le. Central Mexico (amaranth, chia, Opuntia and a race of the avocado) 24 species id. Southern Mexico and lowland Central America (yam bean, indigo and papaya) 27 species le. Guatemalan highlands (lima bean, chayote and a race of the avocado) 8 species 2. Andean area (cotton and tobacco) . . . . 6 species 3. Brazil-Paraguay region (pineapple) . . . 8 species 4. Uncertain (chile peppers and sweet potato) 8 species It will be noted that most of the Mexican cultivated plants are native to Middle America, but that some spe- cies have been received through diffusion from the An- dean area and from lowland South America. The data do not conflict with the hypothesis of several independent centers of origin for New World agriculture. The origin of agriculture as such is discussed and a gradual develop- ment through what may be termed a ‘‘commensal’”’ pat- tern is upheld. The botanical evidences concerning trans- Pacific cultural contact are briefly noted. It is felt that real progress is being made in the study of cultivated plants and that cooperation between different disciplines promises continued and valuable progress in the future. [ 160 ] Or -~l 16. Vis 18. 19. 20. BIBLIOGRAPHY Aellen, Paul. 1929. Chenopodium amaranticolor Coste & Reynier, «ey Ch. Quinoa Willd., ..., und XCh. Reyniert Ludwig & Aellen. Ber. Schweiz. Bot. Ges. 38: 5-28. 1929. Beitrag zur Systematik der Chenopodium—Arten Amerikas, vorwiegend auf Grund der Sammlung des United States National Museums in Washington, DC. Fedde, Rep. Spec. Nov. 26: 31-64, 119-160. —— & Theodor Just. 1943. Key and synopsis of the American species of the genus Chenopodium L. Am. Mid]. Nat. 30: 47-76. Alzate y Ramirez. 1884. Guacamote. La Naturaleza 7 (app.): 3. Ames, Oakes. 1939. Economic Annuals and Human Cultures. Cambridge, Mass. Anderson, Edgar. 1946. Maize in Mexico, A preliminary sur- vey. Ann. Mo. Bot. Gard. 33: 147-247. 1949, Introgressive Hybridization. New York. ——. 1950. Variation in avocados at the Rodiles Plantation. Ceiba 1: 50-55. —, 1950. An Indian garden at Santa Lucia, Guatemala. Ceiba 1: 97-108. 1952. Plants, Man and Life. Boston. — R. H. Barlow. 1943. The maize tribute of Moctezuma’s empire. Ann. Mo. Bot. Gard. 30: 413-420. Bailey, L. H. 1929. The domesticated Cucurbitas Herbarum 2: 63-115. 1943. Species of Cucurbita. Ibid. 6: 266-322. Baker, J. G. 1888. Handbook of the Amaryllideae. London. Baker, Kenneth F. and J. L. Collins. 1939. Notes on the dis- tribution and ecology of Ananas and Pseudananas in South Amer- ica. Am. Journ. Bot. 26: 697-702. Barghoorn, Elso S., Margaret K. Wolfe and Kathryn H. Clisby. Fossil maize pollen from the Valley of Mexico. Bot. Mus. Leafl. Harvard Univ., to be published. Batalla de Rodriguez, M. A. 1942. Estudio de las plantas culti- vadas en la regi6n de Izticar de Matamoros, Puebla. An. Inst. Biol. Mex. 13: 463-491. Beals, Ralph L. 1946. Cheran: A Sierra Tarascan Village. Smithson. Inst. Soc. Anthrop. 2: 1-225. Bennett, Wendell C. and Junius B. Bird. 1949. Andean Cul- ture History. Am. Mus. Nat. Hist. Handbook Ser. 15: 1-319. Bird, Junius B. 1948. Preceramic cultures in Chicama and Viru. Mem. Soc. Amer. Archaeol. 4: 21-28. [ 161 ] I. Gentes 21. 31. 82. 39. 40. Blake, S. F. 1920. A preliminary revision of the North Ameri- can and West Indian avocados (Persea spp.). Journ. Wash. Acad. Sci. 10: 9-21. Brand, Donald D., ed. 1936. Symposium on prehistoric agricul- ture. Univ. N. M. Bull. Anthropol. Ser. 1 (5): 1-72. 1939. The origin and early distribution of New World cultivated plants. Agr. Hist. 13: 109-117. Bravo H., Helia. 1934. Las cactéceas entre los antiguos mexi- canos. An. Inst. Biol. Mex. 5: 139-148. 1934. Estudio botanico acerca de las solanaceas mexicanas del género Capsicum. Ibid. 5: 303-321. 1937. Las Cactéceas de México. Mexico, D.F. Brieger, F. G. 1949. Origem e domesticacao do milho. Lilloa 20: 37-44. Britton, N. L. and J. N. Rose. 1919. The Cactaceae. Carn. Inst. Wash. Publ. 248, 4 vols. Bruman, Henry J. 1944. Some observations on the early history of the coconut in the New World. Acta Americana 2: 220-243, 1948. The culture history of Mexican vanilla. Hispanic Amer. Hist. Rev. 28: 360-376. Bukasov, S. M. 1981. The cultivated plants of Mexico, Guate- mala and Colombia. Bull. Appl. Bot. Gen. Pl. Breed., Suppl. 47: 1-553 (Russian 1-469; English summary 470-553). Burkart, Arturo. 1939. Estudios sistematicos sobre las Legumi- nosas-Hedisareas de la Reptblica Argentina y regiones adya- centes. Darwiniana 3: 117-302. ——, 1942, Las especies de Indigafera de la flora argentina. Darwiniana 4: 145-178. 1952. Las Leguminosas Argentinas, 24 edicién. Buenos Aires. Camp, W. H. 1947. The World in your Garden. Nat. Geogr. Mag. 92: 1-65. de Candolle, Alphonse. 1885. Origin of Cultivated Plants. New York. Carter, George F. 1945. Plant geography and culture history in the American Southwest. Viking Fund Publ. in Anthrop. 5: 1-140. ——, 1946. Origins of American Indian agriculture. Amer. Anthrop. 48: 1-21. 1948. Sweet corn among the Indians. Geogr. Rev. 38: 206-221. 1950. Plant evidence for early contacts with America. S. W. Jour. Anthrop. 6: 161-182. ——, 1958. Plants across the Pacific. Mem. Soc. Amer. Arch- aeol. 9: 62-71. [ 162 | 60. 60a, Castetter, Edward F. and Willis H. Bell. 1942. Pimaand Papa- go Indian Agriculture. Albuquerque. . 1951. Yuman Indian Agriculture. Albuquerque. ——,—— and Alvin R. Grove. 1938. The early utilization and the distribution of Agave in the American Southwest. Univ. N.M. Bull. 335: 1-92. Cheesman, E. E. 1939. Recent botanical researches in cacao. Tropical Agric. 16: 4-7. 1944. Notes on the nomenclature, classification and possi- ble relationships of cacao populations. Jbid. 21: 144-159. Chevalier, A. 1946. Révision du genre Theobroma, d’aprés 1’ Herbier du Muséum national d’ Histoire naturelle de Paris. Rev. Bot. Appl. Agr. Trop. 26: 265-285. 1948, Genése et origine géographique des Arachides cul- tivées et possibilité de leur amélioration. /bid. 28: 514-519. Clausen, Robert T. 1944. A botanical study of the yam beans (Pachyrrhizus), Cornell Univ. Agric. Exp, Sta. Mem. 264: 1-38. Clavigero, Francisco J. [1780] 1882. Breve noticia de las plantas y animales de México (trans. to Spanish by Francisco P. Vasquez). La Naturaleza 6 (app.): 5-97. Cobo, P. Bernabé. [1653] 1890. Historia del Nuevo Mundo, vols. 1 & 2, Soc. de Bibliéfilos Andaluces. Cogniaux. A. and H. Harms. 1924, Cucurbitaceae—Cucurbiteae —Cucumerinae. (in) A. Engler. Das Pflanzenreich 1V. 275 (2): 1-246, Collins, J. L. 1948. Pineapples in Ancient America. Scientific Monthly 67: 372-377. 1949. History, taxonomy and culture of the pineapple. Econ. Bot. 3: 335-359. 1951. Antiquity of the pineapple in America. 5S. W. Journ. Anthrop. 7: 145-155. Costantin [J.] and [D.] Bois. 1910. Sur les graines et tubercules des Tombeaux Péruviens de la Période Incasique. Rev. Gén. de Bot. 22; 242-265. Cook, O. F. 1901. The Chayote: a tropical vegetable. U.S.D.A. Div. Bot. Bull. 28: 1-31. ——, 1925. Peru as a center of domestication. Journ. Hered. 16: 33-46, 95-110. Cooley, J.S. 1951. The sweet potato—Its origin and primitive storage practices. Econ. Bot. 5: 378-386. Correll, Donovan S. 1944. Vanilla: Its history, cultivation and importance. Lloydia 7: 236-264. ——-and ——, 1952. Section Tuberarium of the genus Solanum of North and Central America. U.S.D.A. Monogr. 11: 1-243. ——. 1953. Vanilla—Its botany, history, cultivation and eco- nomic import. Kcon. Bot. 7: 291-358. [ 163 ] 68. 69. Crane, M. B. 1940. The origin and behaviour of cultivated plants. (in) J. Huxley ed. The New Systematics, 529-547. Oxford. ——, 1950. The origin and improvement of cultivated plants. Journ. Roy. Hort. Soe. 75: 427-435, 465-474, Cronquist, Arthur. 1945. Studies in the Sapotaceae—IV. The North American species of Mani/kara. Bull. Torr, Bot. Club 72: 550-562. ——, 1946. Studies in the Sapotaceae—II. Survey of the North American genera. Lloydia 9: 241-292. Cutler, Hugh C. 1951. The geographic origin of maize. Chron, Bot. 12: 167-169. ——, 1951. The oldest corn in the world. Chicago Nat. Hist. Mus. Bull. 22 (2): 4-5. Deevey, Edward S. 1944. Pollen analysis and Mexican archae- ology: an attempt to apply the method. Amer. Antiq. 10: 135- 149, Diguet, Léon. 1928. Les Cactacées utiles du Mexique. Archives d’Histoire Naturelle 4: 1-551. Dixon, Roland B. 1932. The problem of the sweet potato in Polynesia. Amer. Anthrop. 34: 40-66. Ducke, A. 1940. As espécies brasileiras de cacau (Género Theo- broma L.) na botanica sistematica e geografica. Rodriguésia 4: 265-276. ——, 1946. Plantas de cultura precolombiana na Amazonia Brasileira. Notas sdbre as espécies ou formas esponténeas que supostamente lhes teriam dado origem. Boletim técnico do Inst. Agron. do Norte 8: 1-24. Eames, Arthur J. and Harold St. John. 1943. The botanical identity of the Hawaiian Ipu nui or large gourd. Am. Journ. Bot. 30: 255-259. Edwards, E. E. and W. D. Rasmussen, 1942. A_ bibliography on the agriculture of the American Indians. U.S.D.A. Misc. Publ. 447: 1-107. Epling, Carl. 1989. A revision of Salvia, subgenus Calosphace. Fedde Rep. Spec. Nov. Beihefte 110: 1-383, or Pub. Univ. Calif. L. A. Bio. Sci. 2: 1-383. ——, 1949. Revisién del género Hyptis. Revista del Museo de La Plata n.s. 7: 153-497. Emerson, R. A. 1953. A preliminary survey of the milpa system of maize culture as practiced by the Maya Indians of the north- ern part of the Yucatan Peninsula. Ann. Mo. Bot. Gard. 40: 51-62. Ewing, Ann. 1952. 25,000 year old corn pollen. Sci. News Let- ter 61: 335. Foex, Felix. 1908. Algunas anondceas frutales de México. Es- tacion Agricola Central. Boletin 9: 1-38. [ 164 | 84, 96. 97. Freemen, George F. 1913. The Tepary, a new cultivated leg- ume from the Southwest. Bot. Gaz. 56: 395-417. Fries, Robert E. 1931. Revision der Arten einiger Anonaceen- Gattungen II. Acta Horti Bergiani 10: 129-341, Gentry, Howard S. 1942. Rio Mayo Plants. Carn. Inst. Wash. Pub. 527: 1-328. Gilly, Charles L. 1943. Studies in the Sapotaceae, II. The Sapodilla-Nispero Complex. Trop. Woods 73: 1-22. . Goodspeed, T. H. 1934. Nicotiana phylesis in the light of chro- mosome number, morphology and behavior. Univ. Calif. Publ. Bot. 17: 369-398. ——. 1942. The South American genetic groups of the genus Nicotiana and their distribution. Proc. 8th Amer. Sci. Cong. 3: 231-238. ——, 1944, Nicotiana Arentsii—A new, naturally occurring am- phidiploid species. Proc. Calif. Acad. Sci. IV, 25: 291-306. ——, 1945. Cytotaxonomy of Nicotiana. Bot. Rev. 11: 533- 092. —— and Muriel V. Bradley. 1942. Amphidiploidy. Bot. Rev. Si 271-316, ——and R. E. Clausen. 1928. Interspecific hybridization in Nicotiana VIII. The sylvestris-tomentosa- Tabacum hybrid triangle and its bearing on the origin of Tabacum. Univ. Calif. Publ. Bot. 11s 245-256. Heiser, Charles B., Jr. 1945. The Hopi sunflower. Bull. Mo. Bot. Gard. 33: 163-166. ——, 1946. A “‘new’’ cultivated sunflower from Mexico. Ma- drofio 8: 226-229. ——, 1951. The sunflower among the North American Indians. Proc. Amer. Phil. Soc. 95: 432-448. —— and P. G. Smith. 1948. Observations on another species of cultivated pepper, Capsicum pubescens R. & P. Proc. Amer. Soc. Hort. Sci. 52: 331-335. .——. and ——. 1953. The cultivated capsicum peppers. Econ. Bot. 7 +-214-227. Hernandez, Francisco. [1651] 1942-1946. Historia de las Plantas de Nueva Espafia, 3 vols., Mexico, D.F. Hernandez Xolocotzi, Efraim. 1949. Maize granaries in Mexico. Bot. Mus. Leafl. Harvard Univ. 13: 153-211. Hiern. W. P. 1873. A monograph of Ebenaceae. Trans. Cam- bridge Phil. Soc. 12 (1): 27-300. Hitchcock, A.S. and Agnes Chase. 1910. The North American species of Panicum. Contrib. U.S. Nat. Herb. 15: 1-396. Hodgson, Robert W. 1950. The Avocado—A gift from the Mid- dle Americas. Econ. Bot. 4: 253-293. [ 165 | 98. 99. 100. 101. 102. 108. 108. 109. 110. 112. 113. 115. Hoover, L. G. 1923. The Chayote: Its culture and _ uses. U.S.D.A. Dept. Cire. 286: 1-11. Hornell, J. 1946. How did the sweet potato reach Oceania? Jour. Linn. Soc, 53: 41-62. House, Homer D. 1908. The North American species of the genus Ipomoea. Ann. N.Y. Acad. Sci 18 (6): 181-263. Howes, F. N. 1930. Chenopodium Nuttalliae, a Mexican cereal. Kew Bull. 1930: 332-333. Hunziker, Armando T. 1943. Las especies alimenticias de Amaranthus y Chenopodium cultivadas por los indios de América. Rev. Argent. Agron. 10: 297-354. Husted, Ladley. 1936. Cytological studies on the peanut, Arachis. 11. Chromosome number, morphology and behavior, and their application to the problem of the origin of the culti- vated forms. Cytologia 7: 396-423. Hutchinson, J. B. 1951. Intra-specific differentiation in Gos- sypium hirsutum. Heredity 5: 161-193. 5. ——, R. A. Silow and S. G. Stephens. 1947. The evolution of Gossypium. London. . Irish, H. C. 1898. A revision of the genus Capsicum with special reference to garden varieties. Ninth Ann. Report Mo. Bot. Gard., 53-110. . Jenkins, J. A. 1948. The origin of the cultivated tomato. Econ. Bot. 2: 379-392. Kelly, Isabel. 1951. The bottle gourd and Old World contacts. (in) Homenaje al Dr. Alfonso Caso, 207-214. Mexico, D.F. Kobiakova, J. A. 1930. The bottle gourd, Lagenaria vulgaris Sér.—a relict crop. Bull. Appl. Bot. Gen. Pl. Breed. 23 (3): 475-520. (Russian 475-518, English summary 519-520). See also, Obolentzeff. 1943. Review of the above. Rev. Argent. Agron, 10: 289-298. de La Llave, Pablo. 1884. Sobre una especie nueva de Salvia. La Naturaleza 7 (app.): 77-80. . Luckwill, Leonard C, 1943. The genus Lycopersicon. An his- torical, biological and taxonomic survey of the wild and culti- vated tomatoes. Aberdeen Univ. Studies 120: 1-44, ——, 1948. The evolution of the cultivated tomato. Journ. Roy. Hort. Soc. 68: 19-25. Lundell, C. L. 1939. Plants probably utilized by the Old Em- pire Maya of Peten and adjacent lowlands. Pap. Mich. Acad. Sci. Arts and Letters 24 (1): 37-56. . ——. 1945. The genus Cnidosculus in Mexico: new species and critical notes. Bull. Torr. Bot. Club 72: 319-334. McBryde, Felix W. 1947. Cultural and historical geography of southwest Guatemala. Smithson. Inst. Soc. Anthrop. Pub. 4: 1-184. [ 166 | 136. 137. MacCaughey, Vaughan. 1917. The guavas of the Hawaiian Islands. Bull. Torr. Bot. Club 44: 513-524. Mackie, W. W. 1943. Origin, dispersal, and variability of the lima bean, Phaseolus lunatus. Hilgardia 15: 1-29. MeVaugh, Rogers. 1944. The genus Cnidosculus: generic limits and intrageneric groups. Bull. Torr. Bot. Club 71: 457-474. . ——. 1945. The genus Jatropha in America: principal intra- generic groups. Ibid. 72: 271-294. . ——. 1951. A revision of the North American black cherries (Prunus serotina Ehrh., and relatives). Brittonia 7: 279-315. Maldonado Koerdell, Manuel. 1941. Los jardines botanicos de los antiguos mexicanos. Rev. Soc. Mex. Hist. Nat. 2: 79-84. . ——. 1946, La botanica. (in) Jorge A. Vivé, ed., México Pre- hispanico, 709-716, Mexico, D.F. Mangelsdorf, Paul C. 1947. The origin and evolution of maize. Advances in Genetics 1: 161-207. . ——. 1948. The role of pod corn in the origin and evolution of maize. Ann. Mo. Bot. Gard. 35: 377-406. . ——. 1950. The mystery of corn. Scientific American 183: 20-24. . ——. 1952. Evolution under domestication. Amer. Nat. 86: 65-77. . —— and James W. Cameron. 1942. Western Guatemala a secondary center of origin of cultivated maize varieties. Bot. Mus. Leafl. Harvard Univ. 10: 217-255. . —— and Douglas L. Oliver. 1951. Whence came maize to Asia? Ibid. 14: 263-291. . —— and R. G. Reeves. 1939. The origin of Indian corn and its relatives. Tex. Agric. Exp. Sta. Bull. 574: 1-315. and . 1945. The origin of maize: Present status of the problem. Amer. Anthrop. 47: 235-243. . —— and Claude FE. Smith. 1949. New archaeological evidence on evolution in maize. Bot. Mus. Leafl. Harvard Univ. 13: 213-259. . Martin del Campo, Rafael. 1938. El pulque en el México pre- cortesiano. An. Inst. Biol. Mex. 9: 5-23. Martinez, Maximino. 1928. Las Plantas mas Utiles que existen en la Reptiblica Mexicana, Mexico, D.F. . ——. 19387. Catdélogo de Nombres Vulgares y Cientificos de Plantas Mexicanas, Mexico, D.F. . ——. 1950. El ahuehuete (Tazrodium mucronatum Ten.). An. Inst. Biol. Mex. 21: 25-82. ——, 1951. Las Casimiroas de México y Centroamérica. [bid, Ne alae 9 at op OP Menzel, Margaret Y. 1951. The cytotaxonomy and genetics of Physalis. Proc. Amer. Phil. Soe. 95: 132-183. [ 167 | 138. Merrill, Elmer D. 1918. Species Blancoanae. Dept. Agr. and Nat. Resources, Bur. Sci. (Manila) Publ. 12: 1-423. 139. ——. 1988. Domesticated plants in relation to the diffusion of culture. Bot. Rev. 4: 1-20. 140, ——. 1950. Observations on cultivated plants with reference to certain American problems. Ceiba 1: 3-36. 141, Millan, Roberto. 1946. Nuevo mate del Uruguay. Darwiniana 7: 194-197. 142. Muller, Cornelius H. 1940. A revision of the genus Lycopersi- con. U.S.D.A. Mise. Pub. 382: 1-29. 143, Nuttall, Zelia. 1925. The gardens of ancient Mexico. Smithson, Inst. Annual Report for 1923: 453-464, 144, ——. 1937. The gardens of ancient Mexico. Mexico, D.F. 145. Palmer, Edward. 1891. Chia. Zoe 2: 140-142. 146. Pangalo, K. I. 1930. A new species of cultivated pumpkins. Bull. Appl. Bot. Gen. Pl. Breed. 23 (3): 258-265 (Russian 253-263, English summary 264-265). 147. Piper, C. V. 1925. The American species of Canavalia and Wenderothia. Contr. U.S. Nat. Herb. 20: 555-588. 148, —-—. 1926. Studies in American Phaseolinae. Jbid. 22: 663- 701. 149. Popenoe, Wilson. 1919. The useful plants of Copan. Amer. Anthrop. 21: 125-138. 150. ——. 1920. Manual of Tropical and Subtropical Fruits. New York. 151, ——. 1921. The native home of the Cherimoya. Journ. Hered. 12: 331-336. 152, ——. 1941. The avocado—a horticultural problem. Trop. Agr. 18: 3-7. 153, ——. 1952. Central American fruit culture. Ceiba 1: 269-367 ; Also, 1952. Fruticultura Centroamericana. Ceiba 3: 225-338. 154. —— and Abelardo Pachano. 1922. The capulin cherry. Jour. Hered. 13: 51-62; Also, 1923. Pan Amer. Union Bull. 56: 152-168. 155, Prain, David and Edmund Baker. 1902. Notes on Jndigefera. Journ. Bot. 40: 60-67, 136-144. 156. Ramirez Laguna, Antonio. 1936. Distribucién de los agaves de México. An. Inst. Biol. Mex. 7: 17-45. 157. Randolph, L. F. 1952. New evidence on the origin of maize. Am, Nat. 86: 193-202. 158. Reed, Howard S. 1944. An account of sixteenth-century agri- culture on the Mexican Plateau. Journ. Wash, Acad. Sci. 34: 209-213. 159. ——. 1950. A new species of Chenopodium from Mexico. Ma- drofio 10: 139-142, [ 168 ] 160. Reeves, R. G. and Paul C. Mangelsdorf. 1942. A proposed tax- onomic change in the tribe Maydeae (Family Gramineae). Am. Journ. Bot. 29: 815-817. Rick, Charles M. 1950. Pollination relations of Lycopersicon esculentum in native and foreign regions, Evolution 4: 110-122. Rivera Morales, Irene. 1941. Ensayo de interpretacién botan- ica del libro X de la Historia de Sahagtin. An. Inst. Biol. Mex. 12: 439-488. 3. —-— and Faustino Miranda. 1942. Nombres vulgares de plan- tas en el SO. del estado de Puebla. /bid. 13: 493-498. Rose, J. N. 1899. Notes on useful plants of Mexico. Contr. U.S. Nat. Herb. 5: 209-259. . =, 19038. Revision of Polianthes with new species. Ibid. 8: 8-14. Roys, Ralph L. 1931. The ethno-botany of the Maya. Tulane Univ. Middle Amer. Research Ser. 2: 1-359. Rydberg, Per Axel. 1896. The North American species of Phy- saligs and related genera. Mem. Torr. Bot. Club 4: 297-874. . Safford, William E. 1912. Annona diversifolia, a custard-apple of the Aztecs. Journ. Wash. Acad. Sei. 2: 118-125. . ——. 1914, Classification of the genus Annona, with descriptions of new and imperfectly known species. Contr. U.S, Nat. Herb. 18: 1-68. . ——. 1917. Food-plants and textiles of ancient America. Proc. 19th Int. Congr. Americanists 12-30. . ——. 1917. A forgotten cereal of ancient America. Ibid. 286- 297. . ——. 1918. Chenopodium Nuttalliae, a food plant of the Aztecs. Journ. Wash. Acad. Sci. 8: 521-527. . ——. 1919. Notes on the genus Dahlia, with descriptions of two new species from Guatemala. Jbid. 9: 364-373. de Sahagtin, Bernardino. [1829-1830] 1938. Historia General de las Cosas de Nueva Espafia, 5 vols., Mexico, D.F. Sauer, Carl O. 1936. American agricultural origins: a consid- eration of nature and culture. (in) Essays in Anthropology in honor of Alfred Louis Kroeber, 279-297. . ——. 1947. Early relations of man to plants. Geogr. Rev. 37: i209. . ——. 1950. Cultivated plants of South and Central America. (in) Julian H. Steward, ed., Handbook of South American In- dians 6: 487-543. ——, 1952. Agricultural Origins and Dispersals. New York. Sauer, Jonathan D. 1950. The grain amaranths. A survey of their history and classification. Ann. Mo. Bot. Gard. 37: 561— 632 [ 169 | 179. 180. 181. 182, 188. 189. 195. Schery, Robert W. 1947. Manioc—A tropical staff of life. Econ. Bot. 1: 20-25. Schiemann, EK. 1932. Entstehung der Kulturpflanzen. Hand- buch der Vererbungswissenschaft 3: 1-377, Berlin. Sears, Paul B. 1952. Palynology in southern North America. I: Archaeological horizons in the basins of Mexico. Bull. Geol. Soc. Amer. 63: 241-254. Setchell, William A. 1921. Aboriginal tobaccos. Amer. An- throp. 23: 397-413. . Sherff, Earl KE. 1951. Dahlias— New facts about these garden favorites. Chicago Nat. Hist. Mus. Bull. 22 (6): 5-7. . ——-. 1951. Miscellaneous notes on new or otherwise note- worthy dicotyledonous plants. Am. Jour. Bot. 38: 54-783. . Smith, Claude E. 1950. Prehistoric plant remains from Bat Cave. Bot. Mus. Leafl. Harvard Univ. 14: 157-180. . Smith, Lyman B. 1939. Notes on the taxonomy of Ananas and Pseudananas. Bot. Mus. Leafl. Harvard Univ. 7: 73-88. 7. Smith, Paul G. and Charles B. Heiser, Jr. 1951. Taxonomic and genetic studies on the cultivated peppers, Capsicum annuum L. and C. frutescens L. Am, Journ. Bot. 38: 362-368. Solms-Laubach, Hermann. 1889. Die Heimath und der Urs- prung des cultivirten Melonenbaumes, Carica Papaya L. Botan- ische Zeitung 47: 709-720, 725-734, 741-749, 757-767, 773- 781, 789-798. Spinden, Herbert J. 1917. The origin and distribution of agri- culture in America. Proc. 19th Int. Congr. Americanists 269-— 276; Also, 1920. (in) A. L. Kroeber, Source Book in Anthro- pology 245-251. . — >. 1950. Tobacco is American. The Story of Tobacco before the Coming of the White Man. New York. Standley, Paul C. 1920-1926. Trees and Shrubs of Mexico. Contrib. U.S. Nat. Herb. 23: 1-1721. . ——. 1980. Flora of Yucatan. Field Mus. Nat. Hist. Pub. 279: 157-492, . ——. 1946. Food plants of the Indians of the Guatemalan highlands. Journ. Arn. Arb. 27: 395-400. . —— and Julian Steyermark. 1946- . Flora of Guatemala. Fieldiana 24: parts 3-6 published. Stebbins, G. L. 1950. Variation and Evolution in Plants. New York. . Thompson, J. E. S. 1930. Ethnology of the Mayas of southern and central British Honduras. Field Mus. Nat. Hist. Pub. 274: 27-213. 196a. ——. 1946. Some uses of tobacco among the Maya. Carn. Inst. Notes on Middle Amer. Archaeol. and Ethnol. 3 (61): 1-5. [ 170 | 197. 198. 1o7 205. 206. 207; 208. 209. Tioutine, M. G. 1935. Breeding and selection of sweet pota- toes. Journ. Hered. 26: 3-10. Towle, Margaret A. 1952. The pre-Columbian occurrence of Lagenaria seeds in coastal Peru. Bot. Mus. Leafl. Harvard Univ. 15: 171-184. Urbina, Manuel. 1902. Los zapotes de Hernandez. Anales del Museo Nacional 7: 209-234; Also, 1912. La Naturaleza III, 1 (3): 53-79. . ——. 1903. Notas acerca de los ‘‘Ayotli’? de Hernandez, 6 calabazas indigenas. An. Mus. Nac. 7: 353-390; Also, 1912: La Naturaleza III 1 (3): 80-117. . ——. 1904. Plantas comestibles de los antiguos mexicanos. An. Mus. Nac. 2a época 1: 503-591. . ——. 1906. Raices comestibles entre los antiguos mexicanos. Ibid, 3: 117-190. . Vaillant, George C. 1939. An early occurrence of cotton in Mexico. Amer. Anthrop. 41: 170. . Van Eseltine, G. P. 1931. Variation in the lima bean, Phase- olus lunatus L., as illustrated by its synonymy. N.Y. State Agric. Exp. Sta. Tech. Bull. 182: 1-24. Vavilov, N. I. 1926. Studies on the origin of cultivated plants. Bull. Appl. Bot. Gen. Pl. Breed. 16 (2): 1-248 (Russian 1-138, English summary 139-248). ——, 1931. Mexicoand Central America as the principal centre of origin of cultivated plants of the New World. Ibid. 26 (3): 135-199 (Russian 135-178, English summary 179-199). ——, 1951. The origin, variation, immunity and breeding of cultivated plants (selected writings trans. by K. Starr Chester). Chron. Bot. 13: 1-364. Vélez, I. 1946. Wild pineapples in Venezuela. Science 104: 427-428. Waldron, Ralph A. 1919. The peanut (Arachis hypogaea)—its history, histology, physiology, and utility. Bot. Contr. Univ. Pa. 4: 301-338. . Watt, George. 1907. The wild and cultivated cotton plants of the world. London. . Weatherwax, Paul. 1935. The phylogeny of Zea Mays. Am. Midl. Nat. 16: 1-71. . ——. 1950. The history of corn. Scientific Monthly 71: 50- 60. . Wellhausen, E. J., L. M. Roberts and E. Hernandez X. 1952. Races of Maize in Mexico. Bussey Inst., Harvard Univ.; Also, 1951. Razas de Maiz en México. Folleto técnico no. 5, Oficina de Estudios Especiales, Secretaria de Agricultura y Ganaderia, Mexico, D.F. 214. 215. 216. S17, 218. 219. 220. Whitaker, Thomas W. 1947. American origin of the cultivated cucurbits. Ann. Mo. Bot. Gard. 34: 101-111. ——, 1948, Lagenaria, a pre-Columbian cultivated plant in the Americas. S. W. Journ. Anthrop. 4: 49-68. —— and G. W. Bohn. 1950. The taxonomy, genetics, produc- tion and uses of the cultivated species of Cucurbita. Econ. Bot. 4: 52-81. —— and G. F. Carter. 1946. Critical notes on the origin and domestication of the cultivated species of Cucurbita, Am, Journ. Bot. 33: 10-15. Whiting, Alfred F. 1944. The origin of corn: an evaluation of fact and theory. Amer. Anthrop. 46: 500-515. Williams, Louis O. 1950. Two new Perseas from Central Amer- ica. Ceiba 1: 55-58. Zhiteneva, N. E. 1930. The world’s assortment of pumpkins. Bull. Appl. Bot. Gen. Pl. Breed. 23 (3): 157-207 (Russian 157-204, English summary 205-207). EXPLANATION OF THE ILLUSTRATION PLare XX. Reproduction of the illustration of Nicotiana rustica (‘‘ pycielt’’ or ‘‘picietl’’) and a portion of the discussion of tobacco, from Francisco Hernandez: ““Re- rum Medicarum Novae Hispaniae thesaurus, seu plant- arum, animalium, mineralium mexicanorum historia’’ (Rome 1651) p. 173. [ 172 } PLATE XX De PYCIELT, feaTabaco. Cap. LI. LANTAM, quam Mexicenfes Pp Pyciele feu Yelt vocant, ab Huitinis ap- ellacur 7 abacus, a quibus non ad Indos fo- los, fed & ad Hifpanos id defluxir nomé,ed quod fuftumigijs admifceretur , que Taba- cos etiam nuncupare confucuerunt. a Bra- {ilianis Petum, ab alijs Herba facra, a non- nullis Nicotiana dicitur.Non eft autem vna huius planta fpecies. alij namque tres in. hoc antiquo Orbe reperiri afirmant , ac plantam hanc in Tabacwm maiorem , mi- norem, & minimum partiuncur. At quia» maioris, minorisq; Tabaci differentia pu- filla eft (confittit enim in magnitudine , & longitudine., ac adharentia foliorum fine» pediculis cauli, ve in maiori ob{eruatur , in minori vero folium eft paulo minus, lon xe) pediculo ramis inhzrens, ac forum pofitu- ra, cumg; id ob cauflas mutationes in plane tis efficientes, Jatius in proamio explicatas contingere potuerit ) non immerito duz funttantum {pecies , quz obferuantur in hac noua Hifpania. quarum alceram gocieg y Pycielt , alteram vero Quauhyelt appellant. Pycieleergo herba elt, folia ferens Pyciels Jata, oblonga, ac Perfonata quadantenus fimilia. caules, quinque pluresvé do- ginz. drances longos , atque hirfutos, inconditos, {triatos,& leues. flores Hyofcyami lucei fimiles,eisq; decidentibus vafcula predicti Hyofcyami emula, referta fe- mine pufillo, Papaueris minore, ac ex rufonigricante. radicem breuem , non species tt admodum tenuem, fed fibratam . Quawbyele vero in magnam aflurgens alti QUuryst tudinem, Affyriam Malum,Limoné vocata,zquat . Caule recto multos emic- tence ramos, & in cis folia mali Affyrij longiora, hirfuta , colore viridi dilutio- re, vei cora planta diffufa . interdumue folia, foli , & cocli ratione variantur. uandoque enim cubitalem longitudinem , ac pedalem latitudinem aflecuta fine pediculo caulem amplectuntur . nonnumquam vero folia minora , pedi- culis inhzrentia ramis con{piciuntur. Flores Campanulz inftar fert,concauos, ac per extremum fex , feptemué angulis diftinétos , candicantes, medio vero purpurefcentes , ordine per ramulorum longitudinem difpofitos . quibus {uc- cedunt capitula Ocymoidi fimilia,maiora tamé , plena pee pufillo ex cine- 3 rco Species. BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CambBripGre, Massacuvusetts, December 31, 1953 Voi. 16, No. 7 FREDERICK OLIVER THOMPSON, 1883-195: BY Enso S. BARGHOORN Tue death of Frederick Oliver Thompson on the third of January of this vear, brought to an untimely end the life and activities of a man who was justly recognized as one of the country’s leading amateur collectors of pre- historic plant life. Certainly in few scientific fields has a nonprotessional worker been so well known to specialists as was Fred Thompson to American paleobotanists. For the careful and accurate documentation of his collections and for the vast amount of material donated to American universities, his record as an amateur paleontologist has few parallels, particularly when it is recognized that his paleontologic interests were confined to only the last two decades of his lite. Fred Thompson (he always preferred to be called “Fred”” by his scientific friends, regardless of their age) was born in Des Moines, lowa, December 29, 1883, the first son of H. DeVere Thompson and Alice Cooper Thompson, who were among the early leaders in the development of Towa’s capitol. After preparation for college at St. Paul’s School in Concord, N.H., he was graduated trom Harvard College in the Class of 1907. He returned to Cambridge the following autumn for a year’s study in the Harvard Law School and then went back to Des Moines to enter the varied business activities eree in his native state which occupied most of his later years. In 1912, he married Anna Stroh Cram, of Des Moines, who, with three of their four children, survives him. While at Harvard, young Fred was quite active in various social clubs, among them the Hasty Pudding Club, the Phoenix Club and the Institute of 1770. Dur- ing his college years, however, he manifested little of that interest in natural history or science which later be- ‘ame an absorbing interest and intellectual outlet. Fred ‘Thompson first became interested in fossils in 1930 when he accompanied a friend of his on a collecting trip to the great coal-stripping mines near Coal City, Illinois. Here, the beautifully preserved remains of an ancient Carboniferous flora were being unearthed in great quantities, contained in the curious ironstone nodules which feature the sedimentary rocks of this area. Much time and effort were devoted by Mr. Thompson during the following years to the careful selection of these speci- mens which were subsequently donated to many colleges and universities. He was somewhat surprised years later to observe one of his own Illinois specimens, properly credited to the collector, on public exhibit in a museum in Mexico City. Harvard’s collections of fossil] plants were enriched by over ten thousand specimens culled from the spoil piles of the Coal City mines. In 1938, while on a fossil collecting expedition, Fred discovered a number of the curious calcareous concretions ‘coal balls.°° Coal balls, . known among paleobotanists as because of their wealth of organically preserved plant tissues, have been one of the major sources of knowledge of the organization and structure of Carboniferous plants since they were first studied in Europe. The interest of paleobotanists in this new discovery was not long in manifesting itself, and during the following seven years Thompson collected literally dozens of tons of the Lowa [ 17+ | coal balls. The majority of these were sent to the Botan- ical Museum at Harvard, but large numbers were given to other institutions, in particular to Washington Uni- versity in St. Louis. Among Fred Thompson's collections of fossil plants, four species, new to the Carboniferous flora, perpetuate his name in the paleontological literature. In addition, he made available for study a large number of other new species of fossil plants, which have added measurably to our knowledge of the structure and evolution of Carbon- iferous plant life. Indeed, many reterences to ‘‘Collected by Frederick Oliver ‘Thompson’ may be seen in paleo- botanical publications in the United States since 1988. Although his field of special interest was paleobotany, noteworthy contributions were also made to invertebrate paleontology: and, to aid studies in this field, a superb collection of modern Floridian sea shells was donated to the State University of Lowa. He also collected many fossil marine invertebrates from the Pennsylvanian strata of central Lowa and especially from the Upper Devonian deposits of the northeastern part of the state. In this work, as so often when in the field, he was accompanied and ably assisted by his wife, Anna. His collections of invertebrate fossils, as in the case of plants, were invaria- bly made available to specialists. Some of the specimens of Lowa’s Paleozoic faunas he even sent to places as re- mote as Canterbury University College in New Zealand, where they have been used by paleontologists for pur- poses of Comparison. Once, while Fred was splitting fossiliferous nodules near Mazon Creek, [linois, a boulder of Joliet dolomite, which he was using for an anvil, fractured and a striking trilobite pygidium was exposed. It became the holotype of Aretinurus thompsont Miller and Unklesbay, named in honor of its discoverer. Had it not been for his innate [ 175 ] FREDERICK OLIVER THOMPSON 1883-1953 Research Fellow in the Botanical Museum Harvard University 1949-1958 PLATE XX] modesty regarding his finds, several species of fossil ceph- wlopods would also have been named for him. Since his death, however, a fine specimen from the Ordovician of Baffin Island will honor him in this manner. It should also be noted, in the realm of invertebrate paleontology, that at least a half dozen of the Special Papers and two of the forthcoming Memoirs of the Geo- logical Society of America carry acknowledgment. to Frederick Thompson for financial assistance during their preparation. The same is true for a paleontological mono- graph recently published by the Museé du Congo Belge and for more than a dozen papers in the Journal of Paleontology. In addition to his indefatigable collecting, both of fos- sil plants and of invertebrates, Fred was invariably fo- menting and fostering research projects in paleobotany and other aspects of paleontology. His voluminous cor- respondence and lively and stimulating communications with workers in widely scattered institutions attest to his restless striving to make paleontology a more dynamic field of intellectual interest and activity. In many ways he was more successful in these efforts than were his pro- fessional (and scientific) colleagues. Although the writer personally knew Fred Thompson only during the last six years of his life, the memory of the man will stand, not only on his scientific contribu- tions, but on his unforgettable personality as well. His interest in people partook both of their foibles and their serious endeavors and he was always ready with an anec- dote in either vein. This made him a delightful com- panion in the field and eased the disappointments so often experienced in unsuccessful collecting in unfamiliar lo- ‘alities. In pursuit of new finds, despite failing health in his later years, Fred’s energy and cheerful persistence in the field were a source of inspiration to younger com- panions. Perhaps, above all, for those who knew Frederick Thompson as an amateur paleontologist, he will be re- membered most clearly for his selflessness, his generosity and his intense devotion to his chosen field of scientific interest. Species named in honor of Frederick O. Thompson Arerinurus THoMPsoNT Miller & Unklesbay in Journ, Paleont. 18 (1944) 364, Macrosracuya THomesony Darrah in Bot. Mus. Leafl. Harvard Univ. 4 (1986) 53, Meputiosa THompsont Andrews in Ann. Mo. Bot. Gard. 32 (1945) 8324, Mersoxyton Tuompsonir Traverse in Am. Journ. Bot. 37 (1950) 824. SpueNostrRoBUS THompsoni Levittan & Barghoorn in Am. Journ. Bot, 35 (1948) 353, [178 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CaMBRIDGE, MassacuuseETTs, Fespruary 19, 1954 PLANTAE AUSTRO-AMERICANAE IX PLANTARUM NOVARUM VEL NOTABILIUM NOTAE DIVERSAE BY RicHarp Evans SCHULTES' ContTINUED studies of recent collections from northwest- ern South America, chiefly from Colombia, have dis- closed a number of plants which are, for one reason or another, of importance to our growing understanding of the flora of this critical part of the New World tropics. Most of the plants herein discussed are from Amazonian Colombia. I acknowledge with thanks the collaboration of Dr. Lyman B. Smith, of the Smithsonian Institution, who is responsible for the work on the Bromeliaceae, and of Dr. Alicia Lourteig, Guggenheim Fellow in Botany at the Gray Herbarium of Harvard University, who has contributed the section on Cuphea. ARACEAE Anthurium atropurpureum Schultes & Maguire ex Schultes in Bot. Mus. Leafl. Harvard Univ. 16 (1958) 60. Further studies on the aroids of Amazonian Colombia have made it seem advisable to publish a drawing of 4n- 1Botanist, Division of Rubber Plant Investigations, Bureau of Plant Industry, Soils, and Agricultural Engineering, Agricultural Research Administration, U. S. Department of Agriculture; Research Fellow, Botanical Museum of Harvard University. [ 179 ] thurium atropurpureum (Pl. XXII). Later studies may indicate that this species represents a very important complex on the Cretaceous quartzitic hills and savannahs of the Vaupés and Amazonas. Anthurium atropurpureum Schultes & Maguire ‘ar. apertum 3 Schiultes var. nov. Haec varietas ab Anthurio atropurpureo nervis secun- dariis arcuatis apertis vel non confluentibus spatha spa- diceque majoribus differt. It might at first seem that the venation character upon which this newly described variety is based is trivial. Not- withstanding its rather unobtrusive nature on superficial examination, | have found that it is stable. Having no other morphological characters of importance to differen- tiate it as a species, this concept is probably best treated as a variety of Anthurium atropurpureum. The secondary veins of Anthurium atropurpureum, as ‘an be seen in the figure published herewith, are arcuate and run together near the margin of the leaf-blades to form a continuous and stout nerve parallel to the margin. In the new varietal concept here described, the secondary veins are also arcuate, but they are not confluent and, consequently, do not form a continuous nerve along each margin. The varietal epithet has reference to the open appearance of the veins in contrast to the closed appear- ance which the confluent veins of the species exhibit. Like Anthurium atropurpureum, this new variety is an inhabitant of the curious xerophytic islands of savan- nah or scrubby vegetation that are found scattered here and there in the Colombian Amazonia. Anthurium atro- purpureum var. apertum is probably much more abun- dant and perhaps more widespread than is A. atropur- pureum itself. CoLombia: Comisarias del Amazonas-Vaupés, Rio Apaporis, Raudal de Jirijirimo. Extensive white-sand savannah or caatinga on right [ 180 | bank. “‘Spike purple. Leaves stiff. In sterile white sand.’’ March 1951, Richard Evans Schultes 12094 (Tyrer in Herb. Gray).—Same locality. August 12, 1951, Richard Evans Schultes & Isidoro Cabrera 13517.—Comisaria del Vaupés, Rio Vaupés, Raudal de Yurupari (+ 350 kts. arriba de Mitt). Alt. 220 m. “‘Espata verdosa; espadice amarillo-violaceo.’’ September 24, 1939, J. Cuatrecasas 6963. Anthurium crassinervium (Jacq.) Schott Melet. 1 (1832) 22. Anthurium crassinervium appears to be rather wide- spread in northern South America, but in the northwest Amazon basin it is confined to the granitic, pre-Cambrian, dome-shaped mountains, where it occurs often in great abundance under conditions of extreme xerophytism. The collection Cuatrecasas 6885 establishes its occurrence in the Colombian Amazonia. Cotompra: Comisaria del Vaupés, Rio Vaupés, Cerro de Mitt. Alt. 380 m. ““Espadice pardo-violaceo.’? September 17, 1939, J. Cuatre- casas 6885, VenezugELa: Territorio del Amazonas, Rio Negro, Piedra del Cocui. **Terrestrial.’’ December 22-23, 1947, Richard Evans Schultes & Francisco Lopez 942%. Anthurium gracile (Rudge) Engler in Bot. Jahrb. 25 (1898) 370. A widespread plant of the rain tropics of South Amer- ica, Anthurium gracile has apparently not hitherto been reported from eastern Colombia. Cotompra: Intendencia del Meta, Sabanas de San Juan de Arama, margen izquierda del Rio Giejar, alrededores del aterrizaje ‘‘ Los Micos.’’ Alt. ca. 500 m. “‘Epifita. Espata purptreo-grisdcea.’’ Jan- uary 22, 1951, Jesus M. Idrobo & Richard Evans Schultes 1268. Anthurium Idroboanum R. FE. Schultes sp. nov. Herba terrestris, usque ad 24 pedes alta, in coloniis in silvis densis pluviosisque crescens. Caudiculus magno- pere abbreviatus, robustus, internodiis brevissimis. Folia rigidissime erecta, valde coriacea, supra atroviridia, sub- [ 181 ] tus pallide viridia, cum petiolis strictis crassioribus suleate subtetragonisque, basi latiuscule dilatatis, usque ad 31 em. longis, 8-10 mm. in diametro; angulis praecipue api- cem versus inconspicue alatis; lamina adulta glabra, supra atroviridis, subtus pallidior, cordato-ovata, apice rotundato-obtusa, leviter marginata, lobis posticis circiter 10 cm. longis, 2-38 cm. latis, rotundatis, sinu angusto acuto sejunctis, nervo centrali robusto recto (siccitate substram- ineo) usque ad apicem penetrante, supra leviter sed subtus valde elevato, nervis secundaris vel lateralibus arcuatis, non confluentibus, e basi nascentibus utrinque tribus cen- tralibus similibus, transversis paucis paullo tenuioribus conjunctis. Inflorescentia erecta, pedunculo aliquid car- nosulo, usque ad 60 cm. longo, 5-8 mm. in diametro. Spatha late lanceolata, glaberrima, utrinque flavo-viridis, basi cordato-amplexicaulis, apice acuta, usque ad 9 cm. longa, inferne 2 cm. lata, vivo non recurva. Spadix erec- tus, caudiformis vel cylindricus, apice plusminusve trun- catus, apparenter estipitatus, 7-10 cm. longus, atropur- pureus. Anthurium Idroboanum, belonging to Engler’s sec- tion Cardiolonchium, seems to resemble most closely 4. magnificum Linden, native of the eastern cordillera of Colombia, a species which has been important in hybrid- ization for horticultural work in the genus. The former may be distinguished from the latter at once by differ- ences in the size and venation of the leaves, by having an estipitate spadix which is only half as long and by having much shorter and more membranaceous, yellowish green spathes. he leaf of the newly described species is also far more coriaceous than that of Anthurium magnificum, and the plant seems to be humbler. The base of the leaf of Idrobo & Schultes 957 is far more deeply cordate than in the type collection, but all other essential characters exhibit an unusual stability in [ 182 ] the three collections which are at present available. Cotomsia: Intendencia del Meta, Cordillera La Macarena (extremo nordeste), Macizo Renjifo, cumbre y alrededores. Alt. 1800-1900 m. ‘En colonias. Sobre capa de hojas secas. Espddice purptreo.’’ Jan- uary 6-20, 1951, Jess M. Idrobo & Richard Evans Schultes 1162 (Tyre in Herb. Gray).—Same locality and date. ‘“Terrestrial. Spathe green, bent back. Leaf discolourous, very dark green above, lucid above, pale green and punctate beneath. Basic colour of spadix reddish brown. ’’ Jesus M. Idrobo & Richard Evans Schultes 957.—Intendencia del Meta, Sierra de La Macarena, Central Mountains, North Ridge. Alt. 1400 m. ‘‘Dense humid forest. Terrestrial; leaves coriaceous. Spathe green, tinged with red; spike brown.’” December 25, 1949, W. R. Philipson & J. M. Idrobo 1926. Anthurium macarenense I. I’. Schultes & Idrobo sp. nov. Anthurio cabrerensi proxima sed principaliter lamina membranacea (non coriacea), apice abrupte et longe at- tenuata (non acuta), spatha atrosanguinea venis viridibus elongato-lanceolata spadici subaequali (non spatha cuprea ovato-lanceolata spadicis longitudinis # aequanti) differt. Anthurium macarenense would seem to be very closely allied to 4. cabrerense Engl. trom the Departamento del ‘Tolima in Colombia, which is described and well illus- trated in Engler Pflanzenreich LV. 23B (Heft 21) (1905) 208. The former species differs from the latter primarily in having a more delicate leaf-blade which is abruptly and attenuately tipped, and in having a spathe which differs in size, shape and color. Cotompia: Intendencia del Meta, Cordillera La Macarena (extremo nordeste), Macizo Renjifo, cumbre y alrededores. Alt. 1300-1900 m. “*Spathe deep red or maroon both sides with veins greenish. Epiphyte. Fruits pink. Peduncle and petioles red. Spathe arcuate.’’ January 6-20, 1951, Richard Evans Schultes & Jesiis M. Idrobo 954 (Tyre in U.S. Nat. Herb. No. 2029669).—Intendencia del Meta, Sierra La Macarena, selva densa entre los Rios Giiejar y Sansa. Sobre la mesa nordeste. Alt. 500-1,000 m. August 29, 1950, J. M. Idrobo 527. r 183 | Anthurium macrocephalum 2P. 1. Schu/tes sp. nov. Herba terrestris, ut videtur usque ad 24 vel 8 pedes alta, in silvis umbrosis humidisque crescens. Caudiculus probabiliter abbreviatus. Folia rigide erecta, coriacea, supra vivo apparenter atroviridia, subtus pallidiora, cum petiolis crassis, lateraliter compressis, sulcatis, quam lam- ina longioribus, 55 cm. longis vel longioribus, 10 mm. in diametro vel vivo majoribus; lamina adulta glabra, utrinque minutissime (subtus densius) nigropuncticulata, ovata, apice rotundata, basi breviter attenuata, margine aliquid incrassata, plusminusve 44cm. longa, 35 em. lata, cum costa sex-nervia, nervis lateralibus primariis intimis in apicem exeuntibus, duobus extimis in infima media parte margine crassiore approximatis et in eo evanescen- tibus, secundariis parallelis, tenuibus inter se 1.5-2 em. distantibus, tertiis pluribus tenuissimis. Inflorescentiae usque ad 62 cm. altae vel altiores, erectae sed apparenter spadice nutanti, pedunculis petiolis similibus, usque ad 12 cm. in diametro. Spadix cylindricus, apice rotundato- obtusus, 20-22 cm. longus, siccitate 8.5 cm. in diametro, estipitatus, viridis. Spatha coriacea, late lanceolata, apice acuta, 8-9 cm. longa, 2—2.4 cm. lata, glaberrima, viri- dis. Baccae oblongae, tetragonae, in stylum prismaticum productae. Anthurium macrocephalum belongs to Engler’s section Digitinervium and seems to be closely related to A. pangoanum Sod. of Ecuador. The former can be distin- guished from the latter principally by having leaves which, while shorter, are wider, giving them a truly ovate shape; by having a much longer petiole; by having a much thicker spadix which is green, not rose-colored; and by having a much longer and wider spathe. 4nthur- tum pangoanum is described as being ** both surfaces of the leaf, but in A. macrocephalum the nigropunctata’’ on [ 184 | upper surface, although extremely remotely beset with minute black dots, cannot be compared with the lower surtace, which is rather densely ornamented with these glandular (¢) structures. Cotomsia: Departamento del Cauca, El Tambo, Cordillera Occiden- tal, vertiente oriental, La Paz y alrededores, Las Juntas. Alt. 1400- 1700 m. “‘En riachuelo, parte sombrosa. Espata verde, espadice verde.** August 29, 1949, J. M. Idrobo 280 (Typr in U.S. Nat. Herb. No. 2029344). Anthurium nemoricola Schultes & Maguire ex Schultes in Bot. Mus. Leafl. Harvard Univ. 16 (1958) 61. It has been considered advisable to publish the accom- panying drawing (PI. X XII) of this curious little xero- phytic species of Anthurium. Anthurium oblongo-cordatum Hngler in Pflan- zenr. LV, 28B (1905) 110. This species was first described from the Quindio re- gion of Colombia. The collection cited below, in spite of slight differences from the type, would seem to be easily accommodated in Anthurium oblongo-cordatum. Cotomspia: Departamento del Cauca, Cordillera Occidental, verti- ente oriental. Cuchilla del Tambo, Alt. 1750 m. August 1949, Jesis M. Idrobo & Alvaro Fernandez 44. Anthurium panduratum Martius ex Schott in Oestr. Bot. Wochenbl. (1855) 273. Anthurium panduratum was described from material] collected by Martius on the ‘‘Japura’” (Rio Caqueta), probably in Colombian territory. The collections cited below indicate that the concept is relatively widely dis- tributed in Amazonian Colombia. Cotomsia: Comisaria del Vaupés, San José del Guaviare. Novem- ber 12, 1939, J. Cuatrecasas 7444.—Comisaria del Amazonas, Rio Loretoyacu, September 1945, Richard Evans Schultes 607 1.—Comisaria del Amazonas, Rio Loretoyacu, October 21, 1946, Richard Evans Schultes & George Black 8549.—Comisaria del Amazonas, Rio Boiauas- si, October 27, 1946, Richard Evans Schultes & George Black 8606. [ 185 | Anthurium pentaphyllum (4uw/l.) G. Don in Sweet Hort. Brit. ed. 3 (1834) 633. Widespread in tropical South America, Anthurium pentaphyllum has hitherto not been recorded from Ama- zonian Colombia. Cotoms1A: Comisarias del Amazonas-Vaupés, Rio Apaporis, entre el Rio Pacoa y el Rio Kananari. Soratama. Alt. ca. 250 m. ‘‘Climb- ing epiphyte. Spathe greenish, yellow near tip. Spadix green-grey, reflexed.’’ August 27, 1951, Richard Evans Schultes & Isidoro Cabrera 13786. Anthurium popayanense Hngler in Bot. Jahrb. 6 (1885) 274. Anthurium popayanense has hitherto been known only from the region around Popayan in Colombia. CoromsiA: Intendencia del Chocé, Coreovada Region, upper Rio San Juan. Ridge along Yeraciii Valley. Alt. 200-275 m. Dense forest. *‘Ontree. Spathe green. Spadix pinkish yellow.’’ April 24-25, 1939, E. P. Killip 35316, Anthurium reticulatum Bentham P|. Hartweg. (1846) 255. Anthurium reticulatum has previously been known from the regions near Cali and Popayan in Colombia. Cotompia: Intendencia del Chocé, dense forest south of Rio Con- doto, between Quebrada Guarapo and Mandinga. Alt. 120-180 m. ‘Terrestrial. Spathe green. Spadix pink.’’ April 22-28, 1939, E. P. Killip 85439. Anthurium scolopendrinum (Ham.) Kunth Enum. Pl. 3 (1841) 68. Notwithstanding the fact that this species is rather widespread in tropical South America, it has apparently not hitherto been reported from eastern Colombia. Cotomsia: Intendencia del Meta, Sabanas de San Juan de Arama, margen izquierda del Rio Giejar, alrededores del aterrizaje “* Los Micos.’’ Alt. ca. 500m. ‘‘Espata roja. Espadice rojo.’’ January 22, 1951, Jesus M. Idrobo & Richard Evans Schultes 1203. | 186 | ERIOCAULACEAE Paepalanthus Moldenkeanus R. EF. Schultes sp. nov. Planta frutescens, robusta, sublignosa, breviter caules- cens. Folia prope rami simplicis apicem rosulate dispos- ita, firmissime chartacea, plana, attenuato-linearia, 18-21 em. longa, basi 1.5-2.8 cm. (parte centrali 1-1.5 cm.) lata, basi dense longeque albo-pilosa, infra medium mar- gine (et raro in lamina) maxime sparsissime breviterque setosa, demum utrinque glabra nitidaque. Caulis florifer erectus, robustior, sublignosus, usque ad 1.80 m. altus, 1-1.5 cm. in diametro, bracteis multis foliaceis, charta- ceis, lanceolato-triangularibus, usque ad 13.5 em. longis, basi 2.4 em. latis, apicem versus minoribus, margine re- motissime albo-ciliatis, utrinque glabris nitidisque, spiral- iter obtectus. Pedunculi plusminusve centum quadragin- ta, filiformes, 30-35 cm. longi, 1.5—2 mm. in diametro, glabri, obscure bicostati, conspicue contorti, vaginis gra- cillimis, cylindricis, chartaceis, stramineis, apice bifidis, basi dense albo-pilosis, contortis, 4.5 em. longis. Capi- tula perfecte globosa, luteo-albida, 10-12 mm. in diame- tro. Bracteolae stramineae, triangulares, circiter 2 mm. longae, 1-1.5 mm. latae. Flores staminiferi basi dense longeque albo-barbati, sepalis hyalinis, oblongis, apice acutis et barbellatis, 8 mm. longis, 1 mm. latis, petalis in tubum infundibuliformem, 3.5-4 mm. longum connatis, staminibus circiter 1 mm. longis. Flores pistillati basi dense longeque albo-barbati, sepalis hyalinis, late spathu- latis, 4.5 mm. longis, 2-2.5 mm. latis, petalis similibus sed paulo majoribus, apice barbellatis, stigmatibus 2 mm. longis, ovario elongato-ovoideo, 2 mm. longo, 1 mm. in diametro, glabro sed basi densissime albo-barbato. Paepalanthus Moldenkeanus is set apart from all other known species of the genus from Colombia and northern [ 187 | South America by its unusual size. It is a robust plant that normally reachesa height of five and a half or six feet. This extraordinary plant inhabits the immense sand- stone savannahs to the north of the Colombian part of the Rio Vaupés above Mitu. It has been collected at Yapoboda at the headwaters of the Rio Kuduyari and at Kanenda on the Rio Kubiyi. An aeroplane reconnais- sance has shown that these two savannahs are continu- ous. I have seen this plant, but was unable to collect it, on the great savannahs of Goo-ran-hoo-da on the Rio Karurti in the upper Vaupés. The Karurti savannahs may possibly be continuous with Kanenda and Yapoboda. Paepalanthus Moldenkeanus undoubtedly represents another of the curious endemic plants which have turned up during our investigation of these ancient quartzitic savannah formations in the Vaupés. Growing in close proximity to Vellozia lithophila, Bom- bax coriaceum, Hevea nitida var. toxicodendroides, Leit- gebia colombiana and Styrax rigidifolius forma yapoboden- sis,’ Paepalanthus Moldenkeanus is admirably adapted to the extreme conditions of xerophytism which obtain on these savannahs. It occurs in isolated colonies of from ten to fifty individuals in slight depressions or swales on the usually flat savannah. These swales are moist and highly acidic and are repositories of very interesting grasses and sedges, as well as xyridaceous, lentibulariaceous and erio- caulaceous species. There are a number of other species of Paepalanthus in the same localities, but all are diminu- tive plants. It is appropriate that I dedicate this majestic new species to my friend, Dr. Harold N. Moldenke, in recog- nition of his extensive researches in the family 2/riocaul- acede. 'Steyermark (in Fieldiana 28 (1953) 492) considers Slyrax yapo- bodensis (Idrobo & R. E. Schultes) Steyerm. to be a distinct species. [ 188 ] CoLomsiA: Comisaria del Vaupés, Rio Kuduyari (tributary of Rio Vaupés), Yapoboddé. Quartzite savannah near headwaters. Alt. about 900-1000 feet. General location: Lat. 1°20/ N, Long. 70°30! W. ‘‘From 4 to 6 feet tall. Flowers white.’” October 4-5, 1951, Richard Evans Schultes & Isidoro Cabrera 14351 (Tyrer in Herb. Gray ).—Comis- aria del Vaupés, Rio Kuduyari (tributary of Rio Vaupés), Yapoboda, quartzite savannah near headwaters. Alt. about 900-1000 ft. General location: Lat. 1°20’ N, Long. 70°30’ W. ‘‘Up to 6 feet tall. Flowers yellow-white.’’? October 5-6, 1951, Richard Evans Schultes & Isidoro Cabrera 14265.—Comisaria del Vaupés, Rio Kubiyt (tributary of Rio Vaupés), Cerro Kafiendd, savannahs about 15 miles upstream from mouth. Quartzite base. Alt. about 800-900 feet. General location: Lat. 1°0/ N, Long. 70°15’ W. November 10, 1952, Richard Evans Schultes & Istdoro Cabrera 18385,—Comisaria del Vaupés, Rio Kubiyt (tributary of Rio Vaupés), Cerro Kafiendaé, savannahs about 15 miles upstream from mouth. Quartzite base. Altitude about 800-900 feet. General location: Lat. 1°0’ N, Long. 70°15’ W. ‘“‘Four feet tall. Flowers yellow-white.’? May 5, 1958, Richard Evans Schultes & Isidoro Cabrera 19239.—Comisaria del Vaupés, Rio Kuduyari (tributary of Rio Vaupés) Yapoboda. Quartzite savannah near headwaters. Alt. about 900-1,000 feet. ‘“Four feet tall. Common in swales.’’ April 1953, Richard Evans Schultes &§ Isidoro Cabrera 19975.—Comisaria del Vaupés, Rio Kubiya, Cerro Kafianddé. Alt. mis o menos 380 m. ‘‘Dos metros alto, erecto; flores blancas.’’ November 2-4, 1952, Hernando Garcia- Barriga 15088. run - _ 7 THURNIACEAE Thurnia sphaerocephala (Rudge) Hooker filius in Hooker Icon. Pl. (18838) t. 1407. The collections Schultes & Cabrera 17568 and 19858 establish for the first time the occurrence in Colombia of the monogeneric family Thurniaceae, allied to the Jun- caceae and the Rapateaceae. The genus Thurnia comprises two species. Thurnia Jenmant occurs in British Guiana, where the type was found ‘“‘thickly choking the Potaro river above and_ be- low the Kaieteur falls.*” The range of Thurnia sphaero- cephala is greater, including British Guiana and in scat- tered localities in the Brazilian Amazonia south to the Xingu and west to the Solimdes. The range extension [ 189 | herewith reported to Amazonian Colombia greatly am- plifies the known distribution of this rare plant. In the small creeks and rills forming in the headwaters of the Rio Piraparana, Thurnia sphaerocephala grows in extraordinary abundance in the shallow waters, together with a species of Pontederia. It often grows so thickly that it impedes canoe travel completely. This habitat is unusual, for the type and numerous later collections of the species from British Guiana report the habitat as ‘*savannahs. *’ Ihave seen Thurnia sphaerocephala, but did not collect it, in Cano Paca, one of the rills forming the headwaters of the Rio Papuri, and in the very sources of the Rio Dji. Both the Papuri and the Dji, rising in an area near the headwaters of the Rio Piraparand, are affuents of the Rio Vaupés. The westernmost locality of the family is that of Schultes & Cabrera 19853, in the headwaters of the Cano Churruco, a brook of clear water, draining a quartzitic area. It is not abundant, however. Cotompia: Comisarias del Amazonas-Vaupés, Rio Apaporis, Sora- tama (above mouth of Rio Kananari) and vicinity, Cafio Churruco, Alt. about 900 feet. General location: Lat. 0°5’ N, Long. 70°40’ W. “‘In water.’’ January 1952, Richard Evans Schultes & Isidoro Cabrera 19588. —Comisaria del Vaupés, Rio Piraparana (tributary of Rio Apaporis), Cafio Paca. General location between Lat. 0°15’ S, Long. 70°30! W and Lat. 0°25’ N, Long. 70°30’ W. September 19, 1952, Richard Evans Schultes & Isidoro Cabrera 17568. RAPATEACEAE Schoenocephalium Martianum Seubert in Mar- tius Fl. Bras. 8, pt. 1 (1847) 130, t. 19. The curious and beautiful genus Schoenocephalium was based on material collected at Araracuara on the Rio Caqueta, Colombia, by Martius in 1820. Two species were described: Schoenocephalium Martianum and 8. arthrophyllum. [190 | In 1944, I collected for a week at Araracuara, Mar- tius’ westernmost station in the Amazon and the type locality for a large number of endemics, some of which, like Schoenocephalium Martianum, had never subse- quently been found. I found this species to be one of the dominant plants on the high, flat savannah of white sand at the picturesque chasm called ‘‘Angostura.’’ Unfortu- nately, my collections of the plant from that trip were lost in an aeroplane accident. In December 1951, Dr. Garcia- Barriga and I spent an afternoon at Araracuara whilst emergency repairs were being made on our aeroplane. We were able to make an interesting collection of about fifty numbers which will be reported later. Amongst these was an ample topo- typical collection of Schoenocephalium Martianum. As we stepped off the aeroplane at the prison colony at Araracuara, we saw, near the shore, a small portable altar set up by visiting clerics for Christmas. This altar was extravagantly decorated with the beautiful pink-red wax-like inflorescences of Schoenocephalium Martianum, one of the rarest plants of the world! A number of months later, we were astonished to find a florist shop in Bogota with a window full of Schoeno- cephalium Martianum heads. It appears that a lucrative commerce has grown up, Every time a Colombian gov- ernment flight goes to the prison colony at Araracuara to relieve the guards, the returning police bring back enormous bundles of the long-lasting heads for sale to florists in the nation’s capital under the common name estrellitas del sur (‘little stars from the south’’). The retail price in Bogota in January 1953 was three pesos a dozen (approximately $1.20 in U.S. money). I collected seed of Schoenocephalium Martianum and sent it to the Bureau of Plant Industry of the United States Department of Agriculture and the Royal Botanic { 191 | Gardens at Kew, in the hope that this precious little gem might be introduced to horticulture, but both attempts were failures. Cotomsi1a: Comisaria del Amazonas, Rio Caqueté, Araracuara. Sa- bana de Angostura. Alt. 400 m. Suelo pedregoso, con arena blanca. ‘“Yerba erecta. Inflorescencia 0.48 m. Flores blancas en el dpice, en la base rojas.”” December 21, 1951, H. Garcia-Barriga & Richard Evans Schultes 14172. BROMELIACEAE (Contributed by Lyman B. Smith) Pitcairnia macarenensis L. B. Smith sp. nov. Pitcairniae pungentt HBK. in systema Mezii proxima, sed foliis majoribus integerrimis petiolatis, vaginis folio- rum atro-castaneis, laminis foliorum lineari-lanceolatis canaliculatis differt. Stemless, the flowering shoot 55 cm. high; leaves en- tire, dimorphic, some reduced to broadly ovate apiculate dark castaneous sheaths, others over 1 meter long with slender elongate petioles, the blades linear-lanceolate with a strong median channel, filiform-acuminate, 30-35 mm. wide, flat, glabrous; scape erect, slender, sparsely white-flocculose; scape-bracts erect, imbricate, lanceo- late, acuminate, sparsely pale-lepidote; inflorescence simple, dense, 7-9 cm. long, white-flocculose except for the petals; floral bracts narrowly triangular, much ex- ceeding the pedicels, pale green; pedicels slender, 8 mm. long; flowers suberect; sepals lance-oblong, acute, 28 mm. long, the posterior ones strongly carinate; petals slightly zygomorphic, linear, obtuse, 45 mm. long, red, bearing a large oblong truncate scale at base; stamens included; ovary three-fifths superior; ovules caudate. Cotompia: Intendencia del Meta, northeastern end of Cordillera La Macarena, slopes of Cuchillo Palmitas, Macizo Renjifo. Alt. 1,500 m. “On rocks.’’ March 1951, Richard Evans Schultes 12126 (Tyre in U.S. Nat. Herb. No. 2048070). [ 192 ] Pitcairnia patentiflora L. B. Smith in Contrib. Gray Herb. 127 (1939) 18, t. 1, fig. 4. This species, one of those with a distribution on the Guiana-Venezuela land-mass, has been reported from Colombia as far west as Cerro Chiribiquete (in Bot. Mus. Leafl. Harvard Univ. 12 (1946) 121). The collection cited below is intermediate between the type locality, Cerro Duida, and Chiribiquete. CotomsBiA: Comisaria del Vaupés, Rio Naquieni, Cerro Monachi. ee . “ae: . Flowers and axis of inflorescence red.’’ June 1948, Richard Evans Schultes & Francisco Lopez 10082. Brocchinia hechtioides Mez in Fedde Rep. Spec. Nov. 12 (1918) 414. The specimen cited below is the first record for Colom- bia of a species previously known from Mount Roraima and the Amazonas Territory of Venezuela. CoLompraA: Comisaria del Vaupés, Rio Kananari, Cerro Isibukuri. Alt. 250 m. above the river, 700 m. above sea-level. Crest of the cerro in rocky sandstone soil. November 29-30, 1951, H. Garcia- Barriga 13796. Navia caulescens Martius ex Schultes filius var. minor Schultes & Schultes filius in Roem. & Schult. Syst. 7 (1880) 1195. The species and its variety minor were described to- gether and, in default of any designation to that effect, the inference is that the second locality mentioned, Ara- racuara, applies to the variety. The recent collections cited below would tend to confirm this view and also to show a large extension of the range of the variety. There is a considerable variation in the dimensions of the leaves but no notable difference in the flowers. Cotompia: Comisaria del Amazonas, Rio Caqueté, Araracuara: Sa- bana de La Angostura. Alt. ca. 400 m. December 21, 1951, H. Garcia- Barriga & Richard Evans Schultes 14171.—Comisaria del Vau- pés, Rio Kananari, Cerro Isibukuri. November 29-30, 1951, H. [ 193 | Garcia-Barriga 13798, 13816.—Comisaria del Vaupés, Rio Karuri, Savannah of Goo-ran-hoo-da. Alt. ca. 240 m. ““In large mats on sand- stone rocks exposed tosun.’’ April 15-16, 1953, Richard Evans Schultes & Isidoro Cabrera 19168, Navia Garcia-Barrigae L. B. Smith sp. nov. Ab omnibus speciebus adhuce descriptis inflorescentia magna, laxe ampleque tripinnatim paniculata differt. Stemless (! Garcia-Barriga), the flowering plant over 8 dm. high; leaves rosulate, the sheaths unknown, the blade linear, acuminate, 22 cm. long, 8 mm. wide, very laxly serrate with minute curved ascending spines, cov- ered on both sides with white appressed scales, roseate for 2 cm. at base; scape unknown; inflorescence nearly 8 dm. long, amply and laxly tripinnate, glabrous, red- brown when dry; primary bracts narrowly triangular, 16 mm. long, much shorter than the naked flattened sterile bases of the branches, entire, nerved: branches spreading, to 25 cm. long, their axes straight and very slender; racemes to 9 cm. long including the short sterile base, subdensely many-flowered; floral bracts broadly ovate, acuminate, 2 mm. long, thin; flowers obscurely pedicellate, spreading; sepals homomorphic, subtriangu- lar, obtuse, 2 mm. long, ecarinate; petals 3.5 mm. long, white; ovary superior; ovules unappendaged. Cotomsria: Comisaria del Amazonas, Rio Caqueta, Araracuara, Sa- bana de La Angostura. In white sand among rocks. December 21, 1951, H. Garcta-Barriga & Richard Evans Schultes 141387 (Vyrr in U.S. Nat. Herb. No. 2057246). Navia heliophila L. B. Smith sp. nov. Herba plusminusve caulescens sed foliis vivis apice solum praedita; foliis multis, rosulatis, pro genere amplis, dense serrulatis, glabris; scapo brevissimo; inflorescentia densissime paniculata, globosa; sepalis oblongo-lanceo- latis, minimis. Somewhat caulescent; most of the stem covered with [ 194 ] decayed leaf-bases; living leaves many in a spreading rosette at the apex of the stem, the sheaths completely covered, the blades linear, acute, pungent, 25 cm. long, 17 mm. wide, flat, densely serrulate, glabrous; scape very short; inflorescence very densely paniculate, glo- bose, 25 mm. in diameter; primary bracts lance-triangu- lar, pungent, serrate, shorter than the globose short- stipitate spikes; floral bracts ovate, acute, slightly but consistently shorter than the sepals, ferruginous-floccu- lose at base; pedicels short and inconspicuous; sepals lance-oblong, acute, 6 mm. long, the posterior ones sharply carinate and connate for 2 mm.; petals white (! Schultes), the blades spreading, narrowly elliptic. In habit, Navia heliophila closely resembles N. angus- tifolia (Bak.) Mez of Guiana and WN. wyridifiora L. B. Smith of Venezuela, but it differs from the former in its much smaller flowers and from the latter in its sharply acute sepals. CotomsiA: Comisaria del Amazonas, Rio Apaporis, Raudal de Jiri- jirimo. “On exposed ledge in savannah and caatinga.’’ March 1951, Richard Evans Schultes 12085 (Tyrer in U.S. Nat. Herb. No. 2048050). —Same locality. November 25-26, 1951, H. Garcia- Barriga 13712, 138736. Navia Lopezii L. B. Smith ex Schultes in Bot. Mus. Leafl. Harvard Univ. 15 (1951) 40. It is now possible to publish the plate (XX VIII) pre- pared to illustrate this species. Navia Lopezii L. B. Smith ex Schultes var. colom- biana L. B. Smith var. nov. A Navia Lopez inflorescentiae bracteis exterioribus quam sepalis brevioribus, petalis albis differt. Cotoms1a: Comisaria del Vaupés, Rio Kananari, Cerro Isibukuri. Alt. 250-700 m. “*On cliff in shade. Bracts of flowers purple outside, pinkish or yellowish inside. Flowers white, Leaves almost membra- naceous.’’ August 4, 1951, Richard Evans Schultes & Isidoro Cabrera 13342 (Tyee in U.S. Nat. Herb. No. 2120933). [ 195 ] Navia myriantha L. B. Smith ev Schultes in Bot. Mus. Leafl. Harvard Univ. 15 (1951) 41. Herewith is a drawing (PI. X XN VIII) which inadver- tently was not published with the original description. Navia reflexa L. B. Smith sp. nov. A Navia Garcia-Barrigae L. B. Smith, supra des- cripta, omnibus partibus multo majoribus, ramulis ulti- mis pendulis, floribus reflexis, ovario 4 infero differt. Terrestrial, the flowering plant over 8 m. high; leaves (only one known) 9 dm. long, the sheath subquadrate, 5 em. long, barely wider than the blade, white except for the lustrous brown apex, entire, nearly glabrous, the blade linear with a long entire acuminate pungent apex, 5 em. wide, flat, closely sulcate, laxly serrate with dark ascending spines 1.5 mm. long, covered above with a very tine white membrane of coalesced scales, glabrous be- neath; scape erect, over 14 mm. in diameter, glabrous (at least in age); scape-bracts erect, foliaceous but much reduced, much exceeding the internodes, but so narrow us to leave most of the scape exposed; inflorescence am- ple, lax, at least tripinnate (only the apical part known), glabrous (at least in age); primary bracts broadly triangu- lar, several times shorter than the naked flattened sterile bases of the branches, entire; branches spreading with the ultimate divisions pendent; racemes to 20 cm. long including the short naked sterile base, densely many- flowered, the rachis strongly angled ; floral bracts broadly ovate, acuminate, 7 mm. long, entire, subcoriaceous; flowers obscurely pedicellate, reflexed ; sepals free, heter- omorphic, unequal, elliptic, the anterior one shorter than the others, ecarinate, the posterior ones 6 mm. long, alate-carinate; ovary 4 inferior; seeds wingless. Cotomsia: Comisaria del Vaupés, Rio Negro below its confluence with Rio Casiquiare, San Felipe (El Castillo). December 12, 1947, Richard Evans Schultes & Francisco Lopez 9321 (Typr in U.S. Nat. Herb. Nos. 1989460, 1989461). [ 196 |] Vriesia Schultesiana L. B. Smith sp. nov. Acaulis; foliis haud bulbose rosulatis, vaginis distinc- tis, dense ferrugineo-lepidotis, laminis linearibus, planis; scapi vaginis haud vel vix imbricatis; inflorescentia sim- plicissima; rhachi subtereti; bracteis florigeris laxe im- bricatis, late ellipticis, sepala multo superantibus, ecari- natis, nervatis; floribus distichis; sepalis liberis, oblongis, VRIESIA Schultesiana L.B. Smith a, habit, one third natural size. b, inflorescence, about natural size. c, sepal, natural size. d, petal and stamens, natural size. subtruncatis, parvis, sparse lepidotis; petalis basi ligulis binis obtusis auctis; staminibus inclusis. Stemless, 12-23 cm. high; leaves 8-15 in a fasciculate rosette, 22 cm. long, sometimes exceeding the inflores- cence, green, concolorous, the sheaths broadly ovate, covered with coarse subappressed ferruginous scales, the blades linear, acuminate, 3.5 mm. wide, ferruginous- lepidote, soon becoming glabrous; scape erect or ascend- [ 197 ] ing, very slender, glabrous; scape-bracts erect and en- folding the scape, barely or not imbricate and exposing sections of the scape but their caudate apices always ex- ceeding the internodes, very obscurely lepidote; inflores- cence simple, linear, complanate, 4-6 em. long, 1-1.5 cm. wide; rhachis slender, subterete; floral bracts laxly imbricate and exposing parts of the rhachis, broadly el- liptic, subacute, 15-18 mm. long, much exceeding the sepals, ecarinate, subcoriaceous, nerved when dry, gla- brous outside or sparsely lepidote near the apex, pale yellow (! Schultes); pedicels very short, broadly obconic: sepals free, oblong, obliquely subtruncate, 8.5 mm. long, ecarinate, thin, nerved, sparsely lepidote; petals oblong, obtuse, 13 mm. long, bearing 2 large, obtuse, entire scales at base; stamens included; capsules shorter than the floral bracts. In my key to the Tillandsiae (in Contributions from the U.S. National Herbarium 29 (1951) 448-455) this species would fall next to Tv//landsia incurva Griseb. and T. patula Mez, since the arrangement is frankly artificial. Its pale leaf-sheaths and small dimensions throughout easily distinguish it from both without reference to its appendaged petals. The nearest species of Vriesia is V’. Barclayana (Baker) L. B. Smith, but this last differs basically in its alate rhachis, while also being immediately distinguishable by size and brown leaf-sheaths. CotompiA: Comisaria del Amazonas, Rio Apaporis, Raudal de Jiri- jirimo. “On ledge of conglomerate quartzite. Savannah (caatinga).*’ March 1951, Richard Evans Schultes 12088 (Tyrer in U.S. Nat. Herb. No. 2048059).—Same locality. November 25-26, 1951, H. Garcia- Barriga 13710. VELLOZIACEAE Vellozia Maudeana R. i. Schultes sp. nov. Frutex usque ad duos pedes altus. Caudex subligno- sus, erectus vel saepe procumbens, basi ad 6-8 cm. in [ 198 | diametro, quinque- vel sex-furcatus, apparenter fibrosus cum radicibus internis foliorum vaginis circumdatis, vag- inis comparate parvis, griseo-stramineis, persistentibus, arcte adpressis, subspiraliter imbricatim dispositis, ves- tigiorum apicibus laciniatis et valde revolutis obtectus, in parte superiore distincte lineatis. Folia in apice ramor- um subrosulata conferta, non numerosa (plusminusve decem ad quattuordecim), rigide erecta, sicca, firme cori- acea, utrinque glabra, margine infra medium et in foliis juvenilibus omnino pilis conspicuis albis, simplicibus vel irregulariter plurichotome ramosis, usque ad 2 mm. lon- gis armata, utrinque sed subtus spiraliter suleata cum vena centrali atque venis minoribus viginti octo valde elevatis, pallide viridia sed basim versus paullo rubentia, linearia, margine integra, valde stramineo-revoluta, apice longis- sime et sensim attenuata, basi non dilatata, plerumque 27-34 cm. longa (sed vulgo breviora), 6-8 mm. lata. Flores fragrantes, solitarii, pseudoterminales, magni spe- ciosissimique, albised basim versus pallide rosei vel pallide rubri, foliis paullo breviores, longe pedunculati; pedun- culo filiformi, triquetro, inferne glabro sed superne vis- coso-glanduloso, minute squamoso-echinato, rubro-pur- pureo, usque ad 14cm. longo sed saepe breviore. Perigonii tubus tenuis gracilisque, cylindricus, 75 mm. longus et 4 mm. in diametro, extus densissime purpureo-viscoso- glandulosus, ovarii regione valde inflatus; limbus infun- dibuliformis, vivo 6.5—7 cm. in diametro, segmentis mem- branaceis, imbricatis, subspathulato-ellipticis, apice irreg- ulariter rotundatis, vivo 45 mm. longis, basi 3-4 mm. (sed in parte latiore usque ad 12-15 mm.) latis, glabris sed extus in parte inferiore glandulosis. Stamina duodecim, tepalis multo breviora; antherae subaequales, elongissime cylindricae, 9 mm. longae (8-10 mm.), glabrae, flavae, filamentis 1 mm. longis. Ovarium oblongo-clavatum, apice truncatum, paleis illis pedunculi similibus sed multo ( 199 | robustioribus, viscoso-glandulosis, luteis, dense ornatum. Stylus flavus, robustius filiformis, 8.5 cm. longus, 1-1.3 mm. in diametro, trigonus, inclusus, stamina multo su- perans. Stigma flavum, trilobatum, crassissimum, 4-4.5 mm. in diametro. Capsula elongato-ovata, 10 mm. X 6 mm., dense et grossiuscule glanduloso-echinulata, conspi- cue trivalvata, protractione usque ad 6 cm. longa, basi 4—5 mm. in diametro sed apice filiformi, dense echinu- lata, nune aliquantulum arcuata, nune stricta coronata. Vellozia Maudeana, one of the most beautiful species of the genus, can easily be distinguished from other Co- lombian species by the unusual coloration of its flowers. In all other known species of northern South America, the flowers are entirely white, but those of Vellozia Maudeana are a deep pink or red towards the base, with a red-purple peduncle. Some of the flowers are a pale pink for fully half their length. This tendency towards a pink or red coloration can be seen even in the vegeta- tive parts of the plant, for the basal portions of the leaves are likewise of a slightly reddish hue. I have studied a number of individuals at the type locality and find that this is a constant character. There are, nevertheless, other important characters which serve to set Vellozia Maudeana apart. The num- ber of the stamens is twelve, whereas all other species from Colombia have either fifteen or eighteen. The sub- spatulate-elliptic shape of the segments of the tube is also apparently peculiar to this species. The flowers of Vellozia Maudeana have a strawberry-like fragrance which I have never before noted in the genus. The flow- ers of other Colombian species seem to lack any odor. In habit, Vellozia Maudeana approaches perhaps most closely to V’. macarenensis Philipson, for it is small and does not branch so profusely as do most of the other spe- cies. There are, however, no indications that these two [ 200 | concepts are morphologically closely allied. In some re- spects, Vellozia Maudeana resembles V. Dumitiana R. E. Schultes, especially in having a small number of leaves at the apex of each stem; but the latter is, in general, a much more robust plant than the former, and there are significant differences in floral structure. Vellozia Maudeana grows on the grotesquely eroded quartzitic mounds of the extensive savannahs of the Rio Karurti which are known by the Kubeo Indian name of Goo-rdn-hoo-da, meaning ‘‘savannah of the deer.” These savannahs are ecologically the same and, as I have ascer- tained from reconnaissance flights, are continuous with the savannahs of the Rios Kubiyt and Kuduyari-Kanen- di and Yapoboda respectively. Vellozia Maudeana has not been found at Kanenda nor at Yapoboda, where Vl’. lithophila R. KE. Schultes, the only representative of the genus at these localities, is one of the dominant shrubs. These two species, notwithstanding their geographical proximity, are not closely related. Vellozia usually preters rocky habitats where condi- tions of chersophytic or psammophytic drought prevail, but the xerophytism which Vellozia Maudeana with- stands at Goo-ran-hoo-da is extreme even for this genus. The plant seems to prefer craggy exposures or knobs of bare quartzite which erosion has left standing here and there on the flat sandy stretches in the extensive savan- nah. Most closely associated with Vellozia Maudeana, a shrub which usually stands quite alone amidst a harsh or gorsey growth of grasses, sedges and prostrate creepers, are Navia caulescens var. minor (filling in cracks and crev- ices in the rocks) and Leitgebia colombiana, Bombax coria- ceum and a red-flowered Calliandra (on the flat expanses). All of these plants are, like Vellozia Maudeana, adapted to withstand extreme radiation and heat in an area where no shade whatsoever lessens the strength of the tropical sun. [ 201 ] { have named this beautiful novelty in honor of my mother, Maude Bagley Schultes. Cotombia: Comisaria del Vaupés, Rio Karurti (tributary of Rio Vaupés), Mesa de Yambi, quartzite savannah Goo-ran-hoo-da. Alt. about 950 feet. General location: Lat. 1°20’ N, Long. 71°20/ W. **Flowers fragrant, with scent of strawberries, white, base of tube red outside, becoming dark purple at very base, often very pale pink even half way up tube. Stigmas bright yellow. Plants in clusters, much- branched, leaves all short, marginal hairs very prominent. Plants many-flowered. Basal diameter at ground up to 34-4 inches. Kubeo name =dd-ko-ree.’’ April 15-16, 1953. Richard Evans Schultes & Isidoro Cabrera 19120 (Type in Herb. Gray). LORANTHACEAE Psittacanthus peronopetalus Michler in Martius I'l. Bras. 5, pt. 2 (1868) 81, t. 9, fig. 4. This species has not hitherto been collected from Co- lombia. The material cited in the original description ‘ame from Manaos and ‘Teffé (ga) in Amazonian Brazil. Cotombia: Comisaria del Vaupés, Rio Naquieni, Cerro Monachi. June 1948, Richard Evans Schultes & Francisco Lépes 10086, M ALPIGHIACEAE Tetrapteris methystica R. 1. Schultes sp. nov. Frutex scandens robustior, trunco nigro cum cortice. Rami cinereo-fulvi, internodiis 4-10 cm. longis.. Ramuli teretes, obscurissime canaliculati, novellissimi minute incano-sericei vel leviter schistacei, 0.8—8.8 mm. in diam- etro. Folia firme papyracea vel chartacea, ovata, apice longiuscule acuminata, basi plerumque bene rotundata, margine integra sed saepe leviter revoluta, 6-8.5 cm. longa, 2.5-5 em. lata, valde discoloria, supra vivo viridia clara (sed siccitate glaucino-straminea), minute et remote sericea, subtus vivo cinereo-viridia, densius sericea et cerae lamina obtecta; nervis secundariis arcuatis, utrinque sex ad octo, supra prominulis, subtus prominentibus sed non conspicue elevatis, nervis tertiis inconspiculs, densis- [ 202 ] sime reticulatis, petiolo usque ad 5 mm. longo, canalicu- lato, aliquid incrassato, dense cinereo-sericeo. Stipulae mox caducae, parvae. Inflorescentiae pseudocorymbosae, pauciflorae (ut videtur usque ad quattuor- vel quinque- florae), in paniculis axillaribus, foliis multo brevioribus, usque ad 2.5 vel 8 cm. longis, vivo probabiliter plusmin- usve 15 mm. in diametro, apparenter sine foliolis (4); pedunculi internodio inferiore 10 mm., pedicellis plus- minusve 5 mm. longis, pedunculis pedicellisque densius sericeis. Bracteae subulatae, plerumque 1.5 mm. longae. Bracteolae breviter ovato-triangulares vel saepe suborbic- ulares, plusminusve 1.5 mm. longae. Sepala crassa, extus pilosa, ovato-lanceolata, apice subacuta, usque ad circiter 3mm. longa, nigris cum glandulis octo, ovalibus, plerum- que 2.5 cm. longis, basi extus pilosis, plusminusve 0.5 mm. superantibus. Petala patentia, membranacea, max- ima pro parte lutea sed parte centrali fulva vel rubra, limbo elongato-orbiculari vel ovali, apice rotundato, basi obtuso vel rotundato, margine subcrenulato (rarenter sub- integro), parte centrali dorsali aliquid carinato-incrassato, 4mm. longo, plerumque 2.5 mm. lato, ungui crasso, 0.5 mm. longo, aliquid reclinato. Stamina non inclusa, ae- qualia; antherae allantoidae, 1.8 mm. longae, 0.4 mm. in diametro, valde arcuatae, filamentis bene complanatis, 1.8 mm. longis, basi usque ad 0.4mm. latis. Styli aequi- longi, apice leviter recurvi. Ovarium densissime albo- pilosum. Samarae ad nucem sericeae, demum glabrae; nux fere complanato-ovoidea, 5 mm. X6 mm. X2 mm. ;: areola ventralis ovata, circiter 8 mm. alta; alae charta- ceae, fulvae, laterales inferiores obcuneiformes, apice truncato-rotundatae, 12 mm. longae, apicem versus 2.5 mm. latae, superiores similes sed saepe subovoideae et paulo majores; alula dorsalis subsemiorbicularis, 8.5—4 mm. longa, illae intermediae ovatae, 7-8 mm. longae. Tetrapteris methystica is sharply set apart from almost [ 203 ] all other South American species of the genus by its strongly discolorous leaves. In many respects, it seems to approach most closely to Tetrapteris discolor (G. F. W. Meyer) DC., a rather polymorphic species which, with its several varieties, occurs from Guatemala and the West Indies south to Bolivia. Tetrapteris methystica may be distinguished from 7°‘. discolor by its smaller and more long-acuminate leaves which are, even in the adult stage, sericeous on both surfaces and which are covered beneath with a thick layer of wax; by its apparent lack of folio- laceous stipules (which, if they do occur, are extremely caducous); by having a more corymbiform inflorescence ; by its long allantoid (instead of obovoid) and recurved or arcuate anthers; by its rather acute and loosely seri- ceous (instead of rounded and glabrous) sepals with smaller glands; by its basally rounded (instead of strongly sagittate) petal-blades; and by the shape of the upper lateral wings of the samara, which are obcuneiform in outline. Tetrapteris methystica has been so named because it is employed by the Maku Indians of the Ira-I garapé (and possibly by other tribes of the upper Rio Negro-Vaupés area) as the source of astrong narcotic drink. It is known as caapl, the same name which is applied to the related Banisteriopsis Caapi (Spruce ex Griseb.) Morton, the source of the well known narcotic discovered by Spruce a century ago in the same region (cf. Spruce, R. [ed. A. R. Wallace] ‘‘Notes of a botanist on the Amazon und Andes’’ 2 (1908) 414 ff.). Inasmuch as [ am preparing a comprehensive article on the malpighiaceous narcotics of South America, a discussion of the use of Tetrapteris methystica will be de- ferred for that general paper. The use of a member of the genus Vetrapteris as a narcotic was first reported in the literature in 1952 (Hill, A. F. ‘‘ Economie Botany” [ 204 ] ed. 2 (1952) 283, t. 143), on the basis of the collection described above. Brazit: Estado do Amazonas, Upper Rio Negro Basin, Ira-Igarapé, affluent of Rio Tikié, Porto Maktis, June 25, 1948, Richard Evans Schultes & Francisco Lopez 10184 (Tyrer in Herb. Gray). EUPHORBIACEAE Senefelderopsis Steyermark ex Schultes in Bot. Mus. Leafl. Harvard Univ. 15 (1951) 45. Senefelderopsis chiribiquetensis (Schultes & Croizat) Steyermark ex Schultes ibid. 47. Senefelderopsis Croizatii Steyermark ex Schultes ibid. 46, t. 16. Through an editorial error, the generic and specific descriptions cited above were published under the family heading Anacardiaceae. The genus is a member of the Luphorbiaceae. STERCULIACEAE Recent investigations of Herrania in the field and herbarium have led to several additions to our knowledge of this interesting sterculiaceous genus. ‘To the best of my knowledge, the first analysis of the pollen-grains of Herrania is that effected by Dr. Thomas van der Hammen of the Servicio Geol6gico Nacional de Colombia and published herewith. The species studied was Herrania tomentella which is described in this paper. Herrania tomentella R. E, Schultes. Pollen collection Serv. Geol. Nac. IV 86. Collection Col, 34377. Pollen grains: tricolporate, reticulate, subsphaeroidal; granulae of the muri visible but not separated. Lumina of reticulum +irregular of size, rather large, polygonal, smaller near the colpae. In the lumina, rather faint granulae are visible. Colpae clear, edges separated; pores clear, without ectexine ele- [ 205 | ments; sometimes indications of small transversal furrows. Magnitudo pollinis: media (28-33 »). Magnitudo luminum: meso-macra (2-4.7 #) and smaller. Index pollinis; subsphaeroidea (1-1.12); (‘prolate sphaeroidal’ of Erdtman). Index areae poli: middle (0.30-0.35). Index exinae: middle (0.05-0.08). Van der Hammen further reports that Theobroma, Guazuma, Sterculia and Herrania have tricolporate and reticulate pollen-grains, but that the grains of Waltheria and Helicteres are of different types. A comparison of the four genera which have tricolporate and reticulate grains leads to the really unexpected conclusion that, in- sofar as pollen morphology is concerned, there is no evi- dence that Herrania and Theobroma are very close allies. On the contrary, the pollen-grains of Theobroma resem- ble those of Guwazuma rather than those of Herrania. He compares the grains of Theobroma Cacao L. and of Herrania tomentella as follows: ‘*The grains of Theo- broma Cacao are subsphaeroidal (‘oblate sphaeroidal’ of Krdtman); index pollinis 0.8-0.9; magnitudo pollinis = 22 w. The polar area is relatively much larger than in those of Herrania, the colpae are very narrow, unclear and short. Pores are small and not very clear; lumina of ret much smaller than in those of Herrania (greatest size measured 1-1.75 ), and more regular. Exine (including sculpture) relatively thicker than in those of Herrania.” This evidence of relationship drawn from a study of pollen-grains should be of interest, inasmuch as there has been a strong tendency in the past to treat the species of Herrania as constituting a section of Theobroma. The following notes comprise the description of an hitherto unknown species, reports of new range exten- sions for several species and miscellaneous information resulting from the collection of more complete material of several concepts. (206 J Some of the observations have arisen from my own studies in the Amazonian areas of Colombia. Others rep- resent the work of the Anglo-Colombian Cacao Collect- ing Kxpedition which has been exploring the forests of Colombia for material of Theobroma and Herrania since July 1952, and which it has been my good fortune to accompany on several trips into the Comisarias del Am- azonas, Putumayo and Vaupés. Herrania albiflora Goudot in Ann. Sci. Nat. Paris, sér. 8, 2 (1844) 230, t. 5, figs. 1-10. In spite of the reddish hue reported for the flowers by the collector (1 suspect that this refers to the outside of the calyx-lobes), Aristeguieta 1598 is referable to Her- rania albiflora. \t represents the second Venezuelan col- lection of this species, the first being Tejera 268 from near Periaé in the Estado de Zulia. At the present time only two species of Herrania are known from Venezuela: A. albiflora and H. lemniscata (Schomb.) R. E. Schult. A. kanukuensis R. EK. Schult. and H. Camargoana R.K. Schult., from boundary regions of adjacent countries, are also to be expected in the Venezuelan flora. VENEZUELA: Estado de Barinas, Fundo Paiva, Santa Barbara de Barinas. ‘““Arbusto de mds o menos 2 m. de alto. Fls. caulinares, rojizas. Frutos de mas o menos 10 em. de largo, amarillentos, encier- ran una pulpa agri-dulce, comestible. Abunda en el estrato arbusti- forme de laselva. N.v.: cacaito.’? February 1952, L. Aristeguieta 1598. Herrania breviligulata R. LH. Schultes in Caldasia no. 4 (1942) 20. t.p. 21, 24; Bot. Mus. Leafl. Harvard Univ. 16 (1958) 78, t. 12. On a recent trip to the Comisaria del Putumayo ad- ditional material of Herrania breviligulata was found at Mocoa, the type locality of this species. Hitherto, this strikingly distinct concept has been known through but two collections: the type, Schultes & Smith 2050, a flow- [ 207 | ering collection; and Meaia 7328, from Ecuador, which is sterile. The fruit has not been known, but several of the collections made in March 1958 and cited below, are in fruit. We are, therefore, able to offer the following description of the capsule: Capsula (unica visa non matura) oblongo-ovoidea, 10 cm. longa, 5 cm. in diametro, apice abruptissime acu- minata, basi haud indentata, petiolo robusto usque ad 2 cm. longo, quinque cum costis primariis regularibus, grossiuscule hebetato-rotundatis, usque ad 7 mm. altis et basim versus usque ad 8-10 mm. latis, et cum quinque costis secundariis similibus sed paulo minoribus, costis omnibus proximis sine costis fibrosis horizontalibus, peri- carpio duro, omnino densissime atque molliter stellato- velutino. Cotompi1a: Comisaria del Putumayo, Mocoa. ‘‘Suckers, 3 m. tall, from an old tree. One immature fruit, no flowers. In second growth.*’ March 17, 19538, F. W. Cope & P. Holliday (Anglo.-Col. Cacao Coll. Exped.) 78.—Comisaria del Putumayo, Rio Caqueté, Puerto Limén. “Fruit 10 em. long <5 em. in diameter.’’ March 17, 1953, Richard Evans Schultes & Isidoro Cabrera 18720. Herrania Camargoana RP. H. Schultes in Bot. Mus. Leafl. Harv. Univ. 14 (1950) 120, t. 29, 82. Herrania Camargoana, hitherto known only from the Brazilian basin of the upper Rio Negro, has recently been collected in Colombia along the lower course of the Rio Vaupés, on the Rio Papuri and on the Rio Negro, just below the confluence of the Rio Guainia and the Casi- quiare. This species, in Brazil, was always found in association with granitic mountains, occurring most frequently near or at the summits. The Anglo-Colombian Cacao Collect- ing Kxpedition found the tree growing in sandy situa- tions along the river banks of the Vaupés and Papuri, and the material from the Rio Negro was collected along [ 208 | the main river and in abandoned house sites on an afflu- ent creek. All of the fruits of the many Brazilian collections were a brilliant blood-red, a character which was especially emphasized in the original description, for the reason that only one other species— Herrania laciniifolia Goudot ex Triana & Planch. of the Colombian Andes—is known to have capsules of this color. It is important now to note that all of the fruiting Colombian collections cited below have capsules which give definite evidence of ripening yellow. It is obvious, then, that two colors are assumed by the fruit of Herrania Camargoana. Whether the red color is due to some mineral deficiency (inasmuch as the plants with fruits of this unusual color grow on very ster- ile granite rock-soil) or not we are, as yet, unable to say. Cotompra: Comisaria del Vaupés, Rio Negro, Cafio Maiyabo, near San Felipe. “‘Tree 16 ft., first collection of this species in Colombia. Leaves and one ripe fruit (sent to Trinidad); native, in forest. Flowers pale yellow.’’ October 27, 1952, R. E. D. Baker (Anglo.-Col. Cacao oll. Exped.) 38.—Same locality and date. Richard Evans Schultes & Isidoro Cabrera 18050.—Comisaria del Vaupés, Rio Negro, San Felipe. ‘Tree 15 feet, in full flower, no fruit. On old house site; seed said to have been brought from the locality of No. 35.’’ November 5, 1952, R. E. D. Baker (Anglo-Col. Cacao Coll. Exped.) 39.—Comisaria del Vaupés, cafio opposite Inambi. November 18, 1952, B. G. Bartley & P. Holliday (Anglo-Col. Cacao Coll. Exped.) 45. Brazit: Estado do Amazonas, creek near north end of Inambit, Rio Uaupés. ““Tree 4m. Fls. Cal. purple-crimson; pet. deep crimson on dull yellow; ligules whitish and purple-crimson at extremities, 9-12 em. long. Fruit 8-8.7 <6 cm.’’ November 17, 1952, B. G. Bartley & P. Holliday (Anglo-Col. Cacao Coll, Exped.) 56. Herrania nycterodendron RF. Ll. Schultes in Cal- dasia no. 6 (1943) 21, t. p. 22, 26; Bot. Mus. Leafl. Harvard Univ. 14 (1950) t. 35. Known hitherto only from the Putumayo River, the boundary between Perti and Colombia, Herrania nyc- terodendron has recently been collected on the Caqueté [ 209 ] at La Pedrera and on the Rio Caguan, an afHuent of the upper Caqueta. It is relatively frequent in these locali- ties. At La Pedrera, it occurs with Herrania nitida (Poepp.) R. KE. Schult. on flood-banks and islands. From the collection Schultes & Cabrera 17778, we now know that the capsule of Herrania nycterodendron ripens yellow, as in most other species of the genus. The buds of Schultes & Cabrera 17773 had been attacked by insects and were hollow and devoid of most floral parts. We can say, however, that the buds are conic, almost pointed, not globose as in most species. Cotomaia : Comisaria del Amazonas, Rio Caqueta, La Pedrera. ‘“Tree 6-7 m., 10 cm. diam, at base. Flowers not seen.’’ September 25, 1952, R. E. D. Baker & F. W. Cope (Anglo-Col. Cacao Coll. Exped.) 14.—Same locality. ‘‘A young plant without flowers or fruit.’’ Sep- tember 27, 1952, R. FE. D. Baker & F. W. Cope ( Anglo-Col. Cacao Coll. Exped.) 20.—Comisaria del Caqueta, Rio Caucaya, Laguna Primavera. ‘*Tree 6-8 m., unbranched. In forest. Ripe fruit 13.5 4.5 em.”’ April 3, 1953, F. W. Cope & P. Holliday (Anglo-Col. Cacao Coll. Ex- ped.) 93.—Comisaria del Caqueté, Rio Caguan, Cartagena. ‘‘Tree 4-5 m., unbranched. In forest. Ligules of dead flowers 12 cm.’’ April 20, 1953, F. W. Cope & P. Holliday (Anglo-Col. Cacao Coll. Exped.) 108.—Comisaria del Caquetaé, Rio Caguin (camp 2). ‘‘Tree 6-8 m., unbranched. In forest. Flowers and fl. buds, but no fruits. Cushions whole length of trunk, infl. with about 45 fls. Fls. deep dull purplish crimson and cream. Ligules 7 em.’’ April 22, 1953, F. W. Cope & P. Holliday (Anglo-Col. Cacao Coll. Exped.) 109.—Comisaria del Ca- queta, Rio Caguén (camp 2). April 22, 1953, F. W. Cope & P. Hol- liday (Anglo-Col. Cacao Coll. Exped.) 110.—Comisaria del Caqueta, Rio Caqueté (camp 6). May 9, 1953, F. W. Cope & P. Holliday (Anglo- Col. Cacao Coll. Exped.) 128.—Comisaria del Caqueta, Rio Caqueta, Piedra Blanca. May 11, 19538, F. W. Cope & P. Holliday (Anglo-Col. Cacao Coll. Exped.) 131.—Comisaria del Amazonas, Rio Caqueta, La Pedrera. “‘On low flood-bank. Hairs soft, golden brown.’ October 6, 1952, Richard Evans Schultes & Isidoro Cabrera 17771.—Same lo- cality. ‘“Treelet 18 ft. tall. Basal diameter 6 inches. Hairs golden brown. Buds purplish maroon, conic, almost pointed.’’ October 7, 1952, Richard Evans Schultes § Isidoro Cabrera 17773.—Same locality. ‘“Treelet 18 ft. tall.’? October 7, 1952, Richard Evans Schultes & Isi- doro Cabrera 17776.—Same locality. ‘“Fruit yellow. Tree 20 feet tall. Diameter 6 inches.’’ October 7, 1952, Richard Evans Schultes & Isidoro Cabrera 17778. [ 210 ] Herrania pulcherrima Goudot in Ann. Sci. Nat. Paris, sér. 3, 2 (1844) 232, t. 5, figs. 11-12. Theobroma pulcherrimum (Goudot) de Wildeman PI. Trop. Grande Cult. (1902) 89. It is often overlooked that de Wildeman made the combination Theobroma pulcherrimum half a century ago. This combination is usually attributed to Pittier (Man. Pl. Usuales Venez. (1926) 147). The type specimen of Herrania pulcherrima has al- ways been thought to be in Paris, but, as in the case of H. albiflora (ct. Schultes in Bot. Mus. Leafl. Harvard Univ. 16 (1953) 75), there is material in Geneva which may well be the true type from which Goudot’s descrip- tion and illustrations (at least of the flowers) were made. Goudot (loc. cit. 232) spoke of the type as inhabiting the great forests situated between the Rios Ariari and Guay- abero, afHuents of the upper Orinoco system in the Co- lombian Llanos. The Geneva material consists of two sheets. It is labelled in Goudot’s hand: ‘‘Herrania pulcherrima mihi. An. Sci. Nat. 1844. Llanos del Orinoco, pueblo d’ Iraca, San Juan. Flos. Dec.*” One sheet consists of a length of golden-tomentose stalk about one foot long, a very young leaf and young capsules. One of the envelopes on the sheet has the Koregwahe Indian name (reported by Goudot in the original description): ‘cacao cahouai— Llanos.’” Another envelope, on the outside of which Goudot has written ‘‘CN. 2 theobroma affinis Herrania pulcherrima,’’ has a completely and beautifully dissected flower, the separate parts glued to the inside of the en- velope. There can be no doubt but that Goudot made his drawing of the flower of Herrania pulcherrima (loc. cit. t. 5, figs. 11-12) from this same dissection. The leaf which is preserved at Geneva could hardly have served asa basis for Goudot’s excellent description, [ 211 | but a study of the material and the description would seem to indicate that the Paris material represents the type of the leaf. It may be of value to publish a few notes on Goudot’s dissection of the flower. The three sepals are laid flat, the very slightly puberulent inner surface exposed. wo are rather broadly ovate, about 15 mm. long and 5 mm. wide (all measurements taken on dried material), apically rounded; the third, somewhat elliptic, 18 mm. long and 4 mm. wide, apically blunt-pointed. The five petals are all about equal, strongly cucullate, very densely muri- cate-papillose or granulose externally, papillose internally in six longitudinal lines, the ligules up to 90 mm. long, 2.5 mm. wide immediately above the constriction at the junction with the petal. The staminodes are lanceolate- elliptic, 15 mm. long, 4-4.5 mm. wide, muricate-gran- ulose and apically so strongly trifid that the tip appears to be mucronate. This was noted by Goudot, when he described the staminodes as apically ‘‘mucronés et échan- crés’>; but, in this drawing, he indicated the tip as ex- tremely acute. The ovary is very densely yellowish to- mentose. It is unusual to find a species of Herrania which occurs both east and west of the Andes, as well as in the valley between the several Andean chains in Colombia. Yet that is apparently the distribution of Herrania pulcher- rima. Goudot said that he had found it in the deep val- leys of the eastern Andean chain, near Savana Grande and Payme where, however, it seemed to be rare and isolated. I have seen no Goudot specimen from this lo- cality; but it is very significant, I think, to note that all earlier and a number of later collections were made, not in the eastern Llanos, but within the Andean cordillera itself. Vegetatively, Herrania pulcherrima can easily be con- [ 212 ] fused (and has been) with HZ. tomentella, a species grow- ing on the eastern Llanos at the foothills of the Andes where the type of HZ. puleherrima was collected. The differences between these two species are discussed below under Herrania tomentella. The earliest reference to Herrania pulcherrima is Eloy Valenzuela’s minute description of the plant written in Mariquita in the Departamento del Tolima, Colombia, in 1784, while he was engaged in the work of the Mutis Botanical Expedition in New Granada. The specific name pulcherrima, meaning ‘‘very beauti- ful,’* could not be more appropriate. It recalls Valen- zuela’s picturesque remark that the flower of Herrania pulcherrima or cacao esquinado ‘‘could be considered as the greatest marvel of the plant kingdom and one can hardly believe that nature, as frugal and simple as she is, would have used so many ribbons and so much orna- mentation to adorn herself almost as ostentatiously as in the fashions. ”’ ‘ Cotomsia: Intendencia del Meta, Iraca, San Juan [de Arama], Llanos Orientales. December 1844. J. Goudot s.n.—Departamento de Boyaca, region of Mount Chapon, extreme western part of the de- partment, northwest of Bogota. Heavy forest. El] Umbo region. Alt. 3000 feet. ““Tree 10 feet high. Flowers blood red.’’ September 13, 1932, A, BE. Lawrance 437.— Departamento de Cundinamarca, Muni- cipio de El Pefion, Hacienda “‘Curiche.’’ Alt. 1000 meters. **Varas altas y delgadas, de 4 metros.’’ September 2, 1942, R. Jaramillo- Mejia 202. Herrania tomentella Rk. EL. Schultes sp. nov. Herrania nitida (Poeppig) R. E. Schultes var. aspera (Karsten & Triana) R. E. Schultes in Bot. Mus. Leafl. Harvard Univ. 14 (1950) 180, pro parte. Disrrisurion: Eastern foothills of the Andes, in the Orinoco drainage area of Colombia. COMMON NAMES: cacao de monte; cacaoito. e258 Arbuscula parva, tenuis, venusta, vulgo plusminusve octo ad duodecim pedes alta; erecto cum trunco tereti, circiter tria pollices in diametro, cortice nigro obtecto; prope trunci apicem parce ramosa vel eramosa; ramis tomentosis. Ramuli dense villosi, ferrugineo-aureis cum pilis persistentibus. Folia amplissima, digitata, septem- foliolata, longissime petiolata. Petioli teretes, basi ali- quid constricti, densissime et molliter aureo-ferrugineo- tomentelli, usque ad 60 cm. longi, 9-10 mm. in diametro. Stipulae persistentes, lineares, dense aspero-tomentellae, usque ad 8 cm. longae, 2 mm. latae. Foliola sessilia, oblanceolata vel late lanceolato-ovata, erecta, valde in- aequalia, membranaceo-papyracea, apice acuminata, basi attenuata, margine dimidio superiore regulariter et levi- ter (praesertim apicem versus) sinuato-dentata et omnino pilis vel pseudociliis stellatis armata, 80-50 cm. longa, 13-20 cm. lata, supra aspera, sparse cum pilis fuscis soli- tariis, subtus submolliter et dense tomentella cum pilis longis stellatis ferrugineo-aureis. Inflorescentiae fascicu- latae, comparate pauciflorae, e trunci partibus inferiori- bus prorumpentes. Pedicelli articulati, 7 mm. longi, 1.5 mm. in diametro, stellato-pilosi. Alabaster floris globo- sus, 15 mm. in diametro, dense stellato-pilosus. Calyx trifidus, fere usque ad basim divisus, subeymbiformis. Sepala vulgo inaequalia, subcrassa, atropurpurea, in ala- bastro valde valvata, extus grossiuscule stellato-pilosa, intus minutissime granuloso-pulverulenta, duo interiora rotundato-ovata, margine integra, apice perfecte rotun- data, plusminusve 14 mm. longa, 10 mm. lata; sepalum exterius vulgo triangulari-ellipticum, margine integrum, apice subacutum, 13-14 mm. longum, basi 6-7 mm. latum. Petala quinque, basi sessilia, obovata vel ovata, apice valdissime concavo-cucullata, circiter 8 mm. longa, 7 mm. (saepe usque ad 8 mm.) lata, atrosanguinea cum nervis purpureis, extus minute muricato-verrucosa, su- [ 214 ] perne in ligulam extensa. Ligulae lineares, plusminusve 70 mm. longae, basi 8 mm. latae, apicem versus filiformes, atrosanguineae sed apicem versus roseae. ‘Tubus stamin- eus quinque-divisus cum staminibus invicem uno- et duo- antheriferis et cum filamentis brevibus liberisque. Stami- nodia petaloidea, atrosanguinea, membranacea, elliptica, margine integra, apice acuta, 14-15 mm. longa, 6-7 mm. lata, utrinque aliquid subverrucosa. Fructus non numer- osi, ellipsoidei, usque ad 9 cm. longi, 4 cm. in diametro, apice longe attenuati sed prope apicem aliquid constricti, apice ipso obtuso et saepe contorto, basi non indentati, pedunculati, cum sepalorum vestigiis persistentibus, pe- dunculo articulato, 8 cm. longo, 4 mm. in diametro, om- nino dense et minutissime velutino-pilosiusculi, tactu molles et sine pilis urticantibus, profundissime decem- costati, quinque cum costis primariis crassis et hebetato- rotundatis, 8 mm. altis, atque quinque cum costis secun- dariis similibus sed minoribus, 4-5 mm. altis, transverse leviter fibroso-rugosi, pericarpio crassiusculo, sublignoso, maturitate flavi dicitur. Semina plusminusve sexaginta, in pulpa alba inclusa. A small tree, slender and graceful, commonly about eight to twelve feet in height. Trunk erect, about three inches in diameter, covered with blackish bark, sparsely branched near the top or unbranched. Branches tomen- tose. Branchlets densely villose, with golden-rust-colored and persistent hairs. Leaves very large, digitate, seven- foliate, very long-petiolate. Petioles round, somewhat constricted at the base, very densely and softly golden or ferrugineous, tomentellous, up to 60 cm. long, 9-10 mm. in diameter. Stipules persistent, linear, densely rough-tomentellous, up to 38 cm. long, 2 mm. wide. Leaflets sessile, oblanceolate or broadly lanceolate-ovate, erect, strongly unequal, membranaceous to papyraceous, apically acuminate, basally attenuate, the margin both E Sts | regularly and lightly sinuate-dentate in the upper half, but especially towards the apex, and everywhere armed with cilia-like stellate hairs, 830-50 cm. long, 13-20 cm. wide, above rough to the touch with sparse single brown hairs, beneath rather softly and densely tomentellous with long golden-rust-colored stellate hairs. Inflorescence fasciculate, relatively few-flowered, growing from the lower portion of the trunk. Pedicels articulate, 7 mm. long, 1.5 mm. in diameter, stellate-pilose. Buds glo- bose, 15 mm. in diameter, densely stellate-pilose. Calyx three-parted, divided almost to the base, subeymbiform. Sepals commonly unequal, rather carnose in life, dark purplish, strongly valvate in bud, externally rather coarsely stellate-pilose, internally very minutely granu- lose-pulverulent; the two interior sepals round-ovate, the margins entire, apically perfectly rounded, about 14 mm. long, 10 mm. wide; the exterior sepal usually tri- angular-elliptic, the margin entire, apically subacute, 13-14 mm. long, basally 6-7 mm. wide. Petals five, basally sessile, obovate or ovate, apically very strongly concave-cucullate, about 8 mm. long, 7 mm. (often up to 8 mm.) wide, dark blood-red with purple nerves, ex- ternally minutely muricate-verrucose, ligulate. Ligules linear, about 70 mm. long, basally 8 mm. wide, filiform near the apex, dark blood-red, but pinkish near the tip. Staminal tube five-parted with stamens bearing one and two anthers alternately and with short, free filaments. Staminodes petaloid, dark blood-red, membranaceous, elliptic, marginally entire, apically acute, 14-15 mm. long, 6-7 mm. wide, somewhat verrucose on both sur- faces. Fruits not numerous, ellipsoid, up to 9 cm. long, 4 cm. in diameter, apically long-attenuate, but near the tip slightly constricted, the tip itself obtuse and fre- quently twisted, basally not indented, pedunculate, with remnants of the persistent sepals (peduncle articulate, [ 216 | 3 cm. long, 4 mm. in diameter), everywhere densely and very minutely velvety-pilose, soft to the touch and with- out stinging hairs, very deeply ten-costate, the five pri- mary ribs thick and bluntly rounded, 8 mm. high, the five secondary ribs similar but smaller, 4-5 mm. high, transversely lightly fibrous-rugose, the pericarp rather thick, almost woody, reported to ripen yellow. Seeds about sixty, embedded in a white pulp. Cotomsia: Intendencia del Meta, Cordillera La Macarena, trocha entre el Rio Giejar y el Cafio Guapayita. Alt. ca. 500-600 m. ‘‘Small tree, 12 feet tall. Fruit with soft velvety indumentum, ribs bluntly rounded, unequal.’’ January 22, 1951. Jesus M. Idrobo and Richard Evans Schultes 1192 (Vyere in Herb. Gray; Dupticate type in Herb. Nac. Colomb.).—Intendencia del Meta, Sabanas de San Juan de Arama, margen izquierda del Rio Giiejar, alrededores del aterrizaje ““Los Micos.’” Alt. ca. 500m. “‘Small tree. All parts of flowers deep searlet.’’ March 15, 1951, Richard Evans Schultes 11821 (Fiona. type in Herb. Gray ).—Intendencia del Meta, Villavicencio. Alt. 300 m. *“Cacaoito de monte.’’ 1851-1857, J. Triana s.n.—Intendencia del Meta, Villavicencio. January 24, 1899, 7. A. Sprague 135.—No pre- cise locality. ‘‘Colombia.’* February 1899, Rocha s.n.—Intendencia del Meta, forest near Villavicencio. February 1943, M. Bates s.n.— Intendencia del Meta, Sierra de La Macarena. Playa Bonita. Alt. 400 m. ‘‘Dense, humid forest. Erect shrub.’’ November 14, 1947, W. R. Philipson, J. M. Idrobo & A. Fernandez 1420.—Intendencia del Meta, Sierra de La Macarena, Cafio Entrada. Alt. 550 m. ‘‘Dense, humid forest. Unbranched slender tree, 8 m. high; fruits green. ”*’ January 22, 1950, W. R. Philipson, J. M. Idrobo & R. Jaramillo 2199. —Intendencia del Meta, Sabanas de San Juan de Arama, margen iz- quierda del Rio Giiejar, alrededores del aterrizaje ““Los Micos.’’ Alt. ca. 500 m. ““Small tree, 12 feet tall.’ December 9, 1950, Jesis M. Idrobo & Richard Evans Schultes 612.—Same locality. ‘‘Arbolito de 2 m. Frutos caulinares.*’ December 18, 1950, Jess M. Idrobo & Richard Evans Schultes 721.—Intendencia del Meta, Cordillera La Macarena, trocha entre el Rio Giiejar y el Cafio Guapayita. Alt. ca. 500-600 m. ‘*Foliolos muy largos, pubescentes. Pequefio arbolito de 3.5 m.”’ December 24, 1950, Jesis M. Idrobo & Richard Evans Schultes 787.— Intendencia del Meta, Cordillera La Macarena, trocha entre el Rio Giiejar y el Cafio Guapayita, Cafio Yerly. “‘Fruit dull green.’’ March 9, 1951, Richard Evans Schultes 11629. Herrania tomentella resembles, in its foliage, H. pul- [ 217 ] cherrima and H. Cuatrecasana, It differs from the former in having a much more finely sinuate margin, in having a smooth (instead of a rather muricate-subtuberculate) upper surface, in being more finely tomentose beneath, and in being membranaceous (rather than coriaceous) and generally smaller. From the latter, it can be distinguished by differences in the shape and margin of the leaflets. Herrania Cuatrecasana is conspicuously and gradually decurrent towards the base and has a very remotely and obscurely crenate-denticulate margin. In the form of its fruit, Herrania tomentella approaches H. Cuatrecasana more closely than HZ. pulcherrima. The capsule of Herrania pulcherrima has strongly cultriform ribs with stinging hairs, whereas that of HZ. tomentella has broadly rounded ribs without stinging hairs. Further- more, floral differences between HZ. pulcherrima and H. tomentella are marked, especially in the staminodes, which are apically trifid in the former but acute in the latter. Although there are a number of resemblances between the capsules of Herrania tomentella and H. Cuatreca- sana, the soft indumentum and lack of stinging hairs in the former are in sharp contrast to the condition in the latter where, except for stinging hairs along the ribs, the surface is glabrous or glabrescent. There are likewise several floral differences. The leaflets of Herrania tomentella are borne in a partly erect position. This is also true of Herrania Cua- trecasana and H. pulcherrima and possibly of all species which have a noticeably swollen callus at the base of the leaflets. In this erect position of the leaflets, Herrania tomentella differs strikingly in habit from the only other species known in the Macarena, Hi. nitida, which has leaflets which tend to be rather reclinate. A study of the fruit of the material fromthe Macarena has clarified a confusion of long standing. Although ( 218 | specimens of Herrania tomentella have, in the past, been referred to H. pulcherrima or to H. nitida (as H. nitida var. aspera), a study of the capsule, until recently un- known, shows conclusively that 1. tomentella has its re- lationships in other directions. GUTTIFERAE Clusia chiribiquetensis Maguire ex Schultes in Bot. Mus. Leafl. Harvard Univ. 15 (1951) 56, Pl. XVIII, upp. fig. We now publish a drawing (PI. XX XIII) of the type specimen of this interesting endemic Clusia. Clusia Schultesii Maguire ex Schultes in Bot. Mus. Leafl. Harvard Univ. 15 (1951) 65. Clusia Schultesii was not figured when it was described, but the concept is of such phytogeographical interest that the accompanying plate (XX XIV) is now offered. LYTHRACEAE (Contributed by Dr. Alicia Lourteig) Cuphea sunubana Lourteig sp. nov. Suffrutex (alt. ignota), ramis sublignosis, profuse ram- ificatis, flavo- vel ferrugineo-setosis (setis basi incrassatis, adpresso-adscendentibus) et minutissime pubescentibus, glabrescentibus. Internodia 4-6 mm. longa, quam folia minora. Folia decussata, subsessilia, conferta, discoloria, lanceolata (10-20 1.5-5 mm.), acuta, margine reflexa, nervo centrali rubente, subtus conspicuo, nervis secun- dariis plusminusve conspicuis, subtus setis in nervis atque marginem versus sed supra rarius in nervo centrali atque in lamina sparsis. Flores alterni inter folia. Pedunculi 3-5 mm. longi, tenues, setosi, apicem versus bibracteo- lati. Bracteolae ovato-acuminatae, plusminusve pubes- centes. Calyx 5.5-6.5 mm. longus, fauce paulo dilatatus, [ 219 ] caleare recto, nervis Copiosissime setosis (setis illis ramo- rum similibus) atque minutissime pubescentibus in lobu- lis appendicibusque, illo sub staminibus interne pubes- cente sed ceterum glabro. Petala sex, lilacea, obovato- subspathulata, 4-5 X2-2.5 mm. Stamina 11, prope par- tem centralem calycis inserta, duo dorsalia brevissima, alia longiora, horum tria ventralia glabra cetera pubes- centia. Ovarium oblongum, in stylum inclusum, minute pubescentem attenuatum. Stigma capitatum. Ovulum carnosum, grossum, subhorizontale. Semen immaturum suborbiculare, complanatum, carunculatum, margina- tum, minute foveolatum. It is unfortunate that no mature seeds are available for this species, for it is difficult to place the concept, which is quite distinct, in any section of the genus. I believe, nevertheless, that it belongs to Sect. 5 Brachy- andra, Subsect. 2 Melanium, Ser. 3 in Koehne’s system (in Engler Pflanzenr. [V. 216 (1903) 82). It would fall between Cuphea ciliata Ruiz & Pav. (C. microphylla HBK.) and C. fuscinervis Koehne. From the former, it differs in its inflorescence and pubescence and in having a red median nerve; from the latter, a Brazilian species, in its simple inflorescence. The specific epithet refers to the locality where the type was taken—the Rio Stinuba. Cotompia: Departamento de Boyaca, Cordillera Oriental, vertiente oriental. Entre Gueteque y Guayata, margenes del Rio Stinuba. Alt. 1500 m. June 30, 1940, J. Cuatrecasas 9698 (Tyrer in Herb. Field Mus. ). The other new species of Cuphea herewith described belong in Section 6: Huandra, Subsection 2: Hyssopo- cuphea in Koehne’s System. It is regrettable that in almost all of the material avail- able to me the seeds are immature. My studies, how- ever, indicate that it is very probable that they would [ 220 | not measure more than 2 mm. at maturity, a character used in Koehne’s key to the Subsections. The appearance of the seeds, as well as the general characters of the plants and their malpighiaceous type of pubescence, agrees with those for Section Hyssopocuphea. As no species of this group has previously been re- corded from Colombia, it would seem to be of interest to give a key for these five species. Stems and branches only malpighiaceous-pubescent. Calyx only malpighiaceous-pubescent. Ovules 2-4. Seeds usually 3... ....... 2.2... . kubeorum Calyx malpighiaceous and setaceous-pubescent. Ovules 5-6... . .. . . stygialis Stems and branches with another type of pubescence in- termixed with the malpighiaceous. Calyx-appendage long-setaceous. Ovules 12-15 . philombria Calyx-appendage short-ciliate or glabrous. Ovary glabrous with sparse short hairs. Disk deflected, narrowly triangular. Ovules 10-12... .......0.~24.~24~S« «Osucumbiensis Ovary densely pubescent. Disk reflexed, ovoid- complanate. Ovules 4-6 . . . . . . . .. .. beneradicata Cuphea kubeorum Lourteig sp. nov. Suffrutex usque ad 14 pedes altus, carnoso-malpighi- aceo-pubescens (pilorum brachiis in angulo acuto, hirsu- tis), plusminusve glabrescens. Rhizoma cylindricum, quam 20 cm. longius, cum ramulis fibrosis. Caulis basis sublignosa, crassa, brevis, e superficie terrae in caules secundarios numerosos divisa. Caules tenues, decumben- tes, usque ad 20 cm. longi. Internodia variabilia, 8 mm. usque ad 2cm. longa. Folia vulgo conferta, decussata, dis- coloria, parva, linealia, usque ad 8X1 mm., obtusa, supra pubescentia, subtus glabra, subsessilia cum petiolis bre- vissimis et angustissimis, margine reflexa, nervo centrali lato supra impresso sed subtus conspicue elevato. Flores alterni vel decussati. Pedunculi plerumque infrapetio- lares, plusminusve 2 mm. longi, bibracteolati. Bracteolae [221 J ovatae, pubescentes, plusminusve 0.8 mm. longae. Calyx 4.5 mm. longus, parte mediana constrictus, fauce dila- tatus, cum calcare conspicuo, oblique incurvato, obtuso; lobulis latis, acuminatis; appendicibus quam lobulis mi- noribus, incrassatis ; pilis malpighiaceis, uniformibus, bre- vibus; intus in nervis dense pubescens, sub staminibus lanosus. Petala sex, violacea, elliptica vel oblonga, 2—-2.5 <0.25-1 mm., obtusa. Stamina 11, inclusa, duo dorsalia brevissima, alia paulo majora, maturitate vulgo calycis marginem attingentia, horum tria ventralia glabra cetera lanosa. Ovarium asymmetrico-semiovoideum, dense pu- bescens, in stylum tenuem, inclusum, pubescentem at- tenuatum. Stigma capitatum. Discus crassus, deflexus, latus, obtusus. Ovula 2-4. Semina vulgo tria (non ma- tura), suborbicularia, complanata, plusminusve 1.5 mm., leviter marginata, minute foveolata. The specific epithet has reference to the Kubeo Indians who inhabit the general area where Cuphea kubeorum is known to grow. Cotomspia: Comisaria del Vaupés, Rio Kubiyt (tributary of Rio Vaupés), Cerro Kafiendd, savannahs about 15 miles upstream from mouth. Quartzite base. Alt. about 800-900 feet. November 10, 1952, Richard Evans Schultes & Isidoro Cabrera 18306 (Tyrer in Herb. Gray). —Comisaria del Vaupés, Rio Parana Pichuna (tributary of Rio Vaupés). Alt. about 700 feet. General location: Lat. 1°10’ N, Long. 70°30! W. June 1953, Richard Evans Schultes & Isidoro Cabrera 19913. Cuphea stygialis Lourteig sp. nov. Suffrutex usque ad pedem altus. Planta omnino ae- qualiter adpresso-malpighiaceo-pubescens, pilis crassis, albocerosis cum fundamento brevi et ramis inaequalibus rigidis. Rhizoma tortuosum, cylindricum, usque ad plus- minusve 20 cm. longum, 8 mm. in diametro, cum ramis longis fibrosisque. Caules e basi in ramos divisi, recti vel aliquid decumbentes, basi usque ad 4 mm. Internodia quam folia minora, 4-8 mm. Folia decussata, rarenter verticillata (3 in verticillo), discoloria, lanceolata vel line- [ 222 | alia, 8-18X1.5-8 mm., subsessilia, obtusa, marginata, nervo centrali fusco utrinque conspicuo, subtus cum pilis sparsis sed ceterum glabro. Flores alterni vel decussati, inter tolia. Pedunculi 2-4.5 mm. longi, apicem versus bibracteolati. Bracteolae crassae, ovatae vel oblongae. Calyx 4.5-5.5 mm. longus, fauce dilatatus, caleare incur- vato, lobulis triangulari-acuminatis, brevissime ciliatis, appendicibus crassis, quam lobulis brevioribus, dense adpresso-malpighiaceo-pubescentibus et hispido-fusco- glandulosis in parte mediana inferiore ; intus dense pubes- cens. Petala sex, alba et purpureo-striata, obovato-sub- spathulata, 2.5-2.61.5-1.75 mm., obtusa. Stamina 11, inclusa, duo dorsalia brevissima, alia paulo majora, horum tria ventralia glabra, cetera lanosa. Pistillum inclusum, plusminusve 8 mm. longum. Ovarium dense pilosum, semiovoideum, in stylum pilosum attenuatum. Stigma capitatum, Ovula 5-6. Discus crassus, deflexus, compla- nato-ovoideus. Semina parva, 1—1.5 m., suborbicularia, plano-convexa, minutissime foveolata, caruncula con- spicua. Cuphea stygialis grows in dense carpets along the very edge of the great waterfall, where it is constantly under a spray of moisture. The specific epithet stygialis has been chosen because, according to Dr. Schultes, the Raudal de Jirijirimo is a sacred place to the Indians of the Rio Apaporis, who believe that it is the home of the dead leaders of their tribes. Cotomura: Comisarias Amazonas-Vaupés, Rio Apaporis, Raudal de Jirijirimo (below mouth of Rio Kananari). Quartzite base. Alt. about 900 feet. February 14, 1952, Richard Evans Schultes & Isidoro Cabrera 15330 (Typr in Herb. Gray). Cuphea philombria Lourteig sp. nov. Frutex plusminusve 50 cm. altus, pilis setosis malpighi- aceisque obtectus. Radices fibrosae, profuse ramificatae, [ 223 | usque ad 80 cm. longae. Caules basi sublignosi, 5 mm. in diametro, dichotome ramificati. Rami inaequales, rubi- cundulo-setosi et cano-malpighiaceo-pilosi. Internodia 0.5-1 mm. longa, quam folia breviora. Folia discoloria, lineari-lanceolata (15-80X4.5-8 mm.), sessilia vel sub- sessilia, petiolo crasso, uninervia, subtus cum pilis mal- pighiaceis et nervo conspicuo, supra leviter strigosa cum nervo aliquid impresso, apice acuta vel obtusa. Flores alterni vel rarenter decussati, inter folia. Pedunculi intra- petiolares, 4mm. longi, tenues, pubescentes, apicem ver- sus bibracteolati. Bracteolae crassae, parvae (circiter 0.5 mm. longae), suborbiculares, rubicundulae. Calyx atro- violaceus, a fauce longitudinis 4 constrictus, fauce am- pliata, viscido-setulosus (setis violaceis) atque cum pilis canis malpighiaceisque, caleare obliquo subacutoque; lobulis pubescentibus, inter sepala appendicibus subula- tis, setosis, quam lobulis aliquid minoribus; intus ubique minute pubescens et parte superiore sub staminibus lanato-pubescens. Petala sex, violacea, obovato-cuneata (4.5-7X2.5-3.5 mm.), obtusa. Stamina 11: duo dorsalia brevissima, alia majora (sed calycem non excedentia), horum tria glabra, cetera pilosa. Discus carnosus, crassus, reflexus, obtusus. Ovarium asymmetrico-semiovoideum, plusminusve 5 mm. longum, breviter pubescens, in sty- lum inclusum attenuatum, plusminusve pilosum. Stigma subeapitatum, papillosum. Ovula 12-15. Semen casta- neum, complanato-suborbiculare (circiter 1.5 mm. statu immaturo), minute foveolatum. The specific epithet philombria is derived from the Greek meaning “‘lover of rain’’ and refers to the high rain-forest habitat where the plant grows abundantly along fast-running rills. CotomBia: Intendencia del Meta, Cordillera La Macarena, Macizo Renjifo. Alt. 1,300-1,900 m. ‘‘On rocky banks at river’s edge, roots imbedded in moss; in shaded, damp areas.’* January 20, 1951, Jesus M. Idrobo & Richard Evans Schultes 1062 (Tyrr in Herb. Gray ).—Same [ 224 ] locality. ‘‘On rocks, with roots imbedded in moss; very full, in clumps ; in sunny spots.’’ January 20, 1951, Jesus M. Idrobo & Richard Evans Schultes 1061. Cuphea sucumbiensis Lourteig sp. nov. Suffrutex (alt. ignota), omnino cano-malpighiaceo- pubescens, basi sublignosus, ramificatus. Rami tenues, setis ferrugineis, basi incrassatis, adscendentibus copiose obtecti. Internodia quam folia minora, 8 mm. longa. Folia decussata, conferta, sessilia vel subsessilia. Petioli lati, rubicundull, setosi, ramis similes. Lamina linealis vel lineali-lanceolata, 5-14 0.6—2 mm., discolor, obtusa vel acuta, intus pubescentior, glabrescens, nervo centrali subtus prominenter elevato, fusca, setosa, margine re- flexa. Flores decussati, alterni, apice ramorum inter folia positi. Pedunculi infrapetiolares, 2.5 mm. longi, pubes- centes, ramis similes, partem centralem superiorem ver- sus bibracteolati. Bracteolae crassae, ovatae, 0.5 mm. longae, obtusae. Calyx gracilis, 4-5 mm. longus, fauce dilatatus, caleare conspicue producto, lobulis acutis, ap- pendicibus incrassatis, lobulis aequalibus vel paulo longi- oribus, adpresse cano-malpighiaceo-pubescentibus et in- ferne cum setis brevibus fuscis commixtis; intus in nervis dorsalibus pubescens et in ceteris plusminusve minute pu- berulentus, sub staminibus lanosus. Petala sex, lilacina, obovata vel oblonga, duo dorsalia aliquid majora, 2.5—2.75 x<1-2 mm. Stamina 11, in parte mediana superiore caly- cis inserta, inclusa, duo dorsalia brevissima, alia majora, horum tria ventralia glabra, cetera plusminusve grosse pilosa. Ovarium asymmetrico-ovoideum, glabrum vel pilis brevibus sparse ornatum. Stylus tenuis, inclusus, plusminusve pubescens. Stigma subeapitatum. Ovula 10-12. Discus magnus, 0.5-0.6 mm., triangulari-cordi- formis, complanatus, abrupte deflexus. Semina (immaz- tura) parva, suborbicularia, complanata, paulo margi- nata, minute foveolata. Cuphea sucumbiensis is named for the type locality, the Rio Sucumbios, which forms part of the boundary be- tween Colombia and Ecuador. The species should be recorded also for the flora of Ecuador, for, doubtless, it occurs on both banks of this river. Co._omsia: Comisaria del Putumayo, frontera Colombo-ecuatoriana. Selva higr6fila del Rio Sucumbios (San Miguel), entre los afluentes Conejo y Hormiga. Alt. 8300 m. December 15, 1940, J. Cuatrecasas 11069a (Tyee in Herb. Field Mus.). Cuphea beneradicata Lourteig sp. nov. Suffrutex 20-25 cm. altus, basi lignosus, crassus, tor- tuosus, in saxorum fissuris crescens, cano-malpighiaceo- pubescens et in caulibus juvenilibus etiam cum. pilis purpureo-hispido-glandulosis. Internodia quam folia minora, 2-4 mm. longa. Folia lanceolata vel linealia, obtusa, margine reflexa, 8-11 X 1-8 mm., subsessilia, pet- iolis crassis, plusminusve 0.6 mm. longis, nervo centrali lato, fusco, infra conspicuo, subtus densius adpresso-cana, malpighiaceo-pubescentia et margine (statu juvenili) cum ciliis longis, tenuibus et glandulosis, caducis. Flores decussati, in ramorum apice. Pedunculi 2—2.5 mm. longi, in parte mediana superiore bibracteolati. Bracteolae ovato-acuminatae, 0.5 mm. longae, pubescentes. Calyx 5-5.5 mm. longus, malpighiaceo-pilosus (pilis plusmin- usve hirsutis) et copiose purpureo-hispido-glandulosus in parte inferiore; lobulis latis, acuminatis: appendicibus crassis, pubescentibus, lobulis aequalibus; calcare con- spicuo incurvato, obtuso; intus in nervis pubescens, sub staminibus lanosus. Petala sex, roseo-purpurea, 3.51.5 mm., oblonga vel obovato-oblonga. Stamina 11, duo dorsalia brevissima, alia majora, calycis marginem attin- gentia, horum tria ventralia glabra, cetera lanosa. Pistil- lum 3.5 mm. longum, inclusum. Ovarium asymmetrico- semlovoideum, dense pubescens. Pistillum tenue, pubes- cens. Stigma subcapitatum. Discus complanato-ovoid- [ 226 | eus, deflexus, obtusus. Ovula 4-6. Semina adhuce ignota. [ have chosen the specific epithet because, according to the three collectors of the material cited below, this little shrub is anchored so firmly by its fibrous roots in the cracks of the bare sandstone where it grows that it could not be collected with the roots but had to be cut off at the surface of the ground. Cotompra: Comisaria del Vaupés, Mesa La Lindosa, 15-20 km. de San José del Guaviare. Alt. 400-600 m. December 13-15, 1950, Jesis M. Idrobo & Richard Evans Schultes 640 (Tyre in Herb. Gray ).— Comisaria del Vaupés, San José del Guaviare. Alt. 270 m. November 12, 1939, J. Cuatrecasas 7691. Cuphea annulata Woe/ne in Martius FI. Bras. 13, pt. 2 (1877) 304, t. 56, fi 5. Although the type of this species bears a label report- ing the ‘‘calyx coccineus, ’ and the collection cited below states ‘‘flowers basally green-yellow, remainder (about 3 of length) red,’ their identity is unquestionable. The petals and stamens are red. The ovules are five, reducing to two or three seeds at maturity. As Koehne did not see the seeds, having only the type material, it seems to me advisable to describe them: Seeds dark brown, complanate-suborbiculate (+ 8 mm.), slightly marginate, border somewhat concave at the place of the caruncle, apex broadly retuse, surface finely foveolate. The type of Cuphea annulata was collected in 1820 by Martius at Araracuara on the Rio Caqueta in Colombia, not from Brazil nor from the Rio Negro as stated on the label: ‘‘Alto Amazonas, Rio Negro bei Araracoara.”’ It is of interest to note that the locality of Schultes & Ca- brera 17509 is an isolated mountain not far in a straight line from Araracuara and belonging geologically to the same formation. The species is one of the characteristic endemics of the ancient cretaceous sandstone Venezuela- [ 227 | Guiana land-mass and has recently turned up on Cerro Duida, as I shall point out in detail in another paper. CotompiaA: Comisaria del Vaupés, Rio Piraparana (tributary of Rio Apaporis), Cerro E-ree-ké-mee-6-kee. General location: between Lat. 0°15’ S, Long. 70°30’ W and Lat. 0°25’ N, Long. 70°30’ W. ‘‘Somewhat scandent shrub 4 feet tall. Flowers basally green-yellow, remainder (about } of length) red. Leaves very bright green. ”’ Sep- tember 18, 1952, Richard Evans Schultes & Isidoro Cabrera 17509. APOCYNACEAE Couma catingae Ducke in Archivos Inst. Biol. Veg. Rio Janeiro 4 (1988) 59. The collection cited below extends the range of this species, hitherto reported only from the Amazonas of Brazil, into Colombia. CotomBiA: Comisarias del Amazonas-Vaupés, Rio Apaporis, Cachi- vera de Jirijirimo y alrededores. Alt. c. 250 m. ““Small caatinga tree. Latex white. Puinave=hwan-hwan.’’? June 13, 1951, Richard Evans Schultes & Isidoro Cabrera 12461. [ 228 | ILLUSTRATIONS EXPLANATION OF THE ILLUSTRATION Pirate XXII. ANTHURIUM ATROPURPUREUM Schultes & Maguire. 1, habit, one third natural size. ANTHURIUM NEMORICOLA Schultes & Maguire. 2, habit, one half natural size. 3, 4, 5, leaves, one half natural size. Drawn by Dorotuy H. Marsu PuateE XXII ANTHURIUM atropurpureum \ Schultes et Maguire ht ks ANTHURIUM. SSS “nemoricola Schultes et Maguire EXPLANATION OF THE ILLUSTRATION Prare XXIII. The type plant of Parpatanruus Motpenkeants RR, Fb. Schultes. Rio Kuduyari, Yapo- boda, Comisaria del Vaupés, Colombia. Photograph by Ricuarp Evans Scuucres Prare XAT] EXPLANATION OF THE ILLUSTRATION Prare XXIV. (Yop) Habitat of ParpaLanruus Mot- DENKEANUS I. bk. Schultes showing, at the right of the Paepalanthus, two shrubs of VeLiozia Lirne- pHina RR. ke. Schultes, Rio Kubiyd, Cerro Kanenda, Comisaria del Vaupés, Colombia. ( Bottom) Habit of THurnta spHoarrocereHaLa (ludge) Hook f. in Catto Paca, Rio Piraparana, Comisaria del Vaupés, Colombia. Photographs by Ricuarpd Evans Scuuires pO. GS 0 PLATE EXPLANATION OF THE ILLUSTRATION Prare XXV. PrroarRNtA MACARENENSIS L. B. Smith. 1, habit, one twentieth natural size. 2, floral bract and flower, natural size. 3, sepal, natural size. 4, base of petal, natural size. Navid Garcra-Barricar L. B. Smith. 5, branch of inflorescence, natural size. 6, floral bract and flow- er, five times natural size. 7, sepal, five times nat- ural size. 8, petal and filament, five times natural size. 9, diagrammatic cross-section of floral bract and sepals, five time natural size. Drawn by Roserr J. Downs PLAGE. Ax PITCAIRNIA macarenensis L.B. Smith TA aes Bae 9 NAVIA Garcia- Barrigae ITB Smith : a ie 2. 5 i i) ‘9 ei? EXPLANATION OF THE ILLUSTRATION Prare XXVIL Navia reecexa L.B.Smith, 1, apex of leaf, one half natural size. 2, branch of inflores- cence, one half natural size. 3, anterior view of flower, five times natural size. 4, seed, five times natural size. NaviaA neniopHtta LL. B.Smith. 5, apex of leaf, nat- ural size. 6, margin of leaf, five times natural size. 7, Horal bract and flower, natural size. 8, posterior view of flower, five times natural size. Drawn by Roserr J. Downs PLATE \\ ‘\ L.B. Smith XXVI NAVIA reflexa 4 NAVIA NN heliophila L.B. Smith y) YY, Nee EXPLANATION OF THE ILLUSTRATION Pruare NXVIT. Habitat of Navia uenropuita LL. 2B, Smith at Raudal de Jirijirimo, Rio Apaporis, Com- isaria del Amazonas, Colombia. Photograph by Rictarp Evans Scuurres Phare X XVII EXPLANATION OF THE ILLUSTRATION Prare NAVITL. Navia Lovezi L. B.Smith. 1, hab- it, about one half natural size. 2, flower, natural size. 3, schematic cross section of flower, about one and one half times natural size. Navia MyRIANTHA LL. B. Smith. 1, habit, about one half natural size. 2, flower, five times natural size, 8, schematic cross section of flower, about ten times natural size. Drawn by Kimer W. Suir PLATE XXVIII /i/f / 2 At ij f Vij NAVIA Lop ez1l f | YY Vi LB Smith NAVIA myriantha \ \ LB. Smith EXPLANATION OF THE ILLUSTRATION Prare XXIX. Vetrozia Mauprana PR. Ee. Schultes in flower and fruit. Goo-rain-hoo-da, Rio Karurd, Comisaria del Vaupés, Colombia. Photograph by Ricuary Kvans Scuuires ».4 €%.* “4 we 4 PLATE EXPLANATION OF THE ILLUSTRATION Prare XXX. (Top) Habit of Vettozia Mauprana R.E. Schultes at Goo-ran-hoo-da, Rio Karurt, Com- isaria del Vaupés, Colombia. (Bottom) Habit of ‘Trerraprerta Meruystrica R. Ee. Schultes at Porto Makds, Rio ‘Tikié, Kstado do Am- azonas, Brazil. Photographs by Ricuarpy Evans Scuuvres XXX PLATE KXPLANATION OF THE ILLUSTRATION Prare XXXL. Herranta purcurerrma Goudot. 1, leaf, about one quarter natural size. 2, flower, one half natural size. 8, petal, twice natural size. 4, staminode and anthers, twice natural size. Drawn by Kimer W. Suivi PLATE XXXI HERRANIA pulcherrima Goudot EXPLANATION OF THE ILLUSTRATION Prare XXXIT. (Zop) Buds of Herranta TroMEN- venta R. be. Schultes. Cordillera La Macarena, In- tendencia del Meta, Colombia. (Bottom) Fruit of Herranta romentrecta PR. F, Schultes. Cordillera La Macarena, Intendencia del Meta, Colombia. Photographs by Ricuarn Evans Scuvuvres PLATE ANAT EXPLANATION OF THE ILLUSTRATION Phare XXXII. Crusta cuiripiguerensis Maguire. 1, habit, one half natural size. 2, fruit, one half natural size. 3, flower, slightly less than natural size. 4, anther, about six times natural size. Drawn by Eimer W. Suir PLatrE XNNIII CLUSIA chirrbiquetensts Maguire EXPLANATION OF THE ILLUSTRATION Phare XXXIV. Crusia Scuuvresiu) Maguire. 1, habit, one half natural size. 2, fruit, closed, one half natural size. 8, fruit, opened, one half natu- ral size. Drawn by Eimer W. Suivi Prate XA XMLTY =a hg ~ y 5 . Sa oa. Vow 3 past USIA Schult testv Mag ULC EXPLANATION OF THE ILLUSTRATION Prare NXAXNV. Cernea sunubnana Lourteig. A, leat, upper surface, twice natural size. B, leaf, lower surface, twice natural size. C, calyx, about four and one half times natural size. D, calyx, opened, five times natural size. EK, petal, five times natural size. IY, pistil, with disk, five times natural size. G, disk, seven times natural size. H, H : seed, five times natural size. Drawn from the type specimen, Cupnba KUBLORUM Lourleig. A, calyx, six times nat- ural size. B, calyx opened, seven times natural size. C, dorsal petal, seven times natural size. C - ven- tral petal, seven times natural size. D, pistil with disk, seven times natural size. EK, disk, ten times natural size. Drawn from the type specimen. Drawn by Avicix Lourvrerc XXXV PLATE CUPHEA sunubana Lourteig CUPHEA kubeorum Lourteig Lourteig del. EXPLANATION OF THE ILLUSTRATION Prare XXXVI. Cupura puttompria Lourteig. A, leaf, twice natural size. B, flower, twice natural size. C, calyx, four times natural size. D, calyx, opened, four times natural size. EF, pistil, four times natural size. F, disk, ten times natural size. G, seed, five times natural size. Drawn from the type specimen. Cupnba sucUMBIENSIS Lourteig, A, leaf, lower sur- face, three times natural size. B, calyx, five times natural size. C, calyx, opened, with petals, five times natural size. D, pistil, with disk, five times natural size. E, disk, ten times natural size. Drawn from the type specimen. Drawn by Atacia LourrreiG PLATE XXNVI CUPHEA philombria Lourteig | CUPHEA \ sucumbiensis Lourteig EXPLANATION OF THE ILLUSTRATION Prare XNAVIT. Curnea sryatanis Lourteig. A, calyx, five times natural size. B, calyx, opened, five times natural size. C, petal, five times natural size. D, pistil with disk, five times natural size. Kk, disk, ten times natural size. Drawn from the type specimen. Curnba BENERADICATA Lourteig. A, young leaf, six times natural size. B, calyx, six times natural size. C, calyx, opened, six times natural size. D, petal, six times natural size. KE, pistil with disk, six times natural size. F, disk, ten times natural size. Drawn from the type specimen. Drawn by Aticta Lourreic Phare NNNVII CUPHEA stygialis - Lourteig CUPHEA | ae beneradicata Lourteig Lourteig del. BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CamBripGr, Massacuusetts, Juty 12, 1954 VoL. 16, No. FOSSIL MAIZE FROM THE VALLEY OF MEXICO BY Exso S. BarGHoorn’, MArGARET K. WoLre’ AND Katuryn H. Cuissy® WirHIn the past decade studies on the origin, antiquity and evolution of maize have received new stimulus from archaeological and palynological discoveries in the south- western United States and Mexico (Deevey, 1944; Mangelsdorf and Smith, 1949). It now appears well es- tablished that a maize culture developed and flourished in the presently semi-arid basins of New Mexico during a period ranging from approximately 8600 B.C., or ear- lier (Libby, 1951), to approximately 1000 A.D. The morphologically primitive structure of this prehistoric corn, in contrast to its surprisingly recent geologic age, is strong presumptive evidence, though not necessarily proof, that maize originated in the New World and was here developed from its wild state by aboriginal migrants into the American subtropics. Recently, additional paleontological evidence has been obtained from the Valley of Mexico which indicates a far greater antiquity for the existence of maize in the New World than has yet been revealed by archaeologi- ' Department of Biology, Harvard University. ” Radcliffe College. ® Department of Geology and Geography, Oberlin College. [ 229 ] cal discovery. ‘The evidence has been secured from a series of deep cores taken in Mexico City, in connection with extensive studies of the sub-surface. These cores have been described in some detail by Sears (1952), and Sears and Clisby (1952). In the course of analyzing the pollen contents of the cores, Mrs. Clisby observed several unusually large grass- pollen grains in sediments close to the 70 meter level in the Belles Artes boring. The size of the grains (75p to 135p by acetylation method) at first appeared to preclude the reasonable possibility that they were derived from native wild grasses then extant in the Valley of Mexico. However, careful study of these fossil pollen grains, in- cluding detailed comparison with material in the Harvard Pollen Collections and in the Botanical Museum, sug- gested three possibilities for their botanical affinity, viz: Tripsacum, Zea or Euchlaena (teosinte)—all of the tribe Tripsaceae. If the grains actually proved to be those of Zea Mays, as seems most probable on the basis of their size alone, they would extend the fossil record of Indian corn far beyond our presently known chronology for the existence and dispersal of this remarkable genus of the grasses. In order to establish a critical basis for identification of the fossil pollen, an extensive study was made of the size-range exhibited by the pollen of various species of Tripsacum and by varieties of maize and teosinte. To sustain essential uniformity in the data, all preparations of both living and fossil grains were prepared by the same technique (modification of Erdtman, 1948), and perma- nent slides were prepared with glycerin jelly as a mount- ing medium. A total of eight species of Tripsacum, three collections of teosinte and fourteen varieties of modern maize were chosen for purposes of comparison (‘Table I). A total of 84 large grass-pollen grains were ultimately [ 230 ] TABLE I AXIS LENGTH, PORE DIAMETER AND PORE-AXIS RATIO IN TRIPSACUM, TEOSINTE, MODERN AND FOSSIL MAIZE Number Pore-Axis Ratio Averages pollen grains Length | Diameter | Pore-Axis Species measured 3.0 3.2 34 3.6 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 7.0 72 74 76 78 80 90 Grain, #.| Pore, EL. Ratio Tripsacum T. australe No.1... 54 1 §& 15 16.11 «tee 57.4 14.3 1:4.0 me Ge & es 25 1 2 38 5 ? 4 2 57.2 14.3 1:4.0 T. dactyloides. . . . . 50 4 2 8 16 10 6 4 50.2 12.2 1:4.1 T. lanceolatum .... 49 3.4 7 7 13 4 ll 49.9 12.5 1:4.0 T. latifolium .. . . . 50 5 9 15 8 8 4 1 41.9 10.6 1:1.4 T.laxum....... 50 2 16 #16 13+ «3 57.2 14.7 1:3.9 T. maizar ...... 25 1 1341 2 2 6 4 Fea 41.7 10.3 1:4.0 T. pilosum ..... 42 1 2 3 6 1310 5 2 57.1 14.7 13.9 — Teosinte Mexico, “Chalco”. . . 50 1 Bees 14 9 6 SB U1 if 81.8 15.9 1:5.1 Guatemala No.1... 31 23 5 5 5 6 3 2 86.4 16.0 1:5.4 Guatemala No.2... 32 4 4 6 4 3 2 79.3 14.7 1:5.4 Maize United States Thompson Flint . . 50 1 5 7 15 6 6 3 3 2 2 120.7 19.4 1:6.2 Thayer Flint. . . . 50 2 3 8 5 8 8 6 5 1 1 i 2 122.8 18.8 1:6.5 Knobless ....., 30 2 2 2 8 12 2 2 118.8 18.2 1:6.5 Mexico Chapalote ...... 50 6 8 4 6 8 4 8 4 2 94.9 15.4 1:6.2 Vel cs 9 ho 50 11 1 5 8 10 7 6 4 5 1 96.6 15.6 1:6.2 Nal-Tel ...... 47 2 4 10 6 5 2 2 4 4 2 8 2 i 8 87.2 13.2 1:6.6 Tabloncillo .... ~ 50 12 3 6 6 9 8 3 6 oa 98.6 16.0 1:6.2 Tuxpefo ..... 50 2 2 6 6 8 4 16 4 99.0 15.8 1:6.3 Vandefio ..... 50 4 8 7 10 5 5 3 4 1 ie ia 97.1 16.1 1:6.0 Zapalote Chico. . . 50 2 2 6 2 6 18 8 6 94.7 15.9 1:6.0 Costa Rica No.1045. . 50 2 18 4 4 6 2 #4 8 2 104.6 16.0 1:6.5 Peru No.300 ...... 50 5 6 9 10 10 7 38 95.7 15.8 1:6.1 No. 420 ...... 50 2 212 4 14 8 4 .. 2 99.0 16.4 1:6.1 No. 427, “Cuzco” . . 50 1 5 7 16 8 8 2 : ae 92.1 TS 1:6.0 Archaeological Bat Cave, Late... 50 1 1 1 5 3 7 7 #10 6 2 3 4a 7) ada | 88.7 14.1 1:6.3 Bat Cave, Early .. 10 2 I 2 1 1 1 ee 1. 93.1 152 1:6.1 Fossil, Mexico City. . 14 1 ye | 3 1 2 ee 2 95.6 155 6.2 secured from the lower levels of the Belles Artes core. ‘These lower-depth grains received the most careful study, although an additional 44 large grass-pollen grains from the upper levels of the Belles Artes and Madero cores were also examined by the same procedures. In the course of making measurements, it early became evident that a wide range in the size of the pollen grains characterizes each of the three genera, a condition possi- bly indicative of their unusual genetic variability. In Tripsacum, a wide-ranging and common genus of sub- tropical and temperate North America, the average for the long axes of the pollen grain varied between 41.7 in 7. maizar and 57.4 in T. australe, with an extreme range in the genus of 83.6u to 64.0u. In teosinte, of the three forms examined, the average length varied between 79.34 and 86.44, with extremes of 74.0 and 102.0p; whereas in maize the average range fluctuated between 87.24 (Mexico ‘‘Nal Tel’’) and 122.8 with extremes of 72u and 141.7. It is evident that the smaller pollen of some of the varieties of cultivated maize measured in this study fall well within the range of teosinte and close to that of the largest 'Tripsacum grains. It is apparent, therefore, that size alone cannot be utilized for the criti- cal identification of presumed fossil maize pollen, and that of its relatives, unless a sufficient number of intact grains are available to be measured and plotted on size- frequency curves to show the statistical probability of one of three possibilities. Owing to the small number of individual intact grains from the Mexican borings, how- ever, this procedure was not possible. Because of the paucity of structural features and the undistinctive sculpture pattern of the pollen exines of the three genera under consideration, it became necessary to attempt some other means of distinguishing the three pollen-types. Consideration of the problem led to one [ 281 ] other possibility, viz: a comparison of the ratio in size which exists between the pore (including the annulus) of the pollen grain and that of its long axis (Text Fig. 2). In order to establish these ratios and to determine their constancy, if any, approximately 50 additional grains were measured from each preparation with respect to these dimensions. In the case of the fossil grains all those exhibiting intact pores were measured. The meas- urements were averaged and the ratios computed from the averaged value for each species. The results showed encouraging consistency, the ratio of pore to long axis being an unexpectedly conservative value, and, more important to the problem at hand, significantly different among the three species in question. The numerical val- ues computed are shown in Table II. Although the data are limited to approximately 1000 pollen grains, the differences are so consistent, both with respect to individual grains and averages as well, that the pore-long-axis ratio appears to be a valid means of al- ways distinguishing maize pollen from that of Tripsa- cum, and in some instances from that of teosinte. It may be noted also in connection with this analysis, that teo- sinte, a postulated hybrid between maize and Tripsacum (Mangelsdorf and Reeves, 1939), shows an intermediate value both in overall size, and perhaps more significantly, in its pore-ratio. The intermediate value is well in har- mony with the postulated hybrid origin of teosinte. Fossil pollen grains resembling maize, teosinte and ‘Tripsacum in size and in their pore-axis ratios occur in the upper levels of both the Madero and Belles Artes Text Fie. 1 Comite eee | Dine shai the distribution il frequency of maize, teosinte and Tripsacum in samples studied from the Madero and “se Belles Artes cores. Note that the Belles Artes core was sampled throughout, the Madero core only to a depth of 5 meters. Numbers in parentheses indicate the number of individ- ual grains. Depth is indicated in meters. [ 232 | mapero SEES MAIZE (28) Resi Oe =a (MAIZE (9) reo MIE 48 = ERO Ee af ATEOSINTE 2 (2) 6- es TRIPSACUM (3) ee — MAIZE (1) 94 , \2- e 154 364 39-4 42- Tad -- TRIPSACUM (1) A— MAIZE (8); TRIPSACUM (12) my MAIZE (1!) = --TRIPSACUM (3) KXPLANATION OF THE ILLUSTRATION Prare XXXVIIL. Photomicrographs of fossil, archeological and modern maize pollen grains. 1, Belles Artes core sample number 163, slide number 5. Note pore at extreme upper right, and characteristic folds of the exine. 435 2, Belles Artes core sample number 148, slide number 1. Pore is clearly visible in upper central portion of grain, >< 435 3, Belles Artes core sample number 1638, slide number la. Pore is visible at lower right. The grain is nearly round, but deeply folded.» 435 4, Bat Cave, New Mexico. Dated by radio-carbon age de- termination of associated charcoal at 5600+ years. Note large size of grain and conspicuous pore. » 485 5, Bat Cave, New Mexico, Same source and age as grain shown in fig, 4.0» 435 6, Thayer Flint, modern maize, for comparison with fossil maize pollen. Note conspicuous pore in central upper por- tion of figure, * 435 PLatrE X XXVIII cores. ‘The larger grains, 88 in number, from these upper levels (above 6.0 meters) presumably represent the pol- len of cultivated maize, and indeed many grains are well within the upper range of several modern varieties. Three grains of intermediate size and pore-ratio in the upper levels (8.3 and 8.6 meters) possibly represent the pollen of teosinte, a common weed in and around the maize fields in the Valley of Mexico today. Three grains contorm- ing to Tripsacum were secured from the upper levels of the Belles Artes core and, interest- ingly enough, one Tripsacum grain from the 45 meter level of the same core. No pollen — 120 p grains Clearly assignable to teo- Text Fic. 2. Diagram show- sinte were found below the 3.6 ing the pore-axisrelationship. and 3.3 meter levels respec- Note that the outer borders of an te “t] ek : 7 » Y > >»; . %f> » 4 Oo, the annulus (shaded portion Ively in either core (hex Ig . oe — surrounding the pore) are in- a? This May indicate that teo- cluded in the pore measure- ginte did not become. estab- mt. (The: ratio ie deter- ; : : ; ; meet CI 1¢ ratio is deter- Ji hed in the Valley of Mexico mined by dividing the numer- . . a ; ies) value, in antexeng, of the. UNUL Some time alter iname- long axis by the numerical cultivation had begun. How- value,in microns,of the pore. ) ever, the total number of grains involved is small and the differences in levels may repre- sent the product of sampling. By far the most significant discovery is that of large pollen grains, closely resembling modern maize, in the 69 and 70 meter levels in the Belles Artes core. A total of 19 large grains were secured, 14 of which were sufh- ciently preserved to permit a pore-axis measurement. Of the 19, 8 are well outside the extreme size-range for teo- sinte, as determined in this study, and of the remaining 11, four are outside the extreme range of the pore-axis i aaa | 230 | TABLE II DATA ON INDIVIDUAL FOSSIL POLLEN GRAINS OF TRIPSACEAE FROM MEXICO CITY CORES Core Description Sample _ Slide Depth Size » Pore-Axis Core Number Number Meters Condition Axis Pore Ratio Identification Belles Artes 148 1 74,2-74.5 elongated 66.0 13.5 4.8 Tripsacum " 7 . 2 — s good 55.8 12.0 4.6 Tripsacum ag “ . 7 = n.p.v.* 45.0 = —_ Tripsacum a. “ 163 1 70.3-70.5 entire 90.0 16.5 5.4 Maize .* “ = la = good 75.0 13.5 5.8 Maize " _ a 2 _ good 93.0 15.0 6.2 Maize i - - 3 = n.p.v. 132.0 —<— — Maize - ie 4 oe fragment 19.8 13.5 5.8 Maize “ “ 5 — entire 127.0 17.2 7.3 Maize - = - 6 “oo fair 124.0 21.0 5.9 Maize i " 7 ~~" fair 124.0 18.0 6.9 Maize - ™ a 8 ~~ good 87.0 15.0 5.8 Maize 7 “ “ 9 ~~ n.p.v. 129.0 ace aa Maize = a as 10 end poor 135.0 18.0 7.5 Maize = a 165 2 69.7-69.9 good 63.0 16.5 3.8 Tripsacum i ™ “ 3 -_— pore eroded 63.0 rn Tripsacum = " - lla cas eroded 64.5 an = Tripsacum 7 = “ l1b ia elongated 69.0 13.5 5.1 Tripsacum i = _ 16 ao” elongated 60.0 15.0 4.0 Tripsacum ne i. a 19 “oo elongated 67.5 15.0 4.5 Tripsacum - om 23 ——< elongated 64.5 15.0 4,3 Tripsacum = = me 26 See poor Le 15.0 3.7 Tripsacum si = a 39 ee eroded 60.0 12.0 5.0 Tripsacum 7 a sis 41 ~~ fair 75.5 13.5 5.5 Maize om = 7 41a ~~ eroded 54.0 — = Tripsacum “ “ ‘s 43 SS fair 79.5 13.5 5.8 Maize ss e 166 2 69.3-69.5 good 92.5 16.5 5.6 Maize 7 i = 3b =~ good 78.5 12.0 6.1 Maize - i ' 12 = crushed 70.5 10.5 6.7 Maize _ m= ™ 18a id good 75.0 15.0 5.0 Maize _ ms = 18b tte fair 55,4 12.0 4.6 Tripsacum a _ ™ 21 a fair 84.0 12.0 7.0 Maize - - = 26 =_ = pore poor, 54.0 — — Tripsacum fragment ‘i “ . 31 —~- good 105.0 16.5 6.3 Maize o _ a 47 “oe good 109.5 15.0 7.8 Maize . 7 - 63 ~ pore not clear 66.0 15.0 4.4 Tripsacum . .* 189 1 45.1-45.3 pore notclear, 72.0 ae — Tripsacum grain elongated - si 232 1 7.4-7.6 very poor, 135.0 —= = Maize fragment ¥ = 234 3 5.9-6.1 good 103.5 15.0 6.0 Maize 7” = 235 5 5.1-5.2 good 52.5 12.0 4.3 Tripsacum i “ 8 ee good 11235 19.5 aa Maize . ” ge 9 nn fair 82.5 15.0 5.5 Maize 7 “ " 10 _- fair 97.5 16.5 5.9 Maize 7 " 236 4 4.4-4.6 fair 76.5 12.0 6.3 Maize sie - = 5 nce pore notclear 79.5 aaa — Maize 3 ied as 7 2 pore notclear 90.0 act = Maize ? “ i “ 8 ~~ good 45.0 9.0 on Tripsacum = - “ 10 aia n.p.v. 82.5 —— — Maize _ se 237 11 3.6-3.8 pore not clear 55.5 wen mai Tripsacum ee _ 12 — pore notclear 70.5 15.0 4.7 Teosinte ? a - - 13 == pore notclear 69.0 12.0 5.7 Maize = a a 14 ~ porenotclear 64.5 15.0 4.3 Teosinte ? Madero 135 1 6.0 elongated 130.0 15.0 8.8 Maize good i 130 1 5.1 n.p.v. 126.0 aoe = Maize 128 l 4.9 good 132.0 21.0 6.2 Maize = a Q ar Dp: 96.0 — — Maize si i 4 id n.p.v. 87.0 — — Maize ? ae 7 6 _ n.p.v. 99.0 oa = Maize “e " Hi - elongated 144.0 = — Maize - 126 1 4.5 n.p.v. 103.5 a — Maize - ' 3 “ good 120.0 19.5 6.1 Maize - vie 4 ™ n.p.v. 99.0 —o — Maize sia ae 6 . good 105.0 18.0 5.8 Maize Ss 124 1 4.1 pore not clear 140.0 — — Maize * ie 2 om pore not clear 105.0 oe a Maize = % 3 i n.p.v. 105.0 — — Maize ” o 4 _ pore notclear 99.0 — =~ Maize " - 5 bi fair 135.0 16.5 8.1 Maize ay 120 1 3.3 n.p.v. 127.5 as i Maize 7 “i 2 “ good 99.0 16.5 6.0 Maize is . 3 - n.p.v. 112.5 — — Maize - 4 us pore not clear 66.0 12.0 5.5 Teosinte ? ~ = 5 = good 124.0 18.0 6.9 Maize 7 " 6 - n.p.v. 124.5 = = Maize sie 7 7 > n.p.Vv. 105.0 — _ Maize , = 8 = good 124.0 18.0 6.9 Maize ™ 119 1 3.1 fair 90.0 15.0 6.0 Maize - 7 2 “ n.p.v. 100.3 one — Maize *n.p.v. indicates no pore visible for measurement. ratio for teosinte, although within the upper level of the long axis dimension of teosinte. The existence of these large fossil grass-grains at these great depths calls for an explanation and the following possibilities must be con- sidered. ay 4. The fossil grains are those of a wild grass, not re- lated to maize or its relatives. This possibility seems quite remote, since, except for the cultivated cereals of the Old World, no grass pollen approaching this size is known. The grains represent contamination occurring in the laboratory. This possibility has been eliminated from consideration by re-examination and re-isola- tion of grains from additional samples of the core sediments. It may also be ruled out on the grounds of the physical, chemical and optical properties of the fossil grains when compared to modern maize pollen. The grains represent contamination which occurred in Mexico during the core drilling (a) either as at- mospheric contaminants or (b) as stratigraphic con- taminants during the drilling operations. Possibil- ity 8a can be excluded on the same grounds as possibility 2. Possibility 8b can logically be ex- cluded on the ground that if the large fossil grains were carried down from the upper 6 meter levels, they should be found at intermediate depths. How- ever, the large grains have not been found between the 6 meter level and the 69 meter level, with the sole exception of the one Tripsacum grain referred to previously. The grains are those of either Tripsacum or teosinte [ 237 ] which have increased in size and corresponding pore- axis ratio as a result of preservation under rather unusual sedimentary conditions. There is obviously no way to prove that this is not true, but if it were the case, it would be totally inconsistent with pre- vious extensive experience by numerous investi- gators in dealing with Pleistocene and Tertiary microfossils. The pollen grains are those of ancient maize. This appears to be the most reasonable interpretation and certainly from the evidence now at hand the only plausible one. The large fossil grains resemble maize pollen not only in their general appearance but in their size, and when it can be determined, in their pore-axis ratio. They differ somewhat from pollen grains of modern maize in possessing a slight- ly thicker exine and, more significantly, a smoother contour in folding. Although these differences may be due in part to conditions of preservation and post depositional change, it is of interest to note that the prehistoric pollen from Bat Cave, New Mexico shares these properties to some extent; but they are less evident in modern maize (Plate XX XVIII). Only one of the well preserved grains from the lower levels does not fall within the range of Tripsacum or maize. This grain measures 70.5 in length, with a pore- axis ratio of 4.7, which is outside the range of maize as determined by our measurements. In pore-ratio it is just within the extreme for Tripsacum and the lower range for teosinte. Whether this one specimen represents an unusually large grain of Tripsacum, other grains of which have been found at this level, or a slightly atypical grain of teosinte, cannot be determined. There remains a re- [ 238 ] mote possibility, therefore, that both maize and teosinte were growing at the time the deeper sediments were deposited. Although definite palynological and geologic conclu- sions have not yet been drawn as to the antiquity of the lowest portions of the Mexico City cores, the available evidence indicates that the sediments extend well back in Wisconsin time, most probably to the early stages of the Iowan advance of the Wisconsin Ice sheet.’ If this is the case, the fossil maize pollen with which we are concerned almost certainly antedates the practice of agri- culture in North America, and probably precedes the advent of man on this continent. 1 Personal communication from Professor Paul B. Sears. [ 2389 | Lor 6. LITERATURE CITED . Deevey, E.S., 1944, Pollen Analysis and Mexican Archeology: An Attempt to Apply the Method. Amer. Antiquity 10: 185-149. Erdtman, G., 1943, An Introduction to Pollen Analysis. Chronica Botanica, Waltham, Mass. Libby, W. F., 1951, Radiocarbon Dates I]. Science 114: 291-296, Mangelsdorf, P. C. and R. G. Reeves, 1939, The Origin of Indian Corn and its Relatives. Texas Agric. Exper. Station Bull. 574: 1- 315. Mangelsdorf, P. C. and C, Earle Smith, Jr., 1949, New Archaeo- logical Evidence on Evolution in Maize. Bot. Mus. Leaflets, Har- vard University 13: 213-244, Sears, P. B., 1952, Palynology in southern North America. 1. Archaeological Horizons in the Basin of Mexico. Bull. Geol. Soc. Amer. 68: 241-254. Sears, P. B. and K. H. Clisby, 1952, T'wo Long Climatic Records. Science 116: 176-178. [ 240 ] A NEW NARCOTIC SNUFF FROM THE NORTHWEST AMAZON BY RicHarpD Evans SCHULTES i AN interesting new drug has been added to our growing list of native narcotic and stimulant plants of South America. Recent investigations have uncovered the use amongst certain Indians of eastern Colombia and north- western Brazil of several species of the myristicaceous genus Viro/a in the preparation of a violently toxic snuff which is employed by the medicine-men in witchcraft, divination and the diagnosis of illness. II. During the course of exploration of the Rio Apaporis in Amazonian Colombia in 1951 and 1952, I had as helpers several Indians of the Puinave tribe from the little known Rio Inirida. The Inirida is the highest Colombian affluent of the Orinoco and, although botan- ically completely unknown, it represents apparently an area where the Amazon (Rio Negro-Vaupés) and Orinoco floras blend. Consequently, my Puinave helpers, from whom I first learned of this myristicaceous snuff, were familiar with many of the plants encountered in the Apaporis basin. During 1951, the uninhabited Rio Apaporis was opened up for tapping wild rubber, and natives from [ 241 ] various tribes of the Colombian Comisaria del Vaupés and the Rio Uaupés of Brazil were transported into the area by air. Thus, I had an unparalled opportunity of investi- gating the preparation of the narcotic by natives of tribes isolated from one another by long distances. To the pres- ent, the investigation has disclosed the use of the snuff in Colombia amongst the Puinaves of the Rio Inirida, the Kuripakos of the Rio Guainia, the Kubeos of the Rio Vaupés and its Colombian affluents, the Tukanos of the Rios Vaupés and Papuri, the Barasanas and Makunas of the Rio Piraparana and the Taiwanos of the Rio Kan- anari. In Brazil, its use is known amongst the Tukanos of the Rio Uaupés, and there is some indirect evidence that the several tribes of the Rio Issana likewise employ it. The narcotic snuff prepared from species of Virola is called yda-kee in Puinave and ya-to in Kuripako; the Tukanos eall it pa-ree-Ad, which is a loan word from the Nheéngatt or Lingoa Geral parica, the term by which the snuff is known in the Rio Negro-Uaupés area of Brazil. The species of Virola employed in preparing the nar- cotic snuff have been verified as the following :’ 1. Virola calophylla Warburg Nova Acta Acad. Leop.-Carol. 68 (1897) 281. Myristica calophylla Spruce Journ. Linn. Soc. 5 (1860) 4, nomen nudum. Virola incolor Warburg loc. cit. 282. Otoba incolor Karsten ex Warburg loc. cit. 282, in synon. This species is known from Amazonian Brazil and ad- jacent parts of Colombia, Peru and Venezuela. Prior to the collection of the material cited below, the species was known in Colombia only from Villavicencio. Subsequent "I acknowledge gratefully the verification of my identifications by Dr. A. C. Smith of the Smithsonian Institution. [ 242 ] collections and observations have shown it to be rather common in Amazonian Colombia. The type (Spruce 3207) was collected a century ago by Spruce along the Casiquiare in Venezuela. The following collections of Virola calophylla are those upon which the identification of one source of the nar- cotic snuff was made. Cotompia: Comisaria del Amazonas, Rio Apaporis, Soratama, entre el Rio Pacoa y el Rio Kananari. Alt. about 250 m. ““Small tree about 35 feet tall; diameter 8-9 inches. Bark exudes inside a reddish resin when ripped off tree. Externally reddish brown, pebbled. Puinave = ya-kee. Source of narcotic snuff. In flood-forest.’? June 26, 1951, Richard Evans Schultes & Isidoro Cabrera 12855.—Same locality. “‘Large columnar tree. Source of yd-kee snuff. Flood-forest.’’ August 16, 1951, Schultes & Cabrera 13587. 2. Virola calophylloidea Markgraf Repert. Sp. Nov. 19 (1928) 24. Virola lepidota A. C. Smith in Brittonia 2 (1986) 152. This species, the type of which was collected in Mandos by Ule (Ule 8846) half a century ago, is rare in Ama- zonian Brazil in the Rio Negro and Rio Madeira valleys. Hitherto, it has not been known from Colombia, where it appears to be much less common than Virola calophylla in the Amazonian regions. The following collection of Virola calophylloidea is that upon which the identification of the second source of the narcotic snuff was made. CoLtomsi1A: Comisaria del Amazonas, Rio Apaporis, Soratama, entre el Rio Pacoa y el Rio Kananari. Altitude about 250 m. “‘Small tree along flood-bank. Flowers brownish. Puinave name =yd-kee. Source of narcotic snuff.’’ July 3, 1951, Richard Evans Schultes & Isidoro Cabrera 12872. 3. Virola spp. The possibility that other species of Virola are used in this way must not be overlooked. ‘The Tawaino Indians of the Rio Kananari in Amazonian Colombia pointed out [ 243 ] EXPLANATION OF THE ILLUSTRATION PLrare XXXIX, Virota caLopuyiia Warburg. 1, habit, approximately one half natural size. 2, flow- ers, approximately ten times natural size. Drawn by Eimer W. Suivi [ 244 ] Prar XX ATX VLROLA \\2 ge: Wartueg- EXPLANATION OF THE ILLUSTRATION Prare XL. Virota catopnyttoipea Markgraf. 1, habit, approximately one half natural size. 2, flow- ers, approximately ten times natural size. Drawn by Ermer W. Suivi PLATE XL a tree of this genus as representing their source of paricad, but there seemed to be some disagreement amongst the several informers. A sterile collection from the tree has been determined with reservation by Dr. A. C. Smith as Virola elongata (Benth.) Warburg. Cotompia: Comisaria del Vaupés, Rio Kananari, at base of Cerro Isibukuri, August 4, 1951, Richard Evans Schultes & Isidoro Cabrera 18278. III. W hat seems almost certainly to have been this unusual narcotic Virola-snuff was apparently first noticed and re- ported by the famous German ethnologist, Theodor Koch-Griinberg, who explored the northwest Amazon and adjacent areas in the upper Orinoco basin in the early part of this century. The botanical source of the snuff, however, was not ascertained. Writing of the Yekwand (Yecuanda) tribe of the Rio Ventuariin Venezuela, Koch- Griinberg (Koch-Griinberg: ‘‘Von Roraima zum Ori- noco, Ergebnisse einer Reise in Nord-Brasilien und Venezuela in den Jahren 1911-13” 8 (1928) 386) reported the following concerning this narcotic: Of an especial magical importance are the cures during which the witch-doctor inhales hakidyfa. This is a magical snuff used exclu- sively by the witch-doctors and prepared from the bark of a certain tree which, pounded up, is boiled in a small earthenware pot until all the water has evaporated and a sediment remains at the bottom of the pot. This sediment is toasted in the pot over a slight fire and is then finely powdered with the blade of a knife. Then the sorcerer blows a little of the powder through a reed (kuraté) into the air. Next he snuffs, whilst, with the same reed, he absorbs the powder into each nostril successively. The hakudyf/a obviously has a strongly stimulat- ing effect, for immediately the witch-doctor begins singing and yell- ing wildly, all the while pitching the upper part of his body back- wards and forwards. Dr. Adolpho Ducke, profound student of the Amazon flora for more than half a century, has attributed the parica of the upper Rio Negro basin to the leaves of a [ 245 ] species of Virola. In a footnote to a discussion of Pip- tadenia peregrina LL. he wrote (Ducke, A.: ‘‘As legu- minosas da Amazonia brasileira’ (1989) 41): Martius and other writers attribute to this species the source of the narcotic paricd employed by certain Amazonian Indians (the powder of the crushed seeds is inhaled through the nostrils), Notwithstanding, according to information which I obtained from the natives themselves in two localities in the upper Rio Negro, the paricd-powder comes from leaves of species of Virola of the Myristicaceae. So far as I have been able to ascertain, this is the first and only reference in the literature to the use of Virola in the preparation of a narcotic snuff. I have been un- able to substantiate the statement that the leaves are ever used; all of the many reports which I have gathered are in agreement and indicate that the bark is the portion of the plant employed in the preparation of the snuff. Ducke fails to make mention of the identification of pa- rica as Virola in the second edition of his ‘‘As legumi- nosas da Amazonia brasileira’? (Bol. Téen. Inst. Agron. Norte 18 (1949) ). LV. It would seem that /’7rola has not hitherto been re- ported as possessing strong narcotic properties. Never- theless, extremely toxic and narcotic principles have been found in other members of the Iyristicaceae. Perhaps the best known case is Myristica fragrans Houtt., which ‘unites to the medicinal properties of the ordinary aro- matics considerable narcotic properties’; fatal and near- fatal consequences have attended its careless use in India (Wood, H.C., J. P. Remington and S. P. Sadtler: **“The Dispensatory of the United States of America” ed. 18 (1899) 889). This thoroughly studied plant is the source of two spices of commerce: nutmeg and mace—the for- mer derived from the dried, ripe seed without its seed coat and arillode, the latter from the dried arillode. Both [ 246 | spices are employed medicinally as stimulants and car- minatives, but in excessive doses they can produce my- driasis and stupor (Youngken, H. W.: ‘*Textbook of pharmacognosy”’ ed. 5 (1943) 805 ff), and mace has been known to cause ‘‘alarming sensorial disturbances’’ (Wat- son, G.C. in Prov. Med. Surg. Journ. (Jan. 26, 1848) ). The toxic effects of Myristica fragrans are due appar- ently to a volatile oil, myristicine, which can act as a narcotic and which can cause a fatty degeneration of the liver (Finnemore, H.: ‘“The essential oils’’ (1926) 274; Guenther, EK.: ‘‘The essential oils’’ 5 (1952) 78). Ac- cording to Youngken (loc. cit.), nutmeg contains up to 40% of a fixed oil and up to 10% of a volatile oil (4% of which is the narcotic principle myristicine). The whole family Myristicaceae is characterized by the presence of cells which contain a semi-fluid or crystalline oil, the color of which varies from yellow to red or even to a brownish black (Kraemer, H.: ‘‘Scientific and applied pharmacog- nosy’ (1915) 250). It would appear probable, then, that the violent narcotic properties of yd/ce-snuff may be due, in great part if not entirely, to myristicine. Pharmaco- logical studies which are at present under way will, it is hoped, clarify this interesting problem. In the Indian Archipelago and New Guinea, other spe- cies of Myristica yield nutmeg, but there seems to be no record of the purposeful use of this genus amongst native Asiatic peoples as the source of a narcotic. Vv. The preparation of ydkee-snuff* is relatively simple. I have observed the process several times with Puinave Indians and have also seen the powder prepared by Tu- > Because of the confusion which has grown up around the widely used term parica, which will be discussed fully below, I prefer to em- ploy the short, easily-pronounced and specific Puinave Indian name for the Virola-snuff. kano and Kuripako Indians. In all cases, the fabrication was essentially the same and corresponds closely to Koch- Griinberg’s brief description of its manufacture amongst the Yekwanas. Descriptions of the preparation as given to me by several Kubeo, Barasana and Makuna Indians indicate that no appreciable difference marks the process as practised by these tribes. In all instances, the same species were pointed out as sources of the snuff. We are, therefore, justified, I believe, in assuming that, in the area of its distribution in the Comisaria del Vaupés of eastern Colombia, as well as in the upper Rio Negro of Brazil, the preparation of ydhee is well standardized. In this respect, it resembles coca (Avrythrovylum Coca Lam.), the fabrication of which is, for practical purposes, the same throughout the area; on the other hand, it is unlike the famed narcotic caapi or yajé (Banisteriopsis spp.), the preparation of which seems to vary with almost each tribe. The Indians usually strip the bark from the trees dur- ing the early hours of the morning, before the sun has begun to penetrate the forest canopy to heat up the trunk. Large strips of the bark, which peel easily from the cambium, are torn from the trunk and tied into loose bundles. Almost immediately upon separation of the bark from the tree, a profuse exudation or ‘“‘bleeding”’ of a thick reddish resin-like liquid, which soon becomes viscous, oozes forth from the inner surface of the bark in small drops. The active principle is contained in this ex- udation—called oom (latex), or, specifically referring to these species of Vrola, hd-oom-tee-et or yd-kee-oom, in Puinave. According to the Indians, this exudation is greatly reduced in quantity and is weaker in its narcotic effects when the trunk of the /7ro/a tree has received the warmth of the sun’s rays. The bundles of bark are brought in and placed in water [ 248 ] for about half an hour. They are then taken out, and the soft inner layer, on the surface of which the red exuda- tion has congealed, is rasped off with a knife or machete. The shavings or raspings (yd-kee-taa in Puinave) are thrown into an earthen pot or enamel tray, and the rest of the bark is discarded. When enough shavings have been accumulated, a small amount of water is added, and the mass is thoroughly kneaded and squeezed. The water becomes muddy and assumes a brownish or tan hue. This turbid liquid is strained several times, usually through a piece of finely hammered bark-cloth (prepared from a species of Olmedia) into a small-mouthed earthen- ware pot. The residual shavings, when as much of the water has been expressed as possible, are thrown away. Enough water is added to the strained liquid to fill the pot, which is then set to simmer over a slow fire. From time to time, a sordid foam, which rises to the surface, must be scraped off with a piece of bark. The boiling is allowed to continue for three or four hours, more water being added if evaporation be too rapid, until nothing remains except a thick, dark brown syrup at the bottom of the pot. This syrup must not be dried rapidly over a fire; the pot is set in the sun, and the syrup is per- mitted to solidify slowly. When nothing but a dry, brown crust is left, the residue is scraped free from the pot and is ground into a fine powder with a water-smoothed stone as a pestle and the pot or anenamelware tray as a mortar. It is then ready to be mixed with ashes which have been made, the while, from the bark of a small wild cacao tree (Theobroma subincanum Mart.). Usually equal amounts by volume of ashes and ydhee-powder are used. When they are thoroughly mixed, the product is put into asmall bag made of finely hammered bark (Ol- media sp.) or cloth and is sifted through the bag by means of a gentle beating against the side of a small-mouthed [ 249 ] receptacle. The resulting dust is the final snuff. [tis kept either in asmall glass bottle, tightly corked, or else, more traditionally, in a type of jar made, as Koch-Griinberg described, from a large snail-shell to which a hollow bird- bone tube has been fixed with pitch. This tube is stopped with a plug of feathers glued together with pitch at the basal end to form a tight-fitting stopper. The consumption of yakee-snuff is limited to medicine- men and is, therefore, small. Since it is said to lose its in- toxicating properties rather rapidly, even when in a tight container, it is made in small amounts and frequently. VI. It may be of interest to append a few observations which I was able to make personally after taking yakee- snuff. I took about one-third of a level teaspoonful of the drug in two inhalations using the characteristic V-shaped bird-bone apparatus by means of which the natives blow the powder into the nostrils. This represents about one- quarter the dose usually absorbed by a diagnosing medi- cine-man, who takes about one slightly heaped teaspoon- ful in two or three inhalations at close intervals (of ap- proximately fifteen or twenty minutes). The dose was snuffed at five o'clock one afternoon. Within fifteen minutes a drawing sensation over the eyes was felt, followed very shortly by a strong tingling in the fingers and toes. The drawing sensation in the fore- head rapidly gave way to astrong and constant headache. Within one half hour, there was a numbness of the feet and hands and an almost complete disappearance of sen- sitivity of the finger-tips; walking was possible with dif- ficulty, as in a case of beri-beri. Nausea was felt until about eight o’clock, accompanied by a general feeling of lassitude and uneasiness. Shortly after eight, I lay down in my hammock, overcome with a heavy drowsiness [ 250 ] which, however, seemed to be accompanied by a muscu- lar excitation, except in the extremities of the hands and feet. At about nine-thirty, probably, I fell into a fitful sleep which continued, with frequent awakenings, until morning. The strong headache over the eyes lasted until noon. A profuse and uncomfortable sweating, especially of the armpits, and what might have been a slight fever lasted from about six o’clock all through the night. There was a strong dilation of the pupils during the first few hours of the experiment. No food was taken and no to- bacco was smoked from the time the experiment began until one o’clock in the afternoon—that is, for twenty hours during the course of the experiment. Since this experiment was performed under primitive conditions in the jungle, all observations had to be made by myself. In spite of its many and serious shortcomings, the experiment indicates the narcotic strength of the snuff. The dose employed by the medicine-men is sufficient to put them into a deep but disturbed sleep, during which delirious mumblings or, sometimes, shouts are emitted; visual hallucinations or dreams are reported to accompany the narcotic sleep very often. These are “‘interpreted’’ by an assistant who awaits the prophetic or divinatory sounds. Some medicine-men, it is said, are affected more violently than others and uncontrollable twitching of the fingers and facial muscles and a popping of the eyes are not infrequent symptoms. There is one report of the death, about twenty years ago, of a Puinave medicine-man on the Inirida River, whilst he was under the influence of ydkee. Some payés (witch-doctors) are said to take ydkee as frequently as four or five times a month; usually, so far as I have been able to ascertain, one doctor will not undergo the diagnosis-narcosis with V7rola-snuff more than once a month. All reports would seem to indicate that it is a dangerous narcotic. [ 251 | VIL. The use of a snuff commonly called paricd has been known for a century or more, and the source of the nar- cotic has quite generally been attributed to the legumi- nous tree Piptadenia peregrina. This tree has long been recognized as the source of a violently narcotic snuff which is employed by the natives of the Caribbean area and of northern South America, including the basin of the Rio Orinoco, and which is widely called yopo or niopo. Hum- boldt’s account of this snuff (Humboldt: ‘‘Voyages aux régions équinoxiales du nouveau continent. . .”’ 2 (1819) 260) referred to the preparation and utilization of the drug which he had observed in 1802 amongst the Otomaco and Guahibo Indians of the Orinoco in Venezuela and Colombia: ‘‘Ex seminibus tritis calci vivae admixtis fit tabacum nobile quo Indi Otomacos et Guajibos utun- tur.’’ The plant used was identified as Acacia Niopo, now considered a synonym of Piptadenia peregrina. Sir Robert Schomburgk, who first explored British Guiana from 1835 to 1839, referred the narcotic parica or paricarand to Mimosa acacioides Benth. (Schomburgk : ‘Travels in British Guiana’’ [trans]. W. EK. Roth] 1 (1922) 92). This binomial is a synonym of Piptadenia peregrina. In June 1854, during his botanical explorations in the vicinity of the cataracts of the Rio Orinoco at Maipures, Richard Spruce came upon a wandering group of Gua- hibo Indians from the Rio Meta in Colombia preparing yopo-snuff. He described the preparation of the narcotic and attributed it (as P. Miopo) to Piptadenia peregrina (Spruce, R. Led. A. R. Wallace] ‘‘Notes of a botanist on the Amazon and Andes”’ 2 (1908) 427), and he reported the common name as niopo in Venezuela and paricd in Brazil. Bates, who worked in the Amazon at the same time [ 252 ] as Spruce, reported, but without the support of botani- cal material, that parvicd was prepared from a species of Inga (Bates, H. W.: ‘‘A naturalist on the River Am- azon’” 1 (1868) 381). Carl F. P. von Martius (Zur Ethnographie Amerika’s sumal Brasiliens’’ (1867) 390) stated that the Mundurukti Indians of Brazil used paricd, a snuff from the ‘‘seeds of Mimosa acacioides’’, having borrowed the habit from their neighbors, the Muras and Mauhés. He also (Joc. cit. 441, 681) asserted that the Omaguas of Peru use this same snuff, and that it was well known amongst the Paravilhanas of the region north of the Rio Negro in Brazil and in British Guiana. The German ethnologist Koch-Griinberg, who carried out very extensive investigations in the upper Rio Negro and lower Apaporis basins from 1908 to 1905, similarly attributed the paricd of this area to Mimosa acacioides (Koch-Griinberg, T. : ‘‘Zwei Jahre unter den Indianern”’ 1 (1909) 823). There seems to be no evidence that ma- terial for botanical determination supports his identifica- tion, which may have been advanced by Koch-Griinberg because of the extreme similarity of the snuff to the al- ready widely known yopo-snuff. Koch-Griinberg re- ported: It is a grey snuff with strong narcotic properties, known in Lingoa Geral as paricd and prepared from the dried seeds of a species of Mi- mosa. It is kept in small rounded calabashes or in snail-shells, the opening of which usually is closed with a piece of mirror imbedded in pitch and which, as in the case of the calabash, has a bird-bone spout fixed with pitch. . . Snuffing is done through a forked instrument made of two communicating bird-bones, which are glued together with pitch. . . In using it, a bit of the powder is poured from the snuff-box into the palm of the hand and is scooped up into the bird- bone. Then the end of one of the bones is inserted into the nostril, and the other is put into the mouth. With short blows, the fine pow- der is injected to the furthest membranes of the nose. In a footnote, Koch-Griinberg (loc. cit.) gives a speci- [ 253 | fic identification, and he makes an observation which might indicate that he had assumed that this identifica- tion must be the correct one, even though he had no botanical specimens: Mimosa acacioides Benth. This snuff is distributed over a great part of tropical South America, from the Orinoco and the Guianas to the southern afHuents of the Amazon River. Likewise, the instruments which are used in taking the snuff are very similar. Somewhat later, Whiffen (Whiffen, T. : ‘‘The North- west Amazonas”’ (1915) 143) reported that: The Tuyuka and other tribes north of the Japura use as a stimulant paricd or niopo, a wonderful snuff which is a strong narcotic and very similar in its effects to coca. It is made from the dried seeds of a mi- mosa, and like coca, is mixed with quicklime and baked clay. The seeds are roasted and then pounded in a shallow wooden mortar, and the snuff, when made, is packed in snail-shells and is inhaled through hollow bird-bones inserted in both nostrils. It would appear that this reference of Whiffen (who never visited the 'Tuyuka country) to the snuff used by the tribes north of the Caqueta (Japura) may be based upon Koch-Griinberg, for in speaking of the Andoke and Karihona tribes, he wrote (loc. cit.): They all use tobacco-juice, coca, and a white snuff that I thought must be the famous niopo but could not find out anything about it. My own assumption would be that this ‘‘white snuff”’ is actually nothing but tobacco-snuff, which is widely used in the upper Amazon area, where it is always a greyish preparation, because of the large amount of ash mixed with the pulverized tobacco. It cannot be the myristicaceous snuff, for even after the admixture of ashes, this is of a brownish color. Furthermore, the use of the Virola-snuff is restricted to a few practitioners and is not permitted to all members of the tribe. Recently, Mr. Paul H. Allen, who was engaged dur- ing 1944 in botanical work in the region of the Vaupés River and its affuents, reported (Allen, P. H.: ‘‘Indians [ 254 ] of southeastern Colombia’’ in Geogr. Rev. 87 (1947) 579) that the Kubeo Indians of the Kuduyari River employ parica. He gives as the source of paricd the leguminous Piptadenia peregrina: The powdered seeds of paricd (Piptadenia peregrina) are blown force- fully through a bone into the nostrils of the payé, producing a sort of ecstasy, during which he determines the guilty party [i.e. enemies who have sent sickness through the air]. In a letter (January 14, 1952), answering my query as to his designation of the source of the snuff, Allen writes me that he did not see the seeds of Piptadenia himself; that he had been informed that seeds were used in the area; and that, on the basis of the literature, he con- cluded that these represented Piptadenia peregrina. It would seem, therefore, that there is much confusion of parica with yopo. This stems primarily from Spruce’s account, for he recorded (/oc. cit. 427): I first gathered specimens of the paricd (or niopo) tree in 1850, near Santarém, at the junction of the Tapajé6z and Amazon, where it had apparently been planted. In the following year, I gathered it on the little river Jauouari—one of the lower tributaries of the Rio Negro— where it was certainly wild. But I did not see the snuff actually pre- pared from the seeds and in use until June, 1854, at the cataracts of the Orinoco, We know that Spruce was aware of Humboldt’s re- port of yopo-snuff. Upon finding the same tree in an area such as the Rio Negro where, according to local reports, the medicine-men took a snuff called parica, it is possi- ble that Spruce assumed that Piptadenia peregrina was also the source of the paricd-snuff. This is not an easy explanation to accept, for Spruce deservedly enjoys the reputation of one of the most critical and accurate of ethnobotanical students of South America. Nor are we at all certain to this date that parica does not refer to snuff made with Piptadenia seeds in some parts of the Amazon Valley. Furthermore, we have the reports of [ 255 | EXPLANATION OF THE ILLUSTRATION Pirate XLI. Stripping bark from Virota caLo- PHYLLA for the preparation of yakee-snuff. Rio Apaporis, Comisaria del Amazonas, Colombia. Photograph by Ricuarnv Evans Scuuttes [ 256 ] : N LI PLATE EXPLANATION OF THE TLLUSTRATION Prare XLUL. (Zop) A photograph of the inner sur- face of the bark of Viroia caropuyiLa (Schultes & Cabrera 12855), showing the red resin-like exuda- tion which is elaborated into yakee-snuff. Rio Apaporis, Comisaria del Amazonas, Colombia. ( Bottom) The thick syrup resulting from the boiling of the scrapings of the inner bark of Viroia cato- PHYLLA is allowed to dry before being pulverized and mixed with ashes of Throproma scuntincaANUM in the preparation of yakee-snutf, Rio Apaporis, Comis- aria del Amazonas, Colombia. Photographs by Vacuarp Evans Scuunres AN LII PLATE Koch-Griinberg stating that the narcotic snuff of the northwest Amazon is derived from a legume. Neverthe- less, we must emphasize Spruce’s own words that he “did not see the snuff actually prepared from the seeds and in use...’ in the Rio Negro area. Some of the records of a narcotic snuff from the Am- azon Valley might actually refer to Piptadenia, as, for example, Herndon’s report (Herndon, W. L.: ‘‘Explo- ration of the Valley of the Amazon’’ (1854) 819), which he attributed to a trader of snuff amongst the Mundu- ruki Indians of the Rio Tapajéz. This snuff is reported to be made from powdered seeds, not from bark, which would suggest Piptadenia and not Virola. But, the Mun- durukt paricd is composed of several constituents: Several vegetable substances compose paricd: first, the ashes of a vine that I cannot class, not having been able to procure the flowers ; second, seeds of the Acacia angico of the leguminous family; third, juice of the leaves of the abuta (cocculus) of the menispermes family. In the twelve years which I have devoted to the study of the flora of the northwest Amazon, I have never seen Piptadenia peregrina either cultivated or wild in the area. I have never seen any snuff prepared from legumi- nous seeds in this area, nor have I heard of the practice which appears, in Colombia, to be confined to the Ori- noco basin. ‘The Puinaves, who live on the Inirida and Guaviare, where Amazon and Orinoco tribes meet, are acquainted with both the myristicaceous and the legumi- nous snuff and are quick to distinguish between the two. Whether or not paricd-snuff is prepared from a legu- minous plant in some parts of the Amazon, it is quite definite that the term parvicd in most parts of the lower Amazon does refer to certain trees of the Leguminosae. Paricd is employed in Brazil to denote Cassia fastuosa Willd., Cedrelinga catenaeformis Ducke, Parkia spp., Piptadenia spp., Pithecolobium spp., Schizolobium ama- [ 257 ] sonicum (Hub.) Ducke, 8S. parahybum (Vell.) Blake, and Senegalia spp. Numerous derivative terms, such as paricd branca, paricd de cortume, parica da terra firme, paricd grande da terra firme, parica da varzea, paricarand and paricazinho are likewise applied to species of the Legu- minosae (Record, S. J. & R.W. Hess: ‘“Timbers of the New World” (1948) 242 ff.; Le Cointe, P.: ‘‘Arvores e plantas tteis,’’ ed. 2 (1947) 387 ff.; Ducke, A.: ‘‘As leguminosas da Amazonia brasileira’ (1989) 164-165; Penna, M.: ‘‘Diccionario brasileiro de plantas medicin- ais’’ ed. 8 (1946) 295; Huber, J. in Bol. Mus. Para. Hist. Nat. Ethnog. 6 (1909) 218). The only pronouncements which we can, at this mo- ment, make are (1) that the paricad of the Rio Negro- Uaupés basin is prepared from the bark of certain species of the myristicaceous genus Virola and not from the seeds of the leguminous genus Piptadenia; and (2) that Pipta- denia-snuff or yopo is (at least, at the present time) prob- ably unknown and most certainly not employed in this area. As the foregoing discussion has pointed out, the iden- tification of the botanical sources of South American narcotic snuffs in general and of those called paricd in particular is in a state of extreme confusion. ‘To a great extent, this is directly attributable to the habit, in an- thropological and geographical articles, of using the com- mon names and of depending upon common names and not botanical material for plant identification. This has been most forcefully emphasized in a recent and excellent summary of our knowledge of stimulants and narcotics of the South American Indians, in reference to the vari- ous snuffs attributed to Piptadenia (Cooper, J. M.: Bur. Amer. Ethnol. Bull. 148 (1949) 586, map 10): Our tribal records on which the above list and accompanying dis- tribution map (map 10) are based are probably very incomplete. On the other hand, some of the attributions may not be correct, since in some cases the lack of exact botanical identification makes it doubtful [ 258 | whether we have to do with Piptadenia snuff, tobacco snuff, or snuff . ‘ * ‘ from some other plant—as, for instance, the topsayri’ in early Peru —or from an unidentified tree bark among the Yecuana. . . There is, in this problem of the identity of parica, a most curious phase which is, as yet, apparently far from solution. In the kits which contain the witch-doctor’s parapher- nalia amongst certain tribes on the Rio Vaupés and the upper Rio Piraparana, there is always a lump or two of a clear yellow amber-like resin. This is used as a snuff, but seems to have no narcotic properties. Before a ‘‘di- agnosis’’, the medicine-man rasps a small amount of this resin and pulverizes it finely, after which it is taken into the nostrils. The resin, when powdered, is slightly aro- matic. It is my belief that the use of this resin as a snuff is related perhaps to a kind of ‘‘purification’’, prepara- tory to making a diagnosis, but we have very little infor- mation about it. I suspect that the source of the resin might be a species of Clusia, but I have no personal ob- servations to offer in support of this suspicion. There is the barest of possibilities that the resin might be of my- risticaceous origin, for we recall that Myristica Bicuhyba Schott of Brazil yields a balsam which is sometimes sub- stituted for copaiba (Copaifera officinalis L..), (Kraemer, H.:‘* Scientific and applied pharmacognosy”’ (1915) 250). There is no actual evidence, however, for such a belief. I have seen these lumps of paricad amongst the Bara- sana Indians of the Cano 'Tee-mee-fia, an affluent of the Rio Piraparana. It was impossible to procure material or a description of either its preparation or of the source tree—but we know that it is procured from a large tree of the forest. Two of my colleagues, who spent long periods in the Colombian Vaupés, have likewise encountered this ele- [ 259 ] ment of the local doctor’s kit. Mr. Paul H. Allen (in litt., January 14, 1952) has written the following infor- mation to me: Payé | witch-doctor] kits seen in the Bajo Vaupés, but particularly from the lower Querari and vicinity, contained lumps of an amber gum, which local people told me was the active principle of paricd. I doubted it very strongly, and the payés were far too suspicious and resentful of my collecting activities to be very communicative, much less to be willing to stage a demonstration. Dr. Lothar Petersen, a physician in Bogota who spent many months in medical studies amongst the Indians of the Vaupés, likewise found these lumps of resin and re- ports that it is known locally as parica. He procured several lumps from a witch-doctor in the headwaters of the Rio Piraparana, but he also was unable to learn the tree from which the resin was obtained. Dr. Petersen has kindly given me two small lumps which will be stud- ied chemically in an attempt to ascertain their approxi- mate composition and, thus indirectly, perhaps to dis- cover their botanical source. It is clear from the problems raised in the present paper that much botanical work still must be carried on in our study of narcotic and stimulant plants in the Am- azon Valley. It is hardly believable that such a widely used and virulent narcotic snuff as ydkee would have to wait until this late date in the history of botanical ex- ploration of the Amazon for identification. It makes one wonder how many more narcotic plants, up to now ob- scured by better-known ones, still wait to be discovered. The drawings reproduced in this article were made possible through a grant from the American Academy of Arts and Sciences. [ 260 ] THE ORIGIN AND POSSIBLE EVOLUTION OF SUB-TASSEL EARS IN MAIZE BY Watton C. Gauinat! MopeERN maize has many vestiges which may reflect primitive characteristics of wild maize and its relatives. One of the least mentioned and possibly most significant rudiments in modern maize is the remains of a spathe subtending the lowermost tassel branch or pair of tassel branches. Anderson (1951) has described this vestige as ‘‘a kind of little ridge or scar as if a leaf had started to grow out and then had been pulled off."” Sometimes this spathe is well-developed and associated with a small sub- tassel ear, as is evident in at least two published illustra- tions (Hiickel, 1887; Weatherwax, 1916—fig. 10). Such an ear near the base of the tassel may be derived from a single staminate spikelet (Weatherwax, 1925). An examination of over 1000 tassels from varieties of maize from North, Central and South America in the Maize Herbarium of Dr. Paul C. Mangelsdort revealed a series of types ranging from a reduced and sometimes adnate spathe subtending the lowermost tassel branch to a fully developed leaf borne at a node with a small sub- tassel ear. The adnate or vestigial spathe and its deriva- tions were found in over 70 per cent of the population studied. Sub-tassel spathe development was especially Research Fellow at the Bussey Institution of Harvard University. 5 [ 261 ] frequent and pronounced in varieties from) Bolivia and Peru. This feature is enhanced by tassel proliferation resulting from unfavorable photoperiods during floral differentiation of short-day maize and is also character- istic of tunicate, corn-grass and teopod maize. ‘Tassels which were proliferated or associated with any of the above variants were excluded from this study. Specimens of Trvpsacum and Muchlaena trom the Keo- nomic Herbarium of Oakes Ames were examined in an attempt to determine if this rudimentary spathe might be derived from introgression with these near relatives of maize. It was found that in Tripsacum pilosum, T. lanceolatum, T. dactyloides and TY fasciculatum there is a small ridge encircling the base of the inflorescence. In these species, with 7. dactyloides as a possible exception, the ridge is more in the nature of a fracture line for the lowermost rachis-fruit case than that of a vestigial leaf. In Muchlaena mevicana there is often a small leaf-like protrusion subtending the lowermost tassel branch. This structure is probably a vestigial spathe since its homo- logue subtending a lateral pistillate spike is a well- developed spathe. The degree of development of the spathe subtending the tassel of duchlaena appears to be, on the average, intermediate between that of Tripsacum and that of maize. In modern maize, partial development of the sub- tassel spathe and its axillary branch is usually associated with various anomalies. Both the spathe and its axillary peduncle may be adnate to the rachis. The peduncle may adhere to the rachis over a greater distance than its subtending spathe (Figs. 1-3). The auricles of the spathe may be greatly elongated on cither side of a reduced blade (Figs. 4-6). Development of the spathe may be unilateral in relation to the branch, so that it appears as if its blade had been cleaved down the midrib and torn [ 262 ] off on one side (Figs. 7-9). Various twisted configura- tions may distort the spathe, rachis, and peduncle as the branch tends to become opposite rather than adjacent to its associated spathe (Figs. 10-138). If the axillary branch develops to one side of, or on the opposite side of, the node at which a terminal leaf would normally be ex- pected, then both spathe-leaf and branch may achieve optimum development (Fig. 14). Two leaves, or a leat and a sheath (husk), may be borne at this node. Such a pair of leaves (Figs. 15, 17) may have a common origin (Figs. 7-9), although distortion of the node may cause them to appear as being separated by a short and twisted internode (Fig. 13). In extreme cases of spathe develop- ment, a single spikelet or the tassel branch as a whole may be modified to form a small shank terminated by a small ear (Figs. 15-17). The morphological change from either a spikelet or tassel branch to a many-ranked ear involves a change from bilateral to radial symmetry. Such a transformation is common in maize. Depauperate ears frequently exhibit reductions from a radial to a bi- lateral condition. One might expect that, if there were a reduction during evolution of a leaf terminal to the culm, then there might also be a corresponding reduc- tion of its axillary ear to a bilateral tassel branch. Archaeological remains have been found which sug- gest that sub-tassel ear development may have been char- acteristic of primitive maize (Mangelsdorf, 1954 and unpub. ). This evidence, so far as it goes, indicates that the tassel of wild maize was unbranched or sparsely branched and bore a few pistillate spikelets basally. The spathe subtending the tassel may have had a protective function for these pistillate spikelets during the prepolli- nation period. Subsequent to pollination, rapid elonga- tion above the sub-tassel node might elevate the mature grain for dispersal. Mangelsdorf has suggested (unpub. ) C gee 4 2 O35 EXPLANATION OF THE ILLUSTRATION Prare XLII. 1, 2, 8, profile, front and rear views respectively, of the divergence point of the lower- most peduncle from the rachis and its subtending spathe-rudiments which are adnate to the rachis. Natural size. t, 5, 6, as in previous figures, with the addition of abnormally elongated auricles to an otherwise inhibited spathe. Natural size. 7, 8, 9, asin previous figures, showing cleavage of the spathe followed by unilateral development. Natural size. 10, 11, 12, as in previous figures, except for the equal development of the rudimentary parts of the spathe, Natural size. 13, unilateral spathe development at a twisted and elongated node. One half natural size. 14, tassel branch development on the opposite side of the node from its associated spathe. One half natural size. 15 sile staminate spikelet. Two leaves, probably of , sub-tassel ear development from a modified ses- common origin, are borne at this node. One half natural size. 16, sub-tassel ear borne opposite a leaf at the same node. One half natural size. Drawn by Wavron C, GAatinat ALV Ig jd a 4 EXPLANATION OF THE ILLUSTRATION Prare XLIV. 17, photograph of a sub-tassel ear of similar nature to that shown in Fig. 15. Photograph by Paut C. MANGELSDORF a Yr a LI\ PLATE that such alittle sub-tassel ear, borne opposite this spathe, may have attracted man to domesticate an otherwise earless form of primitive wild maize. The present vari- ability in development of this sub-tassel ear or its rudi- ments might be attributed to its presence in only one or a few of several geographical races of wild maize. It may have also been variable in its expression in wild maize, perhaps dependent on growing conditions. The frequent adnation of the sub-tassel spathe and its axillary peduncle to the rachis may be another indication of the important role that adnation has played in the de- velopment of the inflorescence and plant of maize. The cupule, a structure in maize associated with the attach- ment point of a pair of pistillate spikelets to the rachis, has probably been formed by a prophyll adnate to the rachis (Nickerson, 1954). LITERATURE CITED Anderson, Edgar, 1951. The sacred plume. 24 pp. Pioneer Hi-Bred Corn Co., Des Moines, lowa. Hiickel, EK., 1887. Gramineae—in Engler and Prantl. Die Natur- lichen PHanzenfamilien 2, Leipzig. Mangelsdorf, Paul C., 1954. New evidence on the origin and ances- try of maize. Amer. Antiquity 19: 409-410. Nickerson, N. H., 1954. Morphological analysis of the maize ear. Amer. Jour. of Bot. 41: 87-92. Weatherwax, Paul, 1916. Morphology of the flowers of Zea Mays. Bull. Torrey Bot. Club 43: 127-144. , 1925, Anomalies in maize and its relatives—II. Many-flowered spikelets in maize. Bull. Torrey Bot. Club 52: 167-170. [ 264 ] BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CaMBRIDGE, MAssaAcHUsETTS, NOVEMBER 26, 1954 Vor. 16, No. 10 ON THE ORIGIN OF OATS BY Dexter R. Sampson INTRODUCTION Man has long been interested in the origin of his culti- vated plants, for they are the foundations upon which his various civilizations have been built. Primitive peoples frequently regarded their plants as gifts of the gods, but modern man seeks more natural explanations, both to satisfy his curiosity and to provide a sound basis for plant improvement projects. The present paper is an attempt to bring together the available evidence on the origin of oats,the world’s fourth most important cereal crop (Stanton, 1953). It is pro- posed to attack the problem by reviewing: the several systems of classification of the genus Avena L., together with the geographical distributions of the species; the cytological and genetical interrelations; and finally, the impact of man upon the genus. The author wishes to acknowledge his indebtedness to Elizabeth Alden and Tatiana Taitschevsky who trans- lated Russian references and to Drs. Albert F. Hill and Paul C. Mangelsdorf for their valuable criticisms and suggestions. Parr I. CLassIvICATION AND DISTRIBUTION The genus Avena is generally considered as consistin fo] > y [ 265 ] of two sections: Huwavena Griseb., which contains the cultivated oats and related wild annuals; and Avenastrum C. Koch, which comprises about forty species of wild per- ennials, most frequent in northern Eurasia, but with two species native to western North America. By recent authors (Hitchcock, 1950; Clapham et al., 1952) the sec- tion Avenastrum is treated as the genus Helictotrichon Besser. In the present paper only the section Muavena will be considered. Natural Systems of Classification Three students, Cosson (1854), Thellung (1911) and Mal’tsev (1929), have each attempted to devise a natural classification for the section Huavena. Cosson divided the section into subsection Sativae con- taining the cultivated species, and subsection 4 grestes containing the wild. This division is based on the fact that the spikelets of the cultivated species are firmly at- tached to the rachilla and do not fall at maturity, whereas the spikelets of the wild species are articulated to the rachilla and fall spontaneously when ripe. Cosson further divided the Agrestes into the series Biformes in which only the lower floret is articulated so that the spikelet falls as a unit, and the series Conformes with all florets articulated and falling separately. The disposition of species by Cosson is: Subsect. I. Sativae A, sativa, A. orientalis, A. strigosa, A. brevis, A. nuda Subsect. Il. Agrestes Series 1. Biformes A. ventricosa, A. sterilis, A. eriantha Series 2. Conformes A. longiglumis, A. clauda, A. hirsuta, A. fatua The system of 'Thellung is based on the assumption that each of the cultivated species is related to and de- rived from some wild species. Accordingly the Sativae [ 266 ] of Cosson are distributed between the Biformes and Con- Jormes. Unfortunately several wild species are not con- sidered. The disposition of species by Thellung is: Subsect. I. Byformes A, sterilis ssp. 1 macrocarpa ssp. 2 byzantina Subsect. II. Conformes A. fatua ssp. 1 fatua ssp. 2 sativa (incl. A. orientalis) ssp. 3 nuda A. strigosa ssp. 1 barbata ssp. 2 strigosa (incl. A. brevis) ssp. 3 Wiestit ssp. 4 abyssinica The system of Mal’tsev was developed after an exten- sive study of herbarium material and of a series of plant- ings of Avena species in various parts of the U.S.S.R. The material was gathered by the expeditions of Vavilov and associates, the geographical and ecological data col- lected by the expeditions being used in the treatment. On the basis of chromosome numbers Mal’tsev divided the section into subsection Aristulatae and subsection Denticulatae. Subsection A ristu/atae contains the diploid and tetraploid species. It is regarded as the older, per- haps derived from section Avenastrum, with its center of origin in the western part of the Mediterranean region. Subsection Denticulatae contains the hexaploid oats and is believed to be the younger. It is regarded as having arisen from an unknown ancestral complex in the moun- tains of Central Asia, a conclusion which Schiemann (1982) says has not been shown to be based on fact. The characters used by Mal’tsev to define his subsections and series will be given in the key. In the following outline of the system all cultivated oats are marked by asterisks and each is believed to be derived from the wild subspe- cies which immediately precedes it. The division into [ 267 ] chromosome groups is provided for the convenience of the reader and is not part of Mal’tsev’s system. Subsection I. Aristulatae Diploid Group 2n=14 Series 1. Inaequaliglumes 1. A. clauda 2. A. pilosa Series 2. Stipitatae 8. A. longiglumis 4. A, ventricosa sens. ampl. ssp. 1. ventricosa 2. Bruhnsiana Series 3. Eubarbatae 5. A. strigosa sens. ampl. ssp. 1. hAirtula 2. strigosa* Tetraploid Group 2n=28 ssp. 3. barbata 4, Wiestu 5. Vaviloviana 6. abyssinica* Subsection II. Denticulatae Hexaploid Group 2n=42 6. A. fatua sens. ampl. ssp. 1. septentrionalis . nodipilosa* . meridionalis 2 3 4. macrantha* 5. fatua 6. sativa* 7 . cultyformis 8. praegravis* 7. A. sterilis sens. ampl. ssp. 1. Ludoviciana . pseudo-sativa* . trichophylla . nodipubescens* » macrocarpa O& Or mB oo ew . bysantina* [ 268 | In the present paper, following Mordvinkina (1936), binomials are restored to several of the taxa treated as subspecies by Thellung and Mal’tsev. This is scarcely justifiable in the case of A. strigosa sens. ampl. which seems to form a unified although poorly understood group. The division of subsection Denticulatae into the traditional four species is also questionable, it being prob- able that these oats comprise but one highly polymor- phous species. The present treatment follows common practice and is justifiable only for the sake of convenience. Key for the Determination of Species (largely after Mordvinkina, 1936) A. Lemma usually ending in two slender awns; plants mostly slen- der. (Subsect. Aristulatae Malz.) B. Glumes very unequal, upper almost twice as long as lower; 2n= 14 (Series Inaequaliglumes Malz.) C. All florets of spikelet articulated, falling separately at ma- turity A. clauda C. Only lower floret articulated, spikelet falling as a unit A. pilosa B. Glumes equal or subequal D. Callus 5-10 mm. long, subulate; 2n=14 (Series Stipitatae Malz. ) E. All florets articulated A. longiglumis E. Only the lower floret articulated F. Callus+5 mm. long; glumes 25-30 mm. long A. ventricosa F. Callus to 10 mm. long; glumes to 40 mm. long A. Bruhnsiana D. Callus about 2 mm. long; 2n=14 or 28 (Series Eubarbatae Malz. ) G. Lemma ending in 2 awns and having 1 or 2 lateral teeth; glumes with 7-9 nerves H. Lemma with 1 lateral tooth (sometimes none); awns at tip of lemma usually exserted beyond the glumes I. Florets articulated, articulation-scar oblong-linear ; awns at tip of lemma to 12 mm. long A, hirtula I. Florets not articulated; awns 1-5 mm. long A, strigosa [ 269 ] H. Lemma with 2 lateral teeth; awns not exserted J. Florets articulated, scar oval or round K. Awns at tip of lemma 3-6 mm. long A, Wiestii kK. Awns+1 mm. long A. Vaviloviana J. Florets not articulated; awns 1-4 mm. long A. abyssinica G. Lemma ending in 2 awns to 5 mm. long, lateral teeth wanting; glumes with 9(10) nerves; articulation-scar oblong A. barbata A. Lemma usually ending in two small teeth, never produced into two setiform awns; plants mostly robust; 2n=42 (Subsect. Den- ticulatae Malz.) L.. Lower floret articulated with the rachilla M. Upper florets not articulated, spikelet falling as a unit A. sterilis M. All florets articulated A. fatua L. None of the florets articulated, separated by fracture of rachilla N. Most of rachilla segment remains attached to upper floret on threshing; fracture surface at base of lower floret slant- ing A. byzantina N. Rachilla segment remains attached to lower floret; fracture surface not slanting A. sativa Distribution and Relationships of Avena Species 1. A. clauda Dur. A wild oat found in Algeria (Battandier and Trabut, 1904), Tripoli, Palestine (Post, 1983), Asia Minor, Iraq, Transcaucasia and eastward to Turkistan (Mal’tsev, 1929). Reported as rare in Attica, Greece (DeHalacsy, 1904). Mal’tsev says this species more closely resembles the genus Helictotrichon than any other Avena species. He therefore considers it to be the most ancient, and to have given rise to the remainder of the Aristu/atae through A. longiglumis. 2. A. pilosa M. Bieb. A wild oat occurring in north- ern Algeria (Battandier and Trabut, 1904), Tripoli, Syria (Post, 1988), Asia Minor, the Caucasus and eastward to southern Turkmenistan (Mal’tsev, 1929). [ 270 | Battandier and Trabut, and Mordvinkina (1936) state that only the lower floret of 4. pilosa is articulated, while Post (1983) says that none of the florets are jointed to the axis. Haussknecht (1894), according to Thellung (1911), noted the same variation and believed this species to be a variety of A. clauda, differing only in the man- ner of floret attachment. Mal’tsev treats A. pilosa as closely related to and derived from A. clauda by a side branch of the main phylogenetic trunk. Both species, according to him, have unequal glumes, a linear callus about 8 mm. long and a linear-elliptical articulation-sear. 3. A.longiglumis Dur. A wild oat found in southern Spain (Willkomm and Lange, 1870), Algeria (Battandier and Trabut, 1904), Palestine (Post, 19383) and Greece (DeHalacsy, 1904). Mal’tsev believes it to have given rise to A. ventricosa on the one hand, and to A. strigosa sens. ampl. on the other. 4. A. ventricosa Bal. A wild oat restricted to west- ern Algeria (Battandier and Trabut, 1904; Mal’tsev, 1929). Mal’tsev derived this species from the preceding by a side branch of the main phylogenetic trunk. It re- sembles 4. longiglumis in having equal glumes and a long callus with a narrow linear scar. It differs in having only the lower floret articulated, as in A. pilosa. 5. A.Bruhnsiana Grun. A wild oat restricted to the Apsheron Peninsulain eastern Transcaucasia, and treated by Mal’tsev as a subspecies of 4. ventricosa from which it differs by the larger size of its spikelet-parts. Although the two are now separated by some 2500 miles, we do not know what migrations they may have undergone in the past. The wide gap is partially spanned by A. longiglu- mis from which they may have evolved independently. L271. | 6. A. hirtula Lag. = A. strigosa Schreb. ssp. hirtula (Lag.) Malz. A wild oat which is restricted to the west- ern portion of the Iberian Peninsula (Mordvinkina, 1986). According to Thellung(1911)and Mal’tsev(1929), A. hirtula is intermediate in its characters between A. Wiestu and A. barbata. Mal’tsev derives the remaining diploid and tetraploid oats from this species. 7. A. strigosa Schreb.=A. strigosa Schreb. ssp. strigosa (Schreb.) Thell. The hairy or sand oat occurs as a weed throughout northwestern Kurope from Portugal to southern Finland (Vavilov, 1926; Werth, 1944). Thel- lung reports it as sometimes cultivated on sandy soils in Portugal and Spain, western and central France, Belgium, northwestern Germany and Great Britain. Vavilov says it is only met with as a crop in France and Great Britain. In England it is rare and local, being found only as a weed in grain fields (Clapham et al., 1952), while Stan- ton (1986) reports that an improved strain has been de- veloped in Wales. A. strigosa has two variants which are sometimes called species and sometimes proles (races). These are A. brevis and A. nudibrevis. A. brevis Roth=ssp. strigosa prol. brevis (Roth) Haus- skn. The short oat, like the typical stvigosa, is adapted to unfavorable soil conditions and is sometimes cultivated in Portugal, Spain, France and northwestern Germany (Thellung, 1911). It is distinguished by its blunt lemmas which end in two teeth rather than in two awns. A .nudibrevis Vav. =ssp. strigosa prol. nuda (L.) Haus- skn. = A. nuda L. var. biaristata Asch. & Gr. The small- grained naked oat is occasionally cultivated in northwest- ern Germany (Vavilov, 1926). 4. nudibrevis differs from the typical strigosa in having lemmas which do not be- come indurated at maturity, but remain membranous allowing the loosely enclosed grains to be readily threshed [ 272 ] free. It was once thought that all naked oats (the so- called species 4. nuda L..) came from China where hex- aploid naked oats are widely grown. The cytological work of Nikolaewa(1922)revealed that the small-grained naked oat was a diploid and undoubtedly European in origin. Later Mal’tsev showed A. nuda to be a highly diverse assemblage, containing, in addition to the present oat, four proles belonging to 4. sativa and one to A. byzantina. A. strigosa and its variants are the only diploid culti- vated oats and are geographically isolated from all other diploids except A. hirtula. Mal’tsev derived A. strigosa from A. hirtula, a judgment sounder than those of Thel- lung (1911, 1928) and Trabut (1914) who suggested the tetraploid 4. barbata had given rise to A. strigosa. De Candolle (1883) stated that 4. strigosa appeared to be a form of A. sativa, so subtle are the differences between the various cultivated oats. 8. A. Wiestii Steud. = A. strigosa Schreb. ssp. Wiest (Steud.) Thell. A desert plant which is widespread in North Africa (Trabut, 1914), Egypt (Tiickholm et al., 1941), Syria and Palestine (Post, 1933), Iraq and rare in eastern Transcaucasia (Mal’tsev, 1929). It is possible that both diploid and tetraploid oats exist which answer the description of 4. Wresta, but this will be considered later. Mal’tsev treats it as a tetraploid. Thellung (1911) and Trabut (1914) suggest that A. Wriestii is the wild species from which A. abyssinica has been derived, and in 1928 the former gave six transitional forms between the two. 9. A. Vaviloviana (Malz.) Mordv. = A. strigosa Schreb. ssp. Vaviloviana Malz. =? ssp. Wiestit (Steud. ) Thell. var. pseudo-abyssinica Thell. A wild oat restricted to Abyssinia and Eritrea (Mal’tsev, 1929). [ 273 ] A. Vaviloviana, which scarcely deserves the rank of species, appears to be a short-awned variety of 4. Wiestit. It is regarded by Mal’tsev as the wild oat from which A. abyssinica has been derived. 10. A.abyssinica Hochst. =A. strigosa Schreb. ssp. abyssinica (Hochst. ) Thell. = 4. sativa L. var. abyssinica Engler. This oat is restricted to Abyssinia, Eritrea and Yemen (Mal’tsev, 1929) where it is both cultivated and semi-wild (Vavilov, 1926). Harlan reports (Stanton and Dorsey, 1927) observing it rarely in Abyssinia and only as an admixture in barley fields. The natives informed him they had seeded it in the barley fields intentionally. A. abyssinica is the only cultivated tetraploid oat. It is described by Mal’tsev as having the awn- points at the tip of the lemma reduced to =1 mm. as in A. Vavilovi- ana. However, Stanton and Dorsey describe plants of this species as having the awn-points 2-4 mm. long. Evidently Mal’tsev was unaware of the range of varia- tion in A. abyssinica. 11. A.barbata Pott= A. strigosa Schreb. ssp. barbata (Pott) Thell. A wild oat occurring along the Atlantic coast of Europe from Brittany southward, throughout the Mediterranean region (Thellung, 1911) and extend- ing eastward through Transcaucasia and southern Turkis- tanto Afghanistan and the western Himalayas (Mal’tsev, 1929). A. barbata has been introduced into the United States, where it occurs as a common weed from Wash- ington and Oregon to Arizona and California (Hitch- cock, 1950). 12. A. sterilis L. sens. ampl. excl. form. cult. The wild red oat, a hexaploid complex distinguished by the lack of articulation of the upper florets, isan Old World species, ranging from the Atlantic to the Himalayas, and | 274 | may be divided into three subspecies: sterilis, trichophylla and Ludoviciana. Ssp. sterilis =ssp. macrocarpa (Moench) Brig. A stout plant with large grains, spikelets with 3-5 florets, glumes 30-50 mm. long, culm-nodes glabrous. It is strictly Mediterranean in distribution (Mal’tsev, 1929). Ssp. trichophylla (C. Koch) Malz. A more slender plant with medium-sized grains, spikelets with 2-8 florets, glumes 25-85 mm. long, culm-nodes pubescent. It shares the easternmost range of ssp. stevi/is in Palestine, Syria and western Asia Minor, but it extends eastward through Asia Minor to Transcaucasia, western Iran and Iraq (Mal’tsev, 1929). Mal’tsev regards this subspecies as intermediate between ssp. sterilis and ssp. Ludoviciana. Ssp. Ludoviciana (Dur.) G. & M., is a plant with small grains, spikelets with 2 (rarely 3) florets, glumes 25-830 mm. long, and glabrous culm-nodes. It ranges from southern England (Clapham et al., 1952), through France, Switzerland, Italy (Thellung, 1911), Spain (Willkomm and Lange, 1870) to North Africa (Battan- dier and Trabut, 1904). It is scattered throughout the Mediterranean region and becomes more abundant from Asia Minor through southern Russia, the Caucasus, southern Turkmenistan and northern Iran to Afghanistan and Turkistan (Mal’tsev, 1929). This subspecies, which has the general aspect of 4. fatwa, occurs largely in areas where the ranges of A. sterilis and A. fatua meet or overlap. Since the similarities between the two species were pointed out by Trabut (1914), it has been universally be- lieved that A. byzantina has been derived from A. steri- lis. Thellung (1928) has distinguished 15 transitional forms. Recently Coffman (1946) has proposed the theory that A. steriis is the progenitor of all other hexaploid oats. [ 275 ] 13. A. byzantina C. Koch, the cultivated red oat, is grown in southern Spain, southern Italy, Greece, North Africa and Asia Minor (Vavilov, 1926) and in the south- ern part of the United States, South America, Australia and New Zealand (Stanton, 1958). A. byzantina, distinguished from A. sterilis by its non- articulate spikelets, has three subspecies: byzantina, nodi- pubescens and pseudo-sativa. According to Mal’tsev (1929), these are derived from A. sterilis subspecies steri- lis, trichophylla and Ludoviciana respectively. Except for the non-articulate spikelets, the subspecies of A. byzan- tina resemble the subspecies of A. sterilis from which they are supposedly derived and may be determined accordingly. Ssp. byzantina, the one most commonly cultivated, has the rare naked prol. denudata (Hausskn.) Malz. Ssp. nodipubescens Malz. occurs from Asia Minor to Pal- estine, while ssp. pseudo-sativa Tihe//. has been re- ported from Switzerland and Uruguay. That A. sterilis has contributed to the formation of A. byzantina seems logical on the basis of geography and on the facts that both species are resistant to heat and cold and to many of the ordinary diseases of oats (Coff- man, 1946). None of these similarities extends to A. Jatua, the only other wild hexaploid oat. For the deri- vation of A. byzantina from A. sterilis it requires only that the spikelet-articulation of the latter cease to func- tion. Vavilov (1950) placed the center of origin of A. byzantina in the Near East. 14. A. fatua L. sens. ampl. excl. form. cult. is a wild oat with a vast range across Eurasia from the Atlantic to Mongolia. It has been widely introduced into North America and is a troublesome weed in the hard spring wheat areas of Minnesota, the Dakotas, Montana and [ 276 ] adjacent Canada (Stanton, 1936). Thellung (1911) re- ports it to be adventive in South Africa, South America, Australia and New Zealand. Tiickholm et al. (1941) re- port it from Egypt, Cyrenaica, Tunis, Algeria and Mo- rocco. It may be divided into four subspecies: meridion- alis, septentrionalis, fatua and cultiformis. Ssp. meridionalis Ma/z., with glumes to 30 mm. long, elongate-lanceolate lemmas to 25 mm. long and mostly glabrous culm-nodes, is the southernmost sub- species, ranging from the mountains of south central Asia through Afghanistan, Iran, Transcaucasia and Asia Minor to Egypt (Mal’tsev, 1929). Ssp. septentrionalis Ma/z., with glumes to 25 mm., lanceolate lemmas to 20 mm. long and culm-nodes always pubescent, extends from northern Russia and the Ural region through western and central Siberia to the Lake Baikal region and Mongolia, thence southwestward through the mountains of central Asia from the Altai to the Pamir Plateau, the Hindu Kush and the northwest Himalayas. Ssp. fatua, with glumes to 25 mm. long, lanceolate lemmas to 20 mm. long, grains 1.5—2 mm. thick, and culm-nodes always glabrous, is the western subspecies ranging from Turkistan and the Caucasus through south- ern Russia, Poland and the whole of western Europe. Ssp. cultiformis Ma/z., with glumes to 25 mm. long, ovate-lanceolate lemmas to 20 mm. long, grains 2-3 mm. thick, and with culm-nodes always glabrous, appears to have no well-defined geographical range. It is reported by Mal’tsev as being most common in North America, but occurring also in Europe eastward to central Russia. A. fatua is generally believed (Thellung, 1911; Tra- but, 1914; Mal’tsev, 1929) to be the progenitor of A. sativa and Thellung (1928) gave 21 transitional forms. It will be recalled that Thellung (1911) placed A. sterils [ 277 ] and its derivatives in subsection Biformes and A. fatua with its derivatives in subsection Conformes. Similarly Mal’tsev divided the hexaploid oats into two indepen- dent species: A. fatua sens. ampl. and A. sterilis sens. ampl. Coffman (1946), however, is convinced that all hexaploid oats are derived monophyletically from A. sterilis, and regards A. fatua as an aberrant weed. It must be remembered, however, that it is the cultivated oats, with their lack of floret-articulation, which are ‘‘aber- rant.”’ A. fatua, although a weed, is a perfectly normal self-propagating grass. 15. A.sativa L.,the common cultivated oat, is widely grown throughout northern Eurasia and in the northern United States and Canada. According to Stanton (1958), it is the most important cultivated oat. A. sativa may be divided into four subspecies: macran- tha, nodipilosa, sativa and praegravis, derived according to Mal’tsev (1929), from subspecies meridionalis, septen- trionalis, fatua and cultiformis of A. fatua respectively. Kxxcept for the non-articulate florets, the subspecies of A. sativa resemble the subspecies of 4. fatua from which they are supposedly derived and may thus be determined. Ssp. macrantha (Hack.) Ma/lz. occurs as a weed in grain fields over much of the same range as A. fatua ssp. meridionalis. It has the rare naked prol. nudata Malz. Ssp. nodipilosa Malz. is cultivated principally in northern Russia, the Ural region, Siberia and Mongolia. It has the naked prol. decorticata Malz. which is culti- vated in Mongolia and northern China. Ssp. Sativa is the most commonly cultivated oat. It has the naked prol. chinensis (Fisch. ) Malz. which is some- times grown in Europe and North America. Ssp. praegravis (Avrause) Malz. is cultivated prin- cipally in southern Russia, Europe and North America. [ 278 ] Its naked prol. grandiuscula Malz. is sometimes grown in Europe and North America. The so-called 4. orientalis Schreber is any form of A. sativa with a condensed, one-sided panicle, a not infre- quent variation. The similarity between the ranges of A. fatua and A. sativa, which extends to the subspecies, suggests a close relationship between the two, as does the parallel varia- tion of the subspecies. The morphology of the spikelets of the two species also suggests a relationship, for with 4. sativa, when the upper florets are separated by fracture of the rachilla, the rachilla segment remains attached to the lower floret exactly as with A. fatua, and different from A. sterilis and A. byzantina where the rachilla seg- ment remains attached to the upper floret. These similar- ities have previously been interpreted as indicating that A. sativa arose from A. fatua, but Coffman (1946) has suggested the reverse might be true. Vavilov(1950) placed the center of origin of A. sativa in the Near East region. Conclusions From the foregoing discussion it seems clear that the diploid species, with the notable exception of A. strigosa, have limited ranges in the Mediterranean region, a few also extending into the Near East. The tetraploids oc- cupy the same general area as the diploids, but are better represented in the Near East. A. Vaviloviana and A. abyssinica, however, are somewhat isolated to the south. The hexaploids A. stertls and A. byzantina are largely restricted to the Mediterranean and Near East regions, while A. fatua and A. sativa extend from those regions far to the north and northeast. It is also clear that phylogenetic studies have reached an unanimous conclusion on the derivation of the culti- vated species only in the case of 4. byzantina. The inter- relationships among the wild species are equally obscure. [ 279 ] Parr Il. THe EvipeENncrE rrom Cy roLocy Chromosome Numbers Following is a list of the chromosome numbers which have been reported for Avena species, together with authorities for the counts. The list of authorities is in- complete, but in cases of disputed counts it indicates which count has been verified by several authors. 14. me RR Re A, . clauda . pilosa . longiglumis . ventricosa . Bruhnsiana . hirtula . strigosa . Wiestit . Vaviloviana . abyssinica . barbata sterilis . byzantina . fatua . sativa 2n= 14 2n=— 14 2n= 14 2n= 14 2n= 14 2n= 14 In= 14 2n= 14 2n= 28 2n=28 2n= 28 2n= 28 2n= 32 2n=42 In= 44 2n=42 2n = 44 2n=— 42 Q2n=48 Qn=42 2n=48 Nikolaewa, 1922; Emme, 1930. Nikolaewa, 1922. Ellison, 1940. Emme, 1930. Emme, 1930. Emme, 1930; Ellison, 1940. Kihara, 1919; Nikolaewa, 1922; Winge, 1925; Aase and Powers, 1926; Hus- kins, 1927; Nishiyama, 1929; Emme, 1930; Spier, 1934; Ellison, 1940. Dorsey, 1925; Aase and Powers, 1926; Huskins, 1927; Kiharaand Nishiyama, 1932; Spier, 1934; Ellison, 1940. Emme, 1930. Emme, 1930. Stanton and Dorsey, 1927; Emme, 1930; Kihara and Nishiyama, 1932; Spier, 1934; Ellison, 1940. Kihara, 1919; Dorsey, 1925; Huskins, 1927; Nishiyama, 1929; Emme, 1930; Spier, 1934; Ellison, 1940. Nikolaewa, 1922. Kihara, 1919; Huskins, 1927; Nishi- yama, 1929; Emme, 1930; Spier, 1934. Nikolaewa, 1922. Kihara, 1919; Dorsey, 1925; Huskins, 1927; Nishiyama, 1929; Emme, 1930. Nikolaewa, 1922. Kihara, 1919; Huskins, 1927; Nishi- yama, 1929; Emme, 1930. Nikolaewa, 1922. Kihara, 1919; Winge, 1926; Aase and Powers, 1926; Huskins, 1927; Nishi- yama, 1929; Emme, 1930; Spier,1934. Nikolaewa, 1922. [ 280 | The counts obtained by Nikolaewa on the polyploid species differ from those of all other workers and proba- bly represent errors. The case of A. Wiesti is a different matter. Emme (1980), working in conjunction with the taxonomist Mal’tsev, found 4. Wiesti to be a tetraploid, but stated that an oat from Palestine, which Vavilov said was essen- tially 4. Wiestii, was a diploid. The oat found by Spier (1934), Huskins (1927) and Dorsey (1925) to be diploid is described by Stanton and Dorsey (1927) and is said to have been received from Egypt. It also appears to be essentially 4. Wiestii except for usually having 9-nerved glumes. A. Wiestii sensu Mal’tsev has glumes with 7(8) nerves. Similarly 4. barbata may have both diploid and tetraploid forms. Jones (1940) reports having received many years ago, as an impurity in seed of 4. Wiesta, an oat which was identified as 4.barbata until Ellison (1938) found it to be a diploid. This diploid oat goes under the designation of Cc1795. Clearly more work is required before the relationships between the diploid and tetra- ploid species can be understood. Genome Analysis Although the chromosome numbers indicate certain Avena species to be polyploids, every cytologist and ge- neticist who has studied the polyploid species has found them to behave like diploids with complete bivalent for- mation at meiosis and high fertility. In discussing how this situation came about, Philp (1983) suggested the tetraploids may be auto- or allopolyploids arising from one or two diploid species, while the hexaploids probably arose from a tetraploid and a diploid and so ultimately may have been derived from two or three diploid species. The study of chromosome pairing at meiosis in the F1 hybrids of interspecific Avena crosses has shed a certain [ 281 ] amount of light on the origins of the polyploid species, as well as the relationships between species having the same chromosome number. Diploid < Diploid—Genome analyses between diploid Avena species have been restricted to the series Hubar- batae. The F, of A. strigosa brevis XA. strigosa showed normal bivalent formation (Ellison, 1988). The chromo- somes of 4. Wiestiu are very similar to those of A. stri- gosa, as Nishiyama (1988) crossed these two species with a synthetic diploid and 7 normal bivalents were observed in the F; of both crosses. The chromosomes of A. hirtula show slight structural differentiation from those of A. strigosa, for in the F; of A lirtulaX A. strigosa brevis Ellison (1940) found com- plete pairing in most cells, but 1% of the cells examined showed 6 bivalents and 2 univalents while in many cells 1 bivalent was found to consist of heteromorphic homo- logues. Greater differentiation is shown by the oat €c1795 which has been crossed with A. Wiesti and A. strigosa brevis (Ellison, 1940), and in both eases 5 biva- lents and 1 quadrivalent were regularly observed. The above evidence suggests the chromosomes of A. strigosa, A. Wiestu and the synthetic diploid are very similar. Nishiyama (1986) has represented the genome of these species as AA. Although showing some differ- entiation, the genome constitution of A. hirtula and Ce1795 appears to be AA also. Tetraploid< Tetraploid— Ellison (1988) observed mei- otic behavior in the F; of A. barbataX A. abyssinica and found complete bivalent formation in most cells. How- ever, 159% of the cells showed 1 quadrivalent and cells having no quadrivalent sometimes had 2 or 4 univalents or aunivalent and atrivalent. Ellison believed this irreg- [ 282 ] ularity to be due to segmental interchange. Otherwise the genome constitution of 4.barbata, which was repre- sented as AA B’B’ by Nishiyama (1986), seems identical to that of A. abyssinica. Hewxaploid