BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY PRINTED AND PUBLISHED AT THE BOTANICAL MUSKUM CAMBRIDGE, MASSACHUSETTS BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VOLUME XIX BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS 1959-1962 r ‘4 ™ TABLE OF CONTENTS NuMBER 1 (August 18, 1959) The Phytomer in Relation to Floral Homologies in the American Maydeae By Watton C.GauinaT .......... = «21 NumBeEr 2 (January 27, 1960) Prehistoric Bean Remains from Caves in the Ocam- po Region of Tamaulipas, Mexico By L. Kaptan ann R.S. MacNeisH . .. . . 88 NuMBER 8 (May 19, 1960) On the Origin of the Orchidaceae By Lestige A.GaRay ........... 5% NuMBER 4 (June 30, 1960) Cytological Observations on Two Tropical Forms of Tripsacum By ¥,0. TING. 5 cow be KEE ee NuMBER 5 (June 30, 1960) Prestonia: An Amazon Narcotic or Not? By Ricuarp Evans ScHULTES AND Ropsert F. RAFFAUF ............ 2... . .109 A Reputedly Toxic Malouetia from the Amazon By RicHarp Evans ScHuLres. ... . . . . 128 NuMBER 6 (January 9, 1961) How Were the Glass Flowers Made? A Letter by Mary Lee Ware ...... . . 125 [v ] Numper 7 (February 17, 1961) The Hallucinogenic Fungi of Mexico: An Inquiry into the Origins of the Religious Idea among Primitive Peoples By R. Gorpon Wasson... ...... . . 187 Number 8 (April 11, 1961) Further Archaeological Evidence on the Effects of Teosinte Introgression in the Evolution of Mod- ern Maize By Watron C. GALINaT AND REYNOLD J. RurerpE ..... 2... ee ee ee ee 168 NuMBER 9 (June 80, 1961) Carludovica palmata in Broommaking By Metvin LEE Brisro. . . . . . . . . . . 188 Neonelsonia—A. Colombian Folk Medicine By Me.vin LEE Brisro.. ........ . 191 Novelties in the Orchid Flora of the Guayana Highlands By CHARLES SCHWEINFURTH . . 2. 2. «+ 195 Number 10 (May 7, 1962) Edible Fruits of Solanum in South American His- toric and Geographic References By Vicrorn ManuEt Patino. ...... . . 215 Edible Fruits of Solanum in Colombia By Ricuarp Evans ScHULTES AND RAFAEL Romero-CasTANEDA .... . . 2. e285 [ vi | INDEX OF ILLUSTRATIONS PLATE Apostasia nuda R. Br.ex Wallich . . . 2. . 2... XV Carludovica palmata Ruiz & Pavon . .XXV,XXVI Carludovica palmata. Fibres (text fig.), p. 184 Duckeella alticola C. Schweinf. . . . 2... XXVII Epidendrum radiatum Lindl. . . 2. 2. 2...) XIX Habenaria distans Griseb. 2. 2. 2 2... .) 6OXVITI Maydeae. Cross sections of axes. . ...... IV Maydeae. Cupule and rachis segment... ... III Maydeae. Vascular systems in pistillate rachis. . II Octomeria cordilabia C. Schweinf. . . . 2. . . XXX Octomeria dentifera C. Schweinf. . . . . . . XXX Octomeria filifolia C. Schwemf. . . ..... XXX Octomeria flaviflora C. Schweinf. . . 2... . XXX Octomeria lancipetala C. Schweinf. . . ... XXX Octomeria nana C. Schweinf. . . . . . 2... XXX Orchidacese, Crosses. . . 2 .-. «1 & @ % » « IV Orchidaceae. Kmbryogeny .......... =X Orchidaceae. Evolution . ......... XIII Orchidaceae. Floral Diagrams. . . .... . VIII Orchidaceae. Gynostemium. ......... XII Orchidaceae. Pollen grains (text fig.), p. 63 Orchidaceae; Seeds . 4.2444: 24%%4«2 XIE Orchidaceae. Types of Placentation . ..... IX [ vii ] Paphiopedilum Wilhelminiae L. Wms. . . 2. . XVI Phaseolus coccineus LZ. . . ........ . Vil Phaseolus vulgaris LZ. . 2. . 2. . 2... .)6VO,VIT Ponthieva ovatilabia C. Schweinf. . 2. 2. . 2. XXXI Rudolph Blaschka, facing p.1380 . . . . . . XXII Sobralia speciosa C. Schweinf... . . . . XXVITI Solanum alibile R. J. Schultes 2. 2... 2. . XXXII Solanum georgicum R. 17. Schu/tes b. &. So. 2 Oe. O.S. 4 hoe. O.@. G's Solanum liximitante PR. l’. Schultes XXXVI,XXXVII Solanum muricatum 7. XXXVITI,XX XIX, XL, XLI Solanum platyphyllum H.& B.. . . . . . . XLII Solanum quitoense Lam... ...... . XNULIII Solanum Topiro H.& B. XLIV,XLV,XLVI,XLVII Stelis latisepala C. Schweimf. . . . . . . . XXIX Stelis obovata C. Schweinf. . . . . . . . . SMXIX Tripsacum australe Cutler & Anderson . . XX,XXI Tripsacum laxum Nash . 2... 2... XX Vanilla planifolia dndr. . 2... . 2... XVII Zea Mays L. Archaeological maize from Cebollita Cave ............ XXIII Zea Mays L. Cross section of pulvinus. . . . . . V Zea Mays L. Manifestations of the phytomer. . . I Zea Mays L.. Teosinte introgression in Cebollita Cave maize. ............. . XXIV [ viii | INDEX TO GENERA AND SPECIES a-meé-md-ra, 254 ADACTYLUS, 62,70 AGAVE spp., 184 alkaloids, 110,111,116,123,153 AMANITA muscaria, 142,155 Amaranth, 36-38 AMARYLLIDACEAR, sens. lat., 60 APHYLLORCHIS, 77 ApocynacEAk, 109,111,117,118, 124 APOSTASIA, 61,62,64,70,72, 85 nuda FR. Br. ex Wall., 67,71, 89 odorata Bl., 86 papuana, 72,73,96 Apostasioideae, 62,64,65,70, 72,74-76, 78-80, 83,85,86,89 ARCTOSTAPHYLLOS pungens, 165 ARISTOLOCHIA, 64,112 arracacha, 191,221 ARRACACIA, 192,193 elata Wolff, 192 Pennellii Constance, 192,193 Wigginsii Constance, 192,193 xanthorhiza Bancroft, 191 asna-lulo, 216 ayahuasca, 109,110,112-116, 120 badoh, 152 badoh negro, 152 bamboo, 6,7 BANISTERIA, 117 Caapi, 110,116 BANISTERIOPSIS, 109,112, 113,115-118,120 Caapi, 110,113-116,119,124 ba-ra, 284 be-ben, 284 be-td-ka, 223,280 beans, 38,34,37-40,54,55,223 archaeological, 33-56 common, 35,36,38,40,42—45, 52=55 lima, 38,40,42,53 runner, 35,36,38,40,45,48— 51,55 sieva, 40,53,54 tepary, 54 bef, 249 bo-po, 223 BONATEA, 76 BURMANNIA longifolia, 85 BuRMANNIACEAR, 84,85 caapi, 109,112,113,115-117, 119,120 caapi-pinima, 116-119 cabuya, 184 cachon, 225,227 cachuma, 225,229 [ ix ] CALOPOGON pulchellus, 71 cana-antigua, 98 CAPSICUM, 35 CARLUDOVICA palmata Ruiz & Pav., 183-189 CATTLEYA, 83 CKPHALANTHERA alba, 66,67,73,96 rome gies 80 chaqui-lulo, 216 che-how-ke-noo-roo, 249 chicle, 124 chilt pepper, 35-38 chonta-ruro, 216 CLAVARIA truncata Quél., 159 CLAVICEPS spp., 153 CLEISTES, 77 CLEOME serrulata, 165 cobuia, 278 cocona, 224,225,241,280, 284, 286 COIX, 6 COLLETOTRICHUM Lindemuthianum, 43 CONOCY BE siligineoides Heim, 159 CONVOLVULUS indicus Mill., 152 rubrocoeruleus (Hook.) D. Dietr., 152 venustus Spreng., 152 violaceus Spreng., 152 CORDICEPS capitata (Holmsk.) Link., 159 [x | corn, 35-38,52,55 corn-grass, 5,6 CORYCIUM crispum, 73,96 cotton, 36-38 CRANICHIS crumenifera, 72,73,96 cubiu, 276 cubiyu 276 cuchara-caspt, 124 cucumber, 225-230 CUCUMIS sativus L., 225,229 CUCURBITA Pepo, 35,38,52 moschata, 36,37 mixta, 37 CURCULIGO, 85 CYPHOMANDRA betacea (Cav. ) Sendt., 235 Cypripedioideae, 62-—66,68,70, 74-76, 78-80, 82,83,85,86 CYPRIPEDIUM, 64,68,74,80 Calceolus L., 73,86,96 de-twa, 280 DENDROBIUM, 83 Diandrae, 78 DICTYOPHORA phalloidea Desv., 159 DIDYMOPLEXIELLA, 72 DISA unifolia, 82 DUCKEELLA alticola C.Schweinf., 195-198 e-td, 249 e-to-pa-a, 279 ELAPHOMYCES granulatus Fr., 159 variegatus Fr., 159 EPIDENDRUM, 83 ciliare L., 87 radiatum Lindl., 93 EPIPACTIS, 68 EPISTEPHIUM, 70,77 ERIAXIS, 65 ERYTHRODES, 72,75 GALEOLA, 70,72,77 altissima, 71 glass flowers, 125-136 GOODYERA, 68,75 pubescens, 71 gourd, 35-38 HABENARIA, 76 distans Griseb., 92 HAEMADYCTION amazonicum, 109,117 Harmoporates, 80 hallucinogenic fungi, 137-162 HEVEA, 123 ho-moo-mé, 284 huevo de tigre, 224 Hypoxrpacrar, 60,84 HYPOXIS, 60,61,85 ingo-sha-hush, 191,192 IPOMOEA Hookeri G.Don, 152 puncticulata Benth., 152 rubrocoerulea Hook., 152 tricolor Cav., 152,153 violacea L., 152,153 iraca, 183 Kerosphaeroideae, 61,63-65, 68,75,77-80,82-84,87,93 ko-bu-ya, 279 ké-mi-he-ro-ya, 249 le-t6, 284 le-t6-ma-ta, 254 LECANORCHIS, 65,77 javanica, 66,67 LAGENARIA siceraria, 35 LecuMInosak, 39 Liniacear, 60 LittaLes, 80 LIMODORUM, 72 abortivum, 66,67 LISTERA, 68 loo-poo-po-rd-la, 254 llullu, 216 llullu-ruru, 216 lulo, 216,217,220-223, 236,237, 240,249, 27 1,275,280,286 lulo de perro, 275 lulo morado, 275 LYCOPERSICON esculentum Mill., 235 ma-ra, 284 ma-ré-da, 284 ma-sha-kvé, 216,270 maize, 1—8,12,14,16,18,20,22, 23-30,37,97, 100-102, 106, 221,223 archaeological, 163-181 (See corn) MALOUETIA, 123 furfuracea Spruce, 124 nitida Spruce, 124 peruviana Woodson, 124 Tamaquarina, 118,123,124 MavpiGuHiaAcrar, 111 [ xi ] mancadera, 277 MANIHOT, 49,118 dulcis, 38 MASCAGNIA, 112 Maydeae, 1,14,16,26 melon pear, 257 MELONGENA laurifolia, 256,258,259 Microspermae, 68 Monandrae, 78 mushrooms, 137-162 narangitas de Quito, 221 naranja, 219 naranjl, 222 naranjilla 216-224,241,270,271 274 narcotics, 109-122,137-162 NEONELSONIA, 191-193 acuminata ( Benth.) Coult. & Rose, 192-194 ovata Coult. & Rose, 192 NEOTTIA, 68 Neottioideae, 61,63-66,68,70, 72,74,75,77,79-86,91 NEUWIEDIA, 61-63,72,78, 84,85 NEVROPHYLLUM floecosum (Schw.) Heim, 159 NICOTIANA, 38 Tabacum L., 235 *nti sitho, 146 OCTOMERIA cordilabia C. Schweinf, 204, 208,209,210 dentifera C. Schweinf., 205, 208,209 filifolia C. Schweinf. , 206,208, 209 flaviflora C. Schweinf: , 207- 209 lancipetala C. Schweinf. , 208- 210 nana C. Schweinf., 208,209, 21) parvula C, Schweinf., 205 ODONTOGLOSSUM, 88 ololiuqui, 151,152 Ophrydoideae, 61,63-65,68, 74-84,87,92 OPHRYS Nidus-avis L., 87 opuntia, 35,36 OrcuipackaAk, 57-96,195-214 ORCHIS Morio L., 87 ORNITHOGALUM, 60 palm, 223 PANAEOLUS campanulatus var. sphinctrinus (Fr.) Bres, 160 fimicola (Fr.) Quél., 160 sphinctrinus (Fr.) Quél., 160 PANICUM, 36,38 PAPHIOPEDILUM, 68,82,83 Wilhelminiae L. Wms., 90 pee-pee-ka, 249 pepino, 225-229, 255-257 pepino dulce, 225 pepino morado, 255 pepino redondo, 255 pepo, 2595 PHARBITIS rubrocoeruleus (Hook. ) Planch., 152 violacea (L.) Bojer, 152 [ xii | PHASEOLUS, 40 acutifolius var. latifolius Free- man, 54 coccineus, 40,42,44-52,54, 55 lunatus, 40,42,44,46,54,55 polyanthus Greenm., 51,52 spp., 53 vulgaris, 35,39-42,44,46,47, 52-54 PHRAGMIPEDIUM, 65 longifolium, 66,67 plantain, 221 po-ro-la, 284 POGONIA, 77 pomas de perro, 277 PONTHIEVA diptera Linden & Reichbf., 214 ecuadorensis Schltr., 214 ovatilabia C. Schweinf., 211- 214 poom-ka, 284 PRESTONIA amazonica, 109-111,114—-120 PROTORCHIS monorchis Massal, 57 PSATHYRELLA sepulchralis Sing., Smith & Guzm., 160 PSILOCHILUS, 72 PSILOCYBE acutissima Heim, 160 aztecorum Heim, 160 caerulescens Murr. var. mazatecorum Heim, 160 fma. heliophila Hem, 160 fma.ombrophila Heim, 161 var. nigripes Heim, 161 caerulipes (Peck.) Sace. var. Gastonii Sing. & Smith, 161 candidipes Sing. & Smith, 161 cordispora Heim, 161 cubensis (Earle) Singer, 162 fagicola Heim & Caill., 161 Hoogshagenii Heim, 161 isauri Singer, 161 mazatecorum Heim, 160 mexicana Heim, 161 mixaeensis Heim, 162 muliercula Sing. & Smith, 159,162 semperviva Heim & Caill., 162 Wassonii Heim, 159,162 yungensis Sing. & Smith, 162 zapotecorum Heim, 162 var. elongata Heim, 162 pumpkin, 35-37 puscolulo, 216,218,220 puscolulu, 216 RICINUS communis L., 218 RIVEA corymbosa (L.) Hallier f., 152,153 rd-ya, 284 Rudolph Blaschka, 125-136 RuTAaceak, 111 SATYRIUM, 60,61,74,78 saxicolum, 73,96 SELENIPEDIUM, 61,62,65, 70 Chica, 67,71 SOBRALIA speciosa C. Schweinf., 198, 200,201 [ xiii ] Weberbaueriana Ariinzl., 199 STROPHARIA Solanum, 235,236,267 cubensis Karle, 162 alibile R. E. Schult., 236-240 sunflower, 37-38 angulatum Ruiz & Pav., 267 tapaculo, 277 angulosum, 221 sconlaniiien. eee teo-nanacatl, 146 galeatum, 221 teosinte, 1-3,10,11,14,16,18, georgicum R. kh. Schult., 220, 24-29,37,38,163-178 240-247 TERELETRA hyporhodium 4. br. & Bouché, violacea (L.) Raf., 152 284,286 TETRAPTERYS, 112,119 methystica, 119 topiro, 223,224 liximitante R. EF. Schult. , 248- 252,276,279 muricatum Ait, , 225,227,230, 254-266 topiru, 280 var. popayanum Bitt., 225 TRIPSACUM, 1,3,14,16,18, peruvianum, 221 24—27,30,97,99 platyphyllum H. & B., 266- australe Cutler & Anderson, 98 269,276 99,101-106 quitense, HBK., 267 dactyloides L., 2,14,16,18,97 quitoense Lam., 215,216,220, floridanum, 100 221,225,240, 241,267,270 latifolium, 98 at) laxum Nash, 97-103,106 var. septentrionale Schull. pilosum, 98 om Cuair., 217,221,222, spp., 163 274,285 tumo, 275 sessiliforum Dunal, 275,276 sisymbrifolium Lam., 276 straminifolium Jacq., 249, tupiru, 223 276,278,279 uchuba colorado, 277 Topiro A. & B.,220,222-224, 237,240,241, 276, 279-286 tuberosum L., 235 variegatum Ruiz & Pav., 229, 256,260,261 tupiro, 223,224 uva de perro, 277 VANILLA, 68,70,77,81 anomala, 72,73,96 Griffithii var. formosana, 72, 73,96 SPIRANTHES, 68,72,75 planifolia Andr,, 71,91 squash, 36,37,55 VANILLACEAE, 70 latisepala C. Schweinf, 199, ZEA, 97 203 Mays L., 1,163 lentiginosa Lindl., 202 mexicana Reeves 5 Mang., 1, obovata C. Schweinf. , 202,203 163 parvifolia Garay, 202 ZEUXINE, 64 [ xiv | ERRATA Page 2, line 10 for was read were Page 85, line 32 insert a after of Page 97, line 25 for mitotic read meiotic Page 124, lines 4—5 for Dogbone read Dogbane Page 124, line 19 replace period following poison with comma Issued November 16, 1962 [ xvi | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Campringe, Massacnusetts, Aucusr 18, 1959 Voi. 19, No. 1 THE PHYTOMER IN RELATION TO FLORAL HOMOLOGIES IN THE AMERICAN MAYDEAE BY Watton C. GALINAT THE structure of the inflorescences in the American Maydeae (maize or Zea Mays .., teosinte or Zea mexi- cana (Schrad.) Reeves and Mang.', and T'ripsacum spp.) may be the result of developmental modifications to a repetitious pattern of organs, the ‘‘phytomer,’’ which is basic throughout the entire plant. The parts of this pat- tern, as described later, have been recognized in vegeta- tive form as an internode, a leaf, and an axillary bud (Gray, 1879 and others), and, recently (Galinat, 1956), as another organ, the prophyll. If the phytomer and its components have a floral man- ifestation, then their basic homologies might be revealed by anatomical comparisons within any plant and with close relatives. Such comparisons have been successful in demonstrating the evolutionary development and homologies of certain floral structures such as carpels, compound ovaries and inferior ovaries. We have already had some such studies in maize. The arrangement of large and small bundles has been de- 'This new name was proposed in 1942 and has been used regu- larly by its authors in a recent series in these Leaflets (Vol. 18, Nos. 7, 8, 9 and 10). [1] scribed in the ear (pistillate rachis) and in the tassel (staminate rachis) by Reeves (1946, 1949) and Lauben- gayer (1948, 1949); in the tassel alone by Kumazawa (1939); and in the culm (stalk) by Esau (1943) and others. Certain bundles in the ear have been suggested as representing those of a lateral organ (the prophyll) which is fused to the main axis or rachis (Nickerson, 1954). These previous studies are largely descriptive, interpretations being difficult because the tissues and vascularization in maize was not compared with those of its close relatives; nor was the anatomy of the rachis con- sidered as a possible reflection of the structure of the culm. The present study attempts to examine the evo- lution and development of the maize plant in terms of modifications, according to function, of the organs in a single basic pattern, the phytomer. In addition to typical maize (sweet corn inbred Purdue 39) and its relatives, teosinte (race Durango) and T'rip- sacum dactyloides, two special maize-ty pes were included in this study. One of these is a derivative from a maize- teosinte hybrid specifically bred for a simplified version of the vascular system which could be represented in three dimensions. ‘This breeding was done by selecting a slender, four-ranked ear with greatly accentuated cu- pules’ which were free from the usual crowding and distortion. Reduced condensation to remove vertical compression between the cupules was derived from Gua- rany maize. Enlargement of the lateral wings of the cu- pules, a character associated with spikelets oriented in the same plane as that of the rachis (Galinat, 1956), was introduced from teosinte. Pairing of the spikelets, a characteristic of maize, further accentuated the cupules by spreading out the lateral wings. The other special 2 Corneous alveoli of the maize cob immediately above the attach- ment point of each pair of pistillate spikelets (Sturtevant, 1899). [ 2 | maize type was a heterozygous tunicate (7'w/tu) strain of Argentine popcorn. The material was prepared for staining by two tech- niques. One of these is asomewhat unusual method sim- ilar to that suggested by Cutler and Cutler (1948) as follows: Intact spikes (ears) were stained at the time of style emergence by placing their freshly cut bases in an aque- ous solution of safranin. In a few seconds this red stain had traveled up through the vascular system and had colored all of the xylem elements. Attempts to preserve this stained material by fixing it in (3:1) absolute alcohol: glacial acetic acid and then clearing it in a cedar wood oil series were unsuccessful because the color always diffused out from the bundles and into the adjacent, highly lig- nified rind; but, in less lignified material, such as that of typical maize, this clearing technique was successful in revealing the vascularization of specimens stained in- tact. It was later discovered that, if our stained speci- mens were immediately dried by warm, circulating air, the red color then remained in the bundles. Free-hand, three-dimensional drawings were then made from studies of the vascular system, as revealed on the exterior of these dried and intact specimens and as reconstructed from cross-sections of the same material. The other method used is a classical one. The material was fixed eighteen days after pollination, then dehydrated in an ethyl-alcohol series and embedded in paraffin for cross-sectioning and eventual staining. Although no diffi- culty was encountered with microtome sectioning in the case of the maize, the teosinte and T'ripsacum specimens were too highly lignified at this age for easy cutting. However, a few excellent free-hand sections of only about one cell-width in thickness were obtained from the em- bedded material of these relatives of maize. All sections [ 3 | were stained by the safranin-fast-green technique. A projection apparatus was used in making tracings from comparable slides. Pulvini swellings from the axils of tassel branches of P39 maize were also sectioned, stained and projected in a manner similar to that used for the ears, because these small axillary protuberances appeared to represent another possible homologue of the prophyll. They were at maximum swelling when collected at the time of an- thesis. THe NATURE OF THE PHYTOMER Continuity of the phytomers. The phytomer, like the cell, was once considered to be the ‘‘true individual.”’ But now the plant as a whole is usually recognized as the individual, and the term ‘‘phytomer’’ is used to describe the level of organization represented by one repetition of its specialized regions or organs. The boundaries of the phytomer, and of the organs which compose it, are only approximate. Neither vascularization nor disarticulation delimit a discrete phytomer (Arber, 1984). Also such a unit is not necessarily delimited by the order of matura- tion, as in the classical segmentation of the phytomer used by Evans and Grover (1940) and others, because the degree and order of development of its various organs differ during vegetative and floral growth. In order to simplify comparison of its various mani- festations, we have chosen a phytomeric cycle comprising the group of organs which are adjacent to a given node or apparent node, as in the inflorescence where the nodes are usually obscure. This combination includes the leaf borne just below the node and its axillary bud with as- sociated prophyll just above the node, as well as the adja- cent internode (Plate I). The more classical delimitation of the phytomer at the nodes includes a leaf and bud [ +] which are isolated at opposite sides and opposite ends of an internode. Our grouping is more convenient for floral comparisons, especially when the internodes are tele- scoped: the axillary buds are either closely associated with or fused to their subtending leaf or leaf rudiment, and the lateral organs are whorled, as in the maize ear. Repetition of the phytomer. Control over the number of repetitions of the phytomer, as well as their individual manifestation, usually seems to follow a functional pat- tern which is characteristic for a certain portion of the plant. In the lower parts of the plant, all organs of the phytomer are large, photosynthetic structures (Plate I, A), while in the highly compacted inflorescences, their counterparts may be reduced or entirely obliterated (Plate I, B through F). The typical course of repetition by the phytomer in a given area of the maize plant may be changed by un- usual genetic and/or environmental conditions. In short- day maize, as in other photoperiodic plants, the number of repetitions by vegetative-type phytomers and the time of change to a floral-type of manifestation is controlled by length of day. Also the production of vegetative phy- tomers by axillary buds, as well as the abruptness of their ultimate shift to a floral manifestation, seem to be con- trolled by the corn-grass (Cg gene) locus. At least four other genes contro] the production of phytomers at spe- cific points in the inflorescence, as follows: The primary branches (rachids), ramosa 1, 2 (ra1, ae) on chromosomes 7 and 8; The spikelets, branched-silkless (bd) on chromosome ic: The florets, polytypic (Pt) on chromosome 6. Gross STRUCTURE AND MANIFESTATION OF THE PHYTOMER Evidence of homologous relationships based on gross [ 5 ] structure must rely largely on a study of developmental and evolutionary variations in the manifestation of the phytomers. Discussion of such variations will orient the floral expression of the phytomer and, thereby, aid in identifying the vestiges of certain reduced parts. Leaves. Although the leaf of the vegetative phytomer is enlarged for maximum photosynthetic activity, the floral homologue is reduced and modified according to the protective device characteristic of the species, as well as according to the order of the axis on which it is borne. At the base of the maize tassel as a whole, or sometimes at the base of each tassel branch, the subtending leaf is usually reduced (Plate I, D-1), although it may undergo all degrees of development (Galinat, 19542). In the ‘‘cen- tral spike’’ of the tassel or rachis and corresponding axis of the ear, the leaf initials are usually inhibited except for a possible rudimentary leaf, the ‘‘glume cushion,”’ at the base of the glumes. But this leaf may be well- developed in certain bamboos (Holttum, 1956), in Cow (where it has a protective role) and in the corn grass and teopod mutants of maize (Galinat, 1956). On the spike- let axes or rachillas, the blade-parts (laminas) of the first two leaves (glumes) are rudimentary, but in the case of the third and fourth leaves (lemmas), single genes may cause the blades to develop as awns in the ‘‘bearded”’ varieties of small grains, or the blades may be stimulated to complete development in proliferated spikelets. Awvillary buds and internodes. The axillary buds rep- resent the starting points for the internodes of new axes of lesser orders. Certain variations in their derivatives (tillers, ear-shoots, tassel-branches, spikelets, florets) demonstrate the homology of the buds concerned and of the internodes of their ultimate axes. This is especially apparent in the various intergrading branches of the mu- tant ‘‘corn grass,’’ which is characterized by a gradual [ 6 ] transition from a vegetative shoot to a reproductive one rather than the usual abrupt change (Galinat, 1954b). Spikelets may be converted into ‘‘tassel-plantlets’’ as an “‘after-effect,’’ resulting from an insufficient number of short-days during the early floral development of short- day maize (Galinat and Naylor, 1951). The growing point of the spikelet-axis or rachilla may shift from ‘‘cutting-off’’ floret primordia to that of initiating spike- lets as this axis becomes the rachis of an ear enclosed by husks modified from glumes and lemmas (Weatherwax, 1925). Finally, during the evolution of the maize ear, either a tassel branch or a spikelet from the tassel seems to have been modified as a tiny, sub-tassel ear which later descended to a more efficient position on the stalk, where it could increase in size (Mangelsdorf, 1958). Prophylls. The prophyll-part of the phytomer is a two-keeled, leaf-like organ which develops at or near the axil of a lateral bud. Its two-keeled form may result from its being pressed between the branch axis and parent axis during early development (Arber, 1934). Pressure between binding leaf-sheaths and their expanding axil- lary buds and associated prophylls is known to be respon- sible for the initiation of permanent grooves in the inter- node of the parent axis, and, in some bamboos, this channel retains the imprint of the prophyll, even after it has been left behind by the elongation of the internode (Arber, 1934). It is apparent that the prophyll occupies the most crowded position in the phytomer, especially along the rachis, where it is either absent or highly modi- fied and reduced. But when the position of the floral prophyll is moved away from the rachis to a less crowded position in the ultimate branches (florets), it then devel- ops fully as the so-called ‘‘palea.’’ The problem then is to identify the anatomical remains of the prophyll at or near the axil of a branch within the Le phytomers of the rachis. ''wo independent theories have been proposed for the role of the prophyll in the devel- opment of the ear and tassel of maize. Nickerson (1954) suggested that the cupule in the ear was formed by a prophyll depressed into and adnate to the rachis, except for the auricles which produce laterally as ‘‘rachis-flaps. *’ In the tassel of maize, which lacks cupules, as well as in the paniculate rachises of other grasses, the primordial prophyll may have been contained as an axillary swelling (the pulvinus) which has become specialized to function in spreading the primary branches at the time of anthe- sis (Galinat, 1956). Arber (1934) has noted that the in- florescence branches of many grasses have such axillary pulvini, which expand at the time of anthesis so as to force the branches outwards; and, after anthesis, the swellings usually wilt as the branches again rise. INTERNAL ANATOMY AND MANIFESTATION OF THE PHY'TOMER Although the homologies of certain reduced and mod- ified organs in the floral phytomer may not be apparent externally, their basic nature may lie hidden in some part of the internal anatomy, such as that of the vascu- lar system. An anatomical study, therefore, may help to establish the anatomical remains of the prophyll at certain of its potential positions which are occupied by other excrescences, such as the cupule in the ear and the pul- vinus in the tassel. Vascularization in maize. The homology of the floral and vegetative internodes is reflected by a close similar- ity in their vascular systems. Certain modifications in vascularization of the ear are caused by the reduction and compaction of lateral organs. Since vascularization in the tassel is so similar to that of the culm (Kumazawa, 1939), it will be excluded, except in regard to the pul- [8] vinus, from these comparative studies. The bundles of both axes tend to be of two distinct diameters which are separated into two locations (a ‘‘meristele’’ arrange- ment). Those bundles with the smallest diameters lie adjacent to the rind or lignified periphery of the axis which, in the case of the (pistillate) rachis, is repeatedly parted into the wings of numerous cupules (Plate II, fig. 1). The bundles of the large diameter are scattered throughout the pith of the culm, but, in the rachis, they are usually concentrated near the margins of the pith, where they supply the longitudinal rows of traces to the lateral spikelets, even though a few ‘‘cauline’’ bundles may be isolated in the center of the pith. The diameter and position of an individual bundle is different in various parts of the plant. The larger trace- bundles, which extend horizontally from a leaf, curve downward from the leaf-node and then extend through about six internodes as they decrease in thickness and slope outward before connecting to the peripheral bun- dles. In the smaller trace-bundles, on the other hand, such connections occur progressively earlier in the de- scent, the smallest bundles remaining free for only one internode or less. As these leaf-traces descend, they sup- ply the axillary buds along the way by means of lateral connections to a network of horizontal bud-traces slightly above each leaf node. The glume cushions, which seem to be rudimentary leaves of the rachis, are vascularized by small bundles descending to the rind-bundles in the cupule wings be- low. Inasmuch as the apical end of these bundles con- nects with the vascular supply to the outer glumes of its axillary spikelets (Plate II, fig. 1) rather than terminat- ing as stubs, they would appear at first to be ‘‘rind- bundles”’; on the other hand, they could be rudimentary leaf-traces which have become folded inward and fused [9] to the glume supply. In any case, the actual elaboration of the glume cushion into a well-developed leaf under certain conditions mentioned previously seems to leave little doubt about the homology of this rudiment. The suppression of the primary leaves along the rachis is associated with a loss of the nodal plates and differ- ences in trace connections from the axillary buds. ‘These traces from the binate spikelets of the rachis fan out to the nearest group of ‘‘common’’ bundles (Plate II, fig. 1) rather than connecting through a vascular network extending to the entire meristele, as with the axillary buds along the culm. Nature of the bundles in the cupule wings. The evi- dence from vascular anatomy does not support the sug- gestion of Nickerson (1954) that the bundles in the cu- pule wings are those of a prophyll adnate to the rachis. Although the bundles located near the lateral edges of the out-folded wings of a typical cupule do have a xylem- phloem orientation opposite to that of the larger bundles within the rachis, if one follows inward along the series of bundles in such wings, the orientation of each bundle is found to twist gradually, so that the innermost ones have the same orientation as the larger bundles (Plate III, fig. 1). This twisting of bundles suggests that the cupule wings are formed in part by a gradual folding out of flaps dislocated from the rind of the rachis. Further evidence in support of this view comes from the two- ranked spikes of our teosinte-derivative of Guarany maize. ‘The wide spacing of the cupules in this stock re- veals that the rind from the barren rachis has exactly the same vascular pattern as that in the cupule wings, and that the tissue at the back of the cupule is devoid of these small bundles (Plate IV, fig. 2). It seems, there- fore, that the portion of the cupule wings which includes the vascularization is derived from the rind of the rachis. [ 10 ] The cupule lining in relation to the pulvinus. Evidence that the wings of the cupule consist of more than just a flap of the rind comes from the experimental in-folding of these structures. When these wings are bent over into the cavity, as can be seen in the diagram of the teosinte derivative (Plate IV, fig. 2), the reconstituted axis re- sembles more closely the structure of the pulvinus and associated axis (fig. 1) than it does that of the culm (fig. 3). The resemblance of this wing-filled cupule to the structure of the pulvinus is revealed by close similarities in size, position and numbers of cells in the areas con- cerned of a sweet-corn inbred (Purdue 39). Although the total number of cells extending from the large (common) bundles outward through the center of the cupule com- bined with those through a folded-in wing exceeds by about thirty cells the growth which occurs between the large bundles and the epidermis of the barren rachis be- tween the cupules (Plate III, fig. 1), it nonetheless cor- responds almost exactly in number and size of cells with those which occupy the corresponding position through the pulvinus (Plate V). Evidence of a relationship between the pulvinus and cupule lining may also be shown by a hypothetical manipulation of the pulvinus into acupule. Starting with the pulvinus (Plate IV, fig. 1), if one visualizes a central split perpendicular to the epidermis extending inward to a point just beyond the small bundles and then diverging in both directions along a line parallel to the epidermis for the width of the pulvinus, then the flaps therein dis- sected will resemble in-folded wings of a cupule. By folding these wings out laterally and away from the cavity, one may produce the structure of a cupule (such as in fig. 2)in which all of the small bundles are removed to lateral wings and in which there is a layer of small cells (represented by cross-hatching) exposed over the [11 ] EXPLANATION OF THE ILLUSTRATION Piate I. Manifestation of the phytomer in different parts of the maize plant. | | i is) A internode a B_ rachillasegment e ky C rachillasegment 2 & D_ rachis segment 7-2 KE rachid* segment iz eI F rachilla segment l p ab leaf prophyll axillary bud (ear) glume cushion cupule-lining binate spikelets lemma palea pistil + leaf rudiment — pulvinus tassel-branch glume cushion rachid* scab _ paired spikelets lemma palea 3 anthers + * axis of a tassel branch Drawn by Watron C, GALINAT PLATE I EXPLANATION OF THE ILLUSTRATION Piare Il. The vascular systems in the pistillate rachis of the American Maydeae in three-dimensional aspect. For simplicity, only that vascularization which is in the outer glume of the spikelet and in the rachis tissue ad- jacent to the observer is shown. Individual lines rep- resent individual bundles except in the main bundles which are thicker and shaded when overlapped by other tissue, such as that of glumes and rind. 1. Vascularization of the maize-type rachis as repre- resented in simplified form by a derivative from a maize- teosinte hybrid. A slight degree of twisting has been incorporated into the drawing in order to illustrate the vascular arrangement from several angles. 2, 3. Separate front and side views of the fruit case of Durango teosinte. 4, 5. Separate front and side views of the fruit case of Tripsacum dactyloides. All about five times natural size. Drawn by Warton C, GaLinatT [ 14 ] EXPLANATION OF THE ILLUSTRATION Priare III. Projection tracings of cross-sectional views of the cupule and rachis segment in the American Maydeae. 1. The cupule and associated rachis tissue of maize as represented by the sweet corn inbred Purdue 39. 2. The rachis segment of Nobogame teosinte. 3. The rachis segment of T'ripsacum dactyloides. All about twenty five times natural size. Drawn by Wavron C, GALINAT [ 16 ] PLATE III EXPLANATION OF THE ILLUSTRATION Piate IV. Diagrammatic cross-sectional representa- tions of various axes to show structure, vasculariza- tion, pubescence, and distribution of an extremely small type of cell. The last is indicated by cross- hatching, while the phloem-region of the vascular bundles is indicated by solid black. 1, pulvinus and associated axis from a tassel branch of P39 sweet corn. 2, rachis with cupule from the distichous branch of a teosinte derivative of Guarany maize. 3, the shank associated with the spike of fig. 2. 4, the rachis of a 12-rowed ear of tunicate Argentine popcorn showing three cupules. 5, pistillate rachis of Nobogame teo- sinte. 6, pistillate rachis of Tripsacum dactyloides. All about fifteen times natural size. Drawn by Wariron C, GALINAT [ 18 | PuLatTEeE IV EXPLANATION OF THE ILLUSTRATION Prats V. Projection tracing of cross-sectional view of the pulvinus and associated branch from the tas- sel of Purdue 39 sweet corn. Note that the tissues of this mature pulvinus suggest the primordium of an organ (such as the prophyll) in that the swelling results from a proliferation of many small cells,rather than an expansion in individual cell size. About forty five times natural size. Drawn by Warton C, GALinaT [ 20 ] PLatre V surface or lining of the cavity. Such a process of cupule formation, by a hypothetical splitting of the pulvinus, serves to illustrate the similarities of the structures in- volved. It does not represent a plausible explanation for the origin of the cupule in terms of ontogeny, as will be noted later. This relationship between pulvinus and cupule lining is also revealed by intergrades between these structures in tunicate (Zw gene) and other variants of maize. The tunicate cupules are usually shallow because of small wings and, in some cases, the ‘‘cupule’’ may be elevated above the adjacent rachis in such a way as to appear like a flattened pulvinus. One such type of tunicate cupule from a twelve-rowed ear of Argentine popcorn is illus- trated in Plate LV, fig. 4. The hairiness and distribution of small cells in this type of cupule produce a striking resemblance to that of the pulvinus (Plate IV, fig. 1). This condition differs from the typical cupule (Plate ITI, fig. 1) in having all of the rind bundles crowded into cor- ners between the cupules rather than dislocated into flaps. When tunicate cupules lack such vascularized wings, they may be ‘‘peeled’’ from the rachis. Other circumstantial differences may appear during the development of the pulvinus and cupule lining. Fer- tilization is accompanied by a metaxenial stimulation for the deposition of lignin in the small cells of the cupule lining, while the corresponding cells of the pulvinus eventually shrink during aging of the plant. Further, the cupule or its lining is embedded into the main axis or rachis, while the axillary pulvinus usually expands along the axis of the primary branch. But the primary branches of the ear are reduced to binate spikelets which, in certain ears of tunicate maize, may be associated with pulvinus-like swellings rather than cupules. Finally, the pulvini in certain highly condensed and compressed tas- [ 22] sels, found in inbred P89, are partly formed by tissues from the main axis. Such modifications in structure are imposed by the different conditions of development in tassel and ear and do not detract from the important anatomical evidence of a close similarity between the lining of the cupule and the pulvinus. The cupule lining and pulvinus as homologues of the prophyll. The pulvinus resembles a rudimentary prophy]I in phytomeric position, in external appearance and in internal structure. The cell structure of the pulvinus suggests the primordium of an organ, such as the pro- phyll, in that the swelling results from a proliferation of many small cells rather than from an expansion in size of individual cells. It is rudimentary in development in being delimited by a zone of rapidly changing cells rather than by an abrupt boundary of cells such as might occur between elaborated organs which are fused (Plate V). Inasmuch as the anatomy and phytomeric position of the pulvinus is also similar to that of the cupule lining, as discussed previously, we conclude that these formations are different manifestations of the rudimentary prophyll- part of the phytomer. Therefore, it is necessary to mod- ify our previous conception of the cupule lining as an adnate prophyll (Nickerson, 1954; Galinat, 1956) to the extent that we now believe that the cupule lining is only one of several possible manifestations of the rudimentary prophyll-part of the phytomer. Under other develop- mental conditions elsewhere in the plant, the tissue from this region has ultimately developed in the form of a prophyll, a pulvinus or a palea. The role of pressure in cupule formation. The effect of pressure from constricting leaf sheaths upon floral devel- opment and floral evolution in the grasses has been rec- ognized (Arber, 1934, and others). In maize, such pres- sure moulds expansion of a plastic inflorescence from the [ 23 | time the plant is only a few weeks old and it continues throughout the development and maturation of the ear. Its role in cupule formation in maize and its relatives seems obvious. Here, in the position of maximum com- pression between two axes, the central portion of the rudimentary prophyll and associated rachis tissue seems to be depressed inward by penetration of the expanding spikelets. The resulting stresses have apparently inhib- ited development of the small rind bundles, while the larger bundles deeper within the rachis buckle under stress and bend inward in conformity to the depth of the depression (Plate II, fig. 1). Meanwhile, the lateral por- tions of the prophyllar tissue and associated rachis, which are free from pressure, bulge out as wings on either side of the penetrating spikelets. At this point in maize, lack of elongation by the internode or condensation forces the elongating spikelets to bend out and away from the cu- pule, finally diverging at right angles from it. This con- dition is in sharp contrast to that of teosinte and T'rip- sacum, as will be discussed later, where a slight elongation of the internode, as well as the rigid, sessile and solitary condition of the spikelets, leave no alternative for the spikelets but to become embedded more deeply into the rachis segment as a result of pressure from constriction. Teosinte and Tripsacum. The homology of the cupule lining to the prophyll is more obscure in teosinte and Tripsacum than it is in maize, because of reductions re- sulting from an extreme depression of the rachis segments in these relatives of maize. The lining of the hollowed rachis segments consists of very small lignified cells sim- ilar to those which line the cupule of maize, except that they are spread out more thinly over the surface of the cavity. As with the maize cupule, these cells of the lin- ing are smaller and more highly lignified than those of the rind. In the relatives of maize, a reduction in the [ 24] thickness of this lining is probably an effect of increased compression (Plate III, figs. 1, 2, 3). The extreme depression of the rachis segments affects their vascularization as well. In T'ripsacum, the larger bundles adjacent to the cupule lining are twisted into a more space-conserving position so that their longest axis lies parallel to the surface rather than in the usual per- pendicular orientation, as exemplified by both large and small bundles from the opposite or convex side of the segment. In teosinte, where the effects of compression seem even more extreme, there is an actual loss of some vascularization. There are fewer strong bundles, and these lie just lateral to the dorsal position, so that they may serve equally well either of the alternate positions of spikelets from successive rachis segments (Plate III). This increased compression in teosinte and T'ripsacum is likewise apparent in the character of the cells in the ‘pith’? region. In 7ipsacum, the pith cells tend to be flattened in a direction parallel to the surface, whereas, in teosinte, the cells are restricted from expansion and, as in the glumes, small cells become lignified during kernel development (Galinat, 1957). The physical effects of pressure in producing the above differences may be visualized by an extension of the same process used previously to manipulate (hypothetically) the pulvinus into resembling the cupule of maize. These derivations may be seen in the figures of Plate IV as follows: Having depressed the pulvinus (fig. 1) into a cupule (fig. 2), as explained previously, further con- centrated pressure from single spikelets at the center of the cupule would cause the lateral wings to assume a position at right angles to that of the cupule and the cavity to sink more deeply into the pith, as the general structure and anatomical reductions come to resemble those of the rachis segments in T'’ripsacum (fig. 6) or in [ 25 ] teosinte (fig. 5), depending upon the degree of depression. The extreme depression of the rachis segments in teo- sinte and 7'ripsacum seems to require a rigidly erect and sessile condition of solitary spikelets in combination with a thickened rachis. In the staminate rachis, the spikelets are paired, the rachis is more slender and the cupule de- velopment is weak or absent. A condition somewhat similar to that of the staminate rachis may be produced in the pistillate region by introducing the tunicate (7%) gene of maize. Thus, in tunicate teosinte, as in tunicate maize, the pistillate spikelets become more pedicellate and accentuated at the expense of a more slender rachis and they are able, thereby, to bend away from the rachis sufficiently early to leave little or no depression in it. Under such conditions, the would-be cupule lining as- sumes many of the aspects of a flattened pulvinus (Plate IV, figs. 1, 4). The identification of the small bundles from near the outer or convex surface of the rachis segments as rind bundles is more obvious in the relatives of maize than it is in the cupule wings of maize, because the two-ranked condition of the former, as compared to the many-ranked condition of the latter, simplifies comparison with its counterpart in the culm, which is also a two-ranked axis (Plate LV, figs. 2, 3, 5, 6). The course of these rind bun- dles, as well as those of the stronger inner bundles in all members of the American Maydeae, tends to be strictly vertical, even though in the cases of teosinte and T'rip- sacum, the spikelet positions alternate between opposite sides of the rachis. Consequently, in these relatives of maize, the small bundles from the wing area of one seg- ment extend upwards into a dorsal position in the next segment above as they assume a position identical to those in the rind of the culm. Finally, in the third seg- ment, some of these bundles merge with those from the [ 26 ] glume cushion (Plate II). The branching and fusion of small bundles is most frequent in the wing area, but such changes in the degree of vascularization may occur else- where in the rind. There is little doubt that these small wing bundles are actually rind bundles. Certain extreme features of teosinte, as already noted, seem at first to be exceptions to the usually intermediate position of teosinte between its putative parents, maize and T'ripsacum, as might be expected if teosinte be a derivative from a hybrid between these other two spe- cies (Mangelsdorf and Reeves, 1939; Reeves, 1958). These extreme features of teosinte are a more slender peduncle, a shorter spike with less pith, fewer vascular bundles, and deeper, more highly lignified cupules, as well as more numerous pistillate spikes arranged in com- pact clusters. But, on final analysis, all seven of these new characters seem to be a hybrid product of combin- ing two other characters from maize and T'ripsacum. The derivation of these new characters might be as follows; If the erect sessile spikelets of 7'’ripsacum should be combined with increased lateral compression from the tightly binding husks characteristic of the maize ear, the spikelets would become more deeply embedded in the rachis segment. Reductions in the pith, the vascular system and in cell size would follow such extreme com- paction, and the smaller cells would accumulate lignin during kernel development (Plate ILI, fig. 2). The ‘‘clusters’’ of numerous pistillate spikes in teo- sinte may also be explained as a recombination of two other characters: condensation in the shank (peduncle) of maize, which has lateral buds at every node, and the small, two-ranked spike of Tripsacum. Although the potential for the production of ears at every node along the shank occurs in most varieties of maize, it seldom develops, because the energy is concentrated into the [ 27 ] formation of asingle large spike, the ear; but in teosinte, where the individual spikes are small, there is sufficient energy for the development of clusters of spikes. DiscusslION AND SUMMARY The structure of the entire plant of maize and its rela- tives results from various controlled manifestations of a basic pattern of organs, the phytomer. The parts of the phytomer have been recognized in vegetative form as an internode, a leaf, an axillary bud and a prophyll. These organs, as well as their organization in the phytomer, constitute specialized regions without exact boundaries. Nevertheless, the phytomer represents a fundamental design which occurs repetitiously throughout the entire plant. A comparison of its various manifestations is sim- plified, if we select a cycle comprising the group of organs which are adjacent to a given node rather than to use the classical delimitation which includes lateral organs at opposite ends of a given internode. The reduction of parts of the phytomers which have internodes along the rachis is usually extreme. Tor ex- ample, the leaves which potentially subtend the spikelets are usually reduced to glume cushions or swellings at the base of the glumes, although certain genes (7p, Cg) may stimulate their development as typical leaves. ‘These glume cushions are vascularized by small-bundles which might be regarded as rind-bundles rather than the vascu- lar remains of rudimentary leaves, because they connect with the vascular supply to the outer glumes rather than terminating as stubs. More information is needed con- cerning vascularization of swellings associated with un- successful attempts at leaf development. The prophyll is another phytomeric part which is highly reduced along the rachis. In the staminate rachis at the axils of tassel branches, it assumes the functional [ 28 J form of a pulvinus, an axillary swelling which has become specialized for spreading these branches at the time of anthesis. ‘The tissues of the mature pulvinus suggest the primordium of an organ such as the prophyll, in that the swelling results from a proliferation of many small cells rather than from an expansion in size of individual cells. A study of the ontogeny of the pulvinus would be of interest in that leaf primordia (which would complete development) are usually initiated in the dermatogen or outermost layer of cells, although this condition might not necessarily apply to reduced leaves. Vascular devel- opment in the pulvinus, like that of the glume cushion, appears to be a part of the rind of the rachis. Higher up in the staminate rachis or central spike of the tassel where the axillary buds are manifest as binate spikelets rather than as elongated branches, the rudimen- tary prophyll usually appears as a scab-like growth on the rachis or may be entirely reduced. In the pistillate rachis, it is the lining of the cupules which seem to represent the prophyll. The small cells of this lining are very similar to the small cells of the pulvinus, and the physical differences in the external shape of these structures may be attributed to differences in compression during development. The depression of the rudimentary prophyll and asso- ciated tissue is even more extreme in the relatives of maize. ‘The result is a deeply hollowed, cupulate rachis segment which is lined with a thinner layer of ‘‘prophy]- lar’? cells and has a distorted and reduced vascular sys- tem. Such effects are extreme in teosinte rather than intermediate between its putative parents, as might be expected if teosinte is a derivative from a hybrid between these other two species, because the extreme compression responsible is a hybrid product of combining two other characters. Accordingly, when the erect and_ sessile [ 29 ] spikelets of T'ripsacum are subjected to increased lateral compression from tightly binding husks of maize, the spikelets then become more deeply embedded in the rachis, and various reductions in the pith, vascular sys- tem and cell size follow. It is curious that the vascularization associated with all three of the rudiments studied—the glume cushion, the pulvinus and the cupules or hollowed rachis seg- ments—appears to be a part of the rind of the rachis, rather than to represent vascular vestiges of their appar- ent homologues. But the tissues of these rudiments, as well as the effects of unusual genetic or environmental conditions, reveal their homology to organs which cor- respond to that of their apparent phytomeric position. In the ultimate axes or rachillas, the parts of the phy- tomer become obvious. The first two phytomers of the rachilla have reduced axillary buds and produce little more than the leaves or glumes. But, in more distal phytomers, the leaves (now called lemmas) subtend a bud of sexual organs and its associated prophyll or so- called ‘‘palea.”’ The gross structure of the entire plant in maize and its relatives is organized upon a basic pattern of organs, the phytomer. The variations in expression which the phytomer has sought in different portions of the plant may (as Arber has suggested with respect to orders) be compared with the intrinsic beauty of a ‘‘theme with variations’ as expressed in certain musical compositions. Both represent harmonious variations upon a repetitious design, and both obey the dictates of an overall plan. ACKNOWLEDGMENT During the course of the investigation and preparation of the manuscript, many helpful suggestions were made by Professor Paul C. Mangelsdorfof Harvard University. [ 30 | LITERATURE CITED Arber, Agnes, 1934, The Gramineae. Cambridge Univ. Press, Cam- bridge, England. Cutler, H. C., and M. C. Cutler, 1948. Studies on the structure of the maize plant. Ann. Missouri Bot. Gard. 35: 301-316. Esau, K., 1943. Ontogeny of the vascular bundle in Zea Mays. Hil- gardia 15: 327-368. Evans, M. W., and F. O. Grover, 1940. Developmental morphology of the growing point of the shoot and the inflorescence in grasses. Jour. Agr. Res. 61: 481-520. Galinat, W. C., 1954a. The origin and possible evolution of sub-tassel ears in maize. Bot. Mus. Leafl. Harvard Univ. 16: 261-264. ——,, 1954b. Corn grass II. Effect of the corn grass gene on the de- velopment of the maize inflorescence. Amer. Jour. Bot. 41: 803- 806. ——,, 1956. Evolution leading to the formation of the cupulate fruit case in the American Maydeae. Bot. Mus. Leafl. Harvard Univ. 17 ¢- 217-230. ——,, 1957. The effects of certain genes on the outer pistillate glume of maize. Bot. Mus. Leafl. Harvard Univ. 18: 57-76. and A. W. Naylor, 1951. Relationship of photoperiod to inflo- rescence proliferation in Zea Mays L. Amer. Jour. Bot. 38: 38-47, Gray, A., 1879. Structural botany or organography on the basis of morphology. Ivison, Blakeman & Co., New York. Holttum, R. E., 1956. The classification of bamboos. Phytomorphol- ogy 6: 73-90. Kumazawa, M., 1939. On the vascular course in the male inflores- cence of Zea Mays. Vascular anatomy in maize. I. Bot. Mag. Tokyo 53: 495-505. Laubengayer, R. A., 1948. The vascular anatomy of the four-rowed ear of corn. Ann. Missouri Bot. Gard. 35: 337-340. [ 31 | Laubengayer, R. A., 1949. The vascular anatomy of the eight-rowed ear and tassel of Golden Bantam sweet corn. Amer. Jour. Bot. 36: 236-244, Mangelsdorf, P. C., 1958. Ancestor of corn. Science 28: 1313-1320. and R. G. Reeves, 1939. The origin of Indian corn and its rela- tives. Texas Agric. Exper. Sta. Bull. 574. Nickerson, N. H., 1954. Morphological analysis of the maize ear. Amer. Jour. Bot. 41: 87-92. Reeves, R. G., 1946. Methods for studying the maize ear. Bot. Gaz. 107: 425, ——, 1949. Morphology of the ear and tassel of maize. Amer. Jour. Bot. 37: 697-704. ——, 1953. Comparative morphology of the American Maydeae. 2 Texas Agric. Exper. Sta. Bull. 761: 1-26. and P. C. Mangelsdorf, 1942. A proposed taxonomic change in the tribe Maydeae (family Gramineae). Amer. Jour. Bot. 29: 815- 817. Sturtevant, E. L., 1899. Varieties of corn. U.S. Dept. Agr. Off. Exp. Sta. Bull. 57: 108 pp. Weatherwax, P., 1925. Anomalies in maize and its relatives—lII. Many-flowered spikelets in maize. Bull. Torrey Bot. Club 52: 167-170. [ 82 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Campripcr, Massacnusetts, JANUARY 27, 1960 Voi. 19, No. 2 PREHISTORIC BEAN REMAINS FROM CAVES IN THE OCAMPO REGION OF TAMAULIPAS, MEXICO BY L. Kapian! anp R. S. MacNetsH’? Durinec the winter of 1954, under the auspices of the Botanical Museum of Harvard University, the Ameri- can Philosophical Society, the American Academy of Arts and Sciences and the National Museum of Canada, the junior author undertook an archaeological survey and excavations in Tamaulipas, Mexico. The primary pur- pose of this investigation was to obtain specimens and information pertaining to the origin, development and diffusion of prehistoric agriculture in the New World. One of the important foods domesticated and dispersed prehistorically was beans. This report is concerned with the bean remains found in the excavations. Tamaulipas is the northeasternmost state of Mexico and is situated along the Gulf Coast. The southwestern portion of this state was the area surveyed. This region shows considerable range in topography and vegetation from east to west. The eastern boundary is the wide flat meandering Guajalejo River valley which has a trop- ical vegetation extending up from the south. West of this valley, running north and south, lies the first ridge 1 Roosevelt University, Chicago, Illinois. ? National Museum of Canada, Ottawa, Canada. [ 33 ] of the Sierra Madre mountains. This rises to a height of 3,000 feet and is covered by a tropical rain forest with large pines on the summits. Further to the west, the valleys gradually become narrower and more canyon- like, with a more xerophytic vegetation in which maguey, cactus, mesquite and chaparral predominate. The moun- tains are higher (over 5,000 feet) with a thin cover of pines on their rocky summits. In this region we were fortunate enough to find three dry caves with stratified deposits representing eight cultures, associated in the main with abundant, preserved food remains (including beans). We must first describe the archaeological findings in order to present an adequate background for an analysis of the bean remains. ‘The sequence is based upon strati- graphy from three caves which is in part confirmed by Carbon 14 dates (see Table 1). Study of the artifacts from the various occupation levels in the caves revealed similar sequential cultural complexes. Romero’s Cave (Tm ¢ 247) had sixteen occupation levels and a few art- ifacts below them belonging to the Infiernillo, Ocampo, Guerra, Mesa de Guaje, Palmillas, San Lorenzo and San Antonio complexes. Valenzuela’s Cave (Tm c 248) con- tained nine stratified layers belonging to the Infiernillo, Ocampo, Flacco, San Lorenzo and San Antonio com- plexes, all of which had associated vegetal materials. Ojo de Agua Cave (Tm c 274) had twelve occupations and cultural remains representing the Infiernillo, Flacco, Palmillas and San Lorenzo complexes. Only the upper Palmillas level had preserved food materials. ‘The accom- panying chart Table I illustrates the dating and correla- tion of the stratigraphy of the three caves. THE SEQUENCE OF CULTURAL PHASES In briefly summarizing the cultural complexes, we [ 34] TABLE I Tm ¢ 247 Tm c 248 Tm c¢ 274 Estimated Cultural dates in Occ. C. 14 C.14 = =Oce. C. 14 C. 14 Occ. C. 14 C. 14 Phase yearsago layer date No. layer date No. layer date No. San Antonio 260 6 9 Lo 500 San Lorenzo 14. 12 520+200 M-501 900 13 8 11 1100 12 10 Palmillas 9 1800 11 1720150 M506 8 La Florida 2400 10 3440 + 250 M505 Mesa de Guaje 3400 9 8650+ 250 M505a 8 7 6 4730+ 300 M504 Guerra 8800 5 vi 394543384 c 7 6 5 I 3 Flacco 4300 2 4 4580+ 350 M508 3 bs 1 5280+ 350 M502 6 5650+ 350 M497 5 Ocampo 6000 4 TOOO 3 la 2 8200+ 450 M498 1 8540450 M-500 Infiernillo 9000 1 shall emphasize what we know of the subsistence and domesticated plants utilized, with only a mention of the more diagnostic artifact types. The people of the Infiernillo Phase were nomadic family bands of wild plant collectors who did some hunt- ing. Nevertheless, they utilized the domesticated gourd (Lagenaria siceraria) and the pumpkin (Cucurbita Pepo). The seeds of the pumpkin, however, are extremely small and it must have been close to the hypothetical wild form if it were not actually wild. Other plants collected, which could have been domesticated, included runner beans, chili pepper and opuntia. These species, however, represented only a small portion of the diet which was composed mainly of a wide assortment of wild plants. Among the distinctive artifacts we might list Fuegian nets, loop-twine and Fuegian baskets, flake choppers, pebble smoothers, checker-woven mats with oblique corners, short incipient contracting-stemmed and small diamond-shaped projectile points, fire tongs and digging sticks. We could further enumerate a number of more general cultural traits such as several kinds of choppers, side scrapers and scraping planes, round-based points and knives, atlat] and dart fragments, twilled mats and dif- ferent kinds of string with various knots in them. Ocampo had more foodstuffs and many more artifacts. The people of this stage were semi-nomadic plant collec- tors who did a little hunting and who gained asmall part of their sustenance from domesticated plants. Their in- cipient agriculture included pumpkins (Cucurbita Pepo), gourds (Lagenaria siceraria), common beans (Phaseolus vulgaris) of two varieties, chili peppers (Capsicum) and, possibly, corn. We say ‘‘possibly corn,’’ because, al- though we found no cobs or kernels in the refuse, an examination of feces from these levels revealed minute particles of cobs and leaves. This discovery suggests that [ 35 | small primitive or green ears had been masticated and digested, leaving no cob or kernel remains in the refuse. Opuntia and runner beans also served as food. Distinc- tive artifacts for this complex are Fuegian and _ full- turned coiled nets, three-over-three twilled baskets, twilled and plaited mats, large triangular and leaf-shaped atlatl dart points, interlocking loop-coiled baskets, gouges, antler hammers, and a wide variety of larger scrapers (planes) and choppers. There likewise occurred more general traits, such as different kinds of string and knots, mats, bone awls, wooden wedges, dart points, bi- facial knives, shell beads, mortars and gourd containers. The Flacco Phase developed directly from the Ocampo. The people of this phase were also semi-nomadic food-gatherers, but they depended more on incipient agri- culture than on hunting or trapping of game. Squash, gourds, corn, chili pepper, two kinds of common beans (found in the feces), Panicum, amaranths, and runner beans made up apparently about twenty percent of their diet. Long contracting-stem and indented-base points, mullers and mortars, spokeshavers, Fuegian baskets, spring traps and snares, twilled baskets, simple coiled bags and elongate chisels are diagnostic traits. Though most of the heavy stone scrapers and choppers of Ocampo are not present, the more general traits continue to occur. With the development of the semi-sedentary Guerra peoples, there occurred a fundamental shift in diet. While they still collected a vast amount of wild plants, agriculture furnished most of the energy-producing foods. The most prevalent remains are corn cobs (Bat Cave race). In addition, gourds, several varieties of pumpkin, squashes (Cucurbita moschata), peppers, com- mon beans, amaranths, Panicum and cotton occur. The cave occupations, and also the survey, indicate the possi- bility that these people occupied small villages. Split- [ 36 ] stitch, interlocking loop, and simple stitch baskets built on a bundle foundation, small tear-drop, triangular and corner-notched dart points, rabbit sticks, leather hu- arches, and other objects, mullers and manos, a twined woven robe, decorated mats with woven borders, as well as a wide variety of different kinds of string, and more general artifacts give this culture a distinctive aspect. Mesa de Guaje is very similar to the previous culture, but most of the nets and baskets have been replaced by plain weave cotton fabrics and plain black and brown pottery. Scrapers and flint tools are very rare; straight- stemmed points occur for the first time, as do manos and metates made of volcanic tufa, obsidian blades, clay disks, atlat] bunts and knotless netting. The Mesa de Guaje people definitely lived in villages. In terms of nutrition, probably more than half of their food was derived from agricultural products, the rest from wild plants. Corn is the main product and much of it shows teosinte introgression; actual grains of teosinte were found. In addition to beans (two varieties), gourds, squash, pumpkin, amaranths, peppers, cotton and sun- flower seeds were present. The next phase, called La Florida, was recognized in the survey, but did not occur in the caves. It is typical of the Late Formative of Mexico, with hand-modelled figurines, stemmed points, corner-notched points, pris- matic blades, pottery bowls with tripod feet and stone- faced pyramids around plazas. The following culture, Palmillas, represents the cul- tural apogee of the region. Furthermore, it represents the period of greatest diversity in agriculture. The great- est variability of pumpkins occurred at this time, as well as of gourds, warty squash (Cucurbita moschata) and walnut squash (Cucurbita mivta). A number of races of maize are present together with grains of teosinte. Three [ 37 | varieties of common beans are found as well as lima and runner beans. In addition to these plants, Manihot dul- cis, amaranth, chili, Panzewm, sunflower, Nicotiana and cotton occur. In bulk, these species represent almost half the plant material found; in terms of food value, however, they represent a much larger proportion. Arch- aeologically, this phase is represented by a mass of ma- terial and traits. Only a bare minimum of the diagnostic artifacts will be mentioned. These include corner-notched arrow and dart points, serrated corner-notched points, engraved red, brown, and black pottery, packboards with net centers and wooden rims, complicated woven mats and cotton cloth, platform pipes and cane cigarettes, mold-made figurines, polished celts, circular pyramids and house platforms, as well as many other stone archi- tectural features. Though the sedentary agricultural San Lorenzo is ob- viously derived from Palmillas, there seems to be a short gap in the sequence and a degeneration in culture and agriculture. While all the kinds of cucurbits and beans appear, there is less variability. There are also only one or two races of corn, no teosinte, and only amaranths, peppers, cotton and tobacco. The bow and arrow, and a considerable range of arrow-point types appear; the pottery is crude — burnished, brushed and corrugated ware; the mats are decorated with colors; split-stitch bundle foundation baskets are plain; decorated cotton double-cloth occurs, and a number of small crude end- scrapers. The San Antonio culture represents an even further degeneration, though the people of this phase seem to have been sedentary agriculturists living in ‘‘ranchos.”’ Corn was apparently of a single race; there are only four kinds of beans. Cucurbits (Cucurbita Pepo and C. mos- chata), gourds, cotton and tobacco still occur. Many of [ 38 | the artifacts are like those of San Lorenzo, but new point and pottery types occur as do large choppers and scrapers. Historic goods were present in some levels, and the woven twilled mats, coiled nets and cloth were much simpler. Though the culture phases mentioned above seem to represent a unilinear development, the area, situated on the northern peripheries of Meso-America, must have had many influences at different periods from the south. It is believed that this peripheral region reflects, perhaps with some time-lag, the general sequence of agricultural practices for all of Middle America. Its geographical position just north of Meso-America indicates that it is a key area for the understanding of diffusion of agricul- ture into North America. The study of our bean remains from the Tamaulipas caves, which seems to bear out these generalizations, also has considerable bearing on the solution of the problem of the origin and dispersal of prehistoric beans in the New World. IDENTIFICATION OF MATERIALS The materials examined consist primarily of desiccated and uncharred bean pods and fragments of pods. Seed remains are few and often fragmentary. Since most of the materials mentioned in this paper are pods, a brief discussion of the nature of legumes will be helpful. In the Leguminosae, gross morphology of the fruit may be diagnostic in species-determination and is useful in delineating varieties of polymorphic species such as Phaseolus vulgaris. 'The legume or pod is a single carpel with two identical valves dehiscing on dorsal and ventral sutures. At the basal end, these elongate valves diverge from a pedicel and terminate at the apex in a straight or curved tip. Behavior of the pods at dehiscence is related [ 39 | to their anatomy and may differ markedly in different varieties. When the fruit is mature and dry, the valves split along dorsal and ventral sutures after which they may simply separate or they may twist to varying de- grees. Twisting results in dislodgment of seeds and proba- bly aids in their dissemination. The twist of one valve is the mirror image of the other, and the twist of both is the result of a shortening of certain fiber-cells in the pod-wall. In the production of beans which are threshed after the pods and seeds mature and dry on the plants, it is essential that the pods be of the type which do not de- hisce violently with the consequent scattering of the seeds prior to threshing. The manner of dehiscence is of little importance in varieties which are customarily harvested during damp weather or which are used in the green stage. Explosive scattering of seeds probably occurs in all wild species of the genus Phaseolus and the loss of this char- acteristic must have been one of the important features of domestication and variation in beans. The pods, and other vegetal debris examined, had been placed in separate packets, according to the site of collec- tion, culture and occupation level. When the relative quantities of materials present in each occupation level had been estimated, tentative species-identifications were made. Many of the twisted and folded pods were ex- tended and pressed after softening in warm water with a detergent to facilitate their identification. Phaseolus vulgaris (common beans) and P. lunatus (the small seeded or sieva group of lima beans) were rec- ognized on the basis of their gross morphology. P. coc- cineus (runner beans) pod fragments were at first classified only as leguminous remains, probably of a species of Phaseolus or a closely related genus. These remains of P. coccineus were later identified on the basis of the hilum [ 40 ] Plate VI { é Pare VI. a, “Indigenous Phaseolus vulgaris.’* From plants grown at Lombard, Illinois by H. C. Cutler; original collection by O. W. Norvell, 14 km. north of Iguala, Guerrero. Natural size. b, ““Wild Phaseolus vulgaris.”” Collected by G. F, Freytag on slopes of volcano of Colima, Jalisco. Natural size. characteristics of a single seed coat fragment still attached toa pod valve, and the dimensions of one almost complete valve which correlated in form and texture with the pre- viously unidentified leguminous type. Although the hilum in question is within the size range of P. /unatus, the separate ridges of the caruncle excluded the possi- bility of its being a lima bean. The hilum characteristics of this fragment are similar to those of a rather large common bean, but the dimensions and the fibrous to al- most woody nature of the associated valve are similar to P. coccineus pods. The possibility that this material pertained to some other species not known as a domesticate was not over- looked. Through the courtesy of Dr. Frederick G. Meyer, then at the Missouri Botanical Garden, it was possible for the senior author to examine a list of species compiled from the extensive plant collection of Dr. Meyer and Dr. R. L. Dressler. Specimens of legumes appearing in their list and which might have fruits sim- ilar to the archaeological fragments were examined in the herbaria of the Missouri Botanical Garden and the Chicago Natural History Museum. None was found to correspond to the material in question. By elimination, then, as well as by positive characteristics, the identifi- ‘ation of Phaseolus coccineus was verified. Variation inthe Phaseolus vulgaris pods indicated their division into three types, each of which is described be- low. These types are to be considered as of coordinate standing with the typology presented for Southwest United States beans by Kaplan (1956). That is, in the absence of diagnostic characters which can be obtained only from the growing plants and complete herbarium specimens, the infraspecific types are without formal taxonomic standing. DeEscrIPTION oF TyPpEs Pod characteristics are summarized in Table III; the types discussed below have been given descriptive names and are numbered in sequence with archaeological beans described elsewhere (Kaplan, 1956). Yellow seeded bush, C 81. The bush habit is indicated by remains of two plants with fragmentary fibrous root systems diverging from the bases of stems bearing fruit- ing branches. The curved pedicels are 0.8—-1.5 em. long borne in pairs on 5-6 cm. peduncles. The pods would have been borne at 15 to 20 cm. above the soil surface and appear to be fewer than 10 per plant, although this could be highly variable. This variety was probably har- vested by pulling up the entire plant, since many of the pods have remained attached to stem sections. In con- temporary Mexico, bush beans are frequently harvested in this manner; the bundles of dried plants are stored in or near the habitation and threshed by beating on a mat as opportunity or need arises. The yellow to yellow-tan color of the rare seed coat fragments associated with the pods represents possibly a change in hue from cream. This group includes thin-walled variants probably picked when young. Thick-walled variants are more common, particularly in the San Antonio phase; 20 of 26 fragments in one San Antonio packet are thick-walled. It is not possible to say whether this is a genetic varia- tion or the result of growing conditions. Lesions which appear to be anthracnose (Colletotrichum Lindemuthia- num) injuries are found occasionally on all types but are abundant on the thick-walled variants. Black seeded bush, C82. The pods are moderately curved with thin and relatively non-fibrous walls. In the immature condition, the fruits of this type might have been useful as snap beans; displacement of the intersti- [ 43 ] TABLE II DISTRIBUTION OF BEAN REMAINS Culture Tentative age P, coccineus P. vulgaris P, lunatus Total in each years ago valves valves valves culture C31 C32 C33 C33a San Antonio 200-500 ] 40 35 76 San Lorenzo Occu- pations 13,14,15,16 500-900 21 296 45 111 4 3 480 Palmillas 1100-1800 11 206 14 1 232 Mesa de Guaje 2400-3400 16 3 19 Guerra 3400-3800 1 1 Flacco 3800-4300 13 13 Ocampo 4300-6000 4 5 1 10 (Portales) Infiernillo 7500-9000 14 14 tial parenchyma tissue, however, shows that the seeds were mature when harvested. A single pod of this type was found containing three small black seeds (Table III). The non-twining shoots on which pods have persisted in- dicate this to be a bush variety probably harvested in a manner similar to that suggested for the previous type. In contrast to all other beans described here, the valves of almost all of the ‘‘black seeded bush’’ type are joined at the base and attached to the pedicel. Long pod, C 33. The pods are curved but show little variation in breadth from base to tip, that is, there is little tapering; nor is there constriction of the dry pods between the seed positions. Most of these valves are dark reddish brown in color and have a cartilaginous rather than fibrous texture when wet. An occasional pod is encountered with the pedicel and peduncle still joined; these are approximately 1 and 8 cm. long respectively. None of the fruiting branches is attached to a main stem, suggesting that ‘‘long pod”’ is vining and that harvesting involved the pulling of individual pods as is the common contemporary practice with pole beans. No seeds have been found attached to the pods. A few fibrous-walled variants of C 33 (C 38ain Tables [I and III) are present with more seeds than the type and are correspondingly larger. Discussion Runner beans Of greatest interest in this group of legumes is the occurrence of remains of Phaseolus coccineus, constitut- ing the first definite archaeological record of this species (Kaplan, 1956). In contemporary agriculture, these beans are grown mainly in the highlands of Chiapas and the Valley of Mexico. From the Federal District to Querétaro, they are frequent, and further north they are [ 45 | TABLE III SPECIES AND VARIETAL CHARACTERISTICS SPECIES & TYPE POD CHARACTERISTICS SEEDS length dorso-ventral twistof no. seeds tip length width thickness color cm. width cm. valves per pod cm. cm. cm. P. vulgaris C31 ce 0.8-1.2 slight 3-6 _ short- fragments — = yellow to straight only yellow tan C32 6-75 0.7-1.1 slight 5 straight $—1.0 0.6 0.4 black or none to down- curved C33 11-13.5 1 moderate 7 few intact ot — = aad fragile C33a 13-15 ae ee moderate 8 few intact — — _ ag fragile P. coccineus greater 1S or tight — _ none intact — — — dark, than 11 more probably probably short and purple stout P, lunatus about 12 3 slight — _ none intact PLATE VII Prare VIL. a, Phaseolus coccineus. Pod fragments from the Infiernillo and Ocampo cultures. Natural size. b, Phaseolus vulgaris (yellow seeded bush C31), three pod valves. From Palmillas culture. Natural size. c, Phaseolus vulgaris long pod C383), pod fragments. Slightly less than one half natu- ral size. From San Lorenzo Culture. found only occasionally in Mexico and in the southwest- ern United States, where at least two varieties are cul- tivated by the Hopi. Runner bean use by the Hopi is almost certainly an historic introduction in view of their absence from archaeological sites thought to be associated with Hopi prehistory. In July 1957, Kaplan briefly visited Ciudad Ocampo, below the Sierras in which the caves are situated, taking with him pods of the excavated Phaseolus coccineus and seeds of purple variegated and self-colored, white, modern runner beans. During the Sunday market, residents of Ocampo and others from villages higher up in the Sierras were shown the samples and asked to identify them. No one recognized the seeds or pods as belonging to either wild or cultivated plants with which they were familiar. It is safe to conclude that P. coccineus is not now utilized in the area in which the excavations were made. The archaeological runner bean pods, notwithstanding their morphological similarities to modern types, differ from them in certain microscopic features. Some of these differences are quantitative and are related to cell-wall thickness and fiber-cell size. Furthermore, the archaeo- logical pods are more tightly twisted than are those of modern cultivated varieties. Most striking is the abun- dance of sclerids or stone cells in the archaeological ma- terial as contrasted with their absence in modern runner bean pods. Although the reduction of sclerids in the domestication of fleshy fruits is of obvious adaptive value, it would not constitute a selective factor in the domesti- ‘ation of beans. Even if the pods were used in the green state, it is doubtful that the appearance of more edible varieties with fewer stone cells ever occurred as the re- sult of human intervention. Since these stone cells affect qualities of pods which would be used in the immature state, any selection for their reduction would have re- [ 48 | quired the application of rather advanced concepts of plant breeding probably not at the disposal of pre- Columbian peoples in the New, or, for that matter, in the Old World. It may be that the deposition of lignin to form stone cells is related to more general physiological processes which were subject to selective influences. There is no apparent dissimilarity between early and later Phaseolus coccineus pods in this site. There are no pods which could be considered as intermediate in reduc- tion of hard texture and fiber between the early Tamau- lipas material and modern varieties. There are, then, remains of Phaseolus coccineus pods, anatomically more primitive than modern types, in in- habited sites and corresponding at later occupation levels (see introduction and Table IL) with definite evidences of agriculture. Several closely related problems remain to be solved. These are: the absence of P. coccineus from late Flacco to the beginning of Palmillas (Table II); the failure of the species to persist into historic times in this region; and its status as a domestic or wild plant during the periods of cave occupation. The disappearance of runner bean use in Tamaulipas resulted probably from the effect of climatic change on plant distribution. Preliminary studies (MacNeish, in conversation) indicate that this region has passed through at least two warm climatic cycles. The earliest of these was presumably warm and moist, as its duration was marked by the presence of Manihot from late Flacco to about the beginning of Palmillas (see Table IL). Runner bean remains are absent during this thermal period of more than 2000 years, although they are present both before and after. This absence cannot be explained by the prevalence of poor conditions for vegetal preservation in the caves, since cucurbit materials are abundant from this period (Whitaker, et al, 1957). [ 49 ] The reappearance of Phaseolus coccineus after the be- ginning of Palmillas suggests strongly the response of a wild rather than cultivated plant. We would hardly expect a cultigen absent for 2000 years to be reestablished merely with the return of favorable climatic conditions. Reintroduction is a possibility, but from where? The southwestern United States could not have been a source for reasons already noted. Plant migration from regions south of Tamaulipas would have been highly unlikely, since conditions there would have been even less favor- able for the survival of a cultivated plant with cool tem- perature requirements. An indigenous plant, on the other hand, might have formed relic communities in pro- tected locations becoming more generally distributed and available for human use at the end of the thermal period. A second thermal period might well have been the cause of the extinction of P. coccineus from this part of its nat- ural range. The hypothesis that Phaseolus coccineus may have been a wild rather than a cultivated plant in ‘Tamaulipas is sup- ported by the extreme age of the remains and their occur- rence long before the practices of ceramic cooking and agriculture, and by the apparent lack of selection for pod characteristics found in modern cultivated varieties. If runner beans were present, but not domesticated, in Tamaulipas in spite of their being included in the gath- ered plant complex prior to and during agricultural times, another problem arises. Why would so useful a food plant be neglected as a domesticate in Tamaulipas but be brought into cultivation in Chiapas? The answer is to be found probably in the reaction of P. coccineus to the differing photoperiods in these widely separated areas of its range. Allard and Zaumeyer (1944) studied the reaction of various leguminous plants to day-length. A daily expo- [ 50 | sure to less than 13 hours of daylight resulted in an ap- preciable delay in flowering date of Phaseolus coccineus which was accompanied by tuberization of the root sys- tem. Longer day-lengths brought about earlier flowering, and tuberization did not occur. Chiapas, at about 16° North latitude with days of less than 13 hours of light during the growing season, corresponds to the delayed flowering-tuberized root situation, while Tamaulipas at about 23° North latitude has photoperiods of more than 13 hours during most of the growing season. In the highlands of Chiapas, the fleshy roots of runner beans are eaten by Tzeltal Indians (fieldwork, 1957). If this be a retention of an early practice (as suggested by Edgar Anderson in conversation), then it is possible that domestication of runner beans was based upon root as well as seed use. Such a practice could not have occurred in ‘Tamaulipas since tuberization does not occur because of the long-day, short-night condition. The early presence of Phaseolus coccineus remains makes it advisable to examine any wild populations of bean species which are otherwise known only as cultivars. Kaplan has collected P. coccineus in a variety of situations, including pine and oak forests and deep barrancas, in Chiapas. None of these sites has been under cultivation in the memory of local inhabitants, but the strong ten- dency of this species toward perenniality leaves open the possibility that these plants are relics of cultivation. However, Ephraim Hernandez X. has stated (in conver- sation) that P. coccineus, and other closely related spe- cies, Which are not escaped or relic cultivars do occur in Chiapas. Probably the most important related species is Phaseolus polyanthus Greenman which also seems to be planted with P. coccineus and reaches the markets with beans of this latter species. Oliver W. Norvell has noted such market mixtures in seed collections in the herbarium [ 51 | of the Chicago Natural History Museum (herbarium accession numbers 981251 and 1119897). The caruncle ridges of this species are less distinctly separated than those of P. coccineus. Concerning the distribution of Phaseolus polyanthus, Piper (1926) says only that the type specimen is from near Jalapa, Vera Cruz, where it was collected on a railway embankment (Greenman, 1907). Piper further notes that specimens which may be from wild P. coccineus plants have been collected near Puebla, Puebla; Monte Esco- bedo, Zacatecas; Saltillo, Coahuila; Tacuba, Mexico; San Juan Capistrano, Jalisco; Tumbala, Chiapas; and Frajanco Santa Rosa, Guatemala. As yet, there is, un- fortunately, not sufficient information concerning affini- ties and barriers between putative wild P. coccineus and wild possibly interbreeding species, and domesticated P. coccineus to draw any conclusions concerning systematic relationships. Common beans Types of Phaseolus vulgaris not generally distinguish- able from modern cultivated varieties appear in abun- dance beginning with the Mesa de Guaje level, at about the same time that intensive agriculture and fired ceramic wares were introduced. Bean remains in non-agricultural, pre-pottery Ocampo and Guerra cultures are limited in number. This situation is similar to that in the South- west, where in the Mogollon and Basketmaker—Pueblo regions the increase in bean remains is correlated with the introduction of pottery (Kaplan, 1956). Common beans, appearing for the first time in the Ocampo culture, join the already established Cucurbita Pepo (Whitaker, et al., 1957). Both of these plants an- tedate the appearance of corn. This is the only area in which common beans have been shown to occur prior to [ 52 | corn and the early bean types persist into more recent archaeological times. The presence of these bean types in early pre-pottery times followed by an expansion of their use with the in- troduction of pottery agrees with the hypothesis pre- sented (Kaplan, 1956) to account for the remarkable constancy of bean types over long time spans in the Southwest. This hypothesis postulates that beans en- tered as domesticates very early, were sparingly culti- vated until the use of pottery began and then the same, long established, well adapted varieties came into more general use. Collections of specimens determined as wild P. vul- garis have been made by G. F. Freytag, O. W. Norvell, A. Burkart and others. The only published account of wild and cultivar affinities which goes beyond morpho- logical comparison is that of Burkart and Briicher (1953). A wild Phaseolus species collected in Central and South America, and studied by these authors, has proven to interbreed with P. vulgaris and although the floral mor- phology is similar, differences in seeds, pods and leaves exist. Asa result of their studies, revisions in the no- menclature of P. vulgaris L. were proposed by Burkart and Briicher (1953). The various collections identified as wild P. vulgaris have all been vining types with seeds smaller than those encountered in most cultivated varieties. It seems that the determinate growth habit as well as increase in seed size and reduction of pod-shattering have been estab- lished under domestication. Sieva beans Small seeded limas, or sieva beans, were introduced late and never attained much importance. Their paucity here substantiates other evidence indicating that these [ 53 ] beans entered the southwestern United States from the west. Sievas appear in the Verde Valley and at Point of Pines, but not in the more easterly Mogollon and Basket- maker—Pueblo areas. With the study of additional re- mains in the future some relationship between south- eastern United States and northeastern Mexico sievas may be shown. Tepary beans The absence of tepary beans (P. acutifolius var. latifo- lius Freeman) supports our present knowledge of their pre-Columbian and historic distribution. Bukasov (1930) and others have indicated a western distribution for con- temporary tepary cultivation, while Carter (1945) has shown a spread of tepary cultivation from south-central Arizona to the north and east in late pre-Spanish times. SUMMARY 1. A total of 845 bean pod fragments from three arch- aeological cave sites in Tamaulipas, Mexico, were studied and identified. ‘These remains consist of: Phaseolus vulgaris (8 domesticated types), P.coccineus (one nondomesticated type), P. /unatus (one domes- ticated type). 2. These beans are considered to be distinct from those occurring in prehistoric time in the southwestern United States. 3. It is thought that P. coccineus, although gathered as a useful wild plant, was not domesticated in this re- gion because of non-tuberization of the roots under the photoperiod conditions of Tamaulipas. 4. P. vulgaris remains first appear with preagricultural materials 4300-6000 years ago and are the oldest com- [ 54 ] mon beans yet reported. They antedate corn but are later than squash in these sites. P. coccineus remains, the earliest of which are dated 7500-9000 years ago, constitute the only archaeological runner beans on record. P. dunatus remains are few and were present 500-1800 years ago. [ 55 | BIBLIOGRAPHY Allard, H. A. and W.J.Zaumeyer. 1944. Responses of beans ( Phas- eolus) and other legumes to length of day. U.S. Dept. Agric. Tech. Bull. 867: 1-24. Bukasov, S. M. 1930. The cultivated plants of Mexico, Guatemala and Colombia. Bull. Appl. Bot. Gen. Pl. Breed., Suppl. 47: 151- 176. Leningrad. (Mss. translated from the Russian by H. J. Kidd, Ann. Mo, Bot. Gard. 41: 271-299. 1954.) Burkart, Arturo and H. Briicher. 1953. Phaseolus aborigineus Burk- art, die mutmassliche andine Stammform der Kulturbohne. Der Ziichter 23 (3): 65-72. Carter, George F. 1945, Plant geography and culture history in the American Southwest. Viking Fund Publ. Anthrop, 5: 1-140. Greenman, J. M. 1907. New or Noteworthy Spermatophytes from Mexico, Central America and the West Indies. Field Col. Mus. Bot. Ser. 2: 247-287. Kaplan, Lawrence. 1956. The cultivated beans of the prehistoric Southwest. Ann. Mo. Bot. Gard. 43: 189-251. Piper, C. V. 1926. Studies of American Phaseolinae. Smithson. Inst., U.S. Nat. Mus. Contr. U.S. Nat. Herb. 22: 663-701. Whitaker, T. W., H.C. Cutler and R. 5. MacNeish. 1957. Cucurbit materials from three caves near Ocampo, Tamaulipas. Amer. Antiq. 22: 352-358 [ 56 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Campripcr, Massacnusetts, May 19, 1960 Vor. 19, No. 3 ON THE ORIGIN OF THE ORCHIDACEAE! BY Lestik A. Garay INTRODUCTION To discourse on the origin of the Orchid family, which represents the culmination of one of the evolutionary lines of the Monocotyledons, is a rather formidable task. Its existing members display such a high degree of com- plexity of organization in their structures that the primi- tive characters or simple elements, which might have shed light on ancestral origins, are either lost or well masked. When we take into consideration that the Or- chid family, comprising at present some 30,000 species distributed in 800 genera, is devoid of paleobotanical documentation, save Protorchis monorchis Massal. from the Eocene of Monte Bolea, it is impossible to explain or even state the manner in which the various evolution- ary forces have acted upon the primordial organism that served as the prototype for modern species. In addition to these difficulties, information on the anatomy, embry- ology, genetics, cytology, fertilization, ecology, ete., of today’s species is too fragmentary to give us a reliable and coherent picture. In spite of these deficiencies, I shall attempt to pre- sent an outline of the Orchid family and its origin, as I 'This paper will be read on May 31, 1960 during the 3rd World Orchid Conference in London, England. [ 57 | understand it, from an interpretation of the evidence now available, as well as to point out the gaps where intensive research is required for further clarification. It is my hope that this effort may stimulate the students of evolution to turn some of their attention to this intricate and per- plexing group of plants. The problem of the origin and phylogeny of various plant groups and families has occupied the minds and in- terest of a number of outstanding botanists from Lin- naeus to our time. All have recognized that in nature a system prevails within which everything seems to follow a pattern of progressive differentiation from simplicity to complexity. While Linnaeus’ sexual system can hardly be associated with progressive differentiation, the sys- tems of Bentham and Hooker or of Engler and Prantl, to mention but two, in essence convey this idea. How- ever, progressive differentiation in the light of our mod- ern species-concept assumes an entirely different role and meaning from that which was understood and employed by the earlier workers. A glance at any of the proposed ‘‘natural’’ systems is sufficient to make us recognize that they are based on a continuous modification along a straight line of descent. We are informed today, how- ever, that species could also have been derived simul- taneously or consecutively from a common ancestor or from the fusion of one or more common ancestors. Pro- gressive differentiation in the light of these latter princi- ples assumes an entirely new and far reaching significance. If we wish to understand the origin or phylogeny of any group, regardless of its taxonomic status (genus, subfamily, family, ete.), we must first understand the basic unit: THe Species. A lengthy discussion about the nature of species does not, however, lie within the scope of this paper, and I must assume that the reader is fully acquainted with the subject. It is sufficient, there- [ 58 | fore, if we merely state that the species represents a dy- namic, adaptive peak in evolution. Any taxonomic group above this level, in order to express a meaningful expan- sion, must represent further dynamic macro-units or en- larged adaptive peaks (which are the results of the co- action and coaptation of the basic units), because if these higher units are merely categories of convenience (as most of them are today) they will hinder rather than broaden our perspective and comprehension of the evolutionary makeup of our object of study. We have to keep constantly in mind that living species are scanty representations of all the possible modifications and combinations that may have occurred in nature, al- though those types which have been eliminated by selec- tive pressures still contributed to the formation of present species. Therefore, when we study a group of plants, we find that its basic units (i.e. species) are connected by a series of trends which are radiating in various directions, forming areticulate pattern. In visualizing this reticulate pattern in a three-dimensional perspective, we find that the connective trends progressively decrease as we ad- vance towards higher levels, but that this decrease never terminates in complete elimination. Such a dynamic and 3-dimensional structure (vaguely resembling the molecular structure of a crystal) cannot be projected into a 2-dimensional perspective without destroying, or at best distorting, its salient features. All of us, from time to time, have felt and recognized the presence of such a system, but our conventional mind with its categorizing instinct continuously interferes with and obscures our vision. I wish to make it clear that I fully recognize the necessity of categories of con- venience, but I strongly object when these categories of convenience are employed as the basis of a so-called ‘“‘natural’’ system; and this is what has happened in the case of the Orchid family [ 59 ] Keeping constantly in mind the dynamic structure of nature, [ shall attempt to examine the Orchid family and its origin against this background. FLORAL DIAGRAMS AND BASIC GROUPS The Orchid family is an extremely heterogeneous unit with respect to its external and internal composition. The vast array of types of modified structures, many of which combine simultaneously characters of both a primitive and advanced nature, are, however, tied together by a few definite characteristics: 1, the inferior ovary; 2, the production of a large number of seeds without endo- sperm; and 8, the various degrees of fusion between the style and stamens. Notwithstanding the large number of species in the family, it is possible to outline and express the arrange- ment of floral parts by means of three or four basic floral diagrams (Plate VIII). From an inspection of these floral diagrams, it is evident that orchid flowers are built on a trimerous pattern and that they are merely modifications of the liliaceous type. The essential deviation from the liliaceous pattern is to be seen in the staminal circles. In the first diagram (PI. VIII, fig. 1), which may equally serve for Ornithogalum of the Liliaceae or Hypoais of the Hypoxidaceae (A marillidaceae sens. lat.), both staminal circles are fertile and fully developed, i.e., all six anthers are functional. In the Orchidaceae (PI. VIII, figs. 2-4) the anthers opposite the lateral sepals and the median petal are completely suppressed (with the exception of Satyrium, about which more will be said later). ‘The re- duction in number of fertile anthers is always paralleled by a fusion of the stamens and of the stamens with the style. Developmental and anatomical investigations with respect to the origin of these fertile circles indicate that the existing species must have evolved along at least two different lines from an already modified ancestral type, [ 60 | HYPOXIS NEUWIEDIA SELEN/IPED/IUM APOSTASIA YN NEOTTIO/DEAE OPHRYDO/DEAE KEROSPHAERO/DEAE zs y SATYRIUM Iq IIA atv because each of these lines, although still sharing detecta- ble characters, displays an unequal rate of differentiation. These functional circles, or the number of functional an- thers present in these circles, are still employed today to characterize the two major subdivisions of the family. In Neuwiedia (Pl. VIII, fig. 2), there are three fertile anthers, one dorsal and two ventral. This is the only genus in the family Orchidaceae where fertile anthers are present simultaneously in both staminal circles. In this respect Neuwiedia, one of the three genera of the A pos- tastoideae, is perhaps the most primitive orchid: the other two genera, Apostasia and Adactylus are merely modifi- cations of the Neuwiedia-pattern. In Apostasia (PI. VIII, fig. 3), we observe a transformation of the dorsal anther into a staminode, while in Adactylus a further suppression completely eliminates this structure. 4 pos- tasioideae are set apart from the rest of the orchids by an additional series of correlated characteristics which will be discussed in further paragraphs. Although the floral diagram of A postasia is identical with Selenipedium (Pl. VIL, fig. 3) of the group Cypri- pedioideae, this identity is restricted to diagrammatical representation. The vegetative appearance, as well as the floral morphology in both of these groups, is very dis- similar. In A postasioideae, the filaments of the anthers are usually still present and recognizable; while in Cyp7i- pedioideae these structures are completely fused with the style. Furthermore, in Cypripedioideae, the shape of the staminode and the median petal exhibit a major devia- tion from A postasioideae, to such an extent that these two groups can hardly be interpreted as representing two subsequent stages of one evolutionary trend. The discontinuity between Meuwiedia and the monan- drous orchids is even greater. The floral diagram (PI. VILI, fig. 4) brings into focus a single criterion only: the [ 62 ] suppression of a different staminal circle. This suppres- sion seemingly conveys the idea that Neuwiedia gave rise simultaneously to both Cypripedioideae and to the monandrous orchids. However, the sum total of charac- teristics which make up the monandrous orchids suggests that during the course of evolution there were other lines besides that of Neuwiedia which fed into the complex. Although the general pattern of the monandrous orchids (which include 90% of the known species) can be ex- pressed by asingle diagram, the group itself is composed of three distinct units, depending on the manner and de- gree by which the individual pollen grains adhere to each other to form the pollinia. These three groups are: Neottiowdeae, Ophrydoideae and Kerosphaeroideae. In Neottioideae and Ophrydoideae (text fig. 1) the pol- len grains are of a soft consistency and cohere into massu- lae in arelatively loose manner; in the former unit they easily separate into a fine powdery mass (sectile pollinia), while in the latter they form large granules (granular pollinia). In Kerosphaeroideae the cohesion of the indi- vidual grains is so compact that the pollinium may be broken only through the exertion of considerable force. The presence of pollinia is characteristic of the three units of monandrous orchids only. In the 4 postasioideae NEOTTIOIDEAE OPHRYDOIDEAE KEROSPHAEROIDEAE and Cypripedioideae, no pollinia are formed; the anthers contain single pollen grains either dry or embedded in a viscid secretion respectively. ENpDOMORPHIC AND EXOMORPHIC FEATURES Vascularization. The distribution of vascular bundles in the stem and inflorescence-axis of orchids exhibits the same more or less scattered arrangement as is generally observable in the vegetative axis of other Monocotyle- dons. The bundles are enclosed within a sclerenchyma- like ring of perivascular fibers, either in a circular manner (similar to the primitive A7istolochia type in Dicotyle- dons) or in ascattered pattern. A preliminary investiga- tion indicates that the circular pattern is always associated with other primitive characters. Those species which ex- hibit this characteristic have a more or less well developed horizontal rhizome (e.g. A postasia, Cypripedium, Zeux- ine, etc.). Nevertheless, more information is needed be- fore the significance of this association may be fully and definitely evaluated. In the vascular supply of the flower, the bundles which enter the floral axis vary in number, and this variation is correlated with the primitive or advanced stage of the group. Swamy reports the number of vascular bundles in the Cypripedioideae as six; in the Neottiordeae and Ophrydoideae and the less specialized members of the Kerosphaeroideae three; while in the advanced types of the Kerosphaeroideae there is a further reduction in num- ber to two. Our investigation of the Apostasiordeae re- veals six vascular bundles, as is the case in the Cypriped- toideae. In A postasioideae and Cypripedioideae, the six bundles constitute the main traces of the ovary without further differentiation. In both of these groups, there is an ad- ditional seventh bundle which gives rise to the midrib of the bract. [ 64 | In the remaining groups (Neottioideae, Ophrydoideae and Kerosphaeroideae) the entering three or two bundles already differentiate in the pedicel to supply the ovary with its six main traces. The pattern by which these six traces of the ovary proliferate in passing to the floral and sex organs is uniform throughout the family. The three dorsal bundles of the ovary enter directly into the three sepals, with a trace leading from each of them to the gyn- ostemium; an additional branch deviates from the main bundle which supports the dorsal sepal to the anther of the outer whorl. The three ventral bundles of the ovary enter directly into the petals, the two lateral ones giving rise to a trace which supports the anthers in the inner whorl. The main trace of the bract is also derived from one of the three incoming bundles. The presence of six undifferentiated bundles in the A postasioideae and Cypripedioideae, as contrasted with three or two in the monandrous orchids, is a further in- dication of the relative primitiveness of these two groups. Placentation. The ovary in the Orchid family is syn- carpous, either three- or one-carpellate (Plate IX). The three-carpellate condition, i.e. with axile placentation, is present in all species and genera of the 4 postasioideae, in Selentpedium and Phragmipedium of the Cypripedioid- eae and ina few genera of the Neottioideae (Lecanorchis, Kriaxis, etc.). The remaining genera of the Cypripedi- oideae and the monandrous orchids exhibit a monocarpel- late ovary, i.e. the placentae are parietal in origin. It is generally accepted that axile placentation is a more primitive condition than parietal, and its occur- rence in the Orchid family is limited to groups of re- spectively primitive status. Our knowledge with regard to the manner of transi- tion from axile placentation to the parietal type is very meager. ‘wo seemingly aberrant types, i.e. Phragmi- [ 65 ] pedium longifolium (Cypripedioideae) (P1. UX, fig. 8) and Lecanorchis javanica (Neottioideae) (Pl. LX, fig. 4), how- ever, may possibly represent independent steps in the reduction process. Theoretically, the placentae in Phrag- mipedium longifolium are parietal in origin because each of the two placental lamellae, although situated in adja- cent locules, are vascularized by a common strand (note the connecting dotted arrow). The placentation itself is intermediate between the axile and parietal positions. In Lecanorchis javanica, we find another type of modifica- tion of axile placentation. The septa are broken down into separate placental lamellae, and the torus is com- pletely eliminated. The individual lamellae facing the adjacent ventral bundles are united by the margin, thus leaving a Y-shaped empty cavity in the center of the ovary, and the ovules are borne in two rows at the point of junction of the lamellae. These intermediary steps suggest that the reduction from a tricarpellate to a monocarpellate condition might have come about by a longitudinal division of the septa which eliminated the torus, followed by a gradual short- ening of the lamellae (Limodorum abortivum, Pl. LX, fig. 5), until merely traces are found along the inner wall (Cephalanthera alba, Pl. IX, fig. 6). Embryogeny. No other plant family exhibits such an inconsistent embryogeny as the Orchidaceae. Only the first and second cell generations of the zygote are con- sistent; the subsequent divisions apparently take place in a random manner. In the first cell generation (Pl. X, fig. 1), the zygote divides into a basal and terminal cell. During the second cell generation, the basal cell differentiates into a suspen- sor initial cell and middle cell, while the terminal cell divides by a vertical wall. From this step onward, the further divisions are without any definite sequence, but [ 66 ] PuatTE IX J ium Chica Phragmipedium. | Lecanorchis . lo Hg if ollum Javanica Cephalanthera Limodorum - abortiwum alba the development may be oriented in two directions: 1. (Pl. X, fig. 1A) all cells including the suspensor initial cell enter into the formation of the embryo, and the ma- ture embryo is suspensorless; 2. (Pl. X, fig. 1B) the sus- pensor initial cell appears as a distinct structure, either simple or modified. ‘The suspensorless type of embryo is primitive and is to be found in Cypripedium of the Cypripedioideae and some members of the Neottioideae (Spiranthes, Listera, Neottia, ete.), while those with a suspensor are distrib- uted among the rest of the groups. Swamy, after having studied the embryogeny of a number of species, recog- nized five different types of suspensors (PI. X, fig. 2): Tyre I, with a single-celled suspensor, occurring in Cypripedioideae (Paphiopedilum) and also in the Neott- oideae (Vanilla, Epipactis, Goodyera, ete.); Tyrr II, which is unique in the family in developing a haustorium, is limited, so far as is known, to the Ophrydoideae. Tyres III to V are found in various representatives of the Kerosphaeroideae. There is a striking parallelism or correlation between the various types of suspensors and the relatively primi- tive or advanced stage of the main groups of the family. The mature embryo is an ovoid mass of cells without any definite differentiation of the tissues; thus, there is no endosperm in orchids. With respect to the method by which the undifferentiated mass of cells becomes or- ganized into the several organs of the embryo, no infor- mation is available as yet. Ina few instances, endosperm development has been reported, but in all of these this development is not consistent. Whenever it occurs, the tissue is of the nuclear type. This rare and casual occur- rence is noted in the Cypripedioideae and Neottioideae. Seeds. The mature embryo is enclosed in a loose, air- filled, reticulate testa which is characteristic of the Jcro- [ 68 | PuLaTE X 1# Cell 2% Cell | (Intermediate) _ Adult . Generation | Generation step) | Organization Basal cell 3 Prtial cell Os spermae (Pl. XI, A). At maturity the cells of the outer- most layer of the integument lose their protoplast, and the seed coat becomes transparent. This type of seed is observable in every genus of the family, except in A pos- tasia, Adactylus, Selenipedium and Vanilla (Pl. X1, B). In these four genera, all layers of the outer integument and most of the inner integument enter into the formation of the seed coat which tightly surrounds the embryo; the testa becomes highly sclerotic, opaque and sculptured. It is rather remarkable that the presence of a primitive type of seed in Apostasioideae, Cypripedioideae and Neottioideae corresponds to the respective status of these groups.’ Gynostemium. One of the most distinct features of the Orchidaceae is the fusion of the stamens and style into a central organ, the column. This structure has generally been interpreted as an extension of the floral axis, thus being axial in origin. Recent studies, however, have shown it to be only an appendicular structure, since the morphological apex of the flower does not extend to the apex of the gynostemium, but only to the level of inser- tion of the perianth; the ovary contains all traces of the floral whorls and the gynostemium of the reproductive whorls. Even today there is a constant debate about the mean- ing and application of the terms ‘‘column”’ or ‘‘gyno- stemium.’’ The group 4 postasioideae, from time to time, has been kept apart from the Orchidaceae as a distinct 1 It is noteworthy that, in addition to Vanilla, both Selenipedium and Apostasia possess aromatic substances. Epistephium and Galeola (Plate XI, A) of Neottioideae have also been reported to have essential oils in the fruits, but in lesser quantity. Kpistephium and Galeola have the seeds provided with a prominent, transparent wing (probably a dis- persal mechanism), but the embryo itself is enclosed by a sclerotic testa. Vanilla, Epistephium and Galeola p. pt. have been recognized as constituting a distinct family, Vanillaceae, on account of the charac- [ 70 | teristics mentioned above. PLatTE XI Goodyera pubescens Galeola altissima B a ener 7 ss de saa a anifolia family, because in some of its members the fusion be- tween stamens and style is only partial and the length of the adnation is relatively short. Indeed, in some species of Neuwiedia and A postasia, the filaments of the anthers are partially recognizable, but this character is not even constant within a given genus. In A postasia papuana (Plate XII), for instance, the filaments are completely obliterated or reduced to a mere connective tissue. In this respect Cranichis crumenifera (Plate XII) or any species of the genus Spiranthes or EHrythrodes of the Neottioideae might likewise be removed from the Orchid family, because structurally the column is quite homo- logous in these taxa. Should we express the differences between families by a degree of adnation of these organs in millimeters ? The same incomplete fusion is also observable in Va- nilla anomala (Plate X11) (Neottioideae) where the struc- ture of the column is comparable to that of Neuwiedia, although only one anther is fertile, while the other two are expressed by the traces of the filaments as staminodes. Vanilla anomala exhibits a further important feature, viz., the versatile anther. This character, along with others, has also been marshalled to support the separation of the A postasioideae into a distinct family. The occur- rence of a versatile anther is not limited to this single species in the Neottioideae; there are a number of other genera and species which possess the same characteristic : e.g. Cephalanthera, Psilochilus, Limodorum, Gatleola, Didymopleaiclla, ete. In other instances, the anther is basifixed and attached rigidly to the column. Vanilla Griffith var. formosana (Plate XII) clearly illustrates this situation, but a similar method of attachment is also present in several species of A postasioideae. In viewing the other columnar structures as depicted on Plate XII, our attention is focused on another point of [72 ] Apostasia papuana =, Aly LL DD ff Coryclum crispum Cranichis crumenifera alba \ a Cypripedium Vanilla IPP Calceolus VAY. LU Satyrium . saxicolum TX Gv 1g significance, the position of the anthers in relation to the stigmas. Both Apostasioideae and Cypripedioideae have been distinguished from the monandrous orchids by the adnation of the anthers to the style at a level below the stigma. This criterion, however, may not be applied as an absolute rule, as it has been in the past, because, in addition to these groups, an extensive number of genera (ca. 50) in the Neottioideae, as well as the whole Satyrium- complex of the Ophrydoideae, exhibit a position of sub- stigmatic insertion of the anthers. Therefore, the criterion of the occurrence of such an insertion, which was also applied to justify the removal of 4 postasioideae from the Orchid family, is invalidated. The striking structural similarity of the column in both Satyrium and Cypripedium points to the convergent na- ture of the respective branches of the main lines of Ophry- doideae and Cypripedioideae. Satyrium itself represents a departure from the general monandrous orchid type in having two distinct anthers developed in the outer whorl of stamens (PI. VIII, fig. 5). The fact that it is referred constantly to the monandrous orchids is, however, due to the nature of its pollinia. Rostellum. One of the most significant features in the organogenesis of the column is the formation of a new or modified structure, the rostellum. The theoretical ex- planation of the origin of this organ, as postulated by Brown and Darwin, is widely discussed in various text- books; therefore, it is sufficient if we merely state that the median stigma during the reorganization of the flow- er has evolved into a new organ, the rostellum, with a specific function. It is the controlling and ensuring de- vice for fertilization, since its position is located between the anther and the remaining stigmas; the pollinia are attached by a viscid secretion to the tip of this structure. Although this general situation is observable in the great [ 74 | majority of orchids, transitional stages, as well as com- plete absence of it, are also well documented. In 4 pos- tasioideae and Cypripedioideae all three stigmas are fertile ; therefore no rostellum is produced. In the remaining groups (Neottioideae, Ophrydoideae and Kerosphacroid- eae) it is assumed that a rostellum must be present. In some members of the Neottiotdeae (e.g. in the gen- era Spiranthes, Goodyera, Erythrodes, etc.) the style is modified into a wedge-shaped structure with the two separate stigmas situated laterally, while the third one is transformed into an elongate rostellum, all on the same plane. In this situation the nether surface of the rostel- lum is, however, still a functional stigma, as has been demonstrated by experimentation. We may look upon this condition as an intermediate step in the reduction or modification process, because in the more evolved members of the Neottioideae the rostellum ceases to be a functional stigma. There are also a number of species in the Ophrydoideae without any reduction in number of stigmas. Several attempts have been made recently to explain the pres- ence and origin in the Ophrydoideae of a so-called ‘‘ros- tellum”’ in addition to the three fertile stigmas. Since in the other groups (Neottioideae and Kerospheroideae) the gland of the pollinia is attached to the tip of the ros- tellum, it is believed that Ophrydoideae must also possess such a structure. Vermeulen postulated that the rostel- lum in the Ophrydoideae has an independent origin when all three stigmas are fertile, while Hagerup would derive the glands of the pollinia from the aborted lateral stamens of the outer whorl. I am unable to find a rostellum in the Ophrydoideae (comparable to that of the other groups): the structure which is generally called ‘‘rostellum’’ is merely a connective tissue between the two thecae of the anthers. Swamy in his studies of vascularization, has [ 75 shown that in Habenaria (Ophrydoideae) the compound stigma is supported by three vascular strands and in those species where the dorsal stigma is aborted, the supporting strand is obliterated simultaneously. Vermeulen’s sug- gestion that the rostellum in Bonatea (Ophrydoideae) represents an elongation of the receptacle is hardly con- vincing, because of the appendicular nature of the col- umn. To derive the viscid gland from the rostellum as a separate organ, or as a modified stigma, and the pollinia from the anther poses a situation rather difficult to com- prehend. In my opinion, the whole structure of the pol- linium originates as a unit from the anther; the gland itself is a transportation mechanism only. Furthermore, the nature of the anther and the pollinia in the Ophry- doideae is such that self fertilization is hardly possible. In those few species which are known to be autogamous, the presence of the connective tissue (whether or not rep- resenting a rostellum in reality) does not prevent self fertilization. Much research must yet be done with re- spect to developmental anatomy before a final conclu- sion as to the origin of the rostellum and the viscid gland may be drawn. Indeed, at this point, it makes no difference which of the proposed theories is correct, because each of them bears out the same conclusion: the column of the Ophry- doideae is not derived from that of any of the other groups, but it is the product of an independent evolu- tionary line emerging from a polyphyletic complex. Pollen. The pollen grains in orchids, at the time of shedding, are either single or more commonly united into tetrads, with or without aggregation into pollimia. In A postasioideae and Cypripedioideae, as mentioned above, no pollinia are formed, but the suleate or monocolpate microspores (similar to other Monocotyledons, e.g. Hy- powis) are always single at maturity. [ 76 | In Neottiowdeae, the pollen grains cohere loosely into sectile pollinia (first step in specialization); in the ad- vanced species, these are composed of microspores which are united into tetrads, while in the less evolved mem- bers (e.g. Cephalanthera, Aphyllorchis, Lecanorchis, Gal- eola, Pogonia, Cleistes, Epistephium, Vanilla, etc.), the pollinia are formed by single, either monocolpate, ulcer- ate or porate grains. In the Ophrydoideae, with granular pollinia (second step in specialization) and Kerosphaeroideae, with hard, compact pollinia (final step in specialization), the pollinia are always composed of tetrads, which, depending on the position they occupy — whether along the periphery or towards the center—may be one of the five basic types: 1. tetrahedral; 2. isobilateral; 3. decussate; 4. T-shaped, or 5. linear. The occurrence of single pollen grains and their aggre- gation into pollinia are in accord with the primitive or advanced status of the five main groups. Discussion I have attempted to demonstrate, in the diagram on Plate XIII, the correlation of the majority of criteria discussed in the foregoing paragraphs. The numbers be- neath each name summarize, out of the 16 selected char- acteristics, the essential constitution of each group, and the numbers along the lines connecting these groups are those of characters shared. It has previously been stated that the dynamic and 3- dimensional structure of nature cannot be projected into a 2-dimensional perspective without destroying, or at best distorting, its salient features. This statement is relevant also to the diagram on Plate XIII. For the proper interpretation of this projection, we have to visu- alize each group (shaded circles) as a 8-dimensional unit [77 ] and the numbers, connecting each of these groups, as representing independent trends which originate from and link the groups at different points. The sundry orchid systems proposed in the past were based essentially on progressive differentiation in a linear sequence, which assumes that the family is monophyletic in origin. Indeed, one’s first impression is of a linear se- quence from A postasioideae to Kerosphaeroideae, espe- cially when the groups are evaluated individually. When we attempt to assign a definite position and sequence to each of these groups, however, after studying their alli- ances, we find the linear arrangement to be rather absurd and unrealistic, since the groups are constantly in juxta- position with each other. The pattern expressed by these interrelationships is reticulate and indicate that the fami- ly, as a unit, is polyphyletic in origin. The general belief that, during the course of evolution, Neuwiedia (A postasioideae) gave rise to both Cypriped- toideae and the monandrous orchids, is hardly tenable. The so-called ‘‘clear cut’’ differentiation between Dian- drae (A postasioideae and Cypripedioideae together) and Monandrae is obliterated, as was mentioned earlier, by the presence of two anthers in the outer staminal circle of the Satyrium-complex in Ophrydoideae. In addition to this criterion, there are a number of characteristics common to both Diandrae and Monandrae which have been discussed above as well as documented on Plate XIII. L look upon each group as having an equal stand- ing with respect to its adjacent group, and, for the pur- pose of classification, | am recognizing each as a distinct subfamily.' Perhaps some phylogeneticists may object to such a conclusion, but we have to bear in mind that each group is rather well circumscribed in spite of the close interrelationship. ' For the descriptions of these subfamilies see Appendix. [ 78 | CYPRIPEDIO/DEAE 1-2-5-6-7-8-9-10-12-15 APOSTASIO/DEAE NEOTTIO/DEAE J-2 -5-(8)-10-12-14-15 © 1-3~5-6-8-9-10-11-12-13-(15)-16 KEROSPHAEROIDEAE | 1~3-4-6-9~11-13-16 Inferior ovary ; fusion of cand? organs 6 vascular bundles entering floral axis 3 vascular bundles entering floral axis 2 vascular bundles entering floral axis Axile placentation Parietal placentation Intermediate placentation Embryo without a suspensor OA DWAWHS OPHRYDOIDEAE 1-3-6~-9-11-13-15-16 Embryo with a_suspensor (types I-Y) 3 stigmas fertile 2 stigmas fertile, and rostellum formed Pollen grain simple Pollen grain compound 3 fertile anthers 2 fertile anthers 1 Fertile anther TIX Alvi As was stated earlier, 4 postasioideae is considered by most recent phylogeneticists to represent a distinct fam- ily and has consequently been removed to distant alli- ances, such as Haemodorales and Liliales by Hutchinson and Takthajan respectively. It is obvious that if Apos- tasioideae is removed from the Orchid family, this pro- cedure automatically demands also the separation of Cypripedioideae, as was suggested by Mansfeld, because of the absence of pollinia formation, different fertile stam- inal circles, ete. If A postasioideae and Cypripedioideae are removed from this interrelated complex, the remaining eroups, Neottioideae, Ophrydoideae and Kerosphaeroid- eae, would also require a new family status, because the equilibrium between these five groups is destroyed. On the other hand, the relationship between Cypripedium (Cypripedioideae) and Cephalanthera (Neottioideae) is so close that to regard them as members of distinct families would defy our whole evolutionary approach to syste- matics. When we study the origin and phylogeny of the Or- chid family devoid of paleobotanical documentation, our analysis is strictly limited to the uncovering of primitive features in living species. Since our approach is based a priori on such terms as genus and subfamily, these higher categories will aid our investigation only if they represent expanded dynamic units, although it is almost impossible to visualize the occurrence of such units in nature. The various evolutionary forces which have shaped and brought about the present-day orchids ob- viously did not act upon the family or even on a given genus, but rather on the species, because a species is the only tangible unit in nature with potentialities to mutate or evolve. Therefore, it would not be surprising if an absolute delimitation of a family becomes impossible. The structure of an orchid flower is definitively a de- [ 80 ] rived one, and several attempts have been made by sun- dry workers to visualize its prototype as having three free segments in each of the five whorls. This arrangement by itself suggests but one course of evolution: progressive differentiation in a linear sequence. The evolutionary makeup of the Orchid family, on the contrary, is indica- tive of a rather complicated origin, because in certain species characters of a primitive and advanced nature occur simultaneously. 3. Photomicrograph of the pachytene chromosomes in the micro- sporocyte of T’ripsacum australe Cutler and Anderson. Arrow points to the internal knob on the long arm of chromosome 8. 750 >» EXPLANATION OF THE ILLUSTRATION Prare XXII. Diagram of the 18 chromosomes in Tripsacum australe. The lengths, arm ratios, spin- dle fiber attachment regions (broken line), large chromomeres (small dots), and knobs (large dots) of the chromosomes are determined by actual meas- urements and observations at pachytene in the microsporocytes. Chromosome 16 has a nucleolar organizer (circle) in the long arm. The chromo- somes are divided into two groups, A and B: group A has the nine long chromosomes, group B, the nine short chromosomes. { 104 | r 68.8 66.0 57.1 52.5 45.3 2.8 2.0 ho.k 37.5 Arm ratio (Long/Short) 1.7 1.0 2.86 2.8 2.0 3.4 4.86 3.4 1.4 Chromosome 1 2 3 k S 6 7 8 9 Length (u) 34.1 32.6 31.0 26.6 23.1 22.0 16.0 13.2 10.8 1.0 10 2.9 11 2.5 12 1.8 13 1.5 14 303 15 1.2 16 2.7 17 3.8 18 IXX @vTd Sterility of the clone of T77psacum laxum is undoubt- edly due to irregular chromosome behavior at meiosis. Most of the resulting microspores receive more than the regular haploid number of chromosomes. There is no evidence that 7'ripsacum australe was in- volved in the hybridization of the aforementioned triploid form of T’ripsacum lawum; no marked chromosomes of the former were identified in the sporocytes of the latter. The form of diploid 7. australe investigated is apparently different, however, in chromosome constituents from the one previously reported on by Graner and Addison (1944), since they did not find any knobs in their material. CONCLUSIONS AND SUMMARY The tropical form of T’ripsacum laxum from Colombia is triploid with 54 chromosomes. The meiotic chromo- some behavior was extremely irregular, and this phenom- enon may serve as an explanation of the sterility of the plants. It is suggested that this tropical form of T'rip- sacum originated by interspecific hybridization between a tetraploid form of 7. /avum and an unknown diploid. A second tropical form of Tripsacum from Colombia, T. australe, is a diploid with 18 bivalents in the micro- sporocytes and 86 chromosomes in the somatic cells. Meiotic behavior of the chromosomes was regular. It apparently was not involved in the course of evolution of the triploid T'ripsacum lawum. Internal knobs which varied in size were observed in both of these tropical forms of T'’ripsacum. Also, in gen- eral, the short arms of the nine short chromosomes in both forms were more heteropycnotic than in those of maize. A diagram (Plate X XI) of the 18 chromosomes in Tripsacum australe is appended. The arm ratio and length of each chromosome are given. [ 106 ] ACKNOWLEDGMENTS The author wishes to thank Professor Paul C. Man- gelsdorf for his advice and encouragement during the course of this study. He also expresses his appreciation to Dr. W. H. Hatheway, staff member of the Rocke- feller Foundation Agricultural Program in Colombia, for providing the materials. [ 107 ] LITERATURE Anderson, E. 1944. Cytological observations on T'ripsacum dactyloides. Ann. Mo. Bot. Gard. 31: 317-323. Cutler, H. C. 1947. A comparative study of Tripsacum australe and its relatives. Lloydia 10: 229-234. Dodds, K. S. and N. W. Simmonds. 1946. A cytological basis of sterility in Tripsacum larum. Ann. Bot. n.s. 9: 109-116. Graner, E. A. and G. O. Addison. 1944. Meiose em T'ripsacum aus- trale Cutler e Anderson. (7. dactyloides subsp. hispidum Hitchcock. ) Anais Escola supr. Agr. “‘Luis de Queiroz’? 1: 218-224. Hernandez, Xolocotzi, E. y L. F. Randolph. 1950. Descripcién de los Tripsacum diploides de México: Tripsacum maizar y Tripsacum soptlotense spp. Nov. Ofic. Estud. Esp. Soc. Agric. y Grand. Fol. Téch. No. 4: 1-28. Longley, A. E. 1924. Chromosomes in maize and maize relatives. Jour. Agric. Res. 28: 673-682. —, 1987. Morphological characters of teosinte chromosomes. Jour. Agric. Res. 54: 835-862. Mangelsdorf, P. C. and R. G. Reeves. 1931. Hybridization of maize, Tripsacum and Euchlaena. Jour. Hered. 22: 329-343. —. 1939. The origin of Indian corn and its relatives. Bull. Texas Agric. Exp. Sta. No. 574: 1-815. Maguire, Marjorie P. 1957. Cytogenetic studies of Zea hyperploid for a chromosome derived from Tripsacum. Genetics 42: 473-486. —, 1960. A study of homology between a terminal portion of Zea chromosome 2 and a segment derived from T’ripsacum. Genetics 45: 195-209, Prywer L., Czeslawa. 1954. Meiosis en Tripsacum maizar (H. & R.). Revist. Soc. Mex. Hist. Nat. 15: 59-64. Randolph, L. F. and Barbara McClintock. 1926. Polyploidy in Zea Mays L. Amer. Nat. 60: 99-102. Reeves, R. G. and P. C. Mangelsdorf. 1935. Chromosome numbers in relatives of Zea Mays L. Amer. Nat. 69: 633-635. [ 108 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CAMBRIDGE, MAssAcHUSETTS, JUNE 30, 1960 VoL. 19, No. 5 PRESTONIA: AN AMAZON NARCOTIC OR NOT? BY Ricuarp Evans SCHULTES AND Rospert F. Rarraur i. THrovuGcuHout the literature concerning native narcotic plants of South America may be found the statement that Prestonia amazonica (Haemadyction amazonicum), an apocynaceous vine, is the basic ingredient in the hal- lucinogen known as yaje. But many specialists, including most field investigators, have attributed yaje to sundry species of the malpighiaceous genus Banisteriopsis. They have been in essential agreement that yaye (of the west- ernmost Amazon of Colombia, Ecuador and a part of Peru, especially along the eastern slope of the Andes), ayahuasca (of Peru, Bolivia and part of Ecuador) and caapi (of the northwestern Amazon of Brazil and adja- cent parts of Colombia) seem to be identical narcotics prepared from malpighiaceous plants. That a member of the Apocynaceae might be em- ployed in the Amazon as the source of a psychotomim- etic drink is vitally important and would not appear to be an improbability. It was for this reason that Schultes (28), in 1957, reviewed the history of reports concerning Prestonia amazonica, keeping a sharp outlook for any well documented and botanically supported record. He concluded that while Prestonia amazonica is frequently [ 109 | ‘‘named as the source of yaje and caapi.... there is little or no reliable evidence that this vine is ever em- ployed, at least as the prime ingredient, in preparing the narcotic drink. ’”’ A chemical study, published shortly thereafter by Hochstein and Paradies (11), seemed, however, to end all uncertainty. Entitled ‘‘Alkaloids of Banisteria Caapi and Prestonia amazonicum”’ [sic], it reported that “‘the hallucinogenic plant Banisteria Caapi contains in addi- tion to harmine, the alkaloids harmaline and d-tetrahy- droharmine’’ and that ‘‘ Prestonia amazonicum leaves have yielded another psychotomimetic amine, N, N-dimethy]- tryptamine.’’? The plant materials studied by the two chemists were collected by Mr. D. H. Allen who was engaged in commercial activity in the vicinity of Iquitos, Peru. Both the ayahuasca (which is identified as Banis- teriopsis Caapi) and the yaje (for which the determination Prestonia amazonica was offered) ‘‘were collected on the Napo River near Iquitos, Peru.’’ Hochstein and Para- dies state in a footnote that ‘‘the botanical identification was made by Dr. R. Ferreyra of the University of San Marcos, Lima.’’ They do note parenthetically that these two vernacular names have, in the past, been cited as representing the same species, Banisteriopsis Caapi. Nowhere in the paper, however, did the chemists state that voucher herbarium specimens, upon which a defini- tive identification could be based, had been sent in by the collector. Ferreyra (in /itt.) informed us that he was not aware of the existence of any herbarium material in con- nection with these identifications. Faced with the lack of voucher specimens, the botanist often, in an attempt to be as helpful as possible in guiding the chemist, suggests a tentative identification based on a vernacular name, and the botanist’s words of qualification are sometimes disre- garded. This is precisely what has transpired in the pres- [ 110 | ent case. We note that Hochstein and Paradies are care- ful to explain that the ‘‘second plant, ‘yage’. .. . was made available to us as an aqueous extract of the leaves. ”’ This statement, together with the knowledge that the identification was not based upon herbarium material, leads us to believe that the aqueous extract was sent in directly from the field. It is probable, therefore, that the identification was made by tracing the vernacular name yaje which, in much of the literature, has, for some inexplicable reason, often been referred to Prestonia amazonica. A significant observation from the chemist’s point of view was made recently, when Raffauf and Folger (22) stated that the “‘reported occurrence of only one simple indole in the Apocynaceae to date is of sufficient interest to warrant some speculation. The structure looks enough out of place to suggest that the sample studied was not Prestonia at all, and indeed, N,N-dimethyltryptamine was isolated from an aqueous extract of leaves, the bo- tanical origin of which appears to be in doubt. Confir- mation of the presence of this alkaloid in an authentic specimen of the plant is certainly necessary.’ ' Because of the fundamental importance of a thorough and detailed understanding of the botanical sources of the New World narcotics, it seems advisable to us, in view of the existing confusion, to review the whole his- tory of whether or not Prestonia is employed as an hal- lucinogen in the Amazon under the name yaje. In so doing, we realize that only further field work can be defin- itive. Such field work, nevertheless, should be attempted 'The possibility that an authentic specimen of Prestonia amazonica would yield N,N-dimethyltryptamine must not be excluded. Trypta- mine is recognized as a possible intermediate in the biogenetic path- ways to the harman-type alkaloids on the one hand (Malpighiaceae, Rutaceae) and many of the more complicated indole types (Apocyna- ceae) on the other. See R. Hegnauer, Planta Medica 6 (1958) 1). [111] with as clear a picture of the literature and other prior sources as possible. Il. There is a complex of narcotics, usually attributed to malpighiaceous species of the genera Banisteriopsis, Tetrapterys and possibly Mascagnia, which has three widely employed vernacular names. Caapi is the Nhen- gatu or Tupi-Guarani epithet used in the northwestern Amazon of Brazil and in the Comisaria del Vaupés in adjacent Colombia; according to Spruce (29), it is the word for ‘‘grass’’ and here means ‘‘thin leaf.’’ Ayahuas- ca, signifying ‘‘vine of the dead,’’ comes from Kechwa and is the accepted name of the hallucinogenic drink and its source plant in Peru, Bolivia and part of Ecuador. Yaje, a word of obscure linguistic origin and unknown meaning, is the name applied to the drink and the source plant along the eastern slopes of the Andes in Colombia and Keuador and in those parts of Peru near the Colom- bian boundary. In 1905, Rocha (25) published an account of his trip to the headwaters of the Rios Caqueté and Putumayo in Colombia and reported that the natives employed as a narcotic a ‘‘little bush’? or ‘‘leaves’’ called yajé. His account of its properties coincided very closely with those described for ayahuasca, and it was widely assumed that the two were identical as to the source plant. In 1923, the Colombian chemist Fischer (6) reported that the yajé which he had analyzed and which had come from the Colombian Caqueta might, on the basis of ana- tomic and histologic species, be a species of Aristolochia. Botanists who have worked in the Colombian Comi- sarias del Putumayo and Caquetd, where the drink is called yaje, agree that the prime ingredient is Banisteri- opsis. The German collector, Klug (19), studied yaye there in 1929 and found only Banisteriopsis employed. [ 112 ] The same is true of the botanist Cuatrecasas (4), who studied the narcotic in the same area in 1939. The Co- lombian botanist Garcia-Barriga (8), who has met with yaje in this and other areas, mentions only Banisteriop- sis as the principal ingredient. Schultes has seen yaje prepared and has partaken of it on anumber of occasions in the Putumayo and elsewhere in Colombia and has not seen used as the basic plant anything but a species of Banisteriopsis. The Russian botanists, Varanof and Juzepezuk, who studied the problem in the Colombian Caqueté in 1925-26, likewise found several species of Banisteriopsis employed either alone or together in pre- paring yaye (9, 10). In 1956, the Brazilian chemist Costa (8) on the basis of isolation of the alkaloid yageine from species of the Amazon basin identified yaje with Banisteriopsis Caapi. A suggestion that ayahuasca and yaje might be differ- ent plants seems first to have been advanced by the French anthropologist Reinberg (23). His study of tribes living between the Rio Napo and Rio Curaray in Peru led him to publish in 1921 the statement that the nar- cotic drink was an infusion of a few fragments of aya- huasca, a liana the diameter of a man’s thumb, and leaves of yqje, ‘‘un petit arbuste, de Im. 50 de haut, 4 feuilles petiolées (petiole de 15 mm.), entiéres, ovales, longues de 20 cm., larges de 7 cm., reguliéres et terminées par une pointe de 2 cm.’’ This description, of course, could very accurately be applied to a species of Banisteriopsis. Reinberg reported that his determinations were based upon specimens, but a search in the herbarium at Paris failed to disclose the existence of any herbarium material at the present time. He held that his specimens showed that ayahuasca and caapi were conspecific, representing Bamnisteriopsis Caapi, but that the yaje of the Rio Cura- ray could, with reservations, be referred to Prestonia [ 113 | (approaching, in some respects, P. amazonica) or a re- lated apocynaceous genus. In 1922, the Belgian horticulturist-explorer Claes (1,2 studied yaje amongst the Koregwahe Indians of the Co- misaria del Caqueta in Colombia. He learned that the yaje, hitherto usually described as a ‘‘small bush’’ was an enormous forest liana, and he argued quite justifiably that those who had described it as a small bush had seen only young, cultivated individuals and not the vine in its wild state. Claes did not offer a botanical determination of yaje. He mentioned, however, that the Belgian bota- nist De Wildeman believed that it ‘‘might be’? Pres- tonia amazonica. This would be most unusual, since, so far as we know, Prestonia amazonica does not become an enormous jungle liana. No voucher specimens were lo- cated in the Rijksplantentuin in Brussels, and Claes him- self stated that he had not obtained material for identify- ing yaje. With this statement, we must assume that De W ildeman was voicing an opinion which he based on the information he was able to glean in the literature through the use of the vernacular name. The pharmacologists Michiels and Clinquart (17) worked on the stems which Claes had collected. In 1926, they suggested—whether from their own observa- tions or from the opinions of De Wildeman we do not know—that the stems appeared to belong to Prestonia amazonica. The same year saw the French pharmacolo- gist Rouhier dismiss as ‘‘doubtful’’ the possibility that yaje could be referable to Prestonia amazonica. In 1980, the French botanist Gagnepain (7) tried rather unsuccessfully to put some order into the chaos. He pointed out that 1) according to Reinberg, ayahuasca was ‘‘probably”’ Banisteriopsis Caapi but that yaje could not be referred to this species ; 2) yaje seemed to approach Prestonia amazonica; 8) fragments received as yaje by [114 ] Rouhier in 1924 showed the plant to be an ‘‘opposite- leaved vine’’; and 4) both Reinberg and Rivet sent in material which seemed to represent the same malpighia- ceous plant. Somewhat later, Gagnepain received through Rouhier a specimen from the Departamento de E] Valle in Co- lombia, where the plant had been cultivated under the name of yaje. The provenience of the plant was un- known. The specimen had leaves and inflorescences. When Gagnepain discovered that it represented Banis- teriopsis Caapi, he arrived at a most astounding conclu- sion: that yaye of Colombia was the same as caapi of Brazil but was not the same as yaye of Ecuador. He asserted that Ecuadorian yaje represented a different spe- cies of Banisteriopsis than did Colombian yaye. Thus, he appeared to drop Prestonia amazonica as the source of any yaje. Most recently, Fabre (5), in reviewing the historical aspects of the identification of ayahuasca, caapi and yaje, concluded that all three are prepared basically from Ban- wsteriopsis, even though other plants may be used as ad- ditives. Several non-botanical workers, without the benefit of voucher specimens, accepted Prestonia amazonica as the source of the narcotic. Their ‘‘identifications’’ served to focus attention in the literature on the apocynaceous plant without really adding anything new of basic value. Reutter (24), for example, reported in 1927 that he had isolated yageine and yagenine from the vegetal parts of yaje or ayahuasca, which he referred to Prestonia ama- zonica. In a dictionary of Amazon plants, LeCointe (13) indicated his belief that the botanical sources of ayahuasca and yaje were two different plants, pointing out that some writers attributed yaje to Prestonia amazonica. In 1936, Pardal (19) stated that caapi was Banisteriopsis [ 115 ] Caapi and yaje was Prestonia amazonica; but the follow- ing year, he (21) attributed both to Banistertopsis Caapt. The German toxicologist Lewin (14,15,16) named Pres- tonia amazonica as one of the plants possibly employed as an admixture with Banisteriopsis Caapt. In 1947, Sandeman (27) mentioned yaje casually and referred it to Prestonia amazonica. Most recently, the chemists Mors and Zaltzman (18), arguing that the alkaloid yageine is different from harmine, concluded, on the basis of a re- view of the literature, that caapi and ayahuasca are refer- able to Banisteriopsis Caapi but that yajye was not the same plant. IL. All of the reports concerning the use of Prestonia amazonica as a narcotic stem directly or indirectly from the work of the British plant-explorer Richard Spruce. Spruce’s meticulous field notes were written down in 1852 but did not see publication until, after his death, they were edited by Wallace and published in book form in 1908 (29). Spruce was the first to identify the source of the caapi drink of the Rio Uaupés in northwestern Brazil as a spe- cies of Banisteriopsis. 1t was a new species and was origi- inally described as Banisteria Caapt. ‘The correct name is now Banisteriopsis Caapi. The description of this new malpighiaceous species was based upon a flowering speci- men collected by Spruce himself. In his notes, however, Spruce stated that there was another kind of caap? in the same region and that it was called caapi-pinima or ‘‘painted caapi.’’ In his original field notebook, preserved at the Royal Botanic Gardens at Kew, we find the fol- lowing entry under ‘‘2712. Banisteria Caapi Mss. From this is prepared an intoxicating drink known to all the natives on the Uaupés by the name of caapi. The lower part of the stem, which is the thickness of the thumb [ 116 ] swollen at the joints, is the part used. This is beaten in a mortar with the addition of water and a small quantity of the slender roots of the Apocynac. (apparently a Hae- madictyon) called caapi-pinima or painted caapi, from its Ivs. being stained and veined with red.. .. Query? May not the peculiar effects of the caapi be owing rather to the roots of the Haemadictyon (though in such small quantity) than to the stems of the Banisteria? The In- dians, however, consider the latter the prime agent, at the same time admitting that the former is an essential ingredient. The two plants are planted near all mallocas (villages)... .”’ When these notes were published in Spruce’s ‘‘Notes of a botanist on the Amazon and Andes’’ (29), they suf- fered a slight change of emphasis. The terms of qualifi- cation disappeared. The published version states that caapi-pinima “‘is an apocynaceous twiner of the genus Flaemadictyon, of which I saw only young shoots, with- out any flowers. The leaves are of a shining green, painted with the strong, blood-red veins. It is possibly the same species. . .. distributed by Mr. Bentham under the name Haemadictyon amazonicum n. sp. It may be the caapi- pinima which gives its nauseous taste to the caapi. .. . and itis probably poisonous. ... , but itis not essential to the narcotic effect of the Banisteria, which (so far as I could make out) is used without any admixture by the Guahibos, Zaparos and other nations out of the Uaupés.”’ Spruce was one of the most meticulous of all scientist- explorers of South America. A less careful and botani- cally untrained observer might easily have confounded the young shoots of a Prestonia with Banisteriopsis, for the leaves of both are opposite, and the leaves of some species of Prestonia do resemble remarkably those of Banisteriopsis in shape and texture. But Spruce could never have confused an apocynaceous plant, full of a [ 117 ] white latex, with a Banisteriopsis. He might have erred as to genus, for the genera of the Apocynaceae are often hard to distinguish even with flowers. But even this pos- sibility would seem, in the case of Spruce, to be rather remote. Schultes, on his long collection trip along the Colombian and Brazilian course of the Rio Vaupés, searched for an apocynaceous vine growing around In- dian huts, as described by Spruce; although every Indian Manihot-plot boasts its several cultivated plants of Ban- isteriopsis, nothing resembling a Prestonia was ever seen under cultivation. A careful reading of Spruce’s notes reveals the fact that he never claimed more for Prestonia or caapi-pinima than the role of a plant used as an admixture. We know from the reports of later workers that other plants are sometimes added in minute amounts to the drink pre- pared from Banisteriopsis in the belief that they change the attributes or properties of the narcotic drink. Schultes (28) reported the admixture of leaves of an apocynaceous tree, possibly Malouetia Tamaquarina, amongst the Ma- kuna Indians of the Rio Popeyaca of Colombia. Later writers, without herbarium specimens to back their claims, and taking their cue from Spruce whose notes they misread or misunderstood, have proposed that the narcotic drink in one part of the Amazon where it is known as yaje is prepared exclusively from Prestonia amazonica. For this assertion there is absolutely no basis in field work. Prestonia amazonica is known from only one collection, the type collection made by Spruce in 1859 at Trombe- tas on the lower Amazon. In more than a century, the species has never been found again. We are forced, con- sequently, to believe that Prestonia amazonica is either a very rare species or else a strict endemic, confined to the general area where the type was collected. The Rio [118 ] Trombetas lies more than 1200 miles in a straight line from the eastern slopes of the Colombian Andes, where we are expected to believe that the natives are using this rare species in relative abundance as the source of their frequently employed yaye. The chances that Prestonia amazonica is used are, for all practical purposes, non- existent; and there seem to be no indications that any species of Prestonia is so employed along the eastern slopes of the Colombian and Ecuadorian Andes. Even in the area where Spruce reported its possible use a hundred years ago, there is all probability that it was employed solely as an admixture with Banisteriopsis Caapi. In the region through which the Rio Vaupés flows, the Indians distinguish two kinds of caapi. Spruce re- ported the minor caapi to be called locally caapi-pinima. Koch-Griinberg (12) found that the Tukanos of the Vaupés know two kinds, but he could identify only one. In 1948, Schultes (28) discovered the Indians on the Rio Tikié, a Brazilian affluent of the Uaupés, preparing a narcotic drink from the malpighiaceous genus 7'etrap- terys. He described the new species Tetrapterys methy- stica on the basis of a flowering specimen from the forest liana. From the bark a definitely hallucinogenic drink was prepared. The drink was rather yellowish, unlike the usually chocolate-brown of the drink prepared from Banisteriopsis Caapt.. One wonders whether or not the term ‘‘painted caapi’’ could be applied to the kind of caapi that makes the unusual yellow drink. Be that as it may, the drink prepared from Tetrapterys represents probably the second kind of caapi reported by Koch- Grinberg in 1909. SUMMARY While we are careful to point out that further field work, especially in Spruce’s area along the Brazilian Rio [119 ] Uaupés, should be encouraged partly in an attempt to locate the use in caapi of a species of Prestonia, we be- lieve that enough is known at the present time to make, in summary, the following statements: 1) 2) 3) There is no botanical support, nor any reliable sup- port in the literature, for the assumption that any species of Prestonia (least of all Prestonia amazon- ica) is employed as the prime ingredient in the prep- aration of ayahuasca, caapi or yaje. There is no reliable reference except Spruce’s that any species of Prestonia is employed even as one of the minor ingredients or admixtures with Ban- isterlopsis. There is serious doubt that the indole N, N-dime- thyltryptamine occurs in Prestonia and every prob- ability that the recent report of its presence in this genus was due to an erroneous identification of the material under analysis. [ 120 ] 10. 11. 12. 13. 14, REFERENCES eeé ° . Claes, Florent. Quelques renseignements sur les coutumes des corregujaes de Colombie.’’ Bull. Soc. Americ. Belg. (Dec. 1931) 39. . Claes, Florent. “‘Chez les hiutotos et correguajes.’’ Bull. Soc. Roy. Belg. Geogr. 56 (1932) 25. . Costa, Oswaldo de A. Rev. brasil. farm. 37 (1956) 481. Cuatrecasas, José. “‘Prima Flora Colombiana. 2. Malpighiaceae.”’ Webbia 23 (1958) 343. Fabre, René. ““Quelques plantes médicinales de 1’ Amerique La- tine: leur utilisation thérepeutique.’’ Rev. Gen. Sci. Pures Appl. 62 (1955) 49. . Fischer, Cardenas, G. ‘‘Estudio sobre el principio activo del yagé.”’ Unpubl. thesis, Univ. Nac., Bogota (1923). Gagnepain, F. ‘‘Observations sur le yajé.’’ Rev. Bot. Appl. Colon. 10 (1930) 293. , ° 66 . e . . Garcia-Barriga, Hernando. — El yaje, caapio ayahuasca.’’ Univ. Nac. Col., no. 23 (1958) 59. . Hammerman, A. F. Bull. Appl. Bot. Gen. Pl. Breed. (1929) 165. Hammerman, A. F. “‘Le yagé en Amazonie.’’ Rev. Bot. Appl. Agric. Colon. 10 (1930) 600. Hochstein, F. A. and Anita M. Paradies. ‘‘Alkaloids of Banis- teria Caapi and Prestonia amazonicum.’’ Journ. Am. Chem. Soc. 79 (1957) 5735. Koch-Griinberg, Theodor. ‘‘Zwei Jahre unter den Indianern.’’ 1 (1909) 298. LeCointe, Paul A. ‘‘A Amazonia brasileira III. Arvores e plantas uteis.’? (1934) 70; 471. Lewin, Louis. “‘Sur une substance envirante, la banisterine, ex- traite de Banisteria Caapi.’? Comptes Rend. Acad. Sci. 186 (1928) 469, . Lewin, Louis. ‘‘Gifte und Vergiftungen.’’ Ed. 4 (1929) 687. [121 ] 16, i 18, 19, 20, 24, 26. 27 28. 29. Lewin, Louis. ““‘Phantastica— narcotic and stimulating drugs.’’ (1981) 140, Michiels, M. and E. Clinquart. “‘Sur les réactions chimiques d’ identification de la yageine.’’ Bull. Acad. Roy. Méd. Belg., s. 5, 6 (1926) 19. Mors, Walter B. and Perola Zaltzman. “‘Sdbre o alkaldide de Banisteria Caapi Spruce e do Cabi paraensis Ducke.’’ Bol. Inst. Quim. Agric., no. 34 (1954) 17. Morton, C. V. “‘Notes on yagé, a drug plant of southeastern Colombia.’’ Journ. Wash. Acad. Sci. 21 (1931) 485. Pardal, Ram6én. “‘Las drogas estupefacientes e iluségenas del indio americano.’’ Rev. Geogr. Amer. 3 (1936) 1. Pardal, Ramon. ‘‘Medecina aborigen americana.’’ Humanior, Sece. C, 3 (1937) 278. Raffauf, Robert F. and M. B. Folger. ‘‘Alkaloids of the Apocyn- aceae.’’ Econ. Bot. 14 (1960) 37. Reinberg, P. ““Contribution A l’étude des boissons toxiques des indiene du Nord-ouest de l’Amazone, l’ayahuasca, le yajé, le huanto.’’ Journ. Soc. Americ. Paris, n.s., 13 (1921) 25; 197. Reutter, [K.]. ““Du yagé ou aya huesca.’’ Schweiz. Apotheker Zeet. 25 (1927) 289. Rocha, Joaquin. ‘‘Memorandum de viaje (regiones amazonicas).”’ El Mercurio (Bogota) (1905) 43. Rouhier, A[lexander]. ‘“‘Les plantes divinatoires.’’ Rev. Méta- psych. (1926) 325. Sandeman, Christopher. ““Thyme and bergamot.’’ (1947) 24. Pd Schultes, Richard Evans. ““The identity of the malpighiaceous narcotics of South America.’’ Bot. Mus. Leafl. Harvard Univ. 18 (1957) 1. Spruce, Richard. “‘Notes of a botanist on the Amazon and An- des’? [ed. A. R. Wallace]. 2 (1908) 414. [ 122 ] A REPUTEDLY TOXIC MALOUERTIA FROM THE AMAZON BY RicHarp Evans ScHULTES ALONG the inundable forests of the uppermost Amazon River, one of the common understory trees is the apocy- naceous Malouetia Tamaquarina. This species I found to be especially frequent along the small Amazonian tributary, Rio Loretoyacu, in the Colombian ‘“Trapecio Amazonico,’’ inthe vicinity of Leticia. It first attracted my attention during a study of Hevea and other lacti- ciferous trees of that region in 1944, since its copious latex was often added to Hevea-latex as an adulterant. What interested me most in my study of Malouetia Tamaquarina was the reputation which the fruits have as a poison. This reputation is widespread in the Leticia- area, and the many reports which I heard during my three-year stay in Leticia agree strictly in details. According to the natives, the ripened fruit of Malou- etia T'amaquarina is consumed by the pajuil (Nothocrax urumutum (Spix)), a wild bird frequently seen under domestication in this part of the Amazon. The flesh of the pajuil is a great delicacy which may be eaten at any time of the year. During the months of March through June, however, when Malouetia Tamaquarina is in fruit, the bones of the bird must not be thrown to the dogs, lest they poison the animal. This poisoning is of a curious kind: it causes immediate and violent upsetting of the digestive tract and, within four or five hours, a glassy- eyed stare and interference with normal muscular coordi- nation of the legs. It sometimes may be fatal. A recent study of apocynaceous alkaloids (Raffauf, R. F. and M. B. Flagler in Econ. Bot. 14 (1960) 37) indi- cates that alkaloids have not been reported from Malou- etia. | have not seen this poisoning reported in the litera- ture, and I encountered it only in the Leticia-area. An [123 J incidental report appeared in one of my previous papers (Schultes, R. E. in Bot. Mus. Leafl. Harvard Univ. 16 (1953) 90). There would be every reason to give some credence to the reports because the A pocynaceae or Dog- bone Family is known to have highly toxic members. Malouetia Tamaquarina is called cuchara-caspt (‘‘spoon tree’’) in the Leticia-area, as the soft wood was formerly whittled into spoons. Some local rubber tappers refer to it as chicle. The tree may attain a height of fifty feet, averaging between thirty and forty. The usually straight, cylindrical trunk, with a diameter of twenty inches, is covered with a brownish or ashy-purple bark. The crown is light and irregular. The tree blossoms profusely, bear- ing white to yellowish, fragrant flowers. The free-flowing latex has a sweet flavor but causes a slight burning of the tongue. The wood is soft and white. There are several closely related species of Ma/louetia, and these may be similarly poisonous. Malowetia nitida Spruce is reported to be used as an arrow poison. Hare, H. A., B. Caspari and H. H. Rusby ‘‘National Dispen- satory,’’ Ed. 2 (1908) 213). The leaves of what appears to be Malouetia Tamaquarina are sometimes added to the narcotic drink prepared from Banisteriopsis Caapt in the Colombian Vaupés (Schultes, R. E. in Bot. Mus. Leafl. Harvard Univ. 18 (1957) 39). Malouctia Tamaquarina extends from the Guianas across the northwestern Amazon of Brazil, Colombia and Peru. A recent segregate has been described: Malouetia peruviana Woodson (in Ann. Mo. Bot. Gard. 22 (1985) 259), but the differences seem to be trivial. Ma/ouwetia furfuracea Spruce of Amazonian Peru is likewise known by the vernacular name cuchara-caspi. An attempt to study Malouetia Tamaquarina or a related species chemically would seem to be worthy of consideration. Cotomsia : Comisaria del Amazonas, Trapecio Amazoénica, Rio Lore- toyacu. Altitude about 100 m. September—November, 1944. Richard Evans Schultes GO34; GO83; 6112. [ 124 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CampripGr, Massacuusetts, JANUARY 9, 1961 VoL. 19, No. 6 HOW WERE THE GLASS FLOWERS MADE? A Lertrer py Mary LEE WarE A Worpb or ExPLANATION The Ware Collection of Blaschka Glass Models of Plants — popularly called the ‘‘Glass Flowers’? —is un- doubtedly the most widely known and appreciated public attraction at Harvard University. An estimated third of a million people annually visit the Botanical Museum where they are housed. The question which visitors most frequently ask is: ‘How were these beautiful flowers made?’’? Another query often heard ts: ‘‘Has the secret been lost —did it die with the makers?” The truth is that there was no secret process employed by Leopold Blaschka and his son, Rudolph, the creators of the *‘Glass Flowers.’’ Every techniqne used was known to glass workers of the period. According to Mr. Louis C. Bierweiler, former Curator of Botanical Collections at the Museum and for more than fifty years custodian of the models, Rudolph Blaschka expressed to him his regret that many people thought that his handiwork utilized secret processes; he insisted that his work represented artin which there is no room for secrecy or egoism. Although there is no complete information on all steps in the manufacture of the models, we do have the descrip- tion of part of the work in a letter from the late Miss Mary | 125. | Lee Ware of Boston to Professor Oakes Ames, then Di- rector of the Museum. This letter not only contains signi- ficant remarks about the technique involved in the glass work but gives an intimate picture of the artist, Rudolph Blaschka, and his wife in their home. It was written on her last visit to the Blaschkas in 1928. Miss Ware, and earlier her mother, Mrs. Elizabeth C. Ware, financially supported the botanical work of Leopold Blaschka from 1877 to 1895 and subsequently Rudolph until his retirement in 1936. In 1898, the Wares presented the collection to the President and Fellows of Harvard College as a memorial to Charles Eliot Ware, M.D., of Boston, a member of the Class of 1834. Miss Ware’s letter is herewith reproduced almost in its entirety and with only minor editorial alterations. The original is preserved in the Botanical Museum. ih.dt.o. [ 126 ] Dresden, Oct. 3rd., °28 Dear Proressor AMES, It seems easiest in this very long letter to separate the description of the glass work itself from the more per- sonal part; so I have done this, and I will tell you about making the models later. I have been out to Hosterwitz, half an hour by auto, alone four times and have passed the whole afternoons, long ones, looking first at the models. It took two afternoons to see them. Then I in- spected the work room and its contents and was shown all the great improvements made by the Blaschkas in the house since his marriage, and finally accepted their hospi- tality of tea and delicious cakes, a far better arrangement than the old one of spending a whole day and having two solid meals! Both Mr. and Mrs. Blaschka received me most cor- dially at the little garden gate, and we looked at each other to see what time had done; that first day, I was daunted to see what seemed a little old man, legs that were not strong, very rounded, stooping shoulders and an exceedingly white face. He must have dropped nearly two inches in height, his hands were somewhat out of shape from rheumatism and were very trembling. How- ever, I came to the conclusion that this was due partly or largely to excitement at seeing me again and anxiety as to what I should say about the models. When I was ready to leave and said ‘‘Aufwieder- sehen,’ he drew himself up with quiet dignity and said, ‘Well, Miss Ware, are you satisfied ?’’ I said, ‘‘Yes, Mr. Blaschka, 1 am more than satisfied, and I do not see how anyone could feel otherwise.’’ He looked intensely re- lieved at once, and the next time I went out, his color had returned, and, when she saw him, Miss Niklason thought he looked well and strong, barring the stoop f 127 |} which, | suppose, is inevitable with such sedentary work. His eyes are simply marvelous, piercing; and yester- day, while painting a leaf, he worked most of the time without glasses, and he is seventy-one! He speaks freely of his age, of the work which can possibly still be aeeom- plished and of the fact that he is the only one in the world who ean do it; which I think is true. You would never find another man who combined the scientific knowledge of many years’ study of plant and animal life; the study of glass, its component parts and its possi- bilities, not merely book knowledge but derived from experiment as well; together with the power of concen- tration, mental and moral; the artistic ideal as a load- star which has enabled him to forego everything called pleasure, except his wife. She is sweet and devoted to his ideal, too, has softened and broadened him, and so humanized him that he is much better fitted to come in contact with the world than the Rudolph Blaschka who came to America in 1896. He is just as modest and absolutely honorable as he ever was, but now he has a sense of his own worth, his own unusual force of intellect and character; and there is everything to justify that. I asked him one day whether he was still applied to for models to be kept in Germany, and he said, ‘‘Oh yes.’ Professor Neumann, to whom he went for something at your suggestion, asked why he would not give them some of his work. I asked what he replied. ‘‘Oh,’’ he said, ‘‘I told him that I worked for Harvard University, that I was a man of absolute honor so would make no change and was satisfied.” I found that Sunday was his one day of rest and, sus- pected, from previous knowledge, that he was not taking enough time for air and exercise and was perhaps work- ing late in the evening. I extracted the information that lately he had taken little or no time for fresh air and had [ 128 ] been working evenings, sometimes till midnight. I re- monstrated vigorously and told him he must not do so, but he only said Professor Ames wanted the models and tho’t him very slow — that it was impossible tor him to do such work any faster, that no man could. It had evi- dently worried him much; and [I had to work hard to reassure him that you would not feel so if you knew him and could see him work, that [ would explain to you and that he absolutely must stop evening work and get the necessary air and relaxation to keep himself in good condition. He only regrets that all the groups of fungi, ete., are not complete, but he has had to do them as he could and when he could get the specimens, depending more or less on seasons, weather, etc. He hopes to send off some twenty-five models, sprays or plants with their sections, etc., by the middle of the month — I think it will be nearer the end, but perhaps not. I believe this includes all the pears, except the blossoms, and strawberries : whether more I do not know. Apples, plums, apricots peaches, cherries, are for the most part finished and ready for the sprays, with the leaves ready to paint, and the exquisite fruit blossoms ready for the branches. He thinks the rest can be shipped in the spring, some thirty more. The fruits are not so beautiful to look at as the flowers are, but they are marvellous, and how one man can sit hour after hour, putting in the gossamer veinlets, or all the’myriad little dots and irregular brown patches, passes my understanding; you would say that years could not do it, or a lifetime. If he hurried or worked quickly. he would be insane. I sat and watched his movements as he worked. The table is covered not only with implements but with trays of leaves, formed but not colored, bottles in which he ‘an stand the glass stems with leaves while drying or (129) cooling, specimens of fungus-covered fruits or dried leaves for use as guides (for the most part he studies them without glasses), bottles with powdered glass for use as needed, and saucers for the enamel paints he makes of powdered glass. In spite of the slightly unsteady hand, his movements are quiet, deft, soft in laying down or taking up where speed or a miscalculated movement might ruin the work of hours. It is breathless to watch. The first afternoon, I saw the pears in every stage of disease, sometimes the fruit only, sometimes leaves and branches also. The moulds were wonderful, and I think you will be delighted with them all, but, of course, I know nothing of fungi. He had magnified 250 times a section of mould which comes on bread — remarkable. The strawberries were fascinating — plants, fruit and moulds: also the result of frosts on the developing fruit. The apples, also, were good; the peaches in their present stage of development seemed to me less so, but the final varnish was not yet on, and they all looked rather glassy. It all leaves you breathless that anyone can and will do such work. Mr. Blaschka’s head and bearing are very expressive, and L wished I could catch a photograph of his profile as he stood for a few moments, a plaque with a model on it held in both hands. His whole expression of absorbed, concentrated study was worth keeping, had it been possi- ble. His own garden is small but full of fruit trees from which he gets some of his material for work: and the rest he gets from large fruit orchards near by. He also has books on fungi and mushrooms, and is looking forward sagerly to the work for Professor Weston. He says the lamellae will be very difficult but that he can make them. In view of his advancing years and the uncertainty of health, I would suggest that Professor Weston make a list of those subjects which he wants most, those which [ 130 ] Rupo.en BLAsSCcCHKA June 17, 1857— May 1, 1939 Photograph taken in 1938 are most important, although, on account of weather, season, and materials, it may be impossible for Mr. Blaschka to follow the sequence exactly; still, it would be a guide. It troubles me very much that he and his wife cannot come over to see his life’s work now that you have the models so beautifully arranged, and he looks so eager and pathetic when I describe the mise en scene. ‘They are very simple, unaffected, dignified people, and I hope some- time that I may be able to manage it if T can only keep well when I come home. It seems cruel not to, but, of course, they could not travel in our country on what they have. I find that he did not lose all his investments in real estate or mortgages, ete., but a// in government investments. One change in the character of his work and, conse- quently, in the time necessary to accomplish results since I was last here is very noteworthy. At that time, he bought most of his glass and was just beginning to make some, and his finish was in paint. Now he himself makes a large part of the glass and a// the enamels, which he powders to use as paint. This he considers to be practi- so that, if we could ‘ally indestructible, except by force come back ina thousand years, we would find form and color as today. He has dozens and dozens of little bottles with colored powders and little boxes labeled with colored enamels which he makes himself, and powders for paint. The colored enamels are beautiful and fascinating. Some pieces he exposed on his roof or under the eaves for over a year, winter and summer, and they did not change in any way!. My last visit to Hosterwitz on October 6 was most happy. Miss Niklason went with me and enjoyed it as much as I did. Supper was excellent, informal and pleas- ant, and I regaled them with all the Museum gossip that r 132 ] OL I could think up. Mr. Blaschka did some leaf work again and Miss N. felt just as I do, that it is a great experience to watch that man at work. His whole head and hands are a study, and he worked until it was about dark with- out turning on his electric light. She also felt that the work was enough to wear anyone’s nerves to madness, the confined position and closed room being a part of it. I told him again that he must stop evening work with late hours and must get out for air and exercise for a time every day, that it was not fair to his wife or him- self, that it made no difference how long it took to com- plete the models and that I should tell you that I said so. She is 45 only and perfectly devoted to him, but nobody can keep fresh without a little fun. She said they used to come in to Dresden sometimes when first mar- ried but it had been a long time since they were there. I tried to get them in to supper and the opera but un- fortunately had to give it up on account of a cold. I know it has given him a fresh start and fresh courage to see me. I have been out there five times and I am sure that I accomplished what I came for. I wish I could have run in oftener, for so few realize what he is doing and their lives are necessarily secluded tho’ evidently they are on friendly terms with their neighbors. And now a word about the way in which he works. I watched while he painted a peach leaf affected by a fungus. {ach leaf is formed of clear white glass, pulled and worked by simple instruments in the flame, and each point on the margin has to be pulled out separately from the hot glass, to make the crenate edge to the degree characteristic of the species. The leaf remains attached to the long stem of glass, 12 inches perhaps, from which it has sprung, until the coloring is completed and annealed. Then it is separated [ 133 ] and the fine wire, necessary for the permanent stem, is attached, coated with glass, and the leaf is ready to be attached to the branch; this last I could not see, as it took from about three till half after five to color just three leaves and put ribs and veins in one, and then anneal it. A green leaf would be made of green glass; the method of coloring would be the same for both. The colors are made of powdered glass mixed (mois- tened) with a few drops of turpentine or. . . . carefully added to the powder in little china saucers and stirred with a fine camel’s hair brush, which, finely pointed in the moist paint, is used to administer the color to the leaf also. The undulations of the leaf have already been made in the white glass so the buff or yellow paint is brushed on perfectly smooth, several times, and, before it has wholly dried, a strip of pointed whale bone marks the main vein down the centre, and the pointed quill of one of the brushes is used to mark each rib. Then a little of the powdered glass is dusted on by a camel’s hair brush, shaken off and dusted on again in spots. If the leaf is partially healthy, faintly colored green glass is applied to the healthy parts. The camel’s hair brushes, larger or finer, then are drawn down the vein and the ribs over and over and over again to give the necessary strength to the vein of size and color and to emphasize, as needed, the ribs. Many, many times the delicate tip of the brush would only touch tiny spots on the lines which needed a thought more of color. Then, with the most delicate touch of the finest pointed brush, the cob-web veins were drawn into the texture of the leaf, between and at the end of each rib, like fine etching perhaps, but almost more delicate, like a breath rather than a touch and absolutely exhausting to nerves [ 134 | and patience to continue, till the leaf was completed and ready to anneal. Two wicks in two cups of parafine were started in front of him, and the flames driven at each other horizontally, his face, nose and mouth, protected by a piece of asbestos. He took the glass stem in his left hand and inserted the leaf between the flames, where they just met. The tip first, moving it constantly after it had become red hot, till the whole leaf was finished. He keeps his right hand free to manipulate the apparatus or the handle, or guide the leaf if necessary, as he turns and twists it in the flame. Not infrequently, the annealing starts a flaw in the glass, and the leaf breaks so that a great many are necessary to complete a branch. He says that there is far more nervous strain, and it is far slower and more difficult to make a leaf than to make flowers. Annealing the powdered glass, instead of simply painting it, makes the process slower and more dangerous, but the final re- sult is much more permanent; in fact, the color cannot change, and nothing but violence can destroy the model. He can scratch and scrape a leaf with his penknife and it leaves not a mark. There is additional labor which I did not see. Anneal- ing leaves the glass glittering and shining, and that appearance he destroys by the application of a certain varnish; and he applies this to the flowers also after stamens and pistils are set. I think he said that he no longer paints at all except with the powdered colored glass which he can anneal. Another complication is that only a certain kind of glass can be used for the foundation glass, as the others spring and destroy the coat of anneal. All this is new since I was here nearly twenty years ago. He told me then that he had not at all come to the end of the possibilities of glass work, and these latest models show it to have been true. [ 135 ] The fruits are made over a fine but very strong wire, and he says in some of them there is as much as a $ Ib. of glass. That process I have not seen. Most of the fruits I think are done and quantities of leaves on their glass stems ready for painting are in drawers. Some of the fruits are blown, but very few, for he dislikes blowing glass, and the fruits are like eggshells, too perishable for transport. There are one or two sprays, however, with fruit of both kinds on them, and you would never know the difference in their looks. The large pears and apples looked to me hopeless to transport, for no wire could stay on them, and no paper hold against their weight. But he is confident that by pinning the stalk firmly above the fruit stem it will never move. He has ordered a very heavy weight for the card- board boxes. This certainly has been a ‘“‘long letter,’’ and I shall be very anxious to know if you receive it safely, but please do not feel obliged to write when you are tired and busy —much as [ always enjoy your letters. Please remember me to Louis and, with most cordial greetings to you and Mrs. Ames, Very sincerely yours, Mary LEE WaRE [ 136 | BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Campripce, Massacuuserts, Fesruary 17, 1961 VoL. 19, No. 7 THE HALLUCINOGENIC FUNGI OF MEXICO: AN INQUIRY INTO THE ORIGINS OF THE ReExticious LDEA AMONG PRIMITIVE PEoPLEs* BY R. Gorpon Wasson f WHEN I received in Mexico your President’s invitation to speak here today, I knew that your Committee had made an unorthodox choice, for I am not a professional mycologist. As the appointed hour approached my trepidation kept mounting, for I saw myself an amateur about to be thrown to a pack of professionals. But your President’s gracious introductory remarks, however un- merited, have put me at my ease and lead me to hope that we shall all enjoy together a mushroom foray of a rather unusual nature. Those of you who do not know the story will be in- terested in learning how it came about that my wife, who was a pediatrician, and [, who am a banker, took up the study of mushrooms. She was a Great Russian and, like all of her fellow-countrymen, learned at her mother’s knee asolid body of empirical knowledge about the com- mon species and a love of them that are astonishing to us Americans. Like us, the Russians are fond of nature— *The Annual Lecture of the Mycological Society of America, Still- water, Oklahoma. August 30, 1960. Revised. t Research Fellow, Botanical Museum of Harvard University. [ 137 ] the forests and birds and wild flowers. But their love of mushrooms is of a different order, a visceral urge, a pas- sion that passeth understanding. The worthless kinds, the poisonous mushrooms—the Russians are fond, in a way, even of them. They call these ‘worthless ones’ paganki, the ‘little pagans,’ and my wife would make of them colorful center-pieces for the dining-room table, against a background of moss and stones and wood picked up in the woods. On the other hand, I, of Anglo-Saxon origin, had known nothing of mushrooms. By inherit- ance, I ignored them all; I rejected those repugnant fungal growths, expressions of parasitism and decay. Be- fore my marriage, I had not once fixed my gaze on a mushroom; not once looked at a mushroom with a dis- criminating eye. Indeed, each of us, she and I, regarded the other as abnormal, or rather subnormal, in our con- trasting responses to mushrooms. A little thing, some of you will say, this difference in emotional attitude toward wild mushrooms. Yet my wife and I did not think so, and we devoted a part of our lei- sure hours for more than thirty years to dissecting it, de- fining it, and tracing it to its origin. Such discoveries as we have made, including the rediscovery of the religious role of the hallucinogenic mushrooms of Mexico, can be laid to our preoccupation with that cultural rift between my wife and me, between our respective peoples, between the mycophilia and mycophobia (words that we devised for the two attitudes) that divide the Indo-European peoples into two camps. If this hypothesis of ours be wrong, then it must have been a singular false hypothe- sis to have produced the results that it has. But I think it is not wrong. Thanks to the immense strides made in the study of the human psyche in this century, we are now all aware that deep-seated emotional attitudes ac- quired in early life are of profound importance. I suggest [ 138 ] that when such traits betoken the attitudes of whole tribes or peoples, and when those traits have remained unaltered throughout recorded history, and especially when they differ from one people to another neighboring people, then you are face to face with a phenomenon of profound cultural importance, whose primal cause is to be dis- covered only in the well-springs of cultural history. Many have observed the difference in attitude toward mushrooms of the European peoples. Some mycologists in the English-speaking world have inveighed against this universal prejudice of our race, hoping thereby to weaken its grip. What a vain hope! One does not treat a consti- tutional disorder by applying a band-aid. We ourselves have had no desire to change the Anglo-Saxon’s attitude toward mushrooms. We view this anthropological trait with amused detachment, confident that it will long re- main unchanged for future students to examine at their leisure. Our method of approach was to look everywhere for references to mushrooms. We gathered the words for ‘mushroom’ and the various species in every accessible language. We studied their etymologies. Sometimes we rejected the accepted derivations and worked out new ones, as in the case of ‘mushroom’ itself and also of ‘chanterelle.” We were quick to discern the latent meta- phors in such words, metaphors that had lain dead in some cases for thousands of years. We searched for the mean- ing of those figures of speech. We sought for mushrooms in the proverbs of Europe, in myths and mythology, in legends and fairy tales, in epics and ballads, in historical episodes, in the obscene and scabrous vocabularies that usually escape the lexicographer; in the writings of poets and novelists. We were alert to the positive or negative value that the mushroom vocabularies carried, their my- cophilic and mycophobic content. Mushrooms are widely [ 189 ] linked with the fly, the toad, the cock, and the thunder- bolt: and so we studied these to see what associations they conveyed to our remote forebears. Wherever we traveled we tried to enter into contact with untutored peasants and arrive at their knowledge of the fungi—the kinds of mushrooms that they distinguished, their names, the uses to which they put them, and their emotional attitude toward them. We made trips to the Basque country, to Lapland, to Friesland, to the Provence, to Japan. We scoured the picture galleries and museums of the world for mushrooms and we pored over books on archeology and anthropology. | would not have you think that we ventured into all these learned paths without guidance. We drew heavily on our betters in the special fields that we were explor- ing. When we were delving into questions of vocabulary, when we worked out an original etymology for a mush- roomic word, we were always within reach of a philolo- gist who had made of that tongue his province. And soin all branches of knowledge. Sometimes it seems to me that our entire work has been composed by others, with us merely serving as rapporteur. Since we began to publish in 1956, persons in all walks of life have come to us in in- creasing numbers to contribute information, and ofttimes the contributions of even the lowliest informants are of highest value, filling a lacuna in our argument. We were amateurs unencumbered by academic inhibitions, and therefore we felt free to range far and wide, disregarding the frontiers that ordinarily segregate the learned disci- plines. What we produced was a pioneering work. We know, we have always known better than the critics, the flaws in ours, but our main theme, which we adumbrated rather diffidently in Mushrooms Russia and History in 1957, seems to have stood up under criticism. If I live and retain my vitality, you may see published over the [ 140 ] coming years a series of volumes, to be called perhaps Ethnomycological Papers, and, at the end of the road, there may be a new edition of our original work, re- shaped, simplified, with new evidence added and the argument strengthened. It would give me pleasure to enumerate the names of those to whom we are indebted, but how tedious the roll call would be for you who are obliged to listen! There is one name, however, that in this audience I must cite. For more than ten years, we have been collaborating closely with Professor Roger Heim, Membre de |’ Insti- tut, and on all matters mycological he has been our guide and teacher. For these many years, he has been the di- rector in Paris of the Laboratoire de Cryptogamie and, even longer, editor of the Revue de Mycologie. More recently, he has also borne the burden of directing the Muséum National d’ Histoire Naturelle, that renowned center for advanced teaching and research in the biologi- cal studies, one of the glories of French culture. But these titles to academic distinction, though themselves of the highest order, do not tell you the story. Vast as is his learning and his experience in field and laboratory, sound as is his judgment in the vexed problems that you mycologists face every day, formidable as he is in po- lemic, it is as a rare human being that I commend him to you. Patient with the beginner, inspiring as a teacher, model of generosity toward others, prodigious worker in field and laboratory, and classical stylist in the French language, who could be more delightful whether in his published writings, or as correspondent, or as companion in the field? In the presence of Roger Heim, the time- worn conflict between science and the humanities fades away. Onesenses that the field of science for him is merely the New World that civilized man, the exponent of the humanities, is exploring and assimilating. What guardian [141 ] angel had me in his keeping when, after the Second World War, I ascended the steps of his laboratory in Paris to meet him for the first time, astranger, an Ameri- can, an ignoramus in the complex, the vast, the exacting discipline that you and he share together? At once he made me feel at home and it was not long before he was developing enthusiasm for our ethnomycological in- quiries. Later he became our indispensable and beloved partner in our Middle American forays. I do not recall which of us, my wife or I, first dared to put into words, back in the °40’s, the surmise that our own remote ancestors, perhaps 4,000 years ago, wor- shipped a divine mushroom. It seemed to us that this might explain the phenomenon of mycophilia vs. myco- phobia, for which we found an abundance of supporting evidence in philology and folklore. Nor am I sure whether our conjecture was before or after we had learned of the role of Amanita muscaria in the religion of several remote tribes of Siberia. Our bold surmise seems less bold now than it did then. | remember distinctly how it came about that we embarked on our Middle American explorations. In the fall of 1952 we learned that the 16th century writers, describing the Indian cultures of Mexico, had recorded that certain mushrooms played a divinatory role in the religion of the natives. Simultaneously we learned that certain pre-Columbian stone artifacts re- sembling mushrooms, most of them roughly a foot high, had been turning up, usually in the highlands of Guate- mala, in increasing numbers. For want of a better name, the archeologists called them ‘mushroom stones,’ but not one archeologist had linked them with mushrooms or with the rites described by the 16th century writers in neighboring Mexico. ‘They were an enigma, and ‘mush- room stone’ was merely a term of convenience. Some of these stone carvings carried an effigy on the stipe, either [142 | a human face or an animal, and allof them were very like mushrooms. Like the child in the Emperor’s New Clothes, we spoke up, declaring that the so-called ‘mush- room stones’ really represented mushrooms, and that they were the symbol of a religion, like the Cross in the Christian religion, or the Star of Judea, or the Crescent of the Moslems. If we are right—and little by little the accumulating evidence seems to be in our favor—then this Middle American cult of a divine mushroom, this cult of ‘God’s flesh’ as the Indians in pre-Columbian times called it, can be traced back to about B.C. 1500, in what we call the Karly Pre-classic period, the earliest period in which man was in sufficient command of his technique to be able to carve stone. Thus we find a mushroom in the center of the cult with perhaps the oldest continuous history in the world. These oldest mushroom stones are technically and stylistically among the finest that we have, evidence of a flourishing rite at the time they were made. Earlier still, it is tempting to imagine countless generations of wooden effigies, mush- roomic symbols of the cult, that have long since turned to dust. Is not mycology, which someone has called the step-child of the sciences, acquiring a wholly new and unexpected dimension? Religion has always been at the core of man’s highest faculties and cultural achievements, and therefore I ask you now to contemplate our lowly mushroom—what patents of ancient lineage and nobility are coming its way! It remained for us to find out what kinds of mush- rooms had been worshipped in Middle America, and why. Fortunately, we could build on the experience of a few predecessors in the field: Blas Pablo Reko, Robert J. Weitlaner, Jean Bassett Johnson, Richard Evans Schultes, and Eunice V. Pike. They all reported that the cult still existed in the Sierra Mazateca in Oaxaca. [ 143 ] And so we went there, in 1958. In books and articles we have described time and time again our later adven- tures, and some of you, surely, are familiar with them. So far as we know, we were the first outsiders to eat the mushrooms, the first to be invited to partake in the agapé of the sacred mushroom.* I propose here this evening a new approach, and will give you the distinctive traits of this cult of a divine mushroom, which we have found a revelation, in the true meaning of that abused word, but which for the Indians is an every-day feature, albeit a Holy Mystery, of their lives. Here let me say a word parenthetically about the na- ture of the psychic disturbance that the eating of the mushroom causes. This disturbance is wholly different from the effects of alcohol, as different as night from day. We are entering upon a discussion where the vocabulary of the English language, of any European language, is seriously deficient. There are no apt words in them to characterize your state when you are, shall we say, “be- mushroomed.’ For hundreds, even thousands, of years we have thought about these things in terms of alcohol, and we now have to break the bonds imposed on us by the alcoholic association. We are all, willy nilly, confined within the prison walls of our every-day vocabulary. With skill in our choice of words we may stretch accepted meanings to cover slightly new feelings and thoughts, but when astate of mind is utterly distinct, wholly novel, then all our old words fail. How do you tell a man born blind what seeing is like? In the present case, this is es- pecially true because superficially the bemushroomed man shows a few of the objective symptoms of one in- intoxicated, drunk. Now virtually all the words describ- ing the state of drunkenness, from ‘intoxicated’ (which, as you know, means ‘poisoned’) through the scores of * This was on the night of June 29-30, 1955. [ 144 ] current vulgarisms, are contemptuous, belittling, pejora- tive. How curious it is that modern civilized man finds surcease from care in a drug for which he seems to have no respect! If we use by analogy the terms suitable for alcohol, we prejudice the mushroom, and since there are few among us who have been bemushroomed, there is danger that the experience will not be fairly judged. W hat we need is a vocabulary to describe all the modali- ties of a Divine Inebriant. These difficulties in communicating have played their part in certain amusing situations. Two psychiatrists who have taken the mushroom and known the experience in its full dimensions have been criticised in professional circles as being no longer ‘objective.’ Thus it comes about that we are all divided into two classes: those who have taken the mushroom and are disqualified by our subjec- tive experience, and those who have not taken the mush- room and are disqualified by their total ignorance of the subject! As for me, a simple layman, I am profoundly grateful to my Indian friends for having initiated me into the tremendous Mystery of the mushroom. In describing what happens, I shall be using familiar phrases that may seem to give you some idea of the bemushroomed state. Let me hasten to warn you that I am painfully aware of the inadequacy of my words, any words, to conjure up for you an image of that state. [I shall take you now to the monolingual villages in the uplands of southern Mexico. Only a handful of the in- habitants have learned Spanish. The men are appallingly given to the abuse of alcohol, but in their minds the mushrooms are utterly different, not in degree, but in kind. Of alcohol they speak with the same jocular vul- garity that we do. But about mushrooms they prefer not to speak at all, at least when they are in company and especially when strangers, white strangers, are present. [ 145 ] Lf you are wise, you will talk about something, anything, else. Then, when evening and darkness come and you are alone with a wise old man or woman whose confidence you have won, by the light of a candle held in the hand and talking in a whisper, you may bring up the subject. Now you will learn how the mushrooms are gathered, perhaps before sunrise, when the mountain side is caressed by the pre-dawn breeze, at the time of the New Moon, in certain regions only by a virgin. ‘The mushrooms are wrapped in a leaf, perhaps a banana leaf, sheltered thus from irreverent eyes, and in some villages they are taken first to the church, where they remain for some time on the altar, in a,écara or gourd bowl. They are never ex- posed in the market-place but pass from hand to hand by prearrangement. I could talk to you a long time about the words used to designate these sacred mush- rooms in the languages of the various peoples that know them. The Aztecs before the Spaniards arrived called them teo-nandcatl, God’s flesh. I need hardly remind you of a disquieting parallel, the designation of the Elements in our Eucharist: ‘Take, eat, this is my Body... .’; and again, ‘Grant us therefore, gracious Lord, so to eat the flesh of thy dear son... .” But there is one difference. The orthodox Christian must accept by faith the miracle of the conversion of the bread into tod’s flesh: that is what is meant by the Doctrine of Transubstantiation. By contrast, the mushroom of the Aztecs carries its own conviction; every communicant will testify to the miracle that he has experienced. In the language of the Mazatecs, the sacred mushrooms are called ’nti' si*tho®. The first word, ‘nti’, is a particle ex- pressing reverence and endearment. * The second element means ‘that which springs forth.” In 1958 our muleteer * The superscript digits indicate the pitch of the syllable, ' being the highest of four. The initial apostrophe indicates a glottal stop. [ 146 | had travelled the mountain trails all his life and knew Spanish, though he could neither read nor write, nor even tell time by aclock’s face. We asked him why the mush- rooms were called ‘that which springs forth.’ His answer, breathtaking in its sincerity and feeling, was filled with the poetry of religion, and I quote it word for word as he gave it: El honguillo viene por si mismo, no se sabe de dénde, como el viento que viene sin saber de dénde ni porqué. The little mushroom comes of itself, no one knows whence, like the wind that comes we know not whence nor why. When we first went down to Mexico, we felt certain, my wife and I, that we were on the trail of an ancient and holy mystery, and we went as pilgrims seeking the Grail. ‘To this attitude of ours I attribute such success as we have had. It has not been easy. For four and a half centuries the rulers of Mexico, men of Spanish blood or at least of Spanish culture, have never entered sympa- thetically into the ways of the Indians, and the Church regarded the sacred mushroom as anidolatry. The Protes- tant missionaries of today are naturally intent on teaching the Gospel, not on absorbing the religion of the Indians. Nor are most anthropologists good at this sort of thing... For more than four centuries the Indians have kept the divine mushroom close to their hearts, sheltered from desecration by white men, a precious secret. We know that today there are many curanderos who carry on the cult, each according to his lights, some of them consum- mate artists, performing the ancient liturgy in remote huts before minuscule congregations. With the passing years they will die off, and, as the country opens up, the cult is destined to disappear. They are hard to reach, these euranderos. Almost invariably they speak no Span- ish. To them, performing before strangers seems a pro- fanation. They will refuse even to meet with you, much [ 147 ] less discuss the beliefs that go with the mushrooms and perform for you. Do not think that it is a question of money: 70 hicimos esto por dinero, ‘We did not this for money,’ said Guadalupe, after we had spent the night with her family and the cwrandera Maria Sabina. Per- haps you will learn the names of a number of renowned curanderos, and your emissaries will even promise to de- liver them to you, but then you wait and wait and they never come. You will brush past them in the market- place, and they will know you, but you will not know them. The judge in the town-hall may be the very man you are seeking; and you may pass the time of day with him, yet never learn that he is your curandero. After all, would you have it any different? What priest of the Catholic Church will perform Mass to satisfy an unbeliever’s curiosity / The curandero who today, for a big fee, will perform the mushroom rite for any stranger isa prostitute and a faker, and his insincere performance has the validity of a rite put on by an unfrocked priest. In the modern world religion is often an etiolated thing, a social activity with mild ethical rules. Religion in primi- tive society was an awesome reality, ‘terrible’ in the original meaning of that abused word, pervading all life and culminating in ceremonies that were forbidden to the profane. This is what the mushroom ceremony is in the remote parts of Mexico. We often think of the mysteries of antiquity as a mani- festation of primitive religion. Let me now draw your attention to certain parallels between our Mexican rite and the Mystery performed at Eleusis. The timing seems significant. Inthe Mazatec country the preferred season for ‘consulting the mushroom’ is during the rains, when the mushrooms grow, from June through August. The Eleusinian Mystery was celebrated in September or early October, the season of the mushrooms in the Mediter- [ 148 ] ranean basin. At the heart of the Mystery of Eleusis lay asecret. In the surviving texts there are numerous refer- ences to the secret, but in none is it revealed. Yet Mys- teries such as this one at Kleusis played a major role in Greek civilization, and thousands must have possessed the key. From the writings of the Greeks, from a fresco in Pompeii, we know that the initiate drank a potion. Then, in the depths of the night, he beheld a great vision, and the next day he was still so awestruck that he felt he would never be the same man as before. What the initi- ate experienced was ‘new, astonishing, inaccessible to rational cognition.’ * One writer in the 2nd century A.D., by name Aristides, pulled the curtain aside for an instant, with this fragmentary description of the Eleu- sinian Mystery: Eleusis is a shrine common to the whole earth, and of all the di- vine things that exist among men, it is both the most awesome and the most luminous. At what place in the world have more miraculous tidings been sung, where have the dromena called forth greater emo- tion, where has there been greater rivalry between seeing and hearing? And then he went on to speak of the ‘ineffable visions’ that it had been the privilege of many generations of fortunate men and women to behold. Just dwell for a moment on that description. How striking that the Mystery of antiquity and our mushroom rite in Mexico are accompanied in the two societies by veils of reticence that, so far as we can tell, match each other point for point! Our ancient writers’ words are as applicable to contemporary Mexico as they were to clas- sic Greece! May it not be significant that the Greeks were wont to refer to mushrooms as ‘the food of the gods,’ broma theon, and that Porphyrius is quoted as having * For this and the following quotations see Walter F. Otto: The Meaning of the Eleusinian Mysteries, published in The Mysteries, 1955, ed. by Joseph Campbell, Pantheon Books, Bollingen Series XXX, 2; pp. 20 et seq. Italics are mine. [ 149 ] called them ‘nurslings of the gods,’ theotréphos*? The Greeks of the classic period were mycophobes. Was this because their ancestors had felt that the whole fungal tribe was infected ‘by attraction’ with the holiness of some mushrooms and that they were not for mortal men to eat, at least not every day? Are we dealing with what was in origin a religious tabu? In earliest times the Greeks confined the common Euro- pean word for mushroom, which in their language was sp(h)éngos or sp(h)éngé, to the meaning ‘sponge,’ and replaced it by a special word, muhés, for the designation of mushrooms.} Now it happens that the root of this word mukés in Greek is a homonym of the root of the Greek word for ‘Mystery,’ mu. A bold speculation flashes through the mind. The word for ‘Mystery’ comes from a root that means the closing of the apertures of the body, the closing of the eyes and ears. If the mushroom played a vital and secret role in primitive Greek religion, what could be more natural than that the standard word for ‘mushroom’ would fall into disuse through a religious tabu (as in Hebrew ‘Yahweh’ gave way to ‘Adonai’) and * Giambattista della Porta: Villa, 1592, Frankfort, p. 764. { Holger Pedersen in an early paper contended that the basic fungal words of Europe were identical: Old High German swamb, Slavic gomba, Lithuanian gumbas, Latin fungus, Greek sp(h)éngos, sp(h)dngé, and Armenian sung, sunk. (Published in Polish: ‘Przyezynki do gramatyki por6wnawezej jezyk6w slowianskich,’ in Materyaly « Prace Komisyi Jesykowey Akademii Unmieietnosci w Krakowie, Cracow, 1(1): 167-176.) Since then some philologists have declined to accept this thesis as more than a possibility, especially as to the Slavic term, but Professor Roman Jakobsan in a recent personal communication to me says: ‘The etymology of Holger Pedersen, the great Danish specialist in the comparative study of Indo-European languages, seems to me and to many other linguists, e.g., the distinguished Czech etymolo- gist V. Machek, as the only convincing attempt to interpret the fungal name of the European languages. Not one single serious argument has been brought againt Pedersen’s ‘attractive’? explanation, as Berneker defines it, and not one single defensible hypothesis has been brought to replace this one.’ [ 150 ] that the Greeks substituted an alternative fungal term that was a homonym of ‘mystery’? Youcan hear the pun, see the gesture, ‘Mum’s the word,’ with the index finger over the mouth. . .. We must remember, in consider- ing this problem, that in antiquity the ecology of Greece and the Greek isles was different from now. Deforesta- tion and the goats had not wrought the havoc of the in- tervening centuries. They had not left the mountains naked to the sun. On the wooded isles and in the forests of the mainland, there must have been a wealth of mushrooms. Let us consider possibilities other than the mushroom. In the Mazatec country the Indians, when there are no mushrooms, have recourse to alternatives. Thanks to the brilliant work of Dr. Albert Hofmann of Sandoz, the Swiss pharmaceutical firm, we are now sorting out and identifying a whole series of indoles that have remarkable psychotropic properties. As you all know, he has isolated the active agents in some of our Mexican mushrooms, psilocybin and psilocin, two tryptamine derivatives and members of the indole family of substances. He has de- fined their molecular structure. The magic indoles are present in other plants used widely among the Indians of Mexico. With Dr. Hofmann’s permission, I am able to announce to you tonight that, only in July of this year, he has isolated and identified three of the active agents in ololiuqui, the famous seeds, subject of many studies, that have long been used in Mexico for their psychotropic properties.* In the Mazatec country the seeds of ololiuqui are one of the alternatives used when the sacred mushrooms are not available. Imagine our surprise, when we began looking for these seeds in quan- tity last year, to discover that the Zapotec Indians, em- * The Chemistry of Natural Products, paper read by Dr. Hofmann, Aug. 18, 1960, in the I.U.P.A.C. Symposium, Melbourne. [151 ] ploy two seeds: in some villages one, in others the other, and in some both. There is no question which seed was the ololiuqui of the Aztecs. It is a climbing morning- glory known to science as Rivea corymbosa (.) Hallier filius.* The seeds are brown and almost round. The second plant was identified at the National Herbarium in Washington as Ipomoea violacea L..,¢ also a climbing morning-glory but easily distinguished in the field from Rivea corymbosa. The seeds are long, black, and angu- lar, and so far as we now know, they are used only in some parts of the Zapotec country. Both are called in Zapotec badoh, but the black seeds are badoh negro, black badoh, to distinguish them from the true ololiuqui seeds. ° * The best summary of the ololiuqui literature and problem is Rich- ard Evans Schultes’ A Contribution to Our Knowledge of Rivea corym- bosa, the Narcotic Ololiuqui of the Aztecs, Botanical Museum, Harvard University, 1941. Also see Humphrey Osmond’s Ololiuqui: The An- cient Aztec Narcotic, Journal of Mental Science, July 1955, 101 (424): 526-537. Dr. Osmond reports on the effects of the seeds on himself. {Ipomoea violacea Linnaeus Pl. Sp. (1953) 161. Convolvulus indicus Miller Gard. Dict. (1768) No. 5. Ipomoea tricolor Cavanilles Icon. Pl. Rar. 3 (1794) 5. Convolvulus violaceus Sprengel Syst. 1 (1825) 399. Convolvulus venustus Sprengel Syst. 1 (1825) 399. Ipomoea rubrocoerulea Hooker Bot. Mag. (1834) t. 3297. Pharbitis violacea (L.) Bojer Hort. Maurit. (1837) 227. Tereietra violacea (L.) Rafinesque Fl. Tellur. 4 (1889) 124. Ipomoea Hookeri G. Don Gen. Syst. 4 (1838) 274. Pharbitis rubrocoeruleus (Hook.) Planchon FI. des Serres 9 (1854) 281. Convolvulus rubrocoeruleus (Hook.) D. Dietrich Syn. Pl. 1 (1839) 670. Ipomoea puncticulata Bentham Bot. Voy. Sulph. (1945) 136. ° Credit for the discovery of the ceremonial use of Jpomoea violacea seeds goes to Thomas MacDougall and Francisco Ortega (‘Chico’), famous Zapotec guide and itinerant trader. They have not yet de- limited the area of diffusion, but they have found badoh negro seeds in use in the following Zapotec towns and villages in the uplands of southern Oaxaca: San Bartolo Yautepec, San Carlos Yautepec and Santa Catarina Quieri, all in the district of Yautepec; Santa Cruz [ 152 ] Dr. Hofmann found that the alkaloidal components of the two seeds were identical, and they yielded d-lysergic acid amide and d-isolysergic acid amide, in the LSD 25 family of substances and known heretofore only as de- rivatives of ergot. Is it not surprising to find in higher plants such as the Convolvulaceae the same lysergic acid derivatives as in the lower fungi’ The third substance found in these seeds was chanoclavine, also isolated by Dr. Hofmann et al. some years ago from a culture of Claviceps species. * Thus it comes about that, thanks to the achievements of our biological chemists, we may be on the brink of re- discovering what was common knowledge among the ancient Greeks. I predict that the secret of the Mysteries will be found in the indoles, whether derived from mush- rooms or from higher plants or, as in Mexico, from both. I would not be understood as contending that only these substances (wherever found in nature) bring about visions and ecstasy. Clearly some poets and prophets and many mystics and ascetics seem to have enjoyed ecstatic visions that answer the requirements of the ancient Mys- teries and that duplicate the mushroom agapé of Mexico. I do not suggest that St. John of Patmos ate mushrooms in order to write the Book of the Revelation. Yet the succession of images in his Vision, so clearly seen and yet Ozolotepee and San Andrés Lovene, District of Miahuatlan; and finally a settlement called Roalo, between Zaachila and Zimatlan, just south of the city of Oaxaca. In San Bartolo J. violacea is used to the exclusion of Rivea corymbosa, but in the other towns both are used. These data are based on personal correspondence and also Thomas MacDougall: Ipomoea tricolor: A Hallucinogenic Plant of the Zapo- tecs, in Boletin of the Centro de Investigaciones Antropolégicas de México, No. 6, March 1, 1960. Reports from Juquila, to the west of the Zapotec towns mentioned above, indicate that J. violacea seeds may also be used among the Chatino Indians. * A. Hofmann with R. Brunner, H. Kobel, and A. Brack, Helv. Chem. Acta, 1957, 40: 1358. [ 153 ] such a phantasmagoria, means for me that he was in the same state as one bemushroomed. Nor do I suggest for a moment that William Blake knew the mushroom when he wrote this telling account of the clarity of ‘vision’: The Prophets describe what they saw in Vision as real and exist- ing men, whom they saw with their imaginative and immortal organs; the Apostles the same; the clearer the organ the more distinct the object. A Spirit and a Vision are not, as the modern philosophy supposes, a cloudy vapour, or a nothing: they are or- ganized and minutely articulated beyond all that the mortal and perishing nature can produce. He who does not imagine in stronger and better lineaments, and in stronger and better light than his perish- ing eye can see, does not imagine at all, {Italics mine. From The Wrilings of William Blake, ed. by Geoffrey Keynes, vol. II], p.108] This must sound cryptic to one who does not share Blake’s vision or who has not taken the mushroom. The advantage of the mushroom is that it puts many (if not everyone) within reach of this state without having to suffer the mortifications of Blake and St. John. It per- mits you to see, more clearly than our perishing mortal eye can see, vistas beyond the horizons of this life, to travel backwards and forwards in time, to enter other planes of existence, even (as the Indians say) to know God. It is hardly surprising that your emotions are pro- foundly affected, and you feel that an indissoluble bond unites you with the others who have shared with you in the sacred agapé. All that you see during this night has a pristine quality: the landscape, the edifices, the carv- ings, the animals—they look as though they had come straight from the Maker’s workshop. This newness of everything—it is as though the world had just dawned— overwhelms you and melts you with its beauty. Not un- naturally, what is happening to you seems to you freighted with significance, beside which the humdrum events of everyday are trivial. All these things you see with an immediacy of vision that leads you to say to [ 154 | yourself, ‘Now I am seeing for the first time, seeing di- rect, without the intervention of mortal eyes.’ (Plato tells us that beyond this ephemeral and imperfect ex- istence here below, there is another Ideal world of Archetypes, where the original, the true, the beautiful Pattern of things exists for evermore. Poets and philoso- phers for millenniums have pondered and discussed his conception. [t is clear to me where Plato found his Ideas: it was clear to his contemporaries too. Plato had drunk of the potion in the Temple of Eleusis and had spent the night seeing the great Vision. ) And all the time that you are seeing these things, the priestess sings, not loud, but with authority. The Indians are notoriously not given to displays of inner feelings— except on these occasions. The singing is good, but under the influence of the mushroom you think it is infinitely tender and sweet. It is as though you were hearing it with your mind’s ear, purged of all dross. You are lying on a petate or mat; perhaps, if you have been wise, on an air mattress and in a sleeping bag. It is dark, for all lights have been extinguished save a few embers among the stones on the floor and the incense in a sherd. It is still, for the thatched hut is apt to be some distance away from the village. In the darkness and stillness, that voice hovers through the hut, coming now from beyond your feet, now at your very ear, now distant, now actually underneath you, with strange ventriloquistic effect. The mushrooms produce this illusion also. Everyone experi- ences it, just as do the tribesmen of Siberia who have eaten of Amanita muscaria and lie under the spell of their shamans, displaying as these do their astonishing dexterity with ventriloquistic drum-beats. Likewise, in Mexico, I have heard a shaman engage in a most com- plicated percussive beat: with her hands she hits her chest, her thighs, her forehead, her arms, each giving a [ 155 different resonance, keeping a complicated rhythm and modulating, even syncopating, the strokes. Your body lies in the darkness, heavy as lead, but your spirit seems to soar and leave the hut, and with the speed of thought to travel where it listeth, in time and space, accompanied by the shaman’s singing and by the ejaculations of her percussive chant. What you are seeing and what you are hearing appear as one: the music assumes harmonious shapes, giving visual form to its harmonies, and what you are seeing takes on the modalities of music—the music of the spheres. ‘Where has there been greater rivalry be- tween seeing and hearing?’ How apposite to the Mexican experience was the ancient Greek’s rhetorical question! All your senses are similarly affected: the cigarette with which you occasionally break the tension of the night smells as no cigarette before had ever smelled; the glass of simple water is infinitely better than champagne. Else- where I once wrote that the bemushroomed person is poised in space, a disembodied eye, invisible, incorporeal, seeing but not seen. In truth, he is the five senses dis- embodied, all of them keyed to the height of sensitivity and awareness, all of them blending into one another most strangely, until the person, utterly passive, becomes i pure receptor, infinitely delicate, of sensations. (You, being a stranger, are perforce only a receptor. But the Mazatee communicants are also participants with the curandera in an extempore religious colloquy. Her utter- ances elicit spontaneous responses from them, responses that maintain a perfect harmony with her and with each other, building up to a quiet swaying antiphonal chant. In asuccessful ceremony this is an essential element, and one cannot experience the full effect of the role of the mushroom in the Indian community unless one attends such a gathering, either alone or with one or at most two other strangers.) As your body lies there in its sleep- [ 156 | ing bag, your soul is free, loses all sense of time, alert as it never was before, living an eternity in a night, seeing in- finity in a grain of sand. What you have seen and heard is cut as with a burin in your memory, never to be effaced. At last you know what the ineffable is, and what ecstasy means. Kestasy! The mind harks back to the origin of that word. For the Greeks ekstasis meant the flight of the soul from the body. Can you find a better word than that to describe the bemushroomed state? In common parlance, among the many who have not experienced ecstasy, ecstasy is fun, and I am frequently asked why I do not reach for mushrooms every night. But ecstasy is not fun. Your very soul is seized and shaken until it tingles. After all, who will choose to feel undiluted awe, or to float through that door yonder into the Divine Presence? The unknowing vulgar abuse the word, and we must recapture its full and terrifying sense. ... A few hours later, the next morning, you are fit to go to work. But how unimportant work seems to you, by comparison with the portentous happenings of that night! If you can, you prefer to stay close to the house, and, with those who lived through that night, compare notes, and utter ejaculations of amazement. As man emerged from his brutish past, thousands of years ago, there was a stage in the evolution of his aware- ness when the discovery of a mushroom (or was it a higher plant?) with miraculous properties was a revela- tion to him, a veritable detonator to his soul, arousing in him sentiments of awe and reverence, and gentleness and love, to the highest pitch of which mankind is capable, all those sentiments and virtues that mankind has ever since regarded as the highest attribute of his kind. It made him see what this perishing mortal eye cannot see. How right the Greeks were to hedge about this Mystery, this imbibing of the potion, with secrecy and _ surveil- [ 157 lance! What today is resolved into a mere drug, a trypta- mine or lysergic acid derivative, was for him a prodigious miracle, inspiring in him poetry and philosophy and re- ligion. Perhaps with all our modern knowledge we do not need the divine mushrooms any more. Or do we need them more than ever? Some are shocked that the key even to religion might be reduced to a mere drug. On the other hand, the drug is as mysterious as it ever was: ‘like the wind it cometh we know not whence, nor why.” Out of a mere drug comes the ineffable, comes ecstasy. It is not the only instance in the history of humankind where the lowly has given birth to the divine. Altering a sacred text, we would say that this paradox is a hard saying, yet one worthy of all men to be believed. If our classical scholars were given the opportunity to attend the rite at Eleusis, to talk with the priestess, what would they not exchange for that chance? They would approach the precincts, enter the hallowed chamber, with the reverence born of the texts venerated by scholars for millenia. How propitious would their frame of mind be, if they were invited to partake of the potion! Well, those rites take place now, unbeknownst to the classical scholars, in scattered dwellings, humble, thatched, with- out windows, far from the beaten track, high in the mountains of Mexico, in the stillness of the night, broken only by the distant barking of a dog or the braying of an ass. Or, since we are in the rainy season, perhaps the Mystery is accompanied by torrential rains and punctu- ated by terrifying thunderbolts. Then, indeed, as you lie there bemushroomed, listening to the music and see- ing the visions, you know a soul shattering experience, recalling as you do the belief of some primitive peoples that mushrooms, the sacred mushrooms, are divinely engendered by Jupiter Fulminans, the God of the Lightning-bolt, in the Soft Mother Earth. [ 158 ] APPENDIX The following enumeration is the first list published in English of the hallucinogenic mushrooms of Mexico. With each name we give the place of publication of (1) the technical name of the mushroom and (2) the earliest report of its use in Mexico as a divinatory agent. Doubt- less more species will be discovered, but we believe our list is complete through 1960. Not all divinatory mushrooms are hallucinogenic. The Indians consume some kinds for divinatory purposes be- cause of their suggestive shape. This is true of Cordyceps capitata (Holmsk.) Link, as well as its host fungi, H/a- phomyces granulatus Fr. or EKlaphomyces variegatus Vitt., and also of Diuctyophora phalloidea Desvaux. Cordyceps capitata has been found to contain an indolic compound that might cause hallucinations, but only in trace amounts. There are also reports of the use of Cla- varia truncata Quél. and Nevrophyllum floccosum(Schw. ) Heim, but their hallucinogenic virtue remains doubtful, and they are always taken in conjunction with Psilocybe Wassonu Heim. (See Les champ. halluc. du Mexique, 1958, pp. 81, 83, 99, 162.) Psilocybe muliercula Singer & Smith has been reported as an hallucinogen (in Mycologia 50(1958) 145), but this concept is a synonym of Ps. Wassonu (see below). Professor Roger Heim and I accept responsibility for all species and varieties listed that are marked by an asterisk. *Conocybe siligineoides Heim in Rev. Mycol. 22 (1957) 197. Use first reported: in Comptes Rend. 242 (1956) 1891. [ 159 | Panaeolus fimicola (/’r.) Quélet ex Fries Hym. Eu- rop. (1874) 812. Use first reported: in Bol. Soc. Bot. Mex. No. 24 (1959) 23. Panaeolus sphinctrinus (/’7.) Quélet ex Fries Epicr. syst. mycol. seu synops. Hymenomye. (1886-38) 235. Use first reported: in Bot. Mus. Leafl. Harvard Univ. 7 (1939) 37 (as P. campanulatus L. var. sphinctrinus (F'r.) Bresadola). Psathyrella sepulchralis Singer, Smith & Guzman in Lloydia 21 (1958) 26. Use first reported: loc. cit. *Psilocybe acutissima Heim in Rev. Mycol. 24 (1959) 106. Use first reported: in Les champ. halluc. du Mexique (1958) 166. *Psilocybe aztecorum Heim in Rev. Mycol. 22 (1957) 78. Use first reported : in Comptes Rend. 244 (1957) 699. *Psilocybe caerulescens Murri// var. mazatecorum Heim in Rev. Mycol. 22 (1957) 78. Psilocybe mazatecorum Heim in Comptes Rend. 242 (1956) 1392, nomen prov., sine diagn. lat. Use first reported: loc. cit. *Psilocybe caerulescens Murrill var. mazatecorum Heim fma. heliophila Heim in Heim & Wasson Les champs. halluc. du Mexique (1958) 141, sine diagn. lat. Use first reported: loc. cit. [ 160 ] *Psilocybe caerulescens Murri/ll var. mazatecorum Heim. fma. ombrophila Heim in Heim & Wasson Les champs. halluc. du Mexique (1958) 140, sine diagn. lat. Use first reported: loc. cit. *Psilocybe caerulescens Murrill var. nigripes Heim in Rev. Mycol. 22 (1957) 79. Use first reported: in Comptes Rend. 244 (1957) 698. Psilocybe caerulipes (Peck) Saccado var. Gastonii Singer & Smith in Sydowia 12 (1959) 236. Use first reported: loc. cit. Psilocybe candidipes Singer & Smith in Mycologia 50 (1958) 141. Use first reported: loc. cit. 250. *Psilocybe cordispora Heim in Rev. Mycol. 24 (1959) 1038. Use first reported: in Comptes Rend. 242 (1956) 13890. *Psilocybe fagicola Heim & Cailleux in Rev. Mycol. 24 (1959) 438. Use first reported: in Comptes Rend. 249 (1959) 1843. *Psilocybe Hoogshagenii Heim in Rev. Mycol. 24. (1959) 104. | Use first reported: in Les champs. halluc. du Mexi- que (1958) 167. Psilocybe isauri Singer in Sydowia 12 (1959) 237. Use first reported: loc. cit. *Psilocybe mexicana Heim in Rev. Mycol. 22 (1957) Ke Use first reported: in Comptes Rend. 242 (1956) 966. [161 ] *Psilocybe mixaeensis Heim in Rev. Mycol. 24 (1959) 104. Use first reported: in Les champs. halluc. du Mexi- que (1958) 169. *Psilocybe semperviva Heim & Cailleuw in Rev. Mycol. 28 (1958) 352. Use first reported: in Comptes Rend. 245 (1957) 1764. *Psilocybe Wassonii Heim in Rev. Mycol. 23 (1958) 119. Psilocybe muliercula Singer & Smith in Mycologia 50 (1958) 142. Use first reported : in Comptes Rend. 245 (1957) 1763. *Psilocybe yungensis Singer & Smith in Mycologia 50 (1958) 142. Use first reported: in Bol. Soc. Mex. No. 24 (1959) 22. *Psilocybe zapotecorum Heim in Rev. Mycol. 22 (1957) 77. Ise first reported ; in Comptes Rend. 242(1956) 13898. *Psilocybe zapotecorum Heim var. elongata Heim in Comptes Rend. 250 (1960) 1158, nomen prov., sine diagn. lat. *Stropharia cubensis /H/ar/e Inf. An. Establ. Centr. Agron. Cub. 1 (1906) 240. Psilocybe cubensis (Karle) Singer in Lilloa 22 (1949) 507. Use first reported: in Comptes Rend. 242 (1956) 967. [ 162 ] BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Camsripar, Massacnusetts, Apri. 11, 1961 Vor. 19, No. 8 FURTHER ARCHAEOLOGICAL EVIDENCE ON THE EFFECTS OF TEOSINTE INTRO- GRESSION IN THE EVOLUTION OF MODERN MAIZE BY Watton C. GALINAT AND REYNOoLD J. Ruprk’ ANOTHER large and significant collection of prehistoric maize cobs (Zea Mays L.) with Tripsacoid characteristics that are indicative of introgression from either T'ripsa- cum spp. or its maize derivative, teosinte’ (Zea mexicana Reeves and Mangelsdorf), has been provided by the archaeologist, who is the junior author, for botanical analysis. Although there have been about a dozen other collections of prehistoric Tripsacoid cobs from north- western Mexico and southwestern United States, the present material, which comes from Cebollita Cave in New Mexico, is the first large (2575 cobs), stratified (five levels) collection to become available for statistical treatment. Our previous statistical study (Galinat, et a/., 1956) of Tripsacoid cobs involved a large non-stratified collection from two caves in Arizona. At that time we established ‘Arizona State University, Tempe, Arizona. ” For the purposes of discussion and consistency, we shall assume, as we have previously, that the immediate source of the introgression represented by these archaeological specimens is from teosinte rather than from less likely hybridization with Tripsacum. { 163 ] the reliability of scoring for teosinte introgression accord- ing to the degree of induration by showing that the more indurated archaeological cobs are like modern maize- teosinte derivatives in having a higher specific gravity which is also positively correlated, in modern maize, to number of teosinte chromosomes. According to this system, induration is subjectively estimated with an arbitrary key of five grades. At grade-1 the glumes and rachis are non-indurated and somewhat flexible. At grade-5 the glumes and rachis are not only highly indurated, but the glumes are curved upwards and at least some pistillate spikelets are borne singly, features which are common in maize-teosinte hybrids, but absent in typical maize. By applying this method to estimate teosinte intro- gression in the present stratified material, we may now determine the evolutionary effects of such introgression upon the maize from this site. Description of the Site The archaeological maize upon which this study is based was excavated from Cebollita Cave in the Cebol- leta Mesa‘ area in Valencia County, New Mexico, about twenty miles south of the town of Grants. The area is bounded on the west by the McCarthys’ Lava Flow and on the east by the western slope of Cebolleta Mesa. It is in the Upper Sonoran climatic zone at an elevation of about 7000 feet. The terrain consists of broad valley floors and sheer sandstone cliffs. The cave is located in a vertical sandstone cliff in the Zuni sandstone member. It faces south and opens out on a broad valley, which, ' According to the principal maps of the area, the name of the mesa is spelled Cebolleta while the name of the cave which contained the archaeological maize is spelled Cebollita. The latter spelling comes from the Spanish word meaning “‘little onion.”’ [ 164. ] before channel cutting had commenced, must have been an ideal flood-farming area. The flora includes pifion pine, juniper, manzanita (Arctostaphyllos pungens), sage, blue gramma grass, yucca, bee weed (Cleome serrulata Pursh), and several varieties of cacti. Canyon floors in the area normally have a good stand of blue gramma grass mixed with some cacti, yucca, sage brush, and manzanita. Minor depres- sions are covered by athick stand of bee plants and sun- flowers after the beginning of the rainy season. Scattered stands of juniper and pifon pine are found on the valley floors. Deer, coyotes, prairie dogs, rabbits, lizards, and snakes constitute most of the faunal assemblage. The climate is semi-arid and precipitation averages about eleven inches annually. The growing season can only be estimated from reports of government stations near Cebolleta Mesa and is thought to be about 110 days long. It is assumed that the climate at the time the cave was inhabited was approximately the same as today. The site is a fourteen room pueblo situated in Cebol- lita Cave. The pueblo was built piece-meal and abandoned at least once during its existence. The abundant rock fall from the roof attests to the hazards of life in the cave. An enormous block of sandstone fell from the roof at one time and caused a temporary abandonment of the pueblo. W hen the pueblo was reinhabited, it was by a group who had a slightly different culture than the previous occu- pants. Although the cave was inhabited in the Pre- pueblo and Pueblo I periods, the pueblo itself was not constructed until the end of the Pueblo II period. The entire occupation of the pueblo was encompassed within the Early Pueblo III period, from about 1050 to 1200 A.D. as dated by ceramic typology. Preservation of organic material in the pueblo was vari- able due to run-off water from the mesa top which flowed [ 165 ] into some of the rooms. Those rooms which remained dry contained large amounts of vegetal remains such as corn, squash, and other seeds, together with cordage, matting, sandals, basketry, and wooden objects. Most of the corn utilized in the present study was found in Rooms Bb, C, and D in the back of the cave. Room B contained four feet of deposit, the deepest fill in the pueblo. The archaeological value of the maize under discussion lies in the fact that the excellent stratigraphic evidence indicates an interesting history of human occupation in the pueblo. Room B supplied most of the evidence and most of the maize. Level 5 in Room B marks the earliest occupation of the pueblo. Three hard-packed adobe floors were superimposed at the base of this level. The original maize at the site was found on the uppermost floor and occurred in fifteen concentrations of charred, shelled ker- nels, ashes, and heat-warped pot sherds. More than 500 charred ears were also found lying on the floor. ‘The con- centrations of shelled kernels had been stored in pots and the loose ears must have been hung from the roof beams. All of the material found in Level 5 and the lower half of Level 4 was burned. A three-foot thick concentration of spalled sandstone slabs was found above Level 5 and the lower half of Level 4. The slabs had spalled off the cave ceiling as a result of the fire that destroyed Room B. In addition, a huge block of sandstone weighing many tons fell across Rooms D and E. The fire and rock fall terminated occupation of the site for a time, but, pend- ing study of the dendrochronological specimens, the du- ration of the abandonment is unknown. The Ceramic Assemblage Reoccupation of the pueblo was accomplished by a group of Indians using a slightly later variant of the Early Pueblo [II ceramic types (normal Tularosa phase) [ 166 ] than their predecessors. Greater changes in the type of pottery than those observed would be expected if the period of abandonment had been long. A quarter of the total number of sherds from one room occupied by the newcomers was of the brown paste type, an atypical pro- portion for the area at that time, but the other rooms do not show as high a proportion. Perishable Material other than Maize A large amount of perishable material other than maize was found in the upper levels, including nine sandals, five wooden arrow foreshafts, a number of fragments of bas- ketry and matting and several hundred pieces of plain, fur, or feather-wrapped cordage. One of the sandals is a modified fish-tail type of the kind found throughout the Mogollon area (Cosgrove, 1947, fig. 92-9b; Haury, 1934, plate 41; Bluhm, 1952, p. 271). It is interesting to note in this connection that Cosgrove also found Trip- sacoid maize together with this type of sandal in the Hueco Mountain caves. Two of the other sandals were typical of the Four Corners Region in Basketmaker and Pueblo III horizons (Kidder and Guernsey, 1919). A similar type was also found at Bat Cave (Herbert Dick, unpub.) and in Tularosa Cave (Bluhm, 1952, p. 279). The sandals might suggest that the newcomers origi- nated from somewhere in the Mogollon region, but such a conclusion is based on slim evidence at best. Recent researches in the Cebolleta Mesa area have resulted in the conclusion that there is a region of cultural blending between the Anasazi and Mogollon regions. In this case, since the Cebolleta Mesa area is on the southern periphery of the Anasazi region, it might have received influences from the blend region just to its south. Therefore, it seems likely that the possessors of the Tripsacoid maize may have come from this blend region. [ 167 ] Level 5 Maize The maize ears from the lowest stratum (Level 5), re- semble those of the race ‘‘Chapalote,’’ an ancient indi- genous race of Mexico described by Wellhausen ef al., 1952, and the principal, if not the only race of the early cultures from this part of North America (Mangelsdorf and Lister, 1956). At Bat Cave, New Mexico, a primi- tive form of Chapalote remaining from an incipient cul- tivation tradition dates back to between 8500 and 2500 B.C. (Mangelsdorf, 1954). Some of the prehistoric maize from coastal Peru (about 600 B.C.) may also have affini- ties with Chapalote (Grobman and Mangelsdorf, 1959). Identification of the original Cebollita maize as Chapa- lote was possible because most of the ears were perfectly preserved by a carbonization process resulting from in- complete combustion. The original Cebollita maize and Chapalote share the following characteristics. Their ears are cigar shaped, with aslight tapering at both base and tip. Prominent glumes may protrude between the ker- nels. Small, hard kernels are rounded on top and nearly isodiametric in length, width, and thickness. The vertical rows of these kernels, especially those of 10- and 14- rowed ears, have a strong tendency to twist. A relatively high row number in combination with a slender rachis, forces the cupule wings and paired kernels to interlock’ slightly with the lateral rows on either side. The inter- locking of cupule wings creates the illusion of broad cu- pules. The actual cupule width (5.5 mm.) is like the kernel width (6.0 mm. ) in being only about one-half that of other North American races such as the 8-rowed flour and flint types. A comparison of the actual values in Cebollita maize ‘Interlocking of adjacent pairs of kernels, sometimes called tessela- tion, is also found ina primitive Peruvian race, Confite Morocho, and certain of its derivatives (Grobman, unpub. ). [ 168 ] with those of modern Chapalote from the Mexican states of Sonora and Sinaloa and with those of certain early Basket Maker ears (about 800 A.D.) obtained from the studies of Hurst and Anderson (1949) on maize from Cottonwood Cave, Colorado, reveals that the ears from Cebollita Cave are slightly smaller (Table 1). The date of the Cebollita maize (about 1050 A.D.) seems to ex- clude it as a more primitive or inherently smaller type of Chapalote. Rather, reduction may be a depauperate condition resulting from poor growing conditions. The latter suggestion is supported by the fact that the best Level 5 ears compare favorably to those of modern Chap- alote and Cottonwood Cave maize. The termination of Level 5 was marked by a fire which either carbonized or charred all of the original Chapalote cobs and caused a large rock fall from the ceiling, as well as a temporary abandonment of the cave. Level 4 Maize Upon reoccupation of the cave, as designated by Level 4, 859% of the cobs changed abruptly to the Tripsacoid type of maize which was becoming widespread through- out southwestern United States during this period (1000— 1200 A.D.). Three percent of these were almost exact counterparts of modern Fy; hybrids or hybrid segregates from experimental crosses between maize and teosinte in being two-ranked for at least part of their length, and in having highly indurated, upward-curved glumes (Plate XXII, cobs C, D). Such highly Tripsacoid cobs were scored as grade-5 according to our system of estimating the degree of teosinte introgression with an arbitrary key of five grades. On the average, the Level 4 specimens were the most 'Tripsacoid from the entire site, having an average introgression grade of 2.6. At the other extreme, fifteen percent of the Level 4 [ 169 ] EXPLANATION OF THE ILLUSTRATION Piare XXIII. A series of cobs from the various levels in Cebollita Cave. The original Chapalote maize (cob A) of Level 5 has some non-carbonized counterparts (cob B) in Level 4. Note the soft glumes of cob B. Level 4 also contains many small highly Tripsacoid cobs (C, D) similar to segregates of experimental hybrids between maize and teo- sinte. Although Level 3 (cobs E, F, G) marks the beginning of a progressive decrease in the indura- tive effects of the teosinte introgression, the varia- bility which it had introduced continues to increase. Finally in Levels 2 and 1, there is almost complete recovery from the detrimental effects of the intro- gression and many of the cobs (H, I) are larger and probably more productive than the original Chapa- lote maize. About one half natural size. [ 170 | ALVT] IX xX cobs were scored as grade-1, because they had long soft glumes which were structurally similar to the carbonized ones from the original Chapalote in the previous level. Some of the soft-glumed cobs were also identical in all other characteristics to the original specimens except in being non-carbonized. Therefore, the continuity of the population of cobs was not completely broken by the change to Tripsacoid maize. The sudden change to these Tripsacoid cobs of Level 4 does not preclude a change in maize background from that of the original Chapalote. ‘Teosinte introgression had already occurred much earlier in the Chapalote from other parts of New Mexico (Bat Cave in Catron County, Mangelsdort and Smith, 1949); it was well established in this race in northwestern Mexico by 750+250 A.D. (Mangelsdorf and Lister, 1956). Inasmuch as ‘Tripsacoid Chapalote was prevalent then and there is no evidence in type of ceramics or sandals of trade from far outside the area, the new variation is attributed to teosinte intro- gression in Chapalote. The onset of teosinte introgression caused a marked drop in the average size of cob to the lowest values for the site. The average kernel row-number dropped by 11%, the cob diameter by 10%, and the cob length by 22% below that of the original Chapalote. These reduc- tions represent modification toward the spike of teosinte and are correlated to estimated teosinte introgression in the population as a whole (Plate NXNTV, figs. 1, 2, 3). These reductions in average cob size in the Tripsacoid maize may not reflect a corresponding loss in over-all yield per plant or peracre of plants. Some modern maize breeders have found that a reduction in ear size in teo- sinte derivatives of maize tends to be compensated for by an increase in number of ears per plant. However, some of the energy in 'Tripsacoid maize may be diverted away PLatTE XXIV INTROGRESSION AVERAGE ESTIMATE OF TEOSINITE 5.0 2.0 a\S W INTROGRESSION and KERNEL ROWS FIG. | | 1 \ W FIG.4 LEVELS o——~ KERNEL ROWS *-——-* INTROGRESSION 10.5 4 6 8 10 i2 14 NUMBER OF KERNEL ROWS LEVELS 3.0} 2.0 \\ Ww INTROGRESSION and COB DIAMETERS FIG. 2 L | FIG.5 LEVELS o—— COB DIAMETER *----* INTROGRESSION ; : i 12 14 16 18 22 Wo 10.0 o uo AVERAGE KERNEL ROWS 19.0 10 20 24 4 BS 2 | COB DIAMETER(mm) LEVELS FIG. 3 INTROGRESSION FIG.6 LEVELS 3.0 and COB LENGTHS oo COB LENGTH 3.0 *-—-—* INTROGRESSION 8.0 2.0 ir 7.0 T 4/ {6.0 fe) | | ae 2 4 12 14 5 4 2 | 6 8 10 COB LENGTH (cm) 3 LEVELS AVERAGE COB DIAMETER (mm) AVERAGE GOB LENGTH (cm) from grain production and into the production of strong lignification of the cob and stalk tissues. But even so, such an expenditure on development of a stiff stalk may be necessary to keep the ears erect and away from cer- tain vermin. The effect of teosinte introgression on cob length is not always detrimental. The longest, as well as the short- est, intact cobs were the most Tripsacoid (fig. 8). Those of intermediate length tended to resemble the original maize in having soft glumes. The same type of parabolic curvilinear correlation be- tween teosinte introgression and cob-length was also found with the cobs from Richards Cave in Arizona (Galinat ef a/., 1956). In this previous study the para- bolic curvilinear correlation, based on 488 intact cobs, had a value of 0.859, which was highly significant. The nature of the curve was explained by assuming that the long Tripsacoid cobs are the vigorous products of hetero- zy gous teosinte germplasm, while their short counterparts are the detrimental effects of homozygous teosinte germ- plasm. This explanation may be applied equally well to the cobs from Cebollita Cave. The repetition of the so- called ‘tmaize-teosinte heterosis’” at another site, indi- cates that this apparent counterpart of modern hybrid maize may have become widespread at the time just prior to 1200 A.D). in the Southwest. The blending of teosinte germplasm into maize would continue if its presence pro- vided any selective advantage such as that resulting from maize-teosinte heterosis. Under such conditions, the dis- tribution of Tripsacoid maize might become many gener- ations and hundreds of miles removed trom teosinte itself. Level 3 Maize Level 8 marks the start of a progressive reduction in teosinte contamination or at least a modification of its [17+ | expression. In either case, as the indurative effects of this introgression’ declined, the average cob size retro- gressed somewhat toward that of the original pure Chapa- lote. The reduction in “‘introgression’’ was slow at first, being only 8% at this stratum (Level 3) and not in pro- portion to the far greater recovery in kernel row number, cob diameter, and cob length of 18%, 22%, and 21% respectively (Plate XXIV, figs. 4, 5, 6). But even as the direct effects of teosinte introgression were apparently diminishing, the variability in cob size which was introduced by this introgression in the pre- vious Level (4), continued to increase in higher levels. For cob diameter, the standard deviations which measure degree of variability, for Levels 5 through 1, were 1.48 mm., 2.12 mm., 2.56 mm., 2.54 mm., and 2.10 mm., respectively. Increases in diameter variance are signifi- cant up to Level 3. But for cob length, the expanding variation proceeds one level higher, as shown by the standard deviations for Levels 5 to 1, respectively, as follows: 1.01 cm., 1.89 cm., 2.27 cm., 2.39 cm., and 1.86 cm. There are several possibilities which may, as a whole or in part, account for the continued increase in varia- bility after a reduction in teosinte introgression. If there was some variability injected by a new non-Tripsacoid race from elsewhere, its effect must have been insignifi- cant because the continuity of the population was not disrupted by a complete break from the features of Chap- alote. [In addition to an actual reduction in teosinte germ- plasm, the accumulating variability may have brought about some modification of its indurative effects. Man- gelsdorf (1958) suggested, on the basis of experimental "In order to facilitate further discussion, we shall assume that our estimate of teosinte introgression, according to the degree of indura- tion, represents a relative measure of its intensity. [175 ] evidence from modern maize-teosinte derivatives, that much of the variation in modern maize is the product not only of recombination of genes from the two species, but also from the mutagenic effects of teosinte germplasm in maize. Similarly, some of the increased variation in the Cebollita maize may be the result of a mutagenic effect of teosinte germplasm. Level 2 Maize Proceeding to the next Level (2), the same trends con- tinue: teosinte introgression decreases while average sar-size increases. ‘The extremes in cob-length and cob- diameter held about the same as those of the previous level (Table IL). But in the case of cob-length, the vari- ous categories became more equally represented with the result that the standard deviation or variability increased. As mentioned previously, the longest cobs are apparently a product of maize-teosinte heterosis. The fact that cob length did not decline with the apparent reduction in introgression at these higher levels might be explained in terms of a selective elimination of deleterious factors from teosinte and/or a buffering against the effects of such factors while beneficial factors involved in maize- teosinte heterosis were retained and blended into the evolving population. Level 1 Maize The cobs from the uppermost level represent the final evolutionary product from this site. Although the actual quantity of cobs was less than ten per cent of that from any previous level, it yields some of the longest and best specimens. ‘These superior ears represent a combination of butt fasciation descended from the original Chapalote together with the more lignified and heterotic products of teosinte introgression. Some of these specimens re- [ 176 ] semble the present day maize from the Southwest (Plate XXIII, cob I). Summary 1. The method of scoring for teosinte introgression according to the degree of induration has been used to study the archaeological record of the role of such intro- gression in the evolution of 2575 cobs found in five suc- cessive strata in Cebollita Cave in New Mexico. 2. The evolutionary sequence starts at Level 5 with a pure type of Chapalote, the indigenous race from this part of North America. Identification of this original Cebollita maize as Chapalote was possible because its morphological details were perfectly preserved by car- bonization. 3. Aftera period of abandonment of the cave most of the maize in Level 4 changed abruptly to a highly Trip- sacoid type of Chapalote which was becoming prevalent in the Southwest. Some of the more Tripsacoid of these specimens resembled, in induration and appearance, seg- regates from experimental hybrids between maize and teosinte, while other specimens remained identical to the original pure type. 4. The immediate effect of the teosinte introgression was to cause a marked reduction in average cob-size to the lowest values of the site. 5. Although advances to higher Levels (8, 2, 1) were marked by a progressive decrease in the indurative effects of this introgression, the variability in cob size which was introduced by the introgression continued to increase. For cob diameter, increases in variation cease at Level [177 ] 3. But for cob length, the expanding variation proceeds up one level higher. 6. The same type of parabolic relationship between teosinte introgression and cob-length which was found in a previous study of cobs from Richards Cave in Arizona was also found in the Cebollita Cave cobs. The interpre- tation of this type of relationship is that the long Trip- sacoid cobs are the vigorous products of heterozygous teosinte germplasm, while their short counterparts show the detrimental effects of homozygous teosinte germ- plasm. 7. In the final evolutionary products from Levels 2 and 1 at Cebollita, there is almost complete recovery from the detrimental effects of teosinte introgression and many of the cobs are larger and probably more productive than the original Chapalote maize. Some of these superior cobs resemble those of the present day maize from the Southwest. ACKNOWLEDGMENT During the course of the investigation and preparation of the manuscript, many helpful suggestions were made by Professor Paul C. Mangelsdorf of Harvard University. [178 | Taste I, A comparison of the ear characteristics of archaeological maize from Cebollita and Cottonwood Caves to that of modern Chapalote from Mexico. Cebollita Cottonwood’ Chapalote’ External Characters of Ear Length (em. ) 8.7 9.0 11.0 Diameter (cm. ) 2.6 8.3 2.9 Row Number 10.5 14.0 12.3 Width of Kernel (mm. ) 6.0 6.0 6.7 Thickness of Kernel (mm.) 4.0 4.0 4,1 Length of Kernel (mm.) a8 — 7.4 Internal Characters of Ear Diameter of Cob (mm.) 19.0 — 22.0 Diameter of Rachis (mm. ) 12.0 — 11.2 Length of Rachilla (mm. ) 2.0 —- 1.8 Glumes prominent prominent prominent Cupule Wings prominent = prominent Teosinte Introgression 1 — — ‘Data from Hurst and Anderson (1949). ? Data from Wellhausen et al. (1952). [179 ] Taste II. Morphological characteristics of five strata of archaeological maize cobs. roe 5 Level 4 Level: 8 Level 2 Level 1 Total population 200 613 798 903 61 Carbonized Y 85 44 0 ) 0 Charred % 15 1 0) 0) 0) Intact % 7 10 Q7 23 33 Fasciated % Q7 18 20 21 23 Unusual % 3 5 8 4 4 1 (low) 100 15 18 20 25 Teosinte 2 — 37 38 48 50 Introgres- 3 = 30 22 24 12 sion (%) 4 _- 15 19 7 5 5 (high) — 3 3 1 1 4 — 3 2 1 2 No. of 6 saad coed l 1 = Kernel 8 27 48 39 39 38 Rows 10 32 30 35 35 41 (%) 12 33 16 20 21 16 14 5 2 2 3 3 16 3 — — — — 10 — — 1 — — Diameter 12 — 3 4 2 — mm.(%) 14 — 14 11 1] 11 16 10 24 24 20 35 18 40 39 31 30 32 20 48 18 23 29 16 22 5 2 5 6 6 24 — — 1 1 — 26 =~ os — 1 —- 3 = — 3 2 — Length 4 aan 8 7 10 5 em.(%) 5 — 29 15 17 15 6 — 19 22 17 20 7 11 20 17 14 20 8 34 9 13 14. 15 9 25 7 8 9 10 10 30 4 6 8 10 11 — a 4 5 — 12 _ 4 3 4 — 13 — —- 1 — — 14 — — 1 — 5 15 _ — ~~ 1 — — Absent LITERATURE CITED Bluhm, Elaine, 1952. Clothing and textiles, Part VI, Mogollon cul- ture continuity and change. Fieldiana: Anthropology 40, Chicago Nat. Hist. Mus. Cosgrove, C. B., 1947. Caves of the upper Gila and Hueco areas in New Mexico and Texas. Papers of the Peabody Museum, Harvard Univ. 24: 1-181. Galinat, W. C., P. C. Mangelsdorf and L. Pierson, 1956. Estimates of Teosinte introgression in archaeological maize. Bot. Mus. Leafl. Harvard Univ. 17: 101-124. Grobman, A. and P. C. Mangelsdorf, 1959. Evidence for existence of a common prehistoric race in both North and South America. Maize Gen. Coop. News Letter 33: 28. Haury, Emil, 1934. The Canyon Creek ruin and the cliff dwellings of the Sierra Ancha. Medallion Papers 14, Gila Pueblo-Globe, Arizona. Hurst, C. T. and E. Anderson, 1949. A corn cache from western Colorado. Amer. Antiquity 14: 161-167. Kidder, A. V. and S. J. Guernsey, 1919. Archaeological explorations in northeastern Arizona. Bur. Amer. Ethnology Bulletin 65. Washington. Mangelsdorf, P. C., 1954. New evidence on the origin and ancestry of maize. Amer, Antiquity 19: 409-410. ——, 1958. The mutagenic effect of hybridizing maize and teosinte. Cold Spring Harbor Symp. Quant. Biol. 23: 409-421. —— and C, E. Smith, Jr., 1949. New archaeological evidence on evolution in maize. Bot. Mus. Leafl. Harvard Univ. 13: 213-247. and R. H. Lister, 1956. Archaeological evidence on the evolu- tion of maize in northwestern Mexico. Bot. Mus. Leafl. Harvard Univ. 17: 151-178. Reeves, R. G. and P. C. Mangelsdorf, 1942. A proposed taxonomic change in the tribe Maydeae (family Gramineae). Amer. Jour. Bot. 29: 815-817. Wellhausen, KE. J., L. M. Roberts and E. Hernandez X. in collabora- tion with P, C. Mangelsdorf, 1952. Races of maize in Mexico. Bus- sey Institution of Harvard University. [ 181 ] BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Campripcr, Massacuusetrts, Junge 30, 1961 Voi. 19, No. 9 CARLUDOVICA PALMATA IN BROOMMAKING BY MeE.vIN LEE Bristrou Carludovica palmata Ruiz & Pavon is best known as the source material for ‘‘Panama’’ hats, the majority of which are manufactured in Ecuador (2). The leaves are also used to a lesser extent for matting, curtains, roofing, baskets, cigar-cases, purses, fly swatters and brooms (2, 3,4). The petioles, when divided into strips, are used for making brooms in Honduras (1). At the eastern base of the Cordillera Oriental in the Comisaria de] Putumayo in southern Colombia, I re- cently encountered a household industry of broommak- ing from the dried leaf blades of Carludovica palmata, known locally as zraca. The brooms are made sporadically throughout the year at the convenience of the women of the household and are sold to an agent in the nearby town of Mocoa for twenty to twenty-five centavos each. Sent to the markets in the highland city of Pasto, they are resold for fifty to sixty centavos. The procedure of broommaking begins with the collec- tion of young, partially expanded leaf blades from plants in the vicinity of the house. They are spread on the ground near the home to dry in the sun for about four days, after which they are hung over a line strung be- 'See References. [ 183 | tween posts of the porch roof. Here they remain indefi- nitely until the housewife makes or obtains a light cord of cabuya (Agave spp.). When she is ready to begin to make a broom, the remaining two to four centimeters of petiole are cut from the leaves with a machete, but the leaf veins (or fibres) are carefully left coherent at their Ficure 1. Method of tying bundles of fibres to the cord. The center of each bundle is placed against the near side of the cord and the upper half of the previous bundle is brought down in front, looped around the cord and pulled snug. bases and are then pulled apart by hand. Why this is not done in one action with the machete is not apparent, for separating the fibres by hand takes several seemingly unnecessary minutes. Perhaps conserving the tough bases of the fibres affords the product a longer life. The cord is now stretched tautly across a corner of the porch at a height of about three feet. While working in a sitting position, the woman finds this a convenient height for making the broom. Beginning near one end, [ 184 ] she ties small bundles of the fibres at their centres to the cord so that both ends hang down (text fig. 1). Each knot may be tied with from three to about fifteen fibres, but the number is relatively constant for each broom. When many fibres are used, the knots are large, giving the completed product a knobby aspect at the top. It is unlikely that the life of the broom is different with either method, since both types of broom contain an equal number of fibres; possibly the size of bundle used for knotting caters to various aesthetic values of the consu- mers. Certainly knotting with large bundles is a more rapid method. Because some brooms are knotted with small bundles of fibres, it may be that aesthetic considera- tions are of more importance in this area than is economy of time. The fibres are tied closely along the cord for a distance of about one and a half meters, and when finished look like a grass skirt about twenty-five centimeters long. Untied from the porch railing, this ‘‘skirt’’ is rolled spirally on the end of a stick and securely bound. A few fibres which are too long are then trimmed off the end with a machete. When the broom is made for home consumption it is immediately provided with a handle. When it is to be sent to urban centers, however, it is rolled into a bundle without a handle, for the consumer simply unrolls a worn out ‘‘skirt’’ from his old broomstick and replaces it with the new one. [ 185 ] REFERENCES . Corréa, M. P. 1926. Diciondrio das Plantas Uteis do Brasil e das ExGticas Cultivadas. Vol. I, p. 319. Rio de Janeiro. . Harling, G. 1958. Monograph of the Cyclanthaceae. Acta Horti Bergiani 18: 1-428 (p. 117). Lund. . Pérez-Arbeliez, E. 1956. Plantas Utiles de Colombia. Bogota. p. 278. . Portillo, A. 1951. Divulgacién de Conocimientos Cientificos sobre las Plantas mas Utiles y Conocidas en Colombia, su Valor Alimen- ticio, Medicinal e Industrial. pp. 462-464. Pasto. EXPLANATION OF THE ILLUSTRATION Pirate XXV. (Upper) Carludovica palmata Ruiz & Pav. near Mocoa, Comisaria del Putumayo, Co- lombia. (Lower) Housewife with dried leaf blades of C. palmata on porch. Photographs by M. L. Brisror [ 186 ] PLATE XXV EXPLANATION OF THE ILLUSTRATION Prare XXVI. (Upper) Tying bundles of 3-5 fibres to the cord, as in Fig. 1. (Lower left) A new broom before trimming (left), and an old worn one (right). (Lower right) Top of broom securely bound to han- dle. Bundles of 8-10 fibres were tied to the cord in this example. Photographs by M. L. Brisvro. [ 188 ] PLATE XXVI NEONELSONIA—A COLOMBIAN FOLK MEDICINE BY Menvin LEE Brisrot. In August 1960, I studied the varieties of arracacha, the umbelliferous Arracacia wanthorhiza Bancroft, in the Colombian Andes. On one occasion during the study, in the Indian village of Sibundoy (near Pasto) in the Comisaria del Putumayo, asmall boy led me to a growth of what he termed ‘‘wild arracachas.’’ It was a coarse tangle of an umbelliferous plant clambering over shrubs in an area of 15 square meters. Since this ‘‘wild arraca- cha’’ appeared to me to be very closely related to the genus I was studying, if not the same, I collected speci- mens in flower and fruit. When I took them back to one of the older Sibundoy Indians, Juan Pedro Chindoy, he called the plant ‘‘ingo-sha-hush’’ in the Kamsa language. Utilization The man told me that the plant is used as a remedy for swelling and inflammation of the upper region of the intestine (hinchazones intestinales swperiores). Further- more, he said that it is employed by all Sibundoy women immediately after childbirth ‘‘to prevent their death.’ In both cases, the preparation and dosage is the same: the leaves and stems are boiled well, then some sugar and ten drops of a distilled alcoholic beverage (trago) are [ 191 | added. One-half demitasse cupful (15 ce.?) is taken internally. Taxonomy I have identified the Sibundoy ‘‘ingo-sha-hush’’ as Neonelsonia acuminata (Bentham) Coulter & Rose ex Drude. Neonelsonia, described in 1895 by Coulter and Rose (4), comprises two species: the type, Neonelsonia ovata of the mountains of southern Mexico; and the spe- cies under consideration here. Neonelsonia acuminata is a scandent, herbaceous, essentially glabrous perennial with a long, woody taproot. The leaves are ternately compound with ovate to lanceolate, spinulose-serrate leaflets often lobed toward the base. The compound um- bels lack an involucre but possess filiform involucels which frequently surpass the fertile pedicels in length. The greenish yellow petals are obcordate, with a nar- row, inflexed tip. The ellipsoid-cordate fruits have five prominent, fleshy ridges. Neonelsonia acuminata bears many resemblances to various species of Arracacia, particularly, as noted by Constance (8), to_4. Pennellit Constance, A. Wigginsii Constance and 4. elata Wolff. Mathias and Constance (6) recognized six differences between Neonelsonia and Arracacia. Four of these differences—the shape of the petal apex, the degree of reduction of the calyx, the po- sition of the oil ducts, and the shape of the groove on the seed face—seem unimportant, for upon examination these characters are seen to grade from one genus to the other. Of greater importance in distinguishing these two genera is the wrinkled surface of the fruits of Neonelsonia, possibly lacking schlerenchymatous tissue, and more es- pecially, their ellipsoid-cordate form. Examination of sixteen type specimens of Arracacia in the Harvard University Herbarium shows that the [ 192 | fruit varies from lanceolate and oblong to ovate, but that none are ellipsoid-cordate. In view of the many resem- blances between these two genera, however, it is possible that a future monographer of Arracacia might emend the genus-concept to include Neonelsonia. Distribution Specimens of Neonelsonia acuminata in the Harvard University Herbarium indicate that its range is Colom- bian and Ecuadorean, extending from the Departamento del Cauca in the north to the Provincia de Azuay in the south at elevations of 2450 to 3660 meters. My speci- mens from Sibundoy in the Comisaria del Putumayo, Colombia, were collected at 2100 meters, extending the known altitudinal limits of the species. Another collec- tion, from San Diego near Guachucal, Narifio, is within the previously known range. Because Neonelsonia acuminata is not readily distin- guished in the field from Arracacia Pennell and from A. Wigginsii when fruits are not available, it is well to note the distribution of these two species of Arracacia in central Colombia and central Ecuador, overlapping the range of N. acuminata. The specimens of Arracacia Pennellit available to Constance (8) when the species was described came from Cundinamarca, Norte de Santander and Santander in Colombia at elevations of 3000 to 3800 meters. Likewise, the specimens of Arracacia Wigginsu were collected in Cafiar and Azuay (one at 8660 meters) in Ecuador. A later collection, now in the Harvard University Herbarium, is also from Azuay but at 2740 meters. Our present scanty knowledge indicates that of these three species, only Neonelsonia acuminata is found in southern Colombia and northern Ecuador and that its altitudinal tolerance extends to lower elevations than does that of the two species of Arracacia. [ 193 ] Specimens of Neonelsonia acuminata examined Cotomnia. Cauca: 3100-3300 m. alt., June 11-13, 1922, Pennell & Killip 6632, 6676. Putumayo: 2100 m. alt., August 3, 1960, Bris- tol 240. Narifio: 3140 m. alt., August 10, 1960, Bristol 247. Ecua- por. Pichincha: 3000-3600 m., alt., August 13, 1923, Hitchcock 20881; March 31, 1920, Holmgren 449; 12,000-12,600 ft. alt., 1855, Couthouy. Azuay : 8000-9600 ft. alt., July 27-August 12, 1945, Camp KE 4508. ur REFERENCES Bancroft. 1825. Trans. Agric. Hort. Soc. Jamaica 1825, 3. er Bentham, G. 1845. Plantae Hartwegianae. London. 187. . Constance, L. 1949. The South American species of Arracacia (Umbelliferae) and some related genera. Bull. Torr. Bot. Club 76: 39-52, . Coulter, J. M. & J. N. Rose. 1895. Contr. U.S. Nat. Herb. 3: 306, Drude, O. 1898. Umbelliferae in Natirlichen PHanzenfamilien 3 (8): 167. Leipzig. Mathias, M. EK. & L. Constance. 1944. Umbelliferae (Pars) in North American Flora 28B, 43-160. New York. [ 194 | NOVELTIES IN THE ORCHID FLORA OF THE GUAYANA HIGHLANDS * BY CHARLES SCHWEINFURTH THE present paper, which is the first of two articles deal- ing with novelties in the Orchidaceae of the Guayana Highlands, treats eleven new species. The order follows that of the System proposed by Dr. Rudolph Schlechter in Notizblatt des Botanischen Gartens und Museums, Berlin-Dahlem 9, No. 88 (1926) 563-591. Duckeella alticola C. Schweinfurth sp. nov. Herba terrestris, pro genere robusta, nigrescens, cir- citer 50 cm. alta. Caulis glaber, saepissime superne in- aequaliter bifurcatus. Folia valde coriacea, biformia; basalia pauca, oblongo-linearia, acuta, suberecta, usque ad 19 vel 22 cm. longa et 9-16 mm. lata; caulina duo, multo breviora, remotissima, elliptico-oblonga, acuta, 8.7-10 cm. longa, usque ad 1.8 cm. lata. Racemi in ramorum apice, densius pauci- vel pluriflori. Flores mediocres, aurei. Sepala crassa, concava, marginibus in- curvis. Sepalum dorsale oblongo-ellipticum, acutum, circiter 2 cm. longum et 8.2 mm. latum. Sepala lateralia persimilia, dorso carinata. Petala latiora, ovalia vel late * For the privilege of studying the extensive collections recently made in this region, I am indebted chiefly toa grant from the National Science Foundation. [ 195 | ‘XPLANATION OF THE ILLUSTRATION Pirate XXVIT. Duckretia articota C. Schweinfurth. 1, plant, one third natural size. 2, flower expanded, one and one quarter times natural size. 3, column and lip, three quarters view, two and one quarter times natural size. Drawn by E.mer W. Situ [ 196 | Puate XXVIII | | C. Schweinf. cola alti DUCKEELLA, elliptica, submembranacea, circiter 1.8 cm. longa et 1.2 em. lata. Labellum in circuitu oblongum, circiter 1.75 cm. longum et 8 mm. latum, basi trilobatum cum lobu- lis parvis semiorbicularibus, apice truncato abrupte acu- tum; discus inter lobos laterales callo brevi pluristriato emarginato ornatus. Columna comparate parva, circiter 8 mm. alta, cucullata, apice bifida, basi dente oblongo prominenti utrinque praedita. This orchid differs conspicuously from the other spe- cies of Duckeella in having broader leaves, fleshy, concave and acute sepals, a relatively short callus on the lip and a pair of conspicuous teeth on the column. It isa compara- tively robust plant of consistently rather high altitudes. Amazonas: Cerro Duida, moist slopes of Savanna Hills, 1350 meters altitude, August 1928—April 1929, G. H. H. Tate 736; Cerro Sipapo (Paraque), near summit of West Peak, at 1750 meters altitude, infre- quent in bogs, flowers yellow, fragrant, December 20, 1948, Bassett Maguire & Louis Politi 27797. Botivar: Cerro Guaiquinima, Rio Para- gua, occasional near Cumbre Camp, at 2000 meters altitude, flowers dull yellow, December 25, 1951, B. Maguire 82748; Same locality, rare in bog, along west escarpment rim 1 km. west of Cumbre Camp, at 1800 meters altitude, flowers yellow, December 30, 1951, B. Maguire 32880, 328394; Same locality, frequent in bogs, ‘‘North Valley,’’ at 1600-1700 meters altitude, flowers yellow with the outer members [sepals] bronze outside, January 4, 1952, B. Maguire 32973 (Type in Herb. Ames No. 69588). Sobralia speciosa C. Schweinfurth sp. nov. Herba elata, speciosa, terrestris, 1-8 metralis (fide collectoris). Caules plusminusve robusti, vaginis longe tubularibus arctis glabris obtecti. Folia numerosa, dis- ticha, lanceolata vel ovato-lanceolata, longe acuminata, plicata, supra nitentia, usque ad 15 cm. longa et 3.7 cm. lata. Inflorescentiae saepissime laterales (raro terminales), racemosae, laxe pauciflorae, suberectae. Flores magni spectabilesque, purpurei, membranacei. Sepalum dorsale oblongo-oblanceolatum, acutum, circiter 4.5-5.5 em. longum et 1.3 cm. latum. Sepala lateralia similia, an- [ 198 ] guste oblanceolato-oblonga, paulo obliqua, dorso carinata, circiter 5-5.6 cm. longa et 1.3. cm. lata. Petala oblongo- obovata, apice late rotundata, sepalis latiora, circiter 2-3 em. lata. Labellum majus, segmenta cetera superans, in positu naturali convolutum, expansum suborbiculari- ellipticum, circiter 6-7 cm. longum et 5 cm. latum, apice profunde bilobatum; discus carinis pluribus superne la- ceris percursus. Columna circiter 3-3.5 cm. alta, apice falcula incurva utrinque ornata. This species appears to be related to the Peruvian Sobralia Weberbaueriana Kriinzl., but has smooth (not furfuraceous) sheaths, much smaller leaves and a dis- similar lip and column. Amazonas: Cerro de la Neblina, Rio Yatua, near Cumbre Camp, at 1700 meters altitude, occasional on low bushy slopes, “‘roots fleshy ; corolla cerise; lip with median yellow line and several white lines; column whitish, pink-flushed,’’ January 4, 1954, Bassett Maguire, John J. Wurdack & George S. Bunting 37028; Same locality, occasional in scrub forest 1 kilometer north of Cumbre Camp, at 1800 meters alti- tude, “‘fls. rich pink,’’ January 10, 1954, Maguire, Wurdack & Bunt- ing 37202; Same locality, locally frequent in scrub forest near Cumbre Camp, at 1800 meters altitude, ‘‘fls. magenta, the lip with subapical white area,’’ November 19, 1957, Bassett Maguire, John J. Wurdack & Celia K. Maguire 42142 (Tyrer in Herb. Ames No. 69507). Stelis latisepala C. Schweinfurth sp. nov. Herba pusilla, caespitosa, epiphytica, 4-5 cm. alta. Caulis abbreviatus, circiter 83 mm. altus, vaginis laxis tubularibus omnino celatus. Folium erectum, oblanceo- latum vel obovato-spathulatum, apice rotundatum, ses- sile, 8-14 mm. longum, expansum usque ad 4.2 mm. latum. Inflorescentia singula, folium multo superans, supra laxe pluriflora, usque ad 4.7 cm. longa. Flores perparvi, subcarnosi. Sepala transverse rhombico-ovata, acuta, trinervia, basi connata. Sepalum dorsale circiter 2.6 mm. longum et 8 mm. latum. Sepala lateralia similia sed minora, valde obliqua, usque ad 2 mm. longa et 2.5 [199 J EXPLANATION OF THE ILLUSTRATION Pirare XXVIII. Sopracta speciosa C. Schweinfurth. 1, plant, about one half natural size. 2, dorsal sepal, three quarters natural size. 3, petal, three quarters natural size. 4, lateral sepal, three quar- ters natural size. 5, lip, three quarters natural size. 6, column from side, three quarters natural size. Drawn by Ev.mer W. Siri [ 200 ] PLaTtE XXVIII speciosa C Schweinf. mm. lata. Petala minuta, rhombico-ovata, supra multo incrassata, circiter 0.7 mm. longa et 0.9 mm. lata. La- bellum simplex, petalis simile, rhombico-ovatum, circiter 0.6 mm. longum et 0.9 mm. latum; discus callo carnoso transverso bilobato prope medium ornatus. Columna generis, minuta, supra abrupte dilatata. This little plant seems to be allied to the Brazilian Stelis parvifolia Garay, but has larger glabrous flowers. Botivar: Chimanta Massif, elfin forest formation on plateau of southeast-facing upper shoulder of Apdcara-tepui, at 2000 meters al- A ee é . titude, epiphyte on branch; leaves fleshy-coriaceous, pale green; seape brown-purple as are the pedicels; calyx [sepals] copper-brick with tawny-yellow margins,’’ June 19, 19538, Julian A. Steyermark 75712 (Typr in Herb. Ames No. 69505). Stelis obovata C. Schweinfurth sp. nov. Herba parvula, caespitosa, epiphytica, usque ad 12 em. alta. Caules approximati, breves, usque ad 1.3 cm. alti, bivaginati, vaginis arctis, tubulatis obtecti. Folium erectum, spathulatum vel oblanceolatum (raro obova- tum), apice saepissime obtusum vel rotundatum, usque ad 2.8 cm. longum et 9 mm. latum. Inflorescentiae fo- lium multo superantes, usque ad 10.8 cm. longae; race- mus superne dense multiflorus, saepe secundus. Flores minuti, carnosiores. Sepala rotundato-ovata, obtusa vel subacuta, concava, trinervia, subaequalia, basi paulo con- nata, lateralia obliqua. Sepalum dorsale circiter 2 mm. longum et 1.8 mm. latum. Petala multo minora, in posi- tu naturali cuneato-obovata, circiter 0.8 mm. longa et fere lata, parte superiore carnosissima. Labellum sim- plex, cuneato-obovatum, apice subtruncatum et incras- satum, circiter 1 mm. longum et latum; discus medio cum callo lanceolato-pandurato. Columna generis, mi- nuta, supra abrupte dilatata. This species appears to be closely allied to Stelis len- tiginosa Lindl., the peculiar lip being almost identical, [ 202 | Puate XXIX STELIS LATISEPALA C, Schweinfurth Flower expanded, lip and petal, all much enlarged. STELIs opovaTA C. Schweinfurth Flower expanded, lip and petal, all much enlarged. Drawn by Etmer W. Situ but it differs in having a caespitose habit of growth and much smaller flowers. Amazonas: Cerro Huachamacari, Rio Cunucunuma, epiphytic in cumbre near Summit Camp, at 1800 meters altitude, flowers cream- colored, December 14, 1950, Bassett Maguire, R. 8. Cowan & John J. Wurdack 30204. (This collection differs from the type in its much smaller vegetative proportions and somewhat smaller flowers. ) Borivar : Chimanta Massif, Central Section, on rocky slopes of zanjon bordering Upper Falls of Rio Tirica above Summit Camp, at 1950 meters alti- tude, ‘epiphyte on mossy branch, lvs. coriaceous, pale green; pedun- cle and pedicels pale green; fls. greenish yellow,’’ February 7, 1955, Julian A, Steyermark & John J. Wurdack 556 (Tyre in Herb. N.Y. Bo- tanical Garden). Octomeria cordilabia C. Schweinfurth sp. nov. Herba valde variabilis, caespitosa, terrestris vel epiphy- tica. Caules pergraciles, elongati, usque ad 37 em. alti, multiarticulati, vaginis longe tubularibus arctissimis om- nino velati. Folium lanceolato-lineare, sessile, apice mi- nute et oblique bilobatum, crasse coriaceum, usque ad 6.6 cm. longum et circiter 5 mm. latum. Flores in folii axilla fasciculati, parvi, aurei vel aurantiaci, membrana- cel. Sepalum dorsale late elliptico-ovatum, subacutum, saepissime trinervium, usque ad 5.2 mm. longum et 4 mm. latum. Sepals lateralia breviora et latiora, orbiculari- ovata, saepissime trinervia, usque ad 4.83 mm. longa et 4.1 mm. lata. Petala late ovato-elliptica, subacuta, tri- nervia, usque ad 4.1 mm. longa et 8.4 mm. lata. Labellum segmentis ceteris multo minus, suborbiculari-ovatum, antice late rotundatum et minute apiculatum, basi corda- tum, usque ad 1.7 mm. longum et 2.4 mm. latum; dis- cus callis binis carnosis convergentibus ornatus. Columna parva, arcuata, basi apiceque dilatata. This species does not appear at present to have any close allies. The relatively long stems, small leaves and entire lip are distinctive. Amazonas: Cerro Huachamacari, Rio Cunucunuma, between Sum- i ce . . . mit Camp and © East Ridge’’ savanna, at 1800 meters altitude, in [ 204 ] densely wooded valley, December 8, 1950, Bassett Maguire, R. S. Cowan & John J. Wurdack 30035; Cerro de la Neblina, Rio Yatua, near Cumbre Camp, in scrub forest along runlet, at 1700 meters alti- tude, on rocks and tree trunks, January 5, 1954, Bassett Maguire, John J. Wurdack & George S. Bunting 37067; Same locality, along escarpment west of Cumbre Camp, at 1700-1800 meters altitude, occa- sional terrestrial, Maguire, Wurdack & Bunting 37099; Same locality, on Cafio Grande slopes, east of Cumbre Camp, at 1600-1800 meters altitude, on limb of low tree, November 22, 1957, Bassett Maguire, John J. Wurdack & Celia K. Maguire 42183; Same locality, near Cumbre Camp, at 1800 meters altitude, occasional terrestrial in scrub forest, November 29, 1957, Maguire, Wurdack & Maguire 42258 (Tyrer in Herb. N.Y. Botanical Garden). Octomeria dentifera C. Schweinfurth sp. nov. Herba parva, caespitosa, saxicola, usque ad 14 cm. alta. Caules graciles, breves, pauciarticulati, 5—8 cm. alti, vaginis longe tubulatis maxima pro parte obtecti. Folium erectum, lineari-lanceolatum vel elliptico-lineare, acut- um, sessile, crasse coriaceum, 3.5—6.7 cm. longum, usque ad 6 mm. latum. Flores bini ut videtur, axillares, pro planta magni, membranacei. Sepala similia, trinervia, lanceolata, 10.5-11 mm. longa. Sepalum dorsale acumi- natum, usque ad 4.2 mm. latum. Sepala lateralia paulo angustiora, leviter obliqua, longe acuminata, usque ad 3.5 mm. lata. Petala sepalo dorsali similia, ovato-lanceo- lata vel elliptico-lanceolata, acuminata, usque ad 10 mm. longa et 3.7 mm. lata. Labellum segmentis ceteris multo minus, in positu naturali oblongo-ellipticum et 5.4 mm. longum, infra medium trilobatum cum lobis lateralibus parvis auriculiformibus erectis et lobo intermedio com- parate magno, oblongo-ovato, antice subtruncato, mar- ginibus irregulariter dentatis; discus in medio cum cari- nis binis carnosis. Columna gracilis, arcuata, circiter 4 mm. alta, in pedem conspicuum producta. This species is allied to Octomeria parvula C.Schweinf., but is larger throughout with the flower nearly twice as [ 205 ] large. The specific name is in allusion to the dentate margins of the lip. Borivar : Churu-tepui (Muru-tepui), northwest cumbres, occasional on rock ledges in upper cumbre, at 2250-2300 meters altitude, flowers white, January 26, 1953, John J. Wurdack 34218 (Tyrer in Herb. Ames No. 69525); Chimanta Massif, east central portion of summit of Apdcara-tepui, at 2450-2500 meters altitude, ‘‘on moist ledges of high large rock around cave recess, leaves dark purple or dull green with purple; pedicels reddish; sepals, petals and lip pale yellow, nodding,’’ June 21-22, 1953, Julian A. Steyermark 75867. Octomeria filifolia C. Schweinfurth sp. nov. Herba gracillima, epiphytica, caespitosa, usque ad 17 em. alta. Caules tenues, pluriarticulati, circiter 3.5—9 cm. alti, vaginis longe tubulatis arctissimis celati. Folium angustissime lineare vel filiforme, in vivo subteres, usque ad 9 cm. longum et in sicco 1.5 mm. latum. Flores in glomerulis axillaribus, minimi, membranacei. Sepala tri- nervia. Sepalum dorsale oblongo-lanceolatum, acutum vel acuminatum, tubulari-involutum, circiter 3 mm. longum et 1 mm. latum expansum. Sepala lateralia triangulari-lanceolata, acuta, cum pede mentum forman- tia, circiter 8 mm. longa et 1.2 mm. lata. Petala lan- ceolata-linearia, acuta vel acuminata, tubulari-involuta, I-nervia, sepalis paulo breviora, circiter 2.4 mm. longa et 0.6 mm. lata. Labellum parvum, tubulari-involutum, prope medium trilobatum, basi cuneatum, expansum circiter 1.5 mm. longum et fere 1 mm. latum; lobi later- ales oblique semiovati, acuti; Jobus intermedius multo major, suborbicularis, apice rotundatus et minute apicu- latus; discus lobi medii basi obscure bicarinatus. Colum- na antice plana, circiter 1.4 mm. alta. No close ally of this species was noted. Borivar: Chimanta Massif, northwestern part of Abdcapa-tepui, vicinity of Camp 8, at 1300 meters altitude, on forested slopes adja- cent to quebrada, epiphyte on tree trunk, ‘‘leaves purplish with green ; petiole purple; flowers pale green with lavender,’’ April 20, 1953, Julian A. Steyermark 75181 (Tyee in Herb. Ames No. 69526). [ 206 | Octomeria flaviflora C. Schweinfurth sp. nov. Herba elata, caespitosa, terrestris. Caules pluriarticu- lati, usque ad 47 cm. alti, vaginis longe tubulatis mar- cescentibus maxima pro parte obtecti. Folium erectum, lineari-oblongum, apice minute tridenticulatum, basi sessile, valde coriaceum, usque ad 18 cm. longum et 1.4 em. latum. Flores numerosi, in glomerulis axillaribus, flavi, membranacei, cum segmentis patentibus. Sepala petalaque lanceolata, acuminata, trinervia. Sepalum dor- sale longe acuminatum, 8-11 mm. longum, 8-8.5 mm. latum. Sepala lateralia simillima, paulo obliqua, 8-11 mm. longa, 2.2-8.1 mm. lata. Petala paulo breviora, ovato-lanceolata, saepissime latiora, circiter 7-9.2 mm. longa, usque ad 4mm. lata. Labellum segmentis ceteris multo minus, in circuitu ovato-oblongum, usque ad 5 mm. longum et 3 mm. latum, prope basim trilobatum; lobi laterales perparvi, anguste falcato-oblongi; lobus intermedius comparate magnus, oblongo-ovatus, apice saepissime subtruncatus et medio acutus, cum margini- bus denticulato-erosis; discus callis binis humilibus orna- tus. Columna parva, arcuata, usque ad 3-4 mm. alta, in pedem brevem producta. Amazonas: Cerro de la Neblina, Rio Yatua, at 2000 meters altitude, locally frequent in rocky ravine 16 kilometers southwest of Cumbre Camp, December 1-2, 1957, Bassett Maguire, John J. Wurdack & Celia K. Maguire 42280; Same locality, at 1900-2000 meters altitude, locally frequent in upper Cafion Grande basin above Salto Grande, December 13, 1957, Maguire, Wurdack & Maguire 42362; Cerro Sipapo (Paraque), at 1600 meters altitude, occasional along streambanks in Cafio Pro- fundo, January 12, 1949, Bassett Maguire & Louis Politi 28315; Cerro Huachamacari, Rio Cunucunuma, at 1800 meters altitude, occasional on left bank of Cafio de Dios in cumbre near Summit Camp, December 6, 1950, Bassett Maguire, R. S. Cowan & John J. Wurdack 30024 (Tyre in Herb. Ames No. 69523); Same locality and altitude, locally fre- quent in dense woodland in cumbre along right fork of Cafio de Dios near Summit Camp, December 13, 1950, Maguire, Cowan & Wurdack 30194. Borivar: Cerro Guaiquinima, Rio Paragua, at 1500 meters altitude, occasional in open savanna on precipitous slope below west [ 207 ] EXPLANATION OF THE ILLUSTRATION Prare XXX. Ocromeria. 1, O. FLAviIFLoRA C. Schweinfurth. Flower expanded, three times natural size. Lip, three quarters view, five and one half times natural size. 2, O. nana C. Schweinfurth. Flower expanded, eight times natural size. Lip, three quarters view, eleven times natural size. 3, O. pentirera C. Schweinfurth. Flower expanded, two and one half times natural size. Lip, three quarters view, four and one half times natural size. 4, O. LtancrperaLa C. Schweinfurth. Flower ex- panded, five and one half times natural size. Lip, three quarters view, twelve times natural size. 5, O. rivirotia C, Schweinfurth. Flower expanded, nine times natural size. Lip, three quarters view, twenty-two times natural size. 6, O. corDILABIA C. Schweinfurth. Flower expanded, five and one half times natural size. Lip, three quarters view, twelve times natural size. Drawn by Ermer W, Situ [ 208 | PLaTE XXX OCTOMERIA filifolia cordilabia escarpment, December 31, 1951, B. Maguire 32877; Chimanta Massif (Central Section), at 1940 meters altitude, on border of large rocks by large savanna below Upper Falls of Rio Tirica above Summit Camp, “*sepals yellow with dull brick-red margins and tips; lip maroon; column greenish with pale yellow apex,’’ February 7, 1955, Julian A. Steyermark & John J. Wurdack 607. Octomeria lancipetala C. Schweinfurth sp. nov. Herba parva vel mediocris, epiphytica. Caules graciles, usque ad 17.5 cm. alti, pluri- vel multiarticulati, vaginis longe tubularibus arctissime celati. Folium lanceolato- lineare, apice acutum vel minute tridenticulatum, crasse coriaceum, circiter usque ad 9 cm. longum et 4.5 mm. latum. Flores parvi, numerosi, in glomerulis axillaribus, cum segmentis patentibus. Sepalum dorsale ovatum vel lanceolato-ovatum, acutum vel acuminatum, trinervium, circiter 4 mm. longum et 2.1 mm. latum. Sepala later- alia similia, sed paulo longiora et angustiora. Petala an- guste lanceolata, acuta vel acuminata, uninervia, circiter 3.2-3.8 mm. longa et 1 mm. lata. Labellum segmentis ceteris multo minus, simplex, valde geniculatum, antice late rotundatum et aliquando apiculatum, postice cor- dato-truncatum, circiter 1.1 mm. longum et 1.5—1.8 mm. latum, cum ungue brevi cuneato; discus basi incrassatione obliqua utrinque ornatus. Columna minuta, crassa, apice dilatata, circiter 1 mm. alta. This species appears to be allied to Octomeria cordilabia C. Schweinf., but has commonly much shorter stems and very dissimilar sepals and petals. Amazonas: Cerro Huachamacari, Rio Cunucunuma, frequent in cumbre, at 1700 meters altitude, flowers purple, December 4, 1950, Bassett Maguire, R. S. Cowan & John J. Wurdack 29823; Same locality, along right fork of Cafio de Dios near Summit Camp, at 1800 meters altitude, in dense woodland, December 13, 1950, Maguire, Cowan & Wurdack 30170A (Tyre in Herb. Ames No. 69528); Cerro dela Neb- lina, Rio Yatua, in scrub forest 1 to 5 kilometers north of Cumbre Camp, at 1800 meters altitude, occasional on tree trunks, buds ma- roon, January 10, 1954, Bassett Maguire, John J. Wurdack & George [ 210 | S. Bunting 37161; Same locality, in swale 2 kilometers northeast of Cumbre Camp, at 1800 meters altitude, on limb of low tree, flowers maroon, November 20, 1957, Bassett Maguire, John J. Wurdack & Celia K, Maguire 42150. Octomeria nana C. Schweinfurth sp. nov. Herba pusilla, caespitosa, rupicola, usque ad 4 cm. alta. Caules brevissimi, usque ad 18 mm. alti, circiter 2-articulati, vaginis tubulatis marcescentibus celati. Fo- lium oblongo-lineare, valde carnoso-coriaceum et per me- dium sulcatum, acutum, sessile, usque ad circiter 23 mm. longum et 2 mm. latum in sicco. Flores axillares, pauci, glomerati, membranacei. Sepala similia, prope basim trinervia. Sepalum dorsale ovato-oblongum, acutum, concavum, circiter 8.7 mm. longum et 1.5 mm. latum. Sepala lateralia oblique ovato-oblonga, acuta, circiter 3.5 mm. longa et 1.5 mm. lata. Petala oblique oblongo- lanceolata, concava, subacuta, circiter 3.5 mm. longa et 1.2 mm. lata, maxima pro parte uninervia. Labellum segmentis ceteris multo brevius, simplex, suborbiculare, antice rotundatum et apice abrupte acuto, ad basim ro- tundatam angustatum, trinervium, circiter 2.2 mm. lon- gum et 2mm. latum; discus in medio obscure bicallosus. Columna minuta, crassa. This dwarf species seems to have no near allies. The only available flowers occurred at the summit of en- larged, ellipsoid ovaries. Botivar: Chimanta Massif, Torono-tepui, north-facing slopes on summit above Cafio Mojado, on rock in savanna, at 2030-2150 meters altitude, “‘lvs. 3-4 mm. thick,’’ February 21, 1955, Julian A. Steyer- mark and John J. Wurdack 1027 (Tyre in Herb. Ames No. 69524). Ponthieva ovatilabia C. Schweinfurth sp. nov. Herba terrestris, elata, usque ad 87 cm. alta. Radices fasciculatae, fibrosae, lanuginosae. Folia plura, in herbae basi, longe petiolata; lamina ovato-lanceolata vel elliptico- lanceolata, acuminata, basi cuneata, membranacea, tri- [ 211 ] EXPLANATION OF THE ILLUSTRATION Pirate XXXII. Ponruieva ovativasia C. Schweinfurth. 1, plant, one half natural size. 2, flower expanded, with lip and column foreshortened, five times natu- ral size. 3, lip expanded, six times natural size. Drawn by Eimer W. Situ [ 212 ] PLate XXXII ONTHIEV. - EV ° OVAL C. Schwe inf. nervia, 7-13 cm. longa, usque ad 3.8 cm. lata; petiolus canaliculatus, satis gracilis, circiter 5-10 em. longus. Caulis glanduloso-pilosus, usque ad inflorescentiam cir- citer 59.6—-71.4 em. altus, vaginis pluribus remotis orna- tus. Racemus laxe multiflorus, cum rhachide circiter 10-16 cm. alta. Flores parvi, viridi-albidi. Sepala extus sparse glanduloso-pilosa, Sepalum dorsale elliptico-lan- ceolatum, ad apicem subacutum angustatum, trinervium, 6-7 mm. longum, circiter 2 mm. latum. Sepala lateralia oblique ovata, subacuta, 6-7 mm. longa, circiter 3.1-3.4 mm. lata, quinquenervia. Petala unguiculata; lamina obliquissime triangularis, acuta, basi semicordata, circiter 4.7 mm. longa et 3 mm. lata; unguis oblongo-linearis, carnosus, circiter 1.8 mm. longus. Labellum columnae parti superiori adnatum, valde inflexum, cum lateribus involutis, expansum in circuitu ovatum, circiter 4 mm. longum et inferne 2.3 mm. latum, parte anteriore ob- longo-lineari et parte posteriore suborbiculari; discus basi cum callo rotundato-hippocrepiformi. Columna perbrev- is, crassa, supra abrupte dilatata, circiter 2.5 mm. alta. This species is vegetatively similar to Ponthieva dip- tera Linden & Reichb.f., but the petals and lip are very different. The lip somewhat recalls that of P. ecuadoren- sis Schltr. Amazonas: Cerro de la Neblina, Rio Yatua, 700 meters altitude, occasional in Clusia scrub forest just south of Camp 8, December 31, 1957, Bassett Maguire, John J. Wurdack and Celia K. Maguire 42559 (Tyre in Herb. Ames No. 69529). Borivar: Cerro Venamo, north- west slopes, 1100 meters altitude, terrestrial in moist mossy forest, leaves firmly membranaceous, deep green above, dull gray green be- neath, pedicels recurved, nodding, pale green, ovary ivory-white, April 21, 1960, Julian A. Steyermark & S. Nilsson 431. [ 214 ] BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY Campripgr, Massacuusetts, May 7, 1962 Vor. 19, No. 10 EDIBLE FRUITS OF SOLANUM IN SOUTH AMERICAN HISTORIC AND GEOGRAPHIC REFERENCES‘! BY Victor MANUEL PatTINo’” One of the imperfectly understood aspects of economic botany in South America seems to be the use of the edi- ble fruits of sundry species of Solanum. Not only is the extent to which long-known species are employed a ques- tion; but how many species, some perhaps not yet de- scribed, are involved remains for intensive agronomic and taxonomic research to clarify. The history of domestica- tion and geographic dispersal of several of the Solanum- concepts herein considered remains, in some aspects, un- certain. It is hoped that a thorough consideration of historic and geographic reports of these plants may add to our growing understanding of them. Solanum quitoense Lamarck Ulustr. 2 (1797) 16. VERNACULAR NAMES: Lulo in western Colombia. This article is part of a work on the history of cultivated plants in equinoctial America which I have been preparing with the help of the John Simon Guggenheim Memorial Foundation and OAS Fellowship Program. The research has been done mainly in the Library of Con- gress, Washington, D.C. and the Botanical Museum of Harvard Uni- versity, Cambridge, Mass. * Formerly Chief of Colonial Crops, Secretariat of Agriculture, Cauca Valley, Colombia, [ 215 ] Ma-sha-kvé in Kamsa or Koche, Sibundoy, Colombia (Schultes, 1949, 45). Naranjilla in Ecuador, southernmost Colombia. The few authors who have been concerned with this fruit agree that the word lu/o is of Keshwa origin. Some linguists attribute it to the Quitoan or northernmost form of ruru, meaning ‘‘egg,’” “‘fruit’’ (Lira, 1945, 557; ‘Toscano Mateus, 1953, 93). The earliest known Keshwa- Spanish vocabulary states of /u//u: ‘‘unripe thing: soft bud of tree or anything similar’” (Domingo de Santo Tomas, 1560, 147). Another Keshwa vocabulary of the beginning of the 17th Century reports: ‘‘W/udlu-ruru, everything that is tender before becoming hard”; and ‘Uullu-ruru=tender, milky fruit’? (Gonzalez Holguin, 1608, 213; Gonzalez Holguin, 1952 (ed. fascim.) ). The first reported name which was applied to a solanaceous plant with a description corresponding reasonably to Solanum aff. quitoense Lam. is puscolulu, derived, as suspected by Jiménez de la Espada (see below), from ppocheco-ruru: ‘sour or acid fruit.’ Gonzalez Holguin, in fact, says: ‘‘pochcco=yeast or thing acid or sour” (Gonzalez Holguin, 1608, 295). Theoretically, it could also be accepted that puscolulo means ‘‘mucilaginous fruit,’’ from ‘‘pucoco=‘‘foam or slaver’’ (Domingo de Santo Tomis, 1560, 162 v.). While we admit that /uw/o is of Keshwa origin, it is not clear how the term pusco has disappeared from puscolulo, whereas the names of several other fruits have kept it: e.g., asna-lulo and chaqui-lulo, quoted by some colonial sources for fruits from the highlands of the Province of Pasto: and chonta-ruro, a name known since the end of the 16th Century (Patino, 1958, AI, NVIILI, 177-204; 299-332). Since Solanum quitoense is indigenous to the equinoc- tial region, one may justifiably assume that the name [ 216 | lulo belonged to a local language. In the Kolorado lan- guage, formerly spoken on the coast and western Andean slope of Ecuador, the root du refers to ‘‘red’’ and ‘‘yel- low’’ and, as a natural extension, ‘‘ripeness’’ (Jijén y Caamanio, 1941, [I, 249). The repetition /u/o should in- dicate, as stated by Buchwald, ‘‘red, red’’ (ibid., 250). There is a river in that same region, the Rio Lulo, a tributary of the Palenque (Wolf, 1892, 188). In Kamsa and its filial tongue Koayker, the particle sha (thsa, za, scha, cha) is equivalent to ‘‘thing good, ad- mirable’’ and serves to classify the fleshy objects (Jijén y Caamano, 1940, I, 102-108, 105, 107, 109, 117-120, 122-124; 157, 160, 191-192). In Kamsa, be means ‘‘round’’ (ibid., 118, 122). The origin of the word naranjilla is established in the references quoted below. In view of the lack of botani- cal collections, some of the data on several so-called varieties of naranjilla (Gattoni, 1935, 7) may refer per- haps to different species. A few years ago, Schultes and Cuatrecasas described a variety of Solanum quitoense * occurring north of the equator and characterized by the presence of spines on the leaves. In 1652, the naturalist Bernabé Cobo, perhaps making use of information sent to him by Jesuit correspondents, described under different regional names one or more species of Solanum with edible fruits, native of the re- gions of Popayadn and Quito. Living in Lima and Mex- ico, Padre Cobo was never able to visit these two regions. The descriptions are alike, except in a few details which are easily detected by reading the references in double column as follows: * Solanum quitoense Lamarck var. septentrionale Schultes & Cuatre- casas in Bot. Mus. Leafl. Harvard Univ. 16 (1953) 100. [ 217 ] Chapter XVI ON PUSCOLULO In the Province of Popaydan, there grows a bush called pusco- lulo,’ which is like hell’s-little- fig in size, leaf and shape. It bears a fruit very similar to an apple in size, colour and rind; but it is covered all over with tiny spines (hairs) which easily rub off. The flesh is between green and yellow, watery, and full of little seeds like those of the pepper; they are eaten to- gether with the flesh. The flavor tends more to sour than to sweet, and eating too many [fruits] sets the teeth on edge’’ (Cobo, 1890, 1, 461; Cobo, 1956, 1, 209-210). Chapter XXIII ON THE NARANJILLAS In the Province of Quito, there grows a bush more or less as tall as a man; its leaf is like that of hell’s-little-fig, a little larger and spiny along the veins. The fruit which it bears is called naranjil- las (‘‘little oranges’’), because of a resemblance to oranges. It is of the size of a medium-sized peach, round, orange-coloured ; the rind and core are like those of the tomato; the inside is of a watery, bittersweet consistency ; it has many little seeds, like the tomato,and of good taste’’ (Cobo, 1890, 1,470; Cobo, 1956, I, 213). Both descriptions agree that the leaves are sinuate, since they are compared with those of ‘‘hell’s-little-fig, ”” the colonial Spanish name for Ricinus communis L. But, while the fruit of puscolulo is said to be covered with spine-like hairs, naranjilla is described as having a smooth fruit like the tomato. Another difference is the presence of spines on the veins of the leaves of naranjilla, a charac- ter not mentioned for puscolulo. From the references given above, it appears that, by the middle of the 17th Century, there were known in Popaydn and Quito solanaceous plants with edible fruits which differed both in names and morphological charac- teristics. Apparently, Cobo did not notice (or at least he fails to mention it) the affinities of the two plants which he described. The same observation can be made in the case of other plants from regions far away from his resi- ‘Perhaps this must be read ppochcco, ‘‘fruit sour or acid,’’ in Qui- chua, which is, I believe, the ‘“‘naranjita de Quito.’’ (Note by the editor Marcos Jiménez de la Espada. ) [ 218 | dence, and which he described on the basis of reports from other persons. The Ecuadorian Jesuit, Juan de Velasco, describing the naranjlla in 1789, adds nothing new or notable but reports that the leaf is ‘‘broad, rough and somewhat spi- nous’’ (Velasco, 1927, I, 73-74). The last two records refer to the interandean equinoc- tial area from Loja to Popayan. The next two concern the Amazon slope. The Jesuit, Jean Magnin (1740), includes naranjas and naranjillas without explanation amongst the cultivated fruits of the Province of Maynas (Magnin, 1940, 156). During the decade of 1760, the Majorcan missionary, Fr. Juan de Santa Gertrudis Serra, lived and worked in the upper part of the Putumayo and Caqueta Rivers. Speaking of the former mission at Santa Rosa de Caqueta, he said: ‘‘There is in Santa Rosa an orchard with its fence; inside it, the third part is planted to naranjillas. This is a bush of a man’s height, with big leaves, similar to those of egg-plant. But above the leaves have spines, thick and long as a half pin, 15 to 20 on every leaf. It bears fruit at the middle of the plant. Perhaps its resem- blance to the orange is why it is called naranjllas. They are half the size of oranges and covered with tiny, very thin and pointed spines, so thickly crowded that the fruit looks like velvet. When the fruits ripen, the spines de- cay, and the naranjilla assumes a very deep scarlet color. The rind is very thin and inside there is no pip. It is like an orange without sections, being entirely a pulp. The color is between green and orange-colored, and the taste bitter-sweet, very appetitious. The fruit is very fresh to the body, and diluted some of them in water with sugar, makes a refreshing drink of which I may say that it is the most delicious that I have tasted in the world”’ (Serra, 1956, I, 148-149).° [ 219 ] The data of the middle of the 17th Century confine the range of puscolulo and naranjilla to the Provinces of Popayadn and Quito. Those of the middle of the 18th Century quoted above are restricted to naranjilla; they place its area of cultivation to the east, yet it still falls within the equatorial belt. But the references to be con- sidered below indicate that cultivated Solanum quitoense (or some other species mistaken for it) had migrated from the original focus, both to the north and to the south. In the year 1701, Fr. Alonso de Zamora, writing on the plants of the New Kingdom of Grenada, stated: ‘There are growing in the hot parts of the country some trees of the stature of lemon trees, called dulos. These give a fruit like small oranges and with the same color as oranges; their skin is very thin, and they are very agreeably scented, moderately sour and with numerous seeds inside a soft pulp. This fruit, diluted, is, according to Doctor Lugo, very wise physician who had been in this New Kingdom, a healthy cordial for those sick with typhus (tabardillo) and other fevers. Sauces made with this fruit are the most seasoned that the culinary art has discovered’’ (Zamora, 1701, 41: Zamora, 1980, 40). The Gongorist style of that time did not contribute towards accuracy of description. In this case, the name /u/o seems to have been borrowed from the western part of New Grenada, but the quotation might equally well be attri- buted to Solanum Topiro (see below) or other species, because it refers the plant to the ‘‘hot parts of the coun- try.’ Some records indicate that in the Guaviare River basin there is a lulo with a fruit larger than that of the » Since Solanum quitoense does not normally grow well in hot areas, and since the plant in the foregoing description was said to have spiny leaves, it is possible that this naranjilla is referable to a new species, Solanum georgicum, described from the same region by Schultes. [ 220 | lulo from western Colombia.° Only field research and the assembling of enough botanical samples will settle the question of what species of Solanum this may be. In a descriptive report on the District of Vijes in the jurisdiction of Cali, made by José Lorenzo de Reina on July 20th, 1808, we find a statement that the few inhabi- tants of Ciénaga Larga, in the basin of the Bitaco River, tributary of the Dagua, a temperate region, grew plan- tains, maize, some arracachas and lulos (Villaquiran, 1959, 61-66: 232). These lulos are undoubtedly Solanum quito- ense, most probably the variety septentrionale. By the middle of the 18th Century, Solanum quitoense had spread to the south. Amongst the plants which he had collected in Lima and vicinity and in the neighboring valleys in the first half of 1778, the botanist Hipoélito Ruiz in- cludes: ‘‘Solanum angulosum, common name narangitas de Quito, ‘little oranges from Quito,’ because they come from that province and because the fruit has the shape and color of a small orange. Women esteem this fruit for its scent and for the particular taste that they give to the maté beverage, the custom being to drop it in some of the juice. They are also accustomed to include these fruits in the mixed bouquets of flowers for the purpose of beautifying the bouquet and of making the mixture more pleasant with its fragrance’’ (Ruiz, 1952, 1, 30). In another place, Ruiz wrote that he sent to Spain seeds of Solanum peruvianum or naranjitas de Quito with the first parcel of plants and seeds mailed in 1780. The cases with the living plants were lost (ibid., 434, 443). In the National Period (the previous data are from Co- lonial documents), Edouard André saw naranjilla in the Pasto market in 1876 and identified it as Solanum galea- tum. Although he did not describe the plant, he praised ® Verbal report from the agronomist Dr. Camilo Castro, at present Governor of the Department of Meta. [ 221 | the fruit (André, 1879, 20. sem. XX XVIII, 322). Car- denas supports the opinion that lulo is a species on the way to domestication, of recent status as a cultigen; and, in some ways, still wild (Cardenas, 1950, 17-18). In his work on the Keshwa terms used in the Cauca Valley, Leonardo ‘Tascén offers the following references about lulo; they are included here because they establish definitively the presence in that Colombian area of two different, well known and clearly distinguished forms: ‘*Lulo. (from llullu, soft, tender). Fruit round, flattened, orange-colored, of sour taste, employed for preparing very agreeable refreshing beverages. It is borne by a solanaceous plant with large, purple leaves, spinous the same as the stem, and with white flowers in a bunch; called in botany Solanum esculentum. The dog’s lulo, the rind of which serves to make sweets, is the fruit of another species that differs from the former in the green colour of its leaves’? (Tasc6n, 1984/, 101). The first of these two concepts of Solanum would appear to be S. quitoense var. septentrionale. The members of the Russian botanical expedition to Colombia in 1925 found lulos in Manizales (Bukasov, 1980, 488). In the middle of the 19th Century, the physician- geographer Villavicencio remarked upon the excellence of naranjillos grown at Baeza in eastern Ecuador (Villa- vicencio, 1858, 403). Under the abbreviated name of naranji, there is a cul- tivated fruit in Ecuador amongst the Jivaro and Canelo Indians (Karsten, 1985, 123, 568; Sarmiento, 1958, 178). Any doubt that the last two references apply to Sola- num Topiro is dispelled below. Solanum Topiro Humboldt & Bonpland ex Dunal Sol. gen. aff. syn. (1816) 10. [ 222 ] VERNACULAR NAMES: Topiro, tupiro, tupiru (Orinoco-Rio Negro basin). Bo-po amongst the Camaratas Indians, Amazonas Territory, Venezuela (Schultes, 1958, 242). Betdka, in Kubeo, Vaupés River, Colombia; detwa in Tatuya, Apaporis River, Colombia (ibid., loc. cit.). As suggested above, the lulo mentioned by Zamora in 1701 may refer to topiro, from some of the western tribu- taries of the Orinoco River; this region, if not often visited at that time, at least was not unknown through the activity of missionaries of different religious orders in touch with Bogota.’ The Jesuits from the eastern plains of New Grenada mentioned sundry native fruits that are probably Sola- num Topiro or related species. Gumilla, using a name perhaps already spread far from its place of origin (as in the case of lulo), lists amongst the wild fruits date palms (in a generic sense, meaning ‘‘palms’’) and ‘‘naranyillas, of a bittersweet taste and very wholesome; they are of the same color, although something smaller, than ordi- nary oranges’’ (Gumilla, 1841, 197; Gumilla, 1944, I, 266; Gumilla, 1955, 174). During his second survey of the Padamu River, an affluent of the upper Orinoco, in March and April of the year 1760, Apolinar Diez de la Fuente of the staff of the Commission of Boundaries between the Spanish and Portuguese colonies organized a few years before, travel- ling from Guaharibos Falls as far as the Casiquiare- Orinoco confluence, found a cultivated field (conuco), started the year before, in which maize, beans and tupi- ros were almost mature (Ramos Pérez, 1946, 407). This 7 Cuatrecasas collected Solanum Topiro (No. 7558) along the Guaya- bero River, 240 m. altitude, November 8, 1939. (Personal communi- cation). The Guayabero and Ariari Rivers are the principal sources of the Guariare. (See footnote 6.) [ 228 ] field had been made by the Indians in three days (Alto- laguirre y Duvale, 1908, 310), near the fortress of Buena Guardia, of which, one hundred years later, not a rem- nant was to be seen (Michelena y Rojas, 1867, 162, 855). On their trip to the upper Orinoco in 1800, Humboldt and Bonpland found tupiro at San Fernando of Atabapo, the type locality of the material on which the first de- scription of the species was based. Humboldt included Solanum Topiro amongst the common plants in the area between the Javité and Pimichin Rivers (Humboldt, 1942, IV, 178). Making use, probably, of the works of those two authors, Lisandro Alvarado drew up the fol- lowing description: ‘‘7'upiro. Solanum Topiro. Shrub with herbaceous, tomentose stem; leaves subovate, acute, sinuate-angulated, unequal at the base, thickly haired above, lightly grey-tomentose below; flowers extra- axillary, aggregate; berries ovate, tetralocular, edible. Blossoming in May. It is called also tépiro’’ (Alvarado, 1953, 345). This form, accented on the antepenult, is given by Tavera Acosta (1954, 218). In a recent paper, Schultes attributes to Solanum Topiro the cocona from eastern Peru and mentioned by Fennel and other authors (Schultes, 1958, 281-282; Fennel, 1948, 181-182). Ricardo Latcham, listing sev- eral fruit-bearing species used by Amazonian tribes, mentioned—without quoting sources—the ‘‘cocona, that bears a berry similar to an orange’’ (Latcham, 19386, 65-66, 72). The following statement appears on the naranjilla of eastern Ecuador in a recent work: ‘‘There are three kinds: the two acid (known in the east under the names of huevo de tigre (*‘puma testicle’’) and cocona, and the other one common with us [in the highlands], bitter- sweet and very agreeable, especially for beverages, pre- serves, ices, sherbets, not to mention its use as an edible [ 224 | fruit, both alone and with sugar. It is so aromatic that just a single fruit is enough to fill a room with pleasant fragrance. The cocona is also a special insecticide, used by the eastern Indians against head-lice. They also eat the fruit roasted’? (Sarmiento, 1958, 178). In the market of Iquitos, I found for sale in 1951, under the name of cocona, and I have eaten there, a berry larger than that of Solanum quitoense, with a rind of dark purple or murrey, not orange-colored or yellow. It is possible that, in eastern Peru, the term cocona is applied to several species-concepts of Solanum. Along with other fruits, cocona is grown by the Kar- eneiris Indians of the upper Madre de Dios River in eastern Peru (Fejos, 1940-1942, 24). By October 1948, seeds of cocona were introduced from the United States to the Tulio Ospina Farm in Medellin, Colombia. According to official reports, the shrub has but few spines, and the fruits are orange- colored (Granja Tulio Ospina; letter August 1956; let- ter 329, October 5, 1961). Solanum muricatum Aiton Hort. Kew., ed. 1, 1 (1789) 250. Solanum muricatum Aiton var. popayanum Bitt. (Bu- kasov, 1930, 580). VERNACULAR NAMES: Cachon, in Keshwa (Domingo de Santo ‘Tomas, 1560, 112 v.; Domingo de Santo Tomas, 1951, 242; Gon- zalez Holguin, 1608, 258). Cachuma, in Aymara (Bertonio, 1612, 32). Pepino, pepino dulce. The name pepino was given to the fruit by the Span- iards who saw in it some resemblance to the cucurbita- ceous cucumber (Cucumis sativus L.), which they had introduced into the New World. [ 225 | Referring to the irrigated Peruvian coastal plains, Cieza de Leén (1553), described the so-called pepinos as one of the most remarkable of American fruits (Cieza, 1924, 209). While recounting the last campaigns of Huayna Capac across the northernmost coast of Peru, the same author gives this anecdotal passage: ‘And it is said about him, that when he was travelling along the beautiful plain of Chayanta, near Chimo, where the city of Trujillo is now built, an old Indian was at work in an agricultural field. Having heard that the king was pass- ing nearby, he took three or four pepinos and, with earth and all, presented them to him, saying: ‘‘Ancha Atu- napu micucampa”’ (‘‘Great sir, eat thou this’’). In the presence of his knights and others, he took the pepinos and, tasting one of them, said to gratify the old man: “*Xuylluy, ancha mizqui cay’ (‘‘This is indeed very sweet’’). From this act, everybody derived very, very great satisfaction’’ (Cieza, 1880, 250-251). The expedi- tions of Huayna Capac in the Chimti-Mochica region were made after the death of his father, Tupac Inca Yupanqui, towards 1481 (Vazquez de Espinosa, 1948, 541-544). The same Cieza de Leon, speaking of the Chin- cha Valley, lauds the beauty of the orchards there ‘‘and saw what delightful and fragrant pepinos [there are], not like those from Spain, although in form somewhat alike, for those from this region [of Peru] are yellow when peeled, and so appetizing that, in truth, a man needs to eat many before he loses his taste for them’’ (Cieza, 1924, 229). Juan de Salinas Loyola, in his description of Loja, Ecuador, written in Spain in 1571-1572, spoke of these fruits, saying that there is ‘‘a native kind of cucumber there.’” In 1572, in his excellent report on Quito, writ- ten likewise in Spain, he is more explicit: ‘“There is another kind of cucumber, which grow like those of [ 226 ] Spain; it is smooth, white, with some murrey veins; the Indians eat it, and it is believed that it is more whole- some than ours, and not so cool.’’ Sancho Paz Ponce de Leon in his report on Otavalo (1582) claimed that there was the Spanish cucumber and also one ‘‘from these parts’’ in the vicinity of Pizque, along the Guail- labamba River, just as in other places near the Mira or Coangue River (J. de la Espada, 1897, III, 2038, 73, 113). Even at the present time, it is the daily custom to sell excellent pepinos to the passengers stopping for a while at the village El Olivar to the north of the city of Ibarra, Imbabura Province, on the road between Quito and 'Tulean and Pasto. In the second quarter of the 17th Century, Vazquez de Espinosa, listing the productions of Quito, stated that there are ‘‘cucumbers very different and better than ours’’ (Vazquez de Espinosa, 1942, 868; Vazquez de Espinosa, 1948, 840). We do not have such early reports for New Grenada. We cannot know whether or not the growing of this species, carried on now as far north as Popayan and even in Antioquia and on the Bogota plateau, dates from pre- hispanic times or only since the Conquest. But if the cachon were cultivated in the upper Mira River basin, the southernmost boundary of expansion of the Pastos group, it may be assumed that the cultivation of Sola- num muricatum had spread northward at least as far as the basin of the river called now Guaitara. Fr. Alonso de Zamora said that at the end of the 17th Century there was in the New Kingdom of Grenada, an abundance of ‘‘nepinos’’ of several kinds, including the sweet one (Zamora, 1701, 45; Zamora, 1980, 43). In 1590, Acosta reported: ‘‘Neither are the so-called pepinos [Solanum muricatum] trees, but vegetables that are annual in habit. This name was given to them be- cause some of them, even most of them, are long and | 227 | round like the true cucumber; but in everything else they differ sharply, for the color is not green but mur- rey, yellow or white; they are not spiny or rough, but quite smooth. Their taste is very unlike and very su- perior, for when they are wholly ripe they have a very agreeable bitter-sweet flavour, but not so strong as that of the pineapple. They are succulent, fresh and easy to digest. They refresh one well during the hot weather; the skin, which is tender, is peeled off; all the rest is pulp. They grow in warm climates and need irrigation, Although because of their shape they are called pepinos, many of them are wholly round, and there are others of different shapes; so even in their figure they are not really like true cucumbers. This plant I do not remem- ber having seen in New Spain [ Mexico] nor on the islands [Caribbean], but only on the plains of Peru’’ (Acosta, 1940, 275-276; Acosta, 1954, 113). Amongst the common fruits of the plains of Trujillo, Peru, Vazquez de Espinosa included the pepino, adding: “(The Indian village of Mansiche is a quarter league from the city, with delicious vegetables and fruit, particularly Peruvian cucumbers; these are of many kinds and | varie- ties; those from this village [have the reputation all over the Kingdom of] being the best in Peru [since they are among the best and most delicious]. The plant resembles a pepper plant, but the leaf is smaller and more elabo- rate [in its color and the [—] of its shape] is like a tomato leaf. [The cucumber] is [there are] of many sorts—pur- ple [likewise there are] yellow and white (Marg. : and of other colors), and they are very smooth. They must be ripe when eaten for when green [they are worth nothing] they are no good; they come long, round and in [many ] other shapes, small and large. They taste very good when fully ripe; they are very juicy and refreshing, and are good for the kidneys and digestion; you peel off the skin, [ 228 | which is very soft and thin, and then eat it all. This fruit [I never saw in all of New Spain and Honduras, or in the islands; it] only grows in Peru.’’ (Vazquez de Es- pinosa, 1942, 390, 393-894; Vazquez de Espinosa, 1948, 365, 367, 368). Cobo described the pepino with praiseworthy exact- ness. He noted varieties which were murrey, yellow, striped white and others, but states that the commonest is murrey striped with bands of different color. ‘‘The best grew on the valleys along the coast of Peru; those from the valleys of Trujillo, Ica and Chincha are specially famous. They require hot and sandy soil; although they have been taken to New Spain [Mexico], they do not yield as well there as here, since the climate is not so favorable. At the Atrisco Valley, I saw them in the Carmen Convent; I tried them myself and found them tasteless, without the sweetness of those in this King- dom. The juice, mixed with red ointment, is valuable for ‘heat of the kidneys.’ In the Quechua language it is called Cachum and in Aymara, Cachuma’’ (Cobo, 1891, II, 8381-383 ; Cobo, 1956, I, 177). In another place, Cobo asserted that, while it grows very well in America, the Spanish cucumber (Cucumis) is used merely as a vege- table, whereas the native ‘‘cucumber’’ or pepino (Sola- num) is preferred as an edible fruit (Cobo, 1891, I1, 486— 437; Cobo, 1956, I, 418). Miguel Feyjoo, in his description of the Province of Trujillo, Peru, about the middle of the 18th Century, included the pepino amongst the cultivated native fruits (Feyjoo, 1763, 13). Amongst the plants collected by the botanist Hipo- lito Ruiz in Lima and its vicinity and in the Andean valleys, in the first half of the year 1778, he listed pepino as Solanum variegatum. The fruits, he reported, were very commonly consumed in Peru, and he erroneously [ 229 ] attributed to them the characteristic of producing, if eaten in excess, tertian fevers and dysentery with tenes- mus (Ruiz, 1952, I, 29). Describing the products of Lurin, near Lima, he mentioned again the ‘‘native cu- cumber.’’ He wrote that this shrub blossoms profusely and that the fruits are usually yellowish, whitish or spot- ted with murrey, violet and red; he repeats the state- ment about the presumed tendency of the fruit to pro- duce the diseases mentioned above, adding: ‘“‘this plant is propagated by the stems, because by the seeds it takes two years to produce fruit, after having been transplanted from the nursery in which they were sown’”’ (Ruiz, 1952, I, 58-54). Latcham reported that Solanum muricatum is culti- vated in the northern part of Chile; but he assumed that, because of the absence of an Araucanian name, the spe- cies had been introduced by the Incas (Latcham, 1986, 214-216). [ 280 ] BIBLIOGRAPHY Acosta, Joseph de, 1940. Historia natural y moral de las Indias. . . Fondo de Cultura Econémica, Mexico. lxxxv-+6388 pp. 1954. Obras del P..., de la Compafiia de Jests. Estudio preliminar y edicién del P. Francisco Mateos, de la misma Com- pafiia. Biblioteca de Autores Espafioles, Madrid. xlix-++633 pp. Altolaguirre y Duvale, Angel de, 1908. Relaciones geograficas de la gobernacioén de Venezuela (1767-68). Con prélogo y notas de. . . Publicaciones de la Real Sociedad Geografica. Imp. del Patronato de Huérfanos de Admon. Militar. Madrid. li--850 pp. Alvarado, Lisandro, 1953. Glosario de voces indigenas de Venezuela. Voces geograficas (Trabajo inédito complementario) Obras Comple- tas, Vol. I. Tip. La Nacioén. Caracas. 422-+(iii) pp. André, Edouard, 1879. L’ Amérique equinoxielle. Le Tour du Monde. Paris. (1879) 2nd sem. xxxviii. Bertonio, Ludovico, 1612. Vocabvlario de la lengva aymara. Primera parte, donde por abecedario se ponen en primer lugar los Vocablos de la lengua espafiola para buscar los que les corresponden en la lengua Aymara. Compuesto por el P. Ludovico Bertonia Italiano de la Compafiia de Iestis en la prouincia del Pir, de las Indias Occidentales, Natural de la Roca contrada de la Marca de Ancona. 474 pp. Segunda parte del vocabvlario en la qval por orden del ABC se ponen en primer lugar los vocablos de la lengua Aymara para hallar los que le corresponden en la Espafiola. Juli. 399 pp. Bukosov, S.M., 1930. The cultivated plants of Mexico, Guatemala and Colombia. With supplementary articles by N.N. Kulesov, N.E. Zhiteneva, V.I. Mazkievicz and C.M. Popova, with 307 photographs and maps. Composed on the basis of the data and materials of the Expedition of the Institute of Applied Botany in 1925-6. Lenin- grad. 553-+16 (illustr.)-+xxxviitv pp., 3 lam. Cardenas, Martin, 1950. Plantas alimenticias nativas de los Andes de Bolivia. Folia Universitaria, Nos. 2, 3, 4. Cochabamba. 25 pp. Cieza de Leén, Pedro, 1924. La crénica general del Peri. Anotada y concordada con las crénicas de Indias—por Horacio H. Urteaga. Vol. I. Lib. e. Imp. Gil. Col. Urteaga. Historiadores Clasicos del Perd. Lima. Vol. VII, xvi+346 pp. , 1880. Segunda parte de la Crénica del Pert, que trata del Sefiorio de los Incas Yupanquis y de sus grandes hechos y gober- nacion. Publicala Marcos Jiménez de la Espada. Biblioteca Ultra- marina. Madrid. 20+279 pp. Cobo, Bernabé, 1890-91. Historia del Nuevo Mundo. Por el P... de la Compafiia de Jests, publicada por primera vez con notas y otras ilustraciones de D. Marcos Jiménez de la Espada. Sociedad [ 231 J de Biblidfilos Andaluces. Sevilla. 1890—Vol. I—viii+530 pp. 1891, Vol. Il, 467 pp. ——, 1956. Obras del P. . . de la Compafiia de Jesis—Estudio pre- liminar y edicién del P. Francisco Mateos, de la misma Compania. Biblioteca de Autores Espafioles. Madrid. Graficas Orbe. 1956, Vol. I, xlviit+439 pp.; Vol. II, 516 pp. Domingo de Santo Tomas (Fr.), 1560. Lexicon, o vocabulario de la lengua general del Perv. . . Francisco Fernandez de Cordova, Im- presor de la M. R. Con Privilegio. viii+179 folios. ~——, 1951. Id. Facsimile edition published, with a foreword, by Rail Porras Barrenechea. Edicién del Instituto de Historia. Lima. 374 pp. Fejos, Paul, 1940. Report on the exploration of the region of the Rio Colorado, Departamento de Madre de Dios, by the Wenner-Gren Scientific Expedition to Hispanic America, 1940-42. Lima. Mime- ogr. 31-+6 pp. Fennel, Joseph L., 1948. Cocona—a desirable new fruit. Foreign Agriculture. Washington, D.C., Vol. XII. August. No. 8; pp. 181-182, Feyjoo (de Sosa), Miguel, 1763. Relacién descriptiva de la ciudad, y provincia de Truxillo del Pera, con noticias exactas de su estado politico. . . Imprenta del Real y Supremo Consejo de Indias. Ma- drid. xiit+164 pp. y 3 planos. Gattoni, Luis A., 1935. La industria del jugo de la naranjilla ecua- toriana. Hoja de Divulgacién Agricola. Ministerio de O. Publicas, Agricultura y Fomento. Direccién Gral. de Agric., Comercio e Industrias. Imprenta Nal. de Quito. 15 pp. Gonzalez Holguin, Diego, 1608. Vocavulario de la lengva general de todo el Perv llamada lengua Qquichua, o del Inca. Corregido y renovado conforme a la propiedad cortesana del Cuzco. Ciudad de los Reyes. 4+365 +332 pp. ——, 1952. Id. New edition with foreword by Radl Porras Barrene- chea. . . Instituto de Historia. Impr. de Santa Maria. Lima, xliv +697 pp. Gumilla, Joseph, 1741. El Orinoco ilustrado. Historia natural, civil y geografica, de este gran rio, y de sus caudalosas vertientes. . . Manuel Fernandez. Madrid. 580-19 de indice pp. , 1944. El Orinoco ilustrado. Historia natural, civil y geografica de este gran rio. Biblioteca Popular de Cultura Colombiana. Bogota. Vol. I. 1944. 360 pp.; Vol. II. 1944. 334 pp. ——-, 1955. Id. . . . Biblioteca de la Presidencia de Colombia. No. 8. Editorial ABC. Bogota. 427 pp. Humboldt, Alejandro de, 1942. Viajea las regiones equinocciales del Nuevo Continente—hecho en 1799—1800-1801, 1802, 1803 y 1804 por... y A. Bonpland. Biblioteca Venezolana de Cultura, Trans- [ 282 | lated by Lisandro Alvarado and others. Caracas. Vol. IV. 606 pp. Jij6n y Caamafio, Jacinto, 1940-41. El Ecuador interandino y occi- dental antes de la Conquista Castellana. Edit. Ecuatoriana. Quito. Vol. 1. 1940. 556 pp.; Vol. II. 1941. 555 pp. Jiménez de la Espada. Marcos, 1897. Relaciones geogrdaficas de In- dias. Peru. Ministerio de Fomento. Madrid. Vol. III. xl+276+ elxxv pp. Juan de Santa Gertrudis Serra (Fr.), 1956. Maravillas de la natura- leza. T.I. Primera y segunda parte. Bogota. (Biblioteca Presiden- cia de la Reptiblica.) Vol. 28. 423 pp. Karsten, Rafael, 1935. The head-hunters of western Amazons. The life and culture of the Jibaro Indians of eastern Ecuador and Peru. Helsingfors. 598 pp. Latcham, Ricardo E(duardo), 1936. La agricultura precolombina en Chile y los paises vecinos. Edic. Universidad de Chile. Santiago. viii+336 pp. Lira, Jorge A, 1945. Diccionario Kkechuwa-Espanol (Universidad Nacional de Tucuman— Departamento de Linguistica y Folk-lore). xii, Talleres Graficos Miguel Violetto. Tucumdn. 1200 pp. Magnin, Juan, 1940. Breve descripcién de la Provincia de Quito y de sus misiones de Succumbios de religiosos de S. Franco., de May- nas de Pp. de la Compa. de Jhs. a las orillas del gran Rio Maranon, hecha para el Mapa que se hizo el ano 1740, porel P. . . , de dha. Compa., missionero en dichas Missiones. Revista de Indias. Ano I, No. 1. Instituto Gonzalo Fernandez de Oviedo. Madrid. pp. 151-185. Michelena y Rojas, F., 1867. Exploracién oficial por la primera vez desde el norte de la América del Sur. . . En los afios de 1855 hasta 1859, por. . . A. Lacroix. Verboeckhoven & Co. Bruselas. 684 pp. Patifio, Victor Manuel, 1958. El cachipay o pijibay (Guilielma Gasi- paes (HBK.) Bailey) y su papel en la cultura y en la economia de los pueblos indigenas de America intertropical. América Indigena. Mexico. Vol. XVIII. No. 3, pp. 177-204: No. 4, pp. 299-332. Ramos Pérez, Demetrio, 1946. El] tratado de limites de 1750 y la expedicion de Iturriaga al Orinoco. Prol. de Armando Melon y Ruiz de Gordejuela. Cons. Sup. de Invest. Cientif. Inst. Juan Sebastian de Eleano. Grdaficas Versal. Madrid. 537 pp. Ruiz, Hipdlito, 1952. Relacién histérica del viage, que hizo a los Reynos del Pert y Chile el botanico don... en el afio de 1777 hasta el afio de 1788, en cuya época regres6 a Madrid. . . Ed. Dr. Jaime Jaramillo-Arango. Real Academia de Ciencias Exactas, Fisi- cas y Naturales de Madrid. Talleres Graficos de Candido Bermejo. Madrid. Vol. I (Texto): xliv+526 pp. Sarmiento, Alberto, 1958. Monografia cientifica del oriente ecuato- riano. Quito. 318 pp. [238 J Schultes, Richard Evans, 1949. Plantae Colombianae XII. De plan- tis principaliter Amazoniae Colombianae Investigationes. Botanical Museum Leaflets, Harvard Univ. Vol. 14, No. 2, pp. 21-47 and plate X. —-, 1958. A little known cultivated plant from northern South America. Botanical Museum Leaflets, Harvard Univ. Vol. 18, No. 5, pp. 229-244 and plates XLVI to XLIX. and Cuatrecasas, José, 1953. Notes on the cultivated lulo. Bo- tanical Museum Leaflets, Harvard Univ. Vol. 16, No. 5, pp. 97- 105. Serra, Juan de Santa Gertrudis (See Juan de Santa Gertrudis). Tascén, Leonardo, 1934? Quechuismos usados en Colombia. Ed. T.E. Tasc6n and Jorge H. Tascoén. Bogota. 153 pp. Tavera Acosta, Bartolomé, 1954. Rionegro; resefia etnografica, his- torica y geografica del Territorio Amazonas. Ed. 3, Caracas. 309 pp., illustr. Toscano Mateus, Humberto, 1953. El espanol en el Ecuador. Ma- drid. 478 pp. Vazquez de Espinosa, Antonio, 1942. Compendium and description of the West Indies. Translated by Charles Upson Clark. Smith- sonian Institution. Vol. 102, Publ. 3646, Washington, D.C. xii-+ 862 pp. ——, 1948, Compendio y descripcién de las Indias Occidentales. Transcrito del manuscrito original por Charles Upson Clark. Smith- sonian Miscellaneous Collections. Vol. 108, Washington, D.C. xii +801 pp. Velasco, Juan de, 1927. Historia del Reino de Quito en la América meridional—escrita por el Pbro. Don. . . —nativo del mismo reino. Vol I and part 1. 1789. Imprenta Nacional, Quito. 270-+iv pp. Villaquirin, Vicente, 1939. Historia y Antigiiedades. Boletin de His- toria y Antigiiedades. Cali. 61-66: pp. 204-220. Villavicencio, Manuel, 1858. Geografia de la Reptblica del Ecuador. New York. Imprenta de Robert Craighead. 505 pp., 2 maps and plates. Wolf, Teodoro, 1892. Geografia y geologia del Ecuador. Published by order of the Supreme Government of the Republic. With 12 plates, 47 text figures and 2 maps. Leipzig. Tip. de F. A. Brock- haus. xii+671 pp. Zamora, Alonso de, 1701. Historia de la Prouincia de San Antonino del Nuevo Reyno de Granada, del orden de predicadores. . . Bar- celona. Imprenta de Joseph de Llopis. (xx)+537-+(xx) pp. , 1930. Historia de la Provincia de San Antonino del Nuevo Reino de Granada. Prélogo de Caracciolo Parra y notas ilustrativas del mismo y de Fr. Andrés Mesanza, Parra Le6n Hermanos, Cara- eas. 559 pp. Folio. [ 284 ] EDIBLE FRUITS OF SOLANUM IN COLOMBIA BY RicHarp Evans ScHULTES AND RAFAEL ROMERO-CASTANEDA!? One of the largest and most interesting plant families in tropical South America is the Solanaceae. The world has acquired from this continent a number of solanaceous economic plants of outstanding value, such as Lycoper- sicon esculentum Mill. (tomato); Cyphomandra betacea (Cav.) Sendt. (tree tomato); Nicotiana Tabacum L. (tobacco); Solanum tuberosum L. (potato); and sundry narcotics. There are, however, other species of lesser economic importance, grown locally, which have never been extensively adopted by peoples in other regions. This family, especially as it is represented in the Andes, merits much closer taxonomic and agronomic investiga- tion. It is well within the realm of probability that new food or drug plants will be found when such a concerted study is pursued. The genus Solanum contains some of the most poison- ous members of the family, yet a few of the species yield edible berries which are utilized by the local inhabitants as fruits or as the source of refreshing and_ probably Instituto de Ciencias Naturales, Bogota, Colombia. The illustra- tions reproduced in this paper have been made possible through a grant from the National Science Foundation. C235 | vitamine-rich beverages. Our information on these edible species of Solanum is sparse, and, in some cases, its re- liability is suspect. This is due primarily to the difficulty of precise identification of collections of Solanum, one of the largest genera of plants and one which, like many others in the family, suffers woefully from lack of modern taxonomic revision. There are undoubtedly an apprecia- ble number of new species in the genus, in spite of the many concepts already described, and it appears that perhaps some of the species locally cultivated as eco- nomic plants fall into this category. The present paper is offered as a preliminary summary of our knowledge of Colombian species of Solanum cul- tivated tor their edible fruits. Much of the information contained in this summary has resulted from the field investigations of the authors and their botanical col- leagues. It continues partial studies by both of the au- thors in previous articles and in a book: Schultes, R. EF. and J. Cuatrecasas: **‘Notes on the cultivated Julo™ in Bot. Mus. Leafl., Harvard Univ. 16 (1958) 97; Schultes, R. Eo: ‘'A little known cultivated plant from northern South America’ ibid. 18 (1958) 229; and Romero- Castaneda, R.: **Frutas silvestres de Colombia’ 1 (1961) 282-292. We have not felt constrained to include litera- ture references to Solanum species with edible fruit with- out seeing voucher specimens ourselves. ‘There remains open, obviously, an extensive field for future studies of this subject along both academic and practical lines of re- search. We hope that this brief contribution may stimu- late studies of such a nature. Solanum alibile PR. 1. Schultes sp. nov. Krautex usque ad quattuor ped. altus, robustior, sub- scandens. Rami robusti, teretes, lepidoto-pubescentes, cortice brunneo. Ramuli densissime albo-stellato-tomen- [ 236 | tosi. Folia grossiuscule membranacea, circiter ovata vel irregulariter elliptica, usque ad 50 cm. longa, 38-40 cm. lata (probabiliter majora), basi inaequaliter truncata, apice abrupte acuminata vel subacuta, margine distantis- sime et profundissime sinuata, valde petiolata (petiolis usque ad 9 cm. longis, densissime albo-stellato-tomento- sis), supra dense albido-sericea, infra regulariter albido- stellata; venis omnino stellatis. Inflorescentiae cymosae, laterales, breviter pedunculatae, pauciflorae. Flores pedi- cellati usque ad 1 cm. longi, 8 mm. in diametro, den- sissime molliterque stellato-pilosi. Calycis lobi aliquid crassulentes, triangulari-ovati, acuti, usque ad 12 mm. longi, extus maxime densissime molliterque albido- stellati, intus subglabri sed minute Jepidote, fructu per- sistentes. Corolla valde membranacea, Jobis albido- viridibus, oblongis, usque ad 18 mm. longis, 6 mm. latis, intus subglabris, extus dense stellato-pilosis. Anthera flava, erecta, plusminusve 8 mm. longa. Stylus teres, 5 mm. longus. Ovarium globosum, dense albido-sericeum, 2.5 mm. in diametro. Fructus globosus vel subglobosus, 9-9.5 em. in diametro, maturitate rufescens, dense mi- nuteque stellato-tomentulosus: indumento faciliter ca- duco,maturitate subglabrescens. Pulpa acidulosa. Semi- na numerosissima, compressa, in circuiter ovalia, 2-8 mm. longa, plusminusve 1.8-2 mm. lata, straminea. Cotomspia: Comisaria del Putumayo, Rio Putumayo, Puerto Asis, altitude about 200 m. ‘‘Fruit orange-red. Size of very large orange, covered with caducous hairs. Leaves and petioles spineless. Shrub 3-4 feet tall. Cultivated. Lulo.’’ August 2, 1960, Richard Evans Schultes 22571 (Tyrer in Eeon, Herb. Oakes Ames; Dup icatre TYPE in Herb. Nac. Col. ). Solanum alibile, sonamed because of its extensive local use asa fruit and source of arefreshing beverage, is close- ly related to S. Topiro. More extensive field or mono- graphic work might indicate that it should be accorded only varietal rank. The inhabitants readily distinguish [ 237 ] EXPLANATION OF THE ILLUSTRATION Prare XXXII. Sonanem anipite PR. Eb. Schultes. 1, leaf, approximately one half natural size. 2, por- tion of upper surface of leaf, greatly enlarged. 3, portion of nether surface of leaf, greatly enlarged. 4, young leaves and buds, approximately one half natural size. 5, fruit, one half natural size. 6, flow- er, approximately one half natural size. Drawn by Eimer W. Siri [ 238 ] LATE 3 NX LI SOLANUN [nae alibile J. \~ RE. Schuttes \ sf the two, however, and state that the fruits of So/anuwm alibile ave less acidulous than are those of S. Topiro. The truit of So/anwm alibile is normally pertectly glo- bose (instead of being ovoid) and more than twice as large (9-9.5 em. instead of 4.5 cm. in diameter) than those of S. Topiro. There are, likewise, significant differences in the form and coloration of the indument of the leaves and floral segments. Solanum alibile and its fruits are called /u/o by the in- habitants of the Comisaria del Putumayo, most of whom ure recent settlers from the more populous parts of Colombia where, in the highland areas, this vernacular name applies to S. quitoense. Solanum georgicum P. LH. Schultes sp. nov. Krutex usque ad quattuor ped. altus, robustior, erec- tus. Rami robusti, teretes, dense albido-stellati, cortice griselli, partibus omnibus spinis armati. Ramuli similes. Folia grossiuscule membranacea, circuiter ovata vel ir- regulariter lateque elliptica, lamina usque ad 85 cm. longa, 28 cm. lata, basi inaequaliter truncata, apice abrupte acuminata vel subacuta, margine distante pro- fundeque sinuata, valde petiolata (petiolis usque ad 5 cm. longis, maxime densissime albido-stellatis), supra dense regulariterque albido-sericea, infra dense albido- stellata (statu juvenile aureo-sericea), venis utroque latere stellatis et spinis stramineis sed basi fuscis, usque ad 15 mm. conspicue armatis. Inflorescentiae cymosae, later- ales, breviter pedunculatae, pauciflorae. Flores pedicel- lati; pedicelli usque ad 6-7 mm. longi, dense stellati. Calycis lobi aliquid crassulentes, triangulari-ovati, valde acuti, leviter marginati, usque ad 10 mm. longi, extus densissime molliterque albido-stellati, intus dense albido- lepidoto; fructu persistentes. Corolla valde membrana- cea; lobis albido-viridibus, oblongis, subacutis, usque ad [ 240 | ~ 12 mm. longis, 5-7 mm. latis, intus glabris, extus dense stellato-pilosis. Anthera flava, erecta, plusminusve 9 mm. longa. Stylus teres, 4-5 mm. longus. Ovarium globo- sum, dense et longe albido-sericeum, 2.5-8 mm. in dia- metro. Fructus perfecte globosus, plerumque 4.5 cm. in diametro, maturitate rufescens, dense minuteque tomen- tulosus, indumento faciliter caduco, maturitate subglab- rescens. Pulpa aliquid acidulosa. Semina numerosissima, compressa, in circuiter ovalia, plusminusve 2.5 mm. lon- ga, 2 mm. lata, straminea. CoLtompia: Comisaria del Putumayo, between Pepino and Mocoa, altitude about 700 m. ‘‘Cultivated. Flowers greenish white. Leaves very spiny. Height 2-4 ft. Fruit globose, ripening dark red, covered with caducous hairs. Naranjilla.’’ July 28,1960, Richard Evans Schultes 22554 (Tyre in Econ. Herb. Oakes Ames; Dupticare type in Herb. Nac. Col.).—Comisaria del Putumayo, Rio Putumayo, Puerto Asis, altitude about 200 m. “‘Shrub 2-3 feet tall. Fruit orange-red, small. Leaves spiny.’’ August 2, 1960, R. KE. Schultes 22573. This new species is allied to Solanum Topiro, but it is distinguished by the inhabitants of the region who apply different common names to the two plants. Solanum Topiro, which is not found in the Comisaria del Putu- mayo as high as 700 meters, is called cocona, whereas JS. georgicum is known as naranjilla, the term which, in the southern Colombian Andes, refers to S. qguitoense. The inhabitants of the Putumayo are quick to point out two other important differences: S. georgicum is conspicu- ously spiny on the stems and along the nerves of the leaves and has perfectly globose fruits, whereas S. T’opiro is unarmed and usually has ovoid fruits. The berry of Solanum georgicum is eaten directly as a fruit and is employed in the preparation of a refreshing, acidulous beverage. The plant is cultivated, but in a hap- hazard way, growing up along roadsides and on the edge of agricultural plots (whence its specific epithet) without special care. It appears to be a wild species which has (240 J EXPLANATION OF THE ILLUSTRATION Pirate XXXII. Soranum Groraicum PR. FE. Schultes, 1, leaves and buds, approximately one half natu- ral size. 2, portion of upper surface of leaf, greatly enlarged. 3, portion of nether surface of leaf, great- ly enlarged. 4, flower, approximately one half natural size. 5, fruiting branch, approximately one half natural size. Drawn by Ermer W. Siri [ 242 ] PLaTE XXXIII \ / , georgucum i RE Schulte, ae 4 & EXPLANATION OF THE ILLUSTRATION Prare XXXIV. Sovranum Georaicum R. EF. Schultes. Photograph of a flowering branch of the type plant. Photograph by Ricuarp Evans Scuuttes [ 244 ] Pinte ky EXPLANATION OF THE ILLUSTRATION Prare XXXV. Sotanum Georcicum R. E, Schultes, Photograph of a fruiting branch of the type plant. Photograph by Ricuarp Evans Scuuires [ 246 ] PLATE XXXV become but recently only partly domesticated. It is said by the natives, however, never to be found in the wild away from man’s influence. Solanum liximitante RP. 1. Schultes sp. nov. Frutex usque ad plusminusve quattuor ped. altus. Truncus ramique robustiores, basi spinosi.. Ramuli usu- wliter sine spinis, glabri, cinereo-virides. Ramuli termi- nales puberulentes, cinereo-fusci. Folia membranacea, atroviridia, supra conspicue bullata, subtus cinereo- violacea (vivo), in circuiter ovata, maturitate plusminusve 27 cm. longa, 20-22 em. lata, basi cuneata ad subtrun- ‘ata, apice acuta, margine profundissime sinuata, valde petiolata (petiolus usque ad 6-7 cm. longus, dense cinereo-pubescens : supra remotissime stellato-pubescen- tes sed nervum centralium versus densissime stellato- pubescentes, minutissime scrobiculata; subtus molliter et dense stellata: nervis utroque latere conspicuis. In- florescentiae cymosae, laterales, breviter pedunculatae, pauciflorae. Flores pedicellati; pedicellis usque ad 8 mm. longi, dense albido-stellato-pubescentes. Calyx cymbi- formis, aliquid coriaceus, inconspicue 5-dentatus, usque ad 4 mm. longus, intus glaber, extus dense flavo- vel albido-stellatus. Corolla subcarnosa, albido-violacea (vivo), lobis oblongo-ovatis, usque ad plusminusve. 6 mim. longis, apice rotundatis cucullatisque, margine valde reflexis, intus glabris, extus dense stellato-pubescentibus. Antherae erectae, flavae, lineares, quam corolla longae. Stylus plusminusve clavatus, usque ad 8 mm. longus. Ovarium dense stellatum. Fructus globosus, usque ad 1.5 cm. in diametro, maturitate sanguineus, dense et minutissime stellatus sed) maturitate subglabrescens. Semina numerosa, plana, in circuiter ovalia, 2 mm. longa. CotombBia: Comisaria del Vaupés, Rio Vaupés, Mit, at mouth of Rio Kuduyari. “‘Fruit red. Flowers whitish violet. Leaves purplish [ 248 | beneath, dark green above, crinkled. Plant with spines on thick, basal stem, glabrous on younger growth. Cultivated.*’ August 12, 1960, Richard Evans Schultes 22583 (Tyrer in Feon. Herb. Oakes Ames).—Same locality. ‘‘Lu/o. Flores amarillo-verdosas. Fruto es- férico, amarillento, comestible. Yerba cultivada.’* June 22-80, 1958, H. Garcia- Barriga, R. E. Schultes & H. Blohm 1577 1a.—Comisaria del Vaupés, Rio Inirida, vicinity of Santa Rosa. Alt. 220 m. ‘“‘Hierba. Frutos rojos.’’ January 25, 1953, 4. Fernandez 1961. Solanum livimitante appears to be related to S. stra- minifolium, but it has a very different appearance because of its conspicuously bullate leaves. Solanum straminifo- lium bears strong spines usually on all parts of the stem, branches and on the ultimate twigs as well as along the midrib and secondary veins of the leaves, whereas in JS. livimitante the leaves and twigs, and often the branches themselves, are unarmed. There are, furthermore, cer- tain interesting floral characters in Solanum laimitante, such as the very strongly cucullate corolla lobes with in- rolled margins, which are not matched in SS. straminifo- lium. The fruit of Solanum livimitante is usually somewhat smaller than that of 8S. straminifolium. The specific epithet of Solanum laimitante means ‘‘re- sembling a sutler or camp-flower”’ and has been chosen to emphasize the semi-domesticated state in which this species finds itself. It grows with no care near and in cul- tivated Indian plots. Footpaths through the growths of Manihot and Erythroxylon are often thickly populated by bushes of So/anum lhaimitante which have come up from seeds casually spread when pieces of the fruit have been spit out by Indians at work in the fields. In the Amazonas and Vaupés of Colombia, Solanum heimitante is known by the following Indian names: Karijona (Rio Caqueta)—che-how-he-noo-roo; Maku (Rio Piraparana)—ber; Makuna( Rio Apaporis)—e-to; Mirana (Rio Caqueta)—/6é-mi-hé-ro-ya (k6-mé-hé = ‘‘water tur- tle’*); Puinave (living on Rio Apaporis)—pee-pee-hd; [ 249 ] EXPLANATION OF THE ILLUSTRATION Prare XXXVI. Sovanum tiximirantre PR, Eb. Schultes. 1, branch, approximately one half natural size. 2, portion of upper surface of leaf, greatly enlarged. 3, portion of nether surface of leaf, greatly enlarged. 4, flower and buds, approximately one half natural size. 0, portion of stem with fruits, approximately one half natural size. 6, bud, approximately one and one half times natural size. Drawn by Fimer W. Suivi [ 250 ] PLaTtE XXXVI | SOLANUM_ liximitante RE. Schultes EXPLANATION OF THE ILLUSTRATION Prare XXXVII. SoLtanum ciximirantre FP, EF. Schulltes, Photograph of fruiting branches of the type plant. Photograph by Ricuard Evans Scuuvres PuatE XXXVII ‘Tanimuka (Rio Popeyaci)—a-meé-ma-ra; Yukuna (Rio Miritiparana)—/loo-poo-po-ré-la_ (loo-poo = ‘‘cayman””). Karapana (Rio Kananari)-—/e-td-ma-ta. Solanum muricatum 4iton Hort. Kew., ed. 1, 1 (1789) 250. Subshrub unarmed, up to 2-8 feet tall. Branches te- rete, greyish green in life, often verrucose-scabrid. ‘Twigs reddish brown, very sparsely beset with simple, soft white hairs. Leaves membranaceous, oblong-lanceolate, marginally entire or slightly undulate (rarely ternate), apex acuminate, basally usually somewhat cuneate, strongly petiolate (petiole usually 8-6 cm. long, more or less marginate); upper surface white-sericeous (hairs sin- gle, but each comprising a main shaft with a minute, articulated tip); nether surface minutely white-squamate, sericeous (as upper surface): veins subconspicuous; mid- rib very conspicuous beneath. Inflorescence lateral, lax, few-flowered cymes. Flowers about 2 cm. in diameter, pedicellate; pedicels up to about 5 mm. long, sparsely white-sericeous. Calyx lobes triangular, apically attenu- ate, up to 12.5 mm. long, densely sericeous without, sub- glabrous within. Corolla thinly membranaceous, bright blue, deeply lobed, lobes ovate, apically acute, about 20 mm. long, pubescent without, glabrous within. Anthers vellow, erect, 6-7 mm. long. Style filiform, up to 9 mm. long. Ovary ovoid. Fruit very variable under cultiva- tion; usually ovoid or elongate-ovoid, long-pedunculate, drooping, smooth, whitish or yellow, sometimes yellow- green, usually with purplish or reddish purple spots or lines, 5-20 cm. long. Pulp yellow with cucumber-like favor. Seeds usually lacking in cultivated material. Cotompia: Departamento de Caldas (?), ‘‘Quindio’’ (?) 1856, J. Triana 3855 /6.—Departamento del Cauca, El] Tambo, Cordillera Occi- dental, vertiente oriental, altitude about 1700 m. K.von Sneidern 4817. —Departamento de Cundinamarca, Bogota, altitude 2600 m. *‘Nom- bres vulgares: pepino redondo, pepino morado. Cultivado.’* November 1856, J. Triana 3855/65. Bogoté, Chapinero. ‘*‘Nombre vulgare: pepino morado. Cultivado.’’ E, Peres-Arbeldez 2507.— Departamento del Huila, Parque Arqueolégico de San Agustin, altitude about 1700 m. *‘Nombre vulgar: pepino. Corolla blanea.** R. Romero-Castaneda 6709,—Comisaria del Putumayo, road from Sachamates to Mocoa, above road camp “‘El Pepino,’’ altitude about 4200 feet. ‘‘Cultivated. Common name: pepino.’’ December 9, 1941, R. EF. Schultes & C. E, Smith 3046. Solanum muricatum, known from the Andean regions trom Chile north to Colombia, has been cultivated since pre-Columbian times in the temperate highlands between 4200 and 7500 feet. A domesticated plant, it is repre- sented now by several distinct cultivated forms, differing primarily in the fruit. (Popenoe, W.: ‘‘ Economic fruit- bearing plants of Ecuador’’ in Contrib. U.S. Nat. Herb. 24 (1924) 188; U.S. Dept. Agric. Bur. Pl]. Ind.: ‘‘In- ventory of seeds and plants imported by the Office of Foreign Seed and Plant Introduction during the period trom Jan. 1 to March 81, 19177" (Jan. 80, 1922) 17.) It is most commonly known as pepino (*‘cucumber’’) but it has other names, such as pepo in Lima, in reference to the shape of the fruit. In Colombia, it is referred to usually as pepino morado or pepino redondo. Solanum muricatum, although described by Aiton as native to Chile and Peru, is native probably to Ecuador (where it appears to be most variable today) and was spread southward through human activity. In Colom- bia, its range is restricted essentially to the southern parts of the country, although it may occasionally be cultivated in the north (Perez-Arbeliez, E.: ‘‘Plantas utiles de Colombia,” ed. 8 (1956) 709). It now occurs in Middle America, where it has doubtlessly been dis- tributed in relatively recent times (Standley, P. C.: in Ann. Rept. Smithson. Inst. 1922 (1922) 817). It is an erect, usually spineless shrubby herb up to 2 or 8 feet in height, with dense clusters of blue flowers somewhat less than an inch in diameter. The leaves are sometimes en- tire, but usually slightly undulate, occasionally trifoliate, linear-lanceolate to ovate, soft-pubescent. The fruit, an almost globose to ovoid or even elongate berry, measures usually 7 to 16 (sometimes 5 to 20) em. in length. Most forms of the plant bear greenish, yellow or whitish fruits marked with purplish or reddish streaks or spots, but some may be basically pure yellow or pale green. It is juicy, aromatic and scented and somewhat acid, described frequently as resembling in taste ‘tan acid eggplant.” Most cultivated forms are said to produce seedless fruits. The plant is stated to yield fruit for three years, a crop every three or four months, and to begin to produce about five months after planting. Too little is known about Solanum muricatum. The variability in fruit and the fact that some sources state that the plant may have spines might indicate that either more than one species or one or more definite botanical varieties are involved. Herbarium material is very scarce. A taxonomic study of variability in pepino over its whole range is long overdue. This species was discussed as early as 1714 by Feuillée (in Journ. Obs. Phys. Math. Bot. (1714) 735, t. 26), who called it Melongena laurifolia, described it at length and published an illustration of it in his account of travels in Peru. It was first introduced to Europe apparently by the French horticulturist Thouin in 1785 (Aiton: loc. cit.) and, shortly thereafter, in 1789, described and given the name Solanum muricatum by Aiton at the Royal Botanical Gardens, Kew. The third botanical considera- tion of pepino was that of Ruiz and Pavoén, the Spanish plant-explorers of Peru and Chile, who, in 1799, gave a full description of it, calling it Solanum variegatum and publishing an excellent drawing of the plant (Fl. Peruy. 2 (1799) 32, t. 162a). There seems to have been no further horticultural in- terest in the species until 1882, when it was introduced into the United States by Eisen. (Anonymous: ‘‘Sundry investigations made during the year’ in Bull. 87, Agr. Exper. Sta. Cornell Univ. (1891) 389-3894; Bailey, L. H. : “Standard cyclopedia of horticulture’’ 3 (1980) 8182.) Because pepino means cucumber, Eisen thought it ad- visable to give the new introduction a more appropriate English common name. In this connection, he wrote (in Gard. Monthly 29 (1887) 84): ‘“‘] suggested the name melon shrub, but through the error or the wisdom of a printer, the name was changed to melon pear, which I contess is not very appropriate. ... 7’ This name has persisted, however, in the American literature and, when the taste of the fruit—somewhat suggesting an acid musk melon—is taken into consideration, it is not altogether inappropriate. There was some success in cultivating Solanum muricatum in California and Florida (Anony- mous, loc. cit.; in Am. Gard. 9 (1888) 265; Orch. and Gard. 10 (1888) 61), but in more northern states there was difficulty in its setting fruit. It does not do so well at low altitudes in Hawaii (Neal, M. C.: ‘‘In gardens of Hawaii” in Spec. Publ. Bishop Mus. 40 (1948) 657). At the turn of the century, Fairchild found that it had become very popular in the Canary Islands but that it was ‘‘doubtful whether it has found its proper niche there, where it can produce as delicate-flavored fruit as it does in the terraced gardens of Grand Canary.’ In a very interesting and complete review of the cul- ture and history of pepino, issued in 1891, the following recommendations were set forth: ‘The pepino is an un- usually interesting plant, and if it could be made to set fruit more freely in the north, it would be an acquisition for the kitchen garden and for markets. It is a good orna- mental plant. Altogether, it is deserving of a wider repu- [ 257 | EXPLANATION OF THE ILLUSTRATION Prare XXXVIII. Sovranum murticatrum Aiton as il- lustrated by Fueillée in Journ, Obs. Phys. Math, Bot. (1714) t. 26 as Melongena laurifolia. riare SAN IL ‘Planche XXVI Melongena Laurtfoka, - ; ; ie turbinato, Varicqato “ Mee Botan. Reg. deli. a Cfpare Soule EXPLANATION OF THE ILLUSTRATION Puare XXXIX. Sonancm muricarum -fifon (figure a) as illustrated by Ruiz and Pavon in Fl. Peruv. 2 (1799) t. 162 under the name Solanum variegatum. | 260 | Prater XX XTX CLAL - semmessco Fs Chalets we L Brae dialer F Quesovd sncaid. «a SOLANUM RMP VEG ANMIN: 4 SOLANUM ceil follum EXPLANATION OF THE ILLUSTRATION Prare XL. SoLtanum muricarum Aiton, 1, branch, approximately one half natural size. 2, portion of upper surface of leaf, greatly enlarged. 3, flower, approximately one and one half times natural size. Drawn by Ermer W. Siri [ 262 ] PuateE XL iB } y |