BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VOLUME XXV BOTANICAL MUSEUM CAMBRIDGE, MASSACHUSETTS 1976-1977 No. No. No. No. No. No. No. No. No. No. DATES OF PUBLICATION - VOLUME 25 B ocpaacederessrieusuaeiasuteesereceeineass December 7, DS yobas tances canes ages Sed January 25, i, ee eR ee Renn ene ey eno seen September 12, RTE ee ANETTA AST een ERT September 12, So ugidascsaa eeu ea piv acweuasessfesspvusseessuneaad October 14, Go obevas wens keveiareeieeetiniend konaeeeieeeeaias December 12, OD) Sete uvcuanenshe iuveiesssanvoresetrndsa peeaeuns ass January 24, B ercieensicev teens snsond saree seed en penteciioaas April 27, OF i cxucoiceesaiwastaee Mycologia 68, 1976, 1267-1272. 19. Guzman, G., personal communication, 1975. 20. McCawley, E.L., R.E. Brummett, and G.W. Dana, ‘‘Convulsions from Psilocybe mushroom poisoning,’ Proc. Western Pharmacol. Soc. 5, 1962, 27-33. 21. Guzman, G. and J. Ott, ‘“‘Description and chemical analysis of a new species of hallucinogenic Psilocybe from the Pacific Northwest,” Mycologia 68, 1976, 1261-1267. 22. Ibid. 23. Schultes, R.E., op. cit., 1939. Also Verrill, A.E., ‘‘A recent case of mushroom intoxication,’’ Science 40: No. 1029, Sept. 1914, Discus- sion and Correspondence. Pollock, S.H.,op. cit., 1974. Pollock, S.H., ‘*Psilocybian mycetismus with special reference to Panaeolus,”’ J. Psychedelic Drugs 8, Jan.-Mar. 1976, 43-57. 24. Restrepo, J.D., “‘Dos Panaeolus de Antioquia, Colombia,’ thesis, Medellin, Colombia: Univ. Nacional de Colombia, Seccional Medellin (Department of Botany), 1972. 25. Ola’h, G.-M., personal communication, 1975. 26. Wasson, R.G., op. cit., 1968. 27. For example, Puharich, A., The Sacred Mushroom, London: Victor Gollancz, 1959. Allegro, J., The Sacred Mushroom and the Cross, New York: Doubleday, 1970. 28. Schultes, R.E. and A. Hofmann, op. cit., 27. 29. Johnston, G., Psychopharmacologia 22, 1971, 230. 30. Benedict, R.G., Lloydia 29, 1966, 333. 31. Chilton, W.S. and J. Ott, ‘“*Toxic metabolites of Amanita pantherina, A. cothurnata, A. muscaria, and other Amanita species,” Lloydia 39, 1976. 32. Ott, J., ‘‘Psycho-mycological studies of Amanita: from ancient sacra- ment to modern phobia,” J. Psychedelic Drugs 8, Jan.-Mar. 1976, 27-35. 148 BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS, SEPTEMBER 30, 1977 ToL. 25, No. 6 A NATIVE DRAWING OF AN HALLUCINOGENIC PLANT FROM COLOMBIA RICHARD EVANS SCHULTES! and ALEC BRIGHT? There has recently been discovered an interesting and un- doubtedly very significant drawing of an hallucinogenic plant made by an Indian artist in southern Colombia. We have been able to identify this plant as Brugmansia vulcanicola a species only recently described as Datura vulcanicola A. S. Barclay, from the Andes of southern Colombia (Barclay, 1959). The drawing, reproduced here, shows a shrub or small tree with tubular flowers and an Indian woman sitting beneath its branches. It is entitled ‘‘Mujer al pie de borrachero’’ (Woman at the foot of a borrachero tree). The name of the plant in the Indian dialect is given as yas. Drawn a quarter of a century ago in Popaydn, Colombia, by a Guambiano Indian from the region of Silvia — Francisco Tumina Pillimue — it has been published ina book of many interesting drawings by the same native artist entitled Nuestra Gente — Namuy Misag, with a text by Dr. Gregorio Hernandez de Alba (Tumina P., 1949). The drawing has all of the characteristics of indigenous art, especially as to lack of detail and disregard of relative size. Yet we believe that it is possible to identify this drawing with certainty as representative of the solanaceous Brugmansia vulcanicola. The leaves match in shape those of this species. The flowers — with a dentate calyx, almost regularly tubular corolla with a slightly flaring dentate lip — match the charac- ters of this plant. Furthermore, the shape and surface texture of the fruit are the same. The name borrachero is applied to all of ‘Botanical Museum, Harvard University, Cambridge, Mass. *Museo del Oro, Bogota, Colombia. Application to mail at Second-Class Postage Rates is pending in Boston, Mass. the Brugmansias in southern Colombia. Brugmansia vulcani- cola, moreover, is endemic to the highlands of southern Co- lombia, being especially abundant and possibly having origi- nated in the region of the Volcan de Puracé, not far from the location of Silvia in the Departamento del Cauca. We believed at first that the drawing might possibly repre- sent another solanaceous hallucinogen of southern Colombia: lochroma fuchsioides (HBK.) Miers. Careful study, however, indicates that, even though this species is likewise called bor- rachero, such an identification would be extremely tenuous and open to grave doubt. The flowers, for example, are drawn as though borne singly, while in Jochroma fuchsioides they occur in multiflorous clusters. Furthermore, the shape and texture of the fruits are so far from the condition of the bacca of Tochroma fuchsioides with its enlarged, persistent calyx as to present serious problems in attributing the differences to artis- tic license. Brugmansia vulcanicola was described in 1959 on the basis of material collected on the northern slopes of the Volcan de Puracé in Cauca, Colombia, between 7000 and 8400 feet. This locality is not far from Silvia. In this region, the plant is ex- tremely abundant. Natives of the region indicated that it was used ‘‘in olden times’’ as an hallucinogenic narcotic by In- dians, but the acculturated peasants now living in the type locality do not utilize it. Ten or twelve years later, however, when additional botanical collections were made in the type locality, botanists found that local residents were assiduously destroying the population because of the initiation of a new industry, bee-keeping, and the fear that the presence of these toxic flowers could damage the honey produced in the region. It has been thought that Brugmansia vulcanicola might rep- resent an endemic of the region of Purace, but collections have also been made near the Laguna La Cocha, above El Encanto, between 9000 and 11,000 feet, in the Departamento de Narino, east of the city of Pasto in southern Colombia. Whether or not the plants growing in this more southern locality, which is a tourist centre, are the result of introductions from the localities near Puracé cannot be ascertained. It is our suspicion that perhaps they may have been introduced because of their hor- 152 ticultural attraction, since the species has never been found in the area between Puracé and the region of La Cocha, and the occurrence of the plants near La Cocha Is sporadic and along roadsides, where it appears to have been planted. There have been several different botanical interpretations of the concept which we here call Brugmansia vulcanicola. Lockwood, in his revision of Brugmansia (Lockwood, 1973) considered the concept described as Datura vulcanicola to represent a subspecies of Brugmansia sanguinea, but he did not have an opportunity to make the indicated nomenclatural change before his untimely death. It can be differentiated from Brugmansia sanguinea, with which it is obviously allied, on the basis of corolla colour and shape; the pericarp which is woody and externally provided with a warty, corky reticulum; an extremely stout, woody peduncle; very hard wood; the calyx which is conspicuously not persistent on the fruit; the usually smooth surface of the seed; and leaf shape and size. Although Lockwood argued that ‘‘... the hard wood, the thick, woody peduncle; and hard, woody pericarp of the fruit may all represent the pleitropic effects of one mutant gene that controls lignification’’, it seems that there are sufficient characters on which to accept specific status. Bristol (Bristol, 1966) studied a number of species of Brug- mansia employed for hallucinogenic purposes by medicine- men in the Valle de Sibundoy in the southern Colombian Andes and recognized it as a distinct species. He (Bristol, 1969) suggested that this concept might represent an incipient species, especially in view of its limited range in the Andean highlands of southern Colombia between about 9,000 and 11,000 feet. Consideration of its very marked differences from its nearest ally, Brugmansia sanguinea, however, would tend to indicate that it is far from incipient and that it is already a well established species. It would seem advisable, in view of the numerous and stable characters, to continue recognizing it as a distinct species. Consequently, the necessary nomenclatural transfer to Brug- mansia is here made. ee Brugmansia vulcanicola (A.S. Barclay) R.E. Schultes comb. nov. Datura vulcanicola A. S. Barclay in Bot. Mus. Leafl., Har- vard Univ. 18 (1959) 260. All parts of the plant of Brugmansia vulcanicola give very positive spot tests for alkaloids with Dragendorff reagent. Al- though further chemical studies have not yet been carried out on this species, it seems logical to presume that it contains the same tropane alkaloids found in all other species of the genus. Some six species of Brugmansia are recognized, all native to South America and all but two native to relatively high parts of the Andes. All have been employed by native peoples as hal- lucinogens (Schultes, 1976; Schultes and Hofmann, 1973). In prehispanic times the Chibcha-speaking Muisca people of the high plateau of Bogota used Brugmansia aurea Lagerheim to drug the women and slaves who were to be buried with a dead chief. Juan de Castellanos described this custom four centuries ago (de Castellanos, 1589): *‘At Tunja, in the land of the Chib- cha-speaking Muiscas, the dead chief was accompanied to the tomb by his women and slaves, who were buried in different layers of earth .. . of which none was without gold. And so that the women and poor slaves should not fear their death before they saw the awful tomb, the nobles gave them things to drink of inebriating tobacco and other leaves of the tree we call borrachero (‘intoxicator’), all mixed in their usual drink, so that of their senses none is left to foresee the harm soon to befall them.”’ The depiction of the tree — which the Chibcha-speaking Guambianos call yas — as the only plant and its association with the name borrachero lead us to suspect strongly that it was chosen and drawn because of its importance as an hallucinogen in indigenous life. This suspicion is heightened by the very conspicuous association of the large bird with the tree: the bird is a symbol of evil and sorcery amongst these Indians. A translation of Hernandez de Alba’s text which accom- panied the Indian drawing follows. 154 “A WOMAN AT THE FOOT OF A BORRACHERO (The Intoxicator). ISUG YAS GYETA ‘*How pleasant is the perfume of the long, bell-like flowers of the Yas, as one inhales it in the afternoon, following the rural paths. ‘But the tree has a spirit in the form of an eagle which has been seen to come flying through the air, and then to disappear; it vanishes completely in the leaves, between the branches, between the flowers. ‘*The spirit is so evil that if a weak person stations himself at the foot of the tree, he will forget everything and stay in that state, feeling up in the air as if on the wings of the spirit of the Yas. This happens to men and women alike, but if a girl who has evil within her, something dirty* sits resting in the tree’s shade, she will dream about men of the Pdez tribe, about those men who never stop chewing coca, and later, a figure will be left in her womb which will be born six months later in the form of pips or seeds of the tree. ‘Spirit which evilly impregnates women. Spirit which pun- ishes Indians if they uproot all the plants where they live in order to make fields, when at least one plant should be left just for seed. ‘*A spirit so evil, our grandparents tell us, was in these trees with flowers like long bells, which give off their sweet perfume in the afternoon, that they were the food of those Indians at whose name people tremble: the fierce Pijaos.”’ BIBLIOGRAPHY Barclay, A. S. ‘‘New considerations in an old genus: Datura’ in Bot. Mus. Leafl., Harvard Univ. 18 (1959) 245-272. . Studies in the Genus Datura (Solanaceae). 1. Taxonomy of Subgen- us Datura (1959) Ph.D. Thesis, Harvard University, Cambridge, Mass. Bristol, M. L. ‘‘Notes on the species of tree Daturas’’ in Bot. Mus. Leafl., Harvard University 21 (1966) 229-248. *This is a literal translation, but it all means ‘‘a menstruating girl’. 155 . “Tree Datura drugs of the Colombian Sibundoy”’ in Bot. Mus. Leafl., Harvard University. 22(1969) 165-227. de Castellanos, Juan. Elegias de Varones Ilustres de Indias (1589) Madrid, Spain. Lockwood, T. E. ‘‘Generic recognition of Brugmansia”’ in Bot. Mus. Leafl., Harvard Univ . 23(1973) 273-284. _A Taxonomic Revision of Brugmansia (Solanaceae) (1973) Ph.D. Thesis, Harvard University, Cambridge, Mass. . “Brugmansia”’ in Staff of L. H. Bailey Hortorium (Ed.) Hortus Third (1976) 184-185 MacMillan Publishing Co., New York, N.Y. Schultes, R. E. Hallucinogenic Plants (1976) Golden Press, New York. _ ‘A new hallucinogen from Andean Colombia: Jochroma fuchsio- ides’’ in Journ. Psyched. Drugs 9(1977) 45-49. Schultes, R. E. and A. Hofmann. The Botany and Chemistry of Hallucino- gens (1973) Charles C. Thomas, Publishers, Springfield, Illinois. Tumina Pillimue, F. and G. Hernandez de Alba. Nuestra Gente — Namuy Misag. (1949) Editorial Universidad del Cauca, Popayan, Colombia. 156 PLATE 21 Plate 21. Brugmansia vulcanicola (Barclay) R.E. Schultes. Photograph of the plant from which the type-specimen was collected. Photograph by Richard Evans Schultes. 157 PLATE 22 Plate 22. Brugmansia vulcanicola (Barclay) R.E. Schultes. Drawn by Lynda Bates and taken from the unpublished Ph.D. Thesis: A Taxonomic Revision of Brugmansia (Solanaceae) by Tommie Earle Lockwood. 158 PLATE 23 Plate 23. Brugmansia vulcanicola (Barclay) R.E. Schultes. Native drawing of an Indian woman under a borrachero tree. 139 BoraNnicAL MuseuM LEAFLETS VoL. 25, No. 6 HALLUCINOGENIC PLANTS IN CHINESE HERBALS Hu 1-LIn Li* The Chinese literature contains an extensive series of works on pharmaceutical botany, composed mainly of pharmacopoe- ias or materia medicas known as pén-t’sao or herbals. These works deal with drugs of all origins, mineral, animal, but mainly vegetable, hence the name. Then there are also many other treatises on plants and natural products from various parts of the country or neighboring states. All these works, accumulated in the last two thousand years, provide us with a vast store of knowledge on medicinal and economic plants and their uses, as well as on natural history in general, origin and distribution of plants, ethnobotany, agricultural history, and other related subjects. Plants with hallucinogenic properties are the subject of the present study. Since very early times, the Chinese, like many other peoples, have discovered plants with hallucinogenic properties in their native flora, finding them perhaps along with their search for plants for medicinal uses. Plants with hallucinogenic effects were recorded in the earliest herbals nearly two thousand years ago. The special significance of these halluci- nogenic plants was, however, not specifically discussed until the sixteenth century, when Li Shih-chén, the greatest author- ity on Chinese medicinal plants, in his magnus opus Peéen-ts'’ao kang-mu (first published in 1596 after his death) recorded along with details of a criminalistic episode involving the possible use of some hallucinogenic drugs. So far as I know, there has been no report of any use of hallucinogenic plants in China in more modern times. We do not know whether the practice of using some such plants by ‘‘sorcerers’’ or some other peoples as mentioned in earlier works occurred also in recent ages or not. It is not impossible that some use of hallucinogens may be found among the aborigines or other non- Han tribesmen along *Honorary Research Associate in Chinese Economic Botany. ‘Current address: De- partment of Biology, University of Pennsylvania. 161 the remote borderlands in the southwest or elsewhere. There seems to be no such ethnobotanical study or survey ever hav- ing been made. We do come across, however, some records indicating that Cannabis was being used by the Uigurs along the Sinkiang (Chinese Turkestan) border in the remote north- west as late as the early twentieth century (Li 1974b). Li Shih-chén’s encyclopedic work, the Pén-ts’ao kang-mu, upon its publication, became the standard treatise on materia medica in China. Later authors, dealing with medicinal plants, nearly all derive their information from his work. This paper is also largely based on it as a primary source. In this work, he gave the details of an event in connection with a note on the general use of hallucinogens. This episode occurred in the year 1561 A.D. Apparently it was a news event of great nation-wide interest, as afterwards, because of this, the Emperor especially proclaimed an edict of warning throughout the whole country. His account is translated below. ‘‘Lang-tang (Hyoscyamus niger), Ytn-shih (Caesalpinia sepiaria), Fang-k’uei(Peucedanum japonica) and Red Shang- lu(Phytolacca acinosa) all can cause hallucination in peoples. In the past, this significance has not been fully divulged. Plants of this kind are all toxic, which can obscure the mind, alter one’s consciousness, and confuse one’s perception of sight and sound. In the T’ang times, An Lu-shan [a foreign warlord in the Chinese army service] once enticed the Kitan [tribesmen sur- rendered to his command] to drink Lang-tang wine and buried them alive while they were unconscious. Again in the second month of the 43rd year of the Chia-ch’in period (1561 A.D.), a wandering monk, Wu Ju-hsiang of Shensi province, who pos- sessed wizardry, arrived at Ch’ang-li and stopped over at the house of a resident, Chang Shu. Upon finding the latter’s wife being very beautiful, he asked that the entire family sit together at the table with him when he was being offered a meal. He put some reddish potion in the rice and after a while the whole family became unconscious and submitted to his assault. He then blew a magic spell into the ears of Chang Shu and the latter turned crazy and violent. Chang visualized his entire family as all devils and thereby killed them all, sixteen altogether, with- out any blood shed. The local authorities captured Chang Shu 162 and kept him in prison. After ten days, he spat out nearly two spittoonsful of phlegm, became conscious, and found out him- self that those he killed were his parents, brothers, sisters-in- law, his wife, sons, sisters, nephews. Both Chang and Wu were committed to the death sentence. The Emperor, Shih-tsung, proclaimed throughout the country about the case. The par- ticular magic potion must be of the kind of Lang-tang or similar drugs. When the man was under the spell, he saw everyone else as a devil. It is thus very important to find out the remedy that counteracts such a thing.” The four plants mentioned by Li Shih-chén above, as well as other hallucinogenic plants from Chinese works, are given below, in the order of their relative importance. It may be mentioned that the botanical identity of some of these plants are not positive, and others may also be subject to questioning. Among the Chinese drug plants in these old her- bals, certain items appearing under one name may actually involve several different species of the same genus or even several species belonging to different genera. In other in- stances, a Chinese drug under one name may involve several species of the same genus as occurring in different geographical areas. However, in many cases, a plant drug may be positively determined as representing a certain species. It is not the aim of this paper to ascertain the definitive botanical identity of all the plants herein discussed, nor the chemical nature of the alleged hallucinogenic agents involved. These plants are being given here on the basis of their records as such as found in the literature. In some cases, as can be seen, an older record was not being further substantiated by later authors or may be even questioned or disputed. Many attempts have been made by modern authors for the botanical identification of Chinese medicinal plants in the old herbals. Among the chief sources available are the works of Matsumura (1915), Stuart (1911), and Read (1936). However, the botanical identifications as well as the nomenclature in these works are, in some cases, subject to further more critical verification or modification. The Chinese herbals and other related works cited here are given below with the author’s name first and arranged 163 chronologically to give a historical perspective to this record. References to these old works are given by the number in the following text. A Bibliography to modern works is given at the end of the paper with citations to the author and year of publi- cation in the text. It has to be noted that most of the earlier herbals have been lost and existed only as quotations in sub- sequent works. A number of these items have been reconsti- tuted by later workers. The illustrations given here are from Chang Ts’un-hui (20), the 1249 A.D. edition of T’ang Shén-wei’s Chéng-lei pén-ts'ao of 1108 A.D. (16), which is one of the earliest illustrated pén- ts'ao extant. HAN DYNASTY (206 B.C.-220 A.D.) 1. Anonymous Shen-nung pén-ts'ao ching (Classical Pén-ts’ao of the Heavenly Husbandman). (Based on Chou and Ch'in, 1122-206 B.C., material reaching final form ca. 2nd cent. A.D.) 2. Chang Chung-ching Chin-k'uei yao-liieh (Essentials of the Golden Cabinet). ca. 150-219 A.D. CHIN DYNASTY (265-420 A.D.) 3. Chang Hua Po-wu chih (Record of the Investigation of Things). 290 A.D. 4. Ko Hung Pao-p'u tzu (Book of the Preservation-of-Solidarity Masters). ca. 320 A.D. 5. Ko Hung Chou-hou pai-i fang (Remedies for Emergencies). 340 A.D. NORTHERN AND SOUTHERN DYNASTIES (386-589 A.D.) 6. Lei Hsiao Lei-kung p’ao-chih lun (Master Lei’s Treatise on the Decoction and Preparation of Drugs). 470 A.D. 7. T’ao Hung-ching Ming-i pieh-lu (Records of Famous Physicians) ca. 510 A.D. 8. Ch’én Yén-chih Hsiao-p'ing fang (Minor Prescriptions). 164 T’ANG DYNASTY (618-906 A.D.) . Chen Ch’uan Pén-ts’ao yao-hsing (Nature of Drugs in Pén-ts’ao). ca. 620 A.D. . Su Ching (=Su Kung) T’ ang pén-ts’ao (Pén-ts’ao of the T'ang Dynasty). 659 A.D. . Méng Shen Shih-liao pén-ts’ao (Nutritional Therapy Pén-ts’ao). ca. 670 A.D. . Ch’én Ts’ang-ch’i Pén-ts'ao shih-i (A Supplement for the Pen- ts’ao). ca. 725 A.D. FIVE DYNASTIES (907-960 A.D.) . Han Pao-shun Shu pén-ts’ ao (Pén-ts’ao of Szechuan). ca. 934-965 . Tao Ku Ch'ing-i lu (Records of Unworldly and Strange Things). 950 A.D. . Ta Ming (Jih Hua Tzu) Jih-hua chu-chia pén-ts’ao (The Sun-rays Master’s Pén-ts’ao, Collected from Many Authors). ca. 972 A.D. SUNG DYNASTY (960-1279 A.D.) . Tang Shén-wel Chéng-lei_ pén-ts'ao (Reorganized Peén-ts’ao). 1108 A.D. . Su Sung et al. Pén-ts'ao tu-ching (Illustrated Pén-ts’ao). 1061 A.D. . Fan Ch’éng-ta K’uei-hai yii-héng chih (Guide to the Southernmost Region [of China]). 1175 A.D. . Ch’én Jén-yu Chiin p’u (A Treatise on Fungi). 1245 A.D. YUAN DYNASTY (1206-1367 A.D.) . Chang Ts’un-hui Ch'ung-hsiu Chéng-ho ching-shih chéng-let pei-yung pén-ts'ao (Revision of the Pén-ts’ao of the Chéng-ho reign- period). 1249 A.D. MING DYNASTY (1368-1644 A.D.) . Liu Wén-tai er al. Pén-ts’ao p'in-hui ching-yao (Essentials of the Pén-ts’ao Ranked According to Nature and Efficacy) 1505 A.D. . LiShih-chén Pén-t' sao kang-mu (The Great Pén-ts’ao). 1596 A.D. Lang-tang — Hyoscyamus niger L. — Plate 24 This Solanaceous plant is the most famous hallucinogenic drug in the Chinese herbals. The very name, meaning violent delirium, implies its physiological effect. The hallucinogenic property is from the seeds, while the root is used as a medicine in pernicious malaria and in parasitic skin diseases. Lang-tang was identified at first with Scopolia japonica L. by Japanese authors. But this is a species of Japan and the Chinese plant, as noted by Read and Liu (1925) and later followed by Chinese as well as Japanese (Matsumura 1915) authors, should be Hyoscyamus niger L. There is also a species of Scopolia in China, S. sinensis Hemsl., which is confined only to western China in Hupei and Szechuan provinces and only the root is used in medicine (Bot. Inst. 1972). Hyoscyamus niger L. is long known as a hallucinogenic drug as given in Schultes and Hoffman (1973), who locate it in western Asia and Europe. The plant is also native to northern and southwestern parts of China, as well as in Russia and India (Bot. Inst. 1972). Makino (1921) considers the Chinese plant as representing a variety as var. chinensis Makino. The seed is long known in the Chinese herbals to be very poisonous and when taken will produce madness. For use in medicine, the seeds should be properly treated to reduce their toxic properties. It is under this drug that Li Shih-chén (22), mentioning it along with three other plants following this item in this paper, discussed the hallucinogenic plants in detail as translated above. The hallucinogenic nature of this drug is noted in the earliest herbal, the Pén-ts’ao ching (1), which states **[The seeds] when taken [when properly prepared] fora prolonged period enable one to walk for long distances, benefit- ing to the mind and adding to the strength . . . and to com- municate with spirits and seeing devils. When taken in excess, it causes one to stagger madly.”’ In using the seeds as a medicine, the preparation consists of soaking in vinegar and then in milk and afterwards drying in air in the shade. As a drug, it is considered to be tonic and con- structive, and is prescribed in dysentery, mania, toothache and other ailments. 166 In the herbals, it is repeatedly mentioned that the seed should not be broken in its use as a medicine. One early reference to this very fact is Ch’én Ts’ang-ch’i (12), which states “‘Do not let the seeds become broken. Broken seeds [when taken] produce madness.’’ Lei Hsiao (6) states that ‘‘[The seed] is extremely poisonous, and when accidentally taken, it causes delirium and seeing sparks and flashes.”’ Another author (9) states that ‘‘[The seed] should not be taken raw as it hurts people, causing them to see devils, acting madly like picking needles.’ Li Shi-chén’s (22) statement about the seeds is that they produce madness and delirium when taken. Yun-shih — Caesalpinia sepiaria Roxb. — Plate 25 This is a drug plant in the Chinese pharmacopeia from early times. It is a shrubby vine of the Legume family widely distrib- uted in China among the provinces south of the Yangtse River and in other warmer countries of Asia. The stem is hollow and densely beset with backwardly hooked spines. The leaves are doubly pinnate-compound with 6-16 pinnae each with 12-14 elliptical pinnules. The flowers are yellow and arranged in racemes. The flat pods are about 3 inches long, each containing 5 or 6 dark seeds, with a somewhat unpleasant odor. The root, flowers, and seeds are all used in medicine. Ac- cording to Li Shih-chén (22), the root is used to assist removal of a bone in the throat. The seeds are attributed to have astrin- gent, anthelmintic, antipyretic and antimalarial properties. The flowers are attributed in the early herbals as having certain occult properties, and in at least one instance, the seed 1s similarly attributed. The first herbal, Pén-ts’ao ching (1) thus says, ‘‘[The flowers] could enable one to see spirits, and when taken in excess, Cause one to stagger madly. If taken over a prolonged period, they produce somatic levitation and effect communication with spirits.” Tao Hung-ching (7) states that ‘(The flowers] will drive away evil spirits. When put in water and burned, spirits can be summoned.’’ The same author, in another instance, says that ‘‘The seeds are like Lang-tang (Hyoscyamus niger), if burned, spirits can be summoned; but this [sorcery] method has not been observed.”’ 167 Li Shih-chén (22) admits the occult properties attributed to the flowers of these early records but expresses doubts about their beneficial effect on prolonged use. Remarking on the statement given above by the Pén-ts'ao ching, he says that “As the flowers of Yun-shih enable one to see spirits and drive one to madness, how can it be possible to gain somatic levitation by taking it over a long time? This shows that this is an error in these old works.”’ Caesalpinia sepiaria has not been noted as a hallucinogenic plant in modern works. In fact, as far as I am aware, it has not been investigated medicinally or chemically. Fang-k’uei — Peucedanum japonica Thunb. — Plate 26 This Umbelliferous plant has also not been noted as a hallu- cinogenic plant in modern works. The root is used in Chinese medicine. It is considered by most authors as an eliminative, diuretic, tussic and sedative, and regarded as a tonic with prolonged use. Some, however, believe it slightly deleterious in nature. Thus, Tao Hung-ching (7) says that ‘‘Feverish people should not take it, because it causes one to be delirious and see spirits.”’ Ch’én Yén-chih (8) says that **Fang-k’uei, if taken in excess, makes one become delirious and act somewhat like mad.”’ One of the noted characters of this drug is that it decays readily. Li Shih-chén (22), who cites the above quotation, is of the opinion that the hallucinogenic effects attributed to this drug are due to adulteration by Lang-tu. Lang-tu is generally referred to some species of the genus Aconitum, a genus witha large number of species widely distributed in China, many of which enter into the pharmacopeia and all are highly poisonous (Stuart 1911, Read 1936). In the Chinese Materia Medica (Pharm. Inst. 1960), however, Lang-tu is referred to a species of Euphorbia, E. fischeriana Steud. (E. pallasi Turcz.). Both Aconitum and Euphorbia species are poisonous in nature, but in the Chinese herbals, although drugs belonging to these gen- era are noted for their high toxicity, hallucinogenic properties do not seem to have been attributed to them. 168 Shang-lu — Phytolacca acinosa Roxb. — Plate 27 The species of Phytolacca are widely distributed in warm to tropical regions in the northern hemisphere, especially in America, and several species are noted for their edible leaves and poisonous roots. The species Phytolacca acinosa Roxb., extensively distributed in China, also in Japan and India, is a well-known drug plant in China. The leaves are known to be edible. According to the old herbals, there are two kinds of Shang- lu: white with white flowers and white root, and red with red flowers and purplish root. The white root is edible when cooked and that kind is cultivated in some parts of the country for the edible root. The red root is considered to be extremely poisonous. Phytolacca acinosa Roxb. var. esculenta Maxim. (Phytolacca esculenta Van Houtte) is apparently referred to the edible kind which is generally referred to as a synonym. The old herbals named both the flowers and roots as useful for medicinal purposes. The flowers, called Ch’ang-hau, are prescribed in apoplexy. The very poisonous root is, when used as amedicine, generally applied only in external application for inflammation. It is also prescribed in dropsy and as a remedy for abdominal parasites. The deadly poisonous nature and the hallucinogenic effect of this drug was noted in many herbals and apparently it must have been quite commonly used by sorcerers in former times. T’ao Hung-ching (7) says that ‘‘The T’aoists used it. By boiling or brewing and then taken, it can be used for abdominal parasi- tic worms and for seeing spirits.” Su Sung (17) says ‘‘It was much used by sorcerers in ancient times.’’ The two kinds were carefully differentiated. Han Pao- shun (13) thus states, ‘‘The red-flowered kind has reddish roots: the white-flowered kind has white roots.’’ As to the white-flowered kind, which is considered as not poisonous and used as a drug, Ta Ming (15) states that ‘The white root has a very cooling effect; it is better taken with garlic.” Su Ching (10) summarizes in more detail the pharmaceutics of this plant. ‘This drug has two kinds, red and white. The white kind is used in medicine. The red kind can be used to 169 summon spirits; it is very poisonous. It can be only used as external application for inflammation. When ingested, it is extremely harmful, causing unceasing bloody stool. It may be fatal. It causes one to see spirits.”’ Ta-ma — Cannabis sativa L. The hemp, Cannabis sativa L., was the chief textile plant in northern China, and the seed was a leading grain. It was also an important drug plant. There are archaeological and historical records to indicate that it has been found in China since Neolithic times (Li 1974a). The early Chinese records clearly differentiate the male and female plants. The male plants produce better fibers. The edi- ble seeds are enclosed in fruit coverings which contain the toxic substance. The Pén-ts’ao ching (1) states that ‘‘Ma-fén (the fruits of hemp) . . . if taken in excess will produce halluci- nations (literally ‘seeing devils’). If taken over a long time, it makes one communicate with spirits and lightens one’s body.” T’ao Hung-ching (7) says that at his time ‘‘Ma-fén is not much used in prescriptions. Necromancers use it in combination with ginseng to set forward time in order to reveal future events.” As a drug plant, Cannabis was used for various purposes but primarily for its anesthetic effect. The hallucinogenic effect caused by Cannabis, especially the effect of temporal distortion, is mentioned in other later works. T’ang Shén-wei (16) gives a more complete account on the pharmaceutics use of the plant: ‘‘Ma-fén has a spicy taste; it is toxic; itis used for waste diseases and injuries; it clears blood and cools temperature; it relieves fluxes; it undoes rheuma- tism; it discharges pus. If taken in excess, it produces halluci- nations and a staggering gait. If taken over a long term, it causes One to communicate with spirits and lightens one’s body.”’ The stupefying effect of the hemp plant, commonly known from extremely early times, was indicated linguistically as the character ma assumed also a connotation of numbness and senselessness, apparently derived from the medicinal charac- ters of the leaves and fruits. Ma as a radical combines with 170 many other radicals to form characters with such meaning as demon, grinding, waste, rubbing, porridge, etc.; or as a charac- ter it is used in combination with other characters to form bisyllabic words meaning narcotic, numbness, paralysis, etc. (Li 1974b). In a discussion on the possible use of hallucinogenic plants by ancient Taoist practitioners in their search of elixir for immortality, Needham (1974) notes a record of the addition of Cannabis to the contents of incense-burners to generate hallu- cinogenic smokes. This record is found in a Taoist collection Wu-shang pi-yao (Essentials of Matchless Books), a work appeared between 561 to 578 A.D. He also quotes the state- ment of the hallucinogenic properties of Ma-fén in the Pén- ts'ao ching mentioned above. The use of the plant as a hallucinogen persisted for some time before it gradually declined. Méng Shen (11) says that **Those people who want to see spirits use raw ma fruits, Ch’ang-p’u (Acorus graminea), and K’uei-chiu (Podophyllum pleianthum) in equal parts, pound them into pills of the size of marbles and take one facing the sun every day. After one hundred days, one can see spirits.”’ It is suggested that in ancient China the use of Cannabis as a hallucinogen was probably associated with Shamanism. The later belief became more and more restricted in China since the Han dynasty but its extensive practice among the nomad tribes north of China perhaps carried its use westward to central and western Asia and to India (Li 1974b). Man t’o-lo — Datura alba Nees This name is generally identified as the Jimson weed, Datura alba Nees, although the Sanskrit equivalent of the Chinese Man-t’o-lo, Madara, refers to Erythrina indica Lam. Several species of Datura have been introduced into China from India and they were not clearly differentiated from each other in the former literature. These species were introduced to China probably in the Sung to Ming times and thus they were not recorded in the earlier herbals. Only in Li Shih-chén’s Pén- ts’'ao kang-mu (22) that the medicinal uses of the Man-t’o-lo began to be given. The flowers and seeds are used externally 171 for infections and eruptions on the face and internally they are prescribed for colds, nervous disorders and others. Notable ts the fact that the drug is used in combination with Cannabis sativa and taken with wine as an anesthesia for small opera- tions and cauterizations. Cannabis was among the earliest plants used in China for its anesthetic effect. The delirious action produced by the Jimson weed seeds was also known to the Chinese along with its introduction. Li Shih-chén himself experimented with this and recorded his actual experience as follows: ‘‘According to traditions, it Is alleged that when the flowers are picked for use with wine while one is laughing, the wine will cause one to produce laughing movements; and when the flowers are picked while one is dancing, the wine will cause one to produce dancing movements. [I have found out] that such movements will be produced when one becomes half drunk with the wine and someone else laughs or dances to induce these actions.”’ Mao-kén — Ranunculus acris L.? The identity of this plant is uncertain. Mao-kén is the name generally referred to species of the genus Ranunculus. A species ora variety of a species of the genus, growing along the waters edge, is alleged, in some earlier works, to have delirious effects on man. The whole plant is considered poisonous and it is not used as medicine internally but applied only externally for irritation and inflammation. The delirious action, however, noted in earlier works, is not mentioned in later herbals. Li Shih-chén cites Ko Hung (4), an author of the 4th century, in the following account: ‘‘Among the herbs there is the Shui Lang (water Lang, a kind of Mao-kén) a plant with rounded leaves which grows along water courses and is eaten by crabs. It is poisonous to man and when eaten by mistake, it produces a maniacal delirium, appearing like a stroke and sometimes with blood-spitting. The remedy is to use licorice.”’ There is the possibility that there is some mistaken identity about the plant in question as a Ranunculus. A quotation similar to the one given above appears also in Li Shih-chén separately under Lang-tang (Hyoscyamus niger L.). This quo- 172 tation is credited to Chang Chung-ching (2) which names the plant Shui Lang-tang (water Lang-tang). The two quotations are so similar that they are clearly referable to the same plant and perhaps derived from the same source. As it is described as a semi-aquatic species with shiny leaves, the plant in question is more likely a species of Ranunculus rather than a Solanace- ous plant related to Hyoscyamus. Read (1936) gives Ranun- culus acris L. var. japonicum Maxim. as a deliration. Fang-féng — Siler divaricatum Benth. & Hook.? Fang-féng is a drug generally identified as an Umbelliferous plant, Siler divaricatum Benth. & Hooker. However, it could be possibly referable to some species of Peucedanum. The root of the plant is regarded in herbals as an antidote for aconite poisoning and as a remedy for curing many types of rheuma- tism and debility. The leaves, flowers, and seeds are also used for some purposes. It is not certain whether the drug actually causes hallucino- genic effects or not as there is only one sketchy reference referring to this. T’ao Hung-ching (5) was purported to say that ‘‘The root is spicy and non-poisonous. The kind that bifurcates at top produces madness. The kind that bifurcates at the bot- tom causes reversion of old ailments.’’ This quotation was given in Li Shih-chén (22) without any substantiation or additional explanation. Lung-li — Nephelium topengii (Merr.)Lo? There is only one reference to the plant Lung-li having hal- lucinogenic effects. This is in Fang Chéng-ta (18) of the South- ern Sung dynasty, subsequently cited in the imperial commis- sioned pharmacopeia by Liu Wén-tai et al. (21) and by Li Shih-chén (22). ‘‘Lung-li grows in Ling-nan (Kwangtung pro- vince). The shape [of the fruit] is like a small Lychee with the flesh tasting like Longan. The body and foliage of the tree are also similar to these two fruit trees so it is called Lung-li. It blossoms in the third month with small white flowers. The fruit ripens at the same time with Lychee which cannot be eaten raw 173 but only after steaming. The taste is sweet and the nature hot. When eaten raw, it causes one to go mad or see devils.”’ The botanical identity of this plant has never been positively made. The description is too meager for a definitive determina- tion. Judging from the reference given to its characters as intermediate between Lychee, Litchi chinensis Sonn. and Longan, Euphoria longan (Lour.) Steud. and its geographical occurrence, the plant in question must be a variation of either one of these two species or a species of one closely related genus in the Sapindaceae that grows in the southernmost part of China. Euphoria longan and Litchi chinensis, two well-known fruits of southern China, are unique fruits in that the edible part is the fleshy aril of the single seed. They are the only species of their respective genus that are native to China. Euphoria contains about 10 species distributed in southern Asia and Litchi two species, the other being known to the Philippines only. In all probability, the plant in question is a species of the genus Nephelium, which is closely related to these two genera, especially in having arillate seeds. Some botanists regard Euphoria and Litchi as congeneric with Nephelium. Stuart (1911), who considers Nephelium in this broad sense, regards Lung-li as Nephelium sp. Nephelium sens. str. differs from the other two primarily in the aril being united with the seed coat while it is distinct in Euphoria and Litchi. There are two species in southern China; N. lappaceum L., a species of tropical Asia that is cultivated in the southernmost part of K wangtung and the Hainan Island as a fruit tree, and N. topen- gu (Merr.) H.S. Lo (N. lappaceum L. var. topengii (Merr.) How et Ho) (Kwangtung Bot. Inst. 1974) native to Hainan, Kwangtung, Kwangsi and Yunnan. The tree, growing in the forests, has a fruit that is edible but the seed is known to be poisonous (Kwangtung Bot. Inst. 1974). Thus it is quite possi- ble that this is the species in question, especially as the un- ripened fruit is mentioned as being toxic. However, among the several vernacular names of the species known locally, there is no record of the name Lung-li. 174 Hsiao-ch tin — Panaeolus papilionaceus Fr. The earliest record of a Laughing Mushroom appears to be in the early account of natural history by Chang Hua (3) in the Chin dynasty. ‘‘In the mountains south of the Yangtze River, on tall trees, there are mushrooms growing from spring through summer ... which are tasty to eat but often prove fatal. It is said that these mushrooms are mostly poisonous .. . . Those growing on the Féng tree (Liquidambar), when ingested, cause people to laugh unceasingly. The method for treating this is to use soil infusion, which cures it readily.”’ Subsequent authors give many similar records. In the Sung dynasty, T’ao Ku (14) states that ‘‘there is a kind of mushroom which causes one to suffer from a dry-laughing disease . , In the early Treatise on Fungi by Ch’én Jén-yu (19), the mush- room is named Tu-hsin ‘‘which grows in the ground. People believe it to be formed by the air from poisonous vermins, and kills people if taken . . . . Those poisoned by it will laugh. As an antidote, use strong tea, mixed with alum and fresh clear water. Upon swallowing this, it will cure immediately.’’ Ch’én treated 27 species of mushrooms from Taichow, Chekiang province. The mushroom is often identified as growing on Liquidam- bar trees. Ch’én Ts’ang-ch’i (12) states that “‘mushrooms that have poisonous snakes and vermins passing beneath them are all poisonous. Those that grow on Féng trees (Liquidambar) produce an unceasing laughing delirium.” This laughing mushroom was also recorded in old Japanese works, which is called Waraitake or Laughing Mushroom. Kawamura (1918) identified this as Panaeolus papilionaceus. Yu (1959) notes that this mushroom is found not only in Japan and China but also in the United States and that the “‘soil infusion’’ described in early Chinese works is the clear liquid after soil is mixed with water and allowed to settle, an effective antidote for poisons. Sanford (1972), in discussing the laughing mushrooms of Japan, records and translates two accounts from Chinese note- books or pi-chi, one in Yeh Méng-té’s Pi-shu lu-hua (early 12th cent.) of the Sung Dynasty, and one from Hsieh Chao-shua’s H73 Wu tsa-tsu (1619) of the Ming Dynasty which are not repeated here. There may be other mentions of this mushroom in the numerous pi-chi of all dynasties. It may be noted that in San- ford’s translation of the former, the term ‘‘Wen-tai’’ receives a footnote explaining that it is possibly not a place name and Is meant for ‘“‘warm spots.’’ Actually it is an abbreviation for Wenchow and Taichow, two districts in the eastern part of Chekiang province, the same area where Ch’én Jén-yu pre- pared his treastise on the fungi (19). In a study on the search of elixir for immortality by Taoist practitioners in ancient China, Needham (1974) mentions the possible use of hallucinogenic plants, which may include the fly agaric, Amanita muscaria. He quotes Watson that this fungus was known in China, as in several other cultures, by the name of toad mushroom, Ha-ma-ch’un, now often Tu-ying-hsin or fly-killing fungus. The laughing mushroom, Hsiao-ch’un, is identified by him as Panaeolus or Pholiota. Needham remarks that “‘the further exploration of hallucinogenic fungi and other plants in Taoism and in Chinese culture in general will be an exciting task.”’ BIBLIOGRAPHY Botanical Institute, Chinese Academy of Sciences. Iconographia Cormophy- torum Sinicorum. Tomus II. Peking. 1972. Kawamura, S. (Panaeolus papilionaceus, a poisonous fungus.) Journ. Jap. Bot. 1:(275-280). 1918. (In Japanese). K wangtung Botanical Institute. Flora Hainanica. Tome III. Peking. 1974 (In Chinese). Li, H.L. The origin and use of Cannabis in eastern Asia: linguistic-cultural implications. Econ. Bot. 28:293-301. 1974a. Li, H.L. An archaeological and historical account of Cannabis in China. Econ. Bot. 28:437-448. 1974b. Makino, T. Hyoscyamus niger Linn. var. chinensis Makino (Solanaceae). Journ. Jap. Bot. 2(5): 1 p/. 1921. (In Japanese). Matsumura, J. Shoko butsu-mei-i. Revised and enlarged. Part I, Chinese Names of Plants. Tokyo. 1915. Needham, J. Science and Civilization in China. Vol. 5. Chemistry and Chem- ical Technology. Part II. Spagyrical Discovery and Invention: Magis- teries of Gold and Immortality. Cambridge. 1974. Pharmaceutical Institute, Chinese Academy of Medical Sciences, et al. Chung Yao-chi (Chinese Materia Medica) Vol. III. Peking. 1960 (In Chinese). 176 Read, B.E. Chinese Medicinal Plants from the Pen Ts’ao Kang Mu... . A.D. 1596. 3rd Edition of a Botanical, Chemical and Pharmacological Refer- ence List. 1936. Peking. Read, B.E. & J.C. Liu. Chinese Materia Medica. The importance of botani- cal identity. 6th Congr. Tokyo, Far East. Assoc. Trop. Med. 1:987-999. 1925. Sanford, J.H. Japan’s ‘‘Laughing Mushrooms.”’ Econ. Bot. 26:174-181. 1972. Stuart, G.A. Chinese Materia Medica. Vegetable Kingdom. Shanghai. 1911. Schultes, R.E. & A. Hofmann. The Botany and Chemistry of Hallucinogens. Springfield, Ill. 1973. Yii, C.J. Hsiao-tiian (Laughing Mushroom). Ta-lu Tsa-chi 19:203-206. 1959. (In Chinese). 177 PLATE 24 Plate 24. Lang-tang, Hyoscyamus niger (From Chéng-lei pén-ts’ao, 1249 ed.) 178 PLATE 25 es Kl Plate 25. Yiin-shih, Caesalpinia sepiaria (From Chéng-lei pén-ts’ao, 1249 ed.) 179 PLATE 26 _ rAd eth na ve x Ay Gow VR is A r STIS OS At r~ « RE So Plate 26. Fang-k’uei, Peucedanum Japonica (From Chéng-lei pén-ts'ao, 1249 ed.) 180 PLATE 27 Plate 27. Shang-lu, Phytolacca acinosa (From Chéng-lei pén-ts'ao, 1249 ed.) 181 BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 25, No. 7 AN UNUSUAL SPICE FROM OAXACA: THE FLOWERS OF QUARARIBEA FUNEBRIS FREDERIC ROSENGARTEN, JR.* Most spices are to be found in the seeds, buds, fruits, leaves, roots or bark of the spice-producing plant. An extraordinary exception to this rule is the small, white flower (Plate 28) of Quararibea funebris (La Llave) Vischer, a majestic, evergreen tree (Plate 29) of the Bombacaceae (Silk-cotton family), indig- enous to southeastern Mexico and Guatemala. The dried flow- ers of Q. funebris provide a highly pungent spice, suggestive in aroma of Slippery Elm (Ulmus rubra Muhl.), Fenugreek (Tri- gonella foenumgraecum L.) or curry powder. This strong fragrance is found also in the fruits and even in the wood of Q. funebris. The genus Quararibea includes about twenty-nine species (Schultes, 1957), all of which are characterized to a greater or lesser degree by this peculiar aroma. The genus is widespread in the neotropics, comprising shrubs and trees occurring in Middle America, the West Indies and northern South America. The odor is so persistent that botanical specimens of Q. fune- bris collected in 1841 by Liebmann were highly aromatic when examined over a century later (Schultes, 1972). For many centuries, the principal use of the spicy white flower of Q. funebris has been to add pungency to various chocolate-flavored beverages in southeastern Mexico. In Oax- aca, the drink is called ‘‘tejate’’; elsewhere it is known as ‘““*pozonque,” “‘pozonqul,”’ or “‘pozol.’’ Quararibea funebris itself has various local names in Mexico: in Oaxaca I found it is called ‘‘rosita de cacao,”’ or in the Zapotec tongue “‘yieb-die.”’ *Research Fellow in Economic Botany. Published ten times annually by the Bot. Mus. of H.U. Cambridge, Massachusetts 02138. Printed by Harvard University Printing Office. Subscription: $20.00 a year, net, postpaid. Orders should be directed to Secretary of Publications at the above address. Application to mail at Second-Class Postage Rates is pending in Boston, Mass. In Chiapas it is known as ‘‘flor de cacao” or *‘molinillo”’; in Puebla and Veracruz, ‘‘palo copado’’ or *‘madre de cacao”’; in the Nahuatl dialect, “‘cacahoaxochitl’* (Pennington & Saruk- han, 1968). When the Spanish came to Mexico in the sixteenth century, they found three important spices indigenous to that country: the chili peppers (Capsicum annum L. and C. frutescens L.), vanilla (Vanilla planifolia Andr.), and allspice (Pimenta dioica L.). These flavoring agents were destined to become highly popular condiments throughout the world (Rosengarten, 1973). Yet OQ. funebris, for some puzzling reason, never became popu- lar or even known outside of a limited region in Middle America. This minor spice had been mentioned, to be sure, by the renowned Spanish chronicler, Sahagun, during his six- teenth-century travels in Mexico: *‘There are also other trees called cacauaxochitl which bear flowers . . . like jasmine and have a very delicate but pungent fragance.”’ While the world-wide consumption of vanilla in 1976 was more than 3,000,000 pounds, of allspice over 5,000,000 pounds, and of chili peppers more than 150,000,000 pounds, the total consumption of the highly aromatic flowers of Q. funebris was insignificant, probably less than 20,000 pounds, and this use was for the most part limited to southeastern Mexico (Plates 30 and 31). In the pleasant Zapotec valley of Oaxaca, 10 kilometers northeast of Oaxaca City, there lies a sleepy village called San Andrés Huayapan. Ordinarily Huayapan, elevation 5,900 ft., consisting of a church and a few scattered adobe homes, would not be especially worthy of note. The village is unusual, how- ever, in that some two dozen trees of Q. funebris may be found growing in and around it. The largest is about 50 feet tall, with the lowest branches spreading to a diameter roughly equal to its height (Plate 29). This magnificent tree, said to be 120 years old, characterized by its conical symmetry and dense foliage, 1s growing in the back yard of a local, indigenous family. Its abundant flowers provide most of this humble family’s annual income. The flowering takes place all year long, although it Is especially prevalent during the rainy season, between May and August. The flowers, having been dried on mats in the sun 184 (Plate 32), are sold in small lots every Saturday throughout the year as “‘rosita de cacao”’ in the market at Oaxaca City (Plate 33). The local price, although somewhat difficult to calculate, comes out roughly to U.S. $1.50 a pound. The volume of sales is small. There are no commercial plantings of Q. funebris trees in the Oaxaca region. When a young tree is planted, it requires at least five or six years before any noticeable flowering oc- curs: The popular chocolate-flavored beverage called ‘‘tejate’’ is produced in Huayapan in the following manner: first, the necessary ingredients are collected, starting with corn which has been boiled with firewood ashes and which is known as “conesli.’’ Dried flowers of Q. funebris are required, as well as cocoa seeds and a few mamey nuts (Plate 34). These ingred- ients are roasted separately on a ‘‘comal’’, or flat earthenware pan (Plate 35). The ingredients are then ground up separately on a primitive “‘piedra moler’’ or grindstone (Plate 36). In the case of the cocoa beans, the aromatic outer skin (the part utilized) is carefully peeled off prior to grinding. A doughy mass is now formed by kneading and mixing together the four ground-up ingredients. Cold water and sugar are added, as the light brown liquid is vigorously stirred until a froth is developed (Plates 37, 38 and 40). It is said that this sticky foam, which floats on top of the ‘‘tejate’’ (a cold, cocoa-like drink), is due to the presence of the Q. funebris flowers — hence the name “‘rosita de cacao”’ or ‘‘flor de cacao.’ ‘*Tejate’’ is indeed a refreshing, invigorating beverage, popular with field workers during the harvest of the corn crop. In Mexican herbal medicine, various therapeutic uses are attributed to Q. funebris (Monografias Cientificas II, 1976): (1) the fruits may be used as an antipyretic to control and allay fevers; (2) the flowers may be used in the treatment of psychopathic fears; and (3) the flowers may be utilized to regulate menstruation. In the Oaxaca region, a popular Zapotec cough remedy is prepared by adding 5 ‘‘rosita de cacao’’ flowers and | cinna- mon stick to % liter of boiling water. The following chemical analysis of approximately 7 ounces of flowers of Q. funebris from Oaxaca, was reported by Stras- 185 burger & Siegel, Inc., chemists and food technologists of Bal- timore, Maryland, on January 20, 1977: REPORT pH of 10% Suspension oe 6.2% Moisture Content (at 70°C in vacuo) — 8.57% Steam Distillable Volatile Oil — 0.57% Total Oils (Ether Extract) — 2.16% 2.16% Volatile Oils (Ether Extract) — 0.13% Fixed Oils (Ether Extract) —- 2.03% ANALYSIS OF ETHER EXTRACTS Glucosides (Acid Ether Extract) — 0.03% Fats and Waxes (Acid Ether Extract) —— 0.02% Principle Aromatics (Aroma and Oils) (Acid Ether Extract) ae 0.59% Alkaloidal Compounds (Basic Ether Extract) —— 0.95% Oils and Resins (Basic Ether Extract) — 0.23% Resins Insoluble in Acid and Basic Ether Extract —_ 0.09% Total Recovery — 1.91% The botanical history of this plant is simple, with apparently only three binomials ever having been applied to the concept. The combination under Quararibea is sometimes attributed to Standley [Rivera M. in An. Inst. Biol. (Mexico) 13 (1942) 502], who actually did indicate this as a new combination in 1923. Vischer’s publication of 1919, however, has priority. Quararibea funebris (La Llave) Vischer in Bull. Soc. Bot. Geneve, ser II, 11 (1919) 205, t.p. 205 (f1.). Lexarza funebris. La Llave ex La Llave et Lexarza, Nov. Veg. Desc., fasc. 2 (1825) 7. Myrodia funebris (La Llave) Bentham in Journ. Linn. Soc. 6 (1862) 115. ‘*Tree, often 20 meters high, with broad dense crown; leaves oval or elliptic, short-petiolate, 13 to 40 cm. long, obtuse to acuminate, rounded at base, glabrous except for the tufts of hairs in the axils of the veins beneath; flowers short-pedicel- 186 late; calyx bracteolate, tomentulose; petals pure white, linear oblong, the slender claws as long as the calyx; stamen tube twice as long as the calyx; fruit subglobose’’. (Standley, 1923). According to Standley, this species is reported from Oaxaca and Veracruz, Mexico, and occurs also in Guatemala and El Salvador. A map presented by Pennington and Sarukhan (1968) indicates that the tree ranges disjuctively in northern Veracruz, northeastern Oaxaca, northern Chiapas into Guatemala along the Pacific coast of Chiapas. None of these sources indicates that the type locality is in Puebla, although Izucar de Matamoros, the type locality, is in the State of Puebla. The locality of my collection from Huayapan, near Oaxaca City, also appears to be outside of the ranges noted by the above authorities. Consequently, we must presume that the distribu- tion of the tree is wider than hitherto indicated in the literature. We must remember, however, that the trees in Izucar and Huayapan were cultivated. Therefore, these and trees reported from other localities outside of the natural range of the species, may have been taken by man from the areas where it is wild and planted for their usefulness and beauty elsewhere. The collection upon which this article is based (Frederic Rosengarten, Jr. s. n., July 16, 1977, Huayapan, Oaxaca, Mexico) has been deposited in the Botanical Museum of Har- vard University. An early and interesting illustration of Q. funebris is repro- duced herein (Plate 39); this drawing was made by a primitive Mexican artist during the last quarter of the sixteenth century. The artist’s intention was to portray the cacahoaxochitl tree with Indians gathering flowers. As sometimes occurs in primi- tive art, the relative proportions are not realistic in that the size of the flowers — the important part — is highly exaggerated. (Drawn as an illustration for the Historia general de las cosas de Nueva Espana of Bernardino de Sahagun and reproduced from the Paso y Troncoso edition, first published in 1905). The curious specific name “‘funebris’> was chosen by the Spanish botanist, Pablo de La Llave (1825), when he heard that the local inhabitants of the Mexican village of Izucar, near Puebla, were accustomed to mourn their dead under the shelter of the thick foliage of the lower branches of a cacahoaxochitl 187 tree. At the present time, however, the Zapotec Indians in the Oaxaca valley do not associate the tree, Q. funebris with mourning the deceased or with death. Possibly the folk connec- tion of Q. funebris with the idea of life and death may be attributed to the fact that this vigorous tree is evergreen. It is not uncommon in primitive societies to associate trees which do not shed their leaves at one time with the concept of ever- lasting life. With increasing emphasis on natural, rather than synthetic flavorings, the long-neglected, highly aromatic flowers of Q. funebris may some day become more important in the world spice trade. Furthermore, the handsome, conical, evergreen tree, characterized by its symmetry and abundant foliage, might be of interest for planting in southern Florida, southern California and Hawaii — as an attractive, ornamental shade tree for parks, gardens and homes. Upon my return from Oaxaca in July 1977, seeds of Q. funebris were sent to the Fairchild Tropical Garden, Miami, Florida and to the Royal Botanic Gardens, Kew, England. These may represent the first intro- ductions of Q. funebris to the horticultural world. ACKNOWLEDGMENT I wish to express my appreciation to Miss Bodil Christensen of Oaxaca, Mexico, for her helpful orientation in locating Q. funebris trees in Huayapan; and to Dr. Richard Evans Schultes for his thoughtful editorial guidance. Illustrations: Plates 28, 29, 32, 33, 34, 35, 36, 37, 38 and 40 by Lynn Rosengarten; plates 30 and 31 by Elmer W. Smith. 188 LITERATURE CITED Instituto Mexicano para el Estudio de las Plantas Medicinales. 1976. (Ed.) José Luis Diaz. Usos de las Plantas Medicinales de México. Monografias Cientificas II. Pennington, T. D. & Sarukhan, José. 1968. Arboles Tropicales de México. United Nations, F. A. O. y José Sarukhan. México. Rosengarten, Frederic, Jr. 1973. The Book of Spices. Pyramid Books, New York. Schultes, Richard Evans. 1957. The genus Quararibea in Mexico and the use of its flowers as a spice for chocolate. Botanical Museum Leaflets, Harvard University. Vol. 17, No. 9, 247-264. 1972. Quararibea funebris: A curious spice for chocolate drinks. The Bulletin, The Horticultural Society of New York. Vol. III, No. 4, 1-4. Standley, Paul C. 1923. Trees and Shrubs of Mexico. Smithsonian Institu- tion. United States National Museum. Contributions from the United States National Herbarium. Vol. 23, Part 3, 787-789. 189 PLATE 28 Plate 28. White flower of Quararibea funebris approximately | in. inlength. 190 PLATE 29 Plate 29. 120-year old tree of Quararibea funebris growing at Huayapan, near Oaxaca, Mexico. Approximately 50 ft. tall. 19] PLATE 30 WIV WALDO : vnsumod unqwonf, (SNOILWIISIGOW HLIM ‘NYHXNYS ONY NOLONINNGd waisv) ODIXEW NI SIUSINNS * VIDINWYVNO 4O NOlingiuisid - e ¢ (NeSeKYNH) i Jf ‘ s ak gee rs ‘ ‘ £ QR ’ = ! \ ru ¥ c t wre sy x a] i) ' ev 2DNZ/ 7 - 3 OxKLyD> ODIXIW 192 PLATE 31 QUARARIBEA _funebris (Llave) Vischer- g 193 PLATE 32 ot ft an ye a f os F ‘ a 5 is ¥ I a 4 PF ihe : ? " i" ‘ Rh 3 ) ' a \ . ‘Az 6° i ' Plate 32. Drying flowers of Quararibea funebris ona straw matat Huayapan. 194 PLATE 33 Plate 33. Every Saturday throughout the year, the flowers of Quararibea funebris and the beverage ‘‘tejate’’ are sold in the market at Oaxaca (see woman at left). 195 PLATE 34 ar. * ie Pe ini. Plate 34. Ingredients used in the preparation of tejate at Huayapan: left, roasted mamey nuts; top, cocoa beans; foreground, corn; and on top of leaf, roasted flowers of Quararibea funebris. 196 PLATE 35 Plate 35. Roasting the ingredients for tejate. 12} PLATE 36 Plate 36. Grinding the ingredients for tejate. 198 PLATE 37 Plate 37. A bowl of fresh tejate, ready for drinking. The sticky froth on top of this chocolate-flavored beverage is due to the mucilage in the flowers of Quararibea funebris: hence the names *‘flor de cacao,”’ or ‘‘rositade cacao.” 199 PLATE 38 Plate 38. In foreground, fruit, flowers and leaves of Quararibea funebris: in background, a bowl of tejate. 200 PLATE 39 Plate 39. Quararibea funebris, reproduced from the Paso y Troncoso edition of Bernardino de Sahagun: Historia general de las cosas de Nueva Espana. 201 PLATE 40 Plate 40. Bud, flower, fruit and seed (in husk) of Quararibea funebris. 202 Botanical MUSEUM LEAFLETS VoL. 25, No. 7 CANNABIS FOLKLORE IN THE HIMALAYAS G. K. SHARMA* Perhaps searching for truth and reality in folklore is well summarized by the Chinese proverb: “‘It is better to journey than to arrive’. Cannabis folklore in the Himalayas is full of mystery and awe. To add further to the tangle, it is shrouded in the metaphysical outlook of the East. The Himalyas offer a great challenge to the ingenuity of the ethnobotanist, ecologist, biochemist, and anthropologist. The remoteness and inacces- sibility of human populations, the confusing and long history of wild and cultivated Cannabis in the area, the antiquity of local culture, the mythological folklore of the area, and the stark immutability of the Himalayas make these mountains an ideal area for investigations such as those herein reported. Cannabis is wild in the area, although cultivation is permit- ted in certain parts of the Himalayan arc, extending from Afghanistan to Burma — a distance of 1,500 miles. The use of Cannabis here was reported in some of the oldest Aryan scrip- tures, including the sacred Vedas (4,000-5,000 B.C.), where it has been called ‘‘joy-giver’’, ‘‘victory’’, and “‘liberator’’. The oldest known Vedic description of Cannabis is found in Book XI, 6, 15 of the Atharva-Veda (3 ,000-4,000 B.C.). In the trans- lation (Whitney, 1905) of the original Sanskrit text, dedicated to many different gods for relief, the following is stated: ‘‘The five kingdoms of plants, having soma as their chief (crestha), we address; the darbha, hemp, barley, saha — let them free us from distress.”’ Other references alluding to the antiquity and the special niche of Cannabis in the ancient Indian culture can be found in classics such as Re-Veda, Susrita and the Mahabharata (Ma- jumdar, 1952; Watt, 1889). The Susrita (600 A.D.) offers hemp as an anti-phlegmatic. Nedkarni (1954) characterizes Cannabis *Research Fellow in Economic Botany, 1976-1977, Botanical Museum of Harvard University. Present address: Biology Department, University of Tennessee, Martin, Tennessee, 38238. 203 ee as ‘*. . . asource of great staying-power under severe exercise or fatigue.’ Sarangadhara (1,500 A.D.), describes Cannabis as an excitant; while in Bhavapakash (1,600 A.D.), the plant is known for its exhilarating properties and for curing leprosy. The Muslims regard bhang as a holy plant and in Tibbi (the Muslim system of medicine) it is used for treating numerous diseases: asthma, dandruff, and urinary disorders. In the Zoroasterian scriptures of ancient Persia (with close resem- blance to Rg-Veda), references to bhang for producing miscar- riage and for euphoria (Darmesteter, 1883) can be found. It would not be an exaggeration to call Cannabis the penicillin of Ayurvedic medicine — the indigenous medical system of India. Even the generally accepted findings of the Hemp Drugs Commission in India (1893-94) did not oppose the moderate use of Cannabis for social and medicinal practices in the Indian sub-continent, although excessive consumption was regarded as injurious. Many methods of using Cannabis in these mountains have been deeply rooted in the cultural, social, and economic lives of the local peoples. A casual observer is not likely to see or hear much Cannabis folklore because of cultural and social differences. Furthermore, folklore is a way of life for some people in the area; hence one finds no glittering pronounce- ments or exaggerations typical of urban societies. Only a seri- ous student of ethnobotany purposefully seeking it out could penetrate the mystery of Cannabis folklore in these mountain fastnesses. The average inhabitant questioned about Cannabis use or folklore, expresses indifference or little interest in it, making research even more difficult. Near the India-Tibet border in the northern Himalayas in India, I met several Tibetan refugees (now Indian citizens), known for their frequent use of ‘‘momea’’ or “‘solaradsa’’ (Cannabis) for its medicinal, food and narcotic properties (Plate I). Mountain caravans travelling through narrow passes at 17,000 feet carry normal supplies of Cannabis preparations for their long and adventurous jour- neys. Even a discreet gesture to befriend them can, at times, arouse a bellicose response — adding to the difficulties of the probing ethnobotanist. Notwithstanding these difficulties, it 204 must be done to preserve the folk medicine associated with Cannabis, else the onslaught of modern medicine and technol- ogy is sure to bury this fascinating heritage. In 1973 and in 1976, I spent several months in the northern and central Himalayas, visiting towns, villages, hamlets and mud-huts to observe the preparation and folk use of Cannabis. I interviewed several local people, representing a variety of professions and social strata. This area of mountain peaks is covered with perpetual snow, ranging up to 29,000 feet or more, although my studies did not take me above 15,000 feet. Human populations are scattered and small. Cannabis grows up to 10,000 feet in a wide variety of microhabitats. The plant exhibits great phenotypic plasticity and, as suggested by folk- lore and my preliminary studies (Sharma, 1975), possibly cor- responding narcotic strengths. Dry conditions of microhabitat have generally been considered more conducive to narcotic potency. There are areas at different elevations (but with a similar humidity range) known to have strongly narcotic Can- nabis. 1 visited several of these places known to the local people for ‘‘strong’’ Cannabis: the folklore was confirmed by users and non-users alike. At certain localities, semi-religious gatherings or fairs are arranged, and the use of Cannabis prod- ucts is common among the religious mendicants, the so-called ascetics, and fakirs. It is important to remember that, with some exceptions, the use of Cannabis is confined usually to the economically lower strata. Cannabis is not generally accepted in the educated society. While narcotic use of Cannabis is more prevalent in the economically lower groups, its medicinal properties are valued throughout the area and in all segments of society. Cough, diarrhea, asthma, malaria, excessive bleeding, and high blood pressure are some of the ills for which preparations of Can- nabis have been used in the past in the Indian sub-continent and in many other parts of the world (Bouquet, 1950; Chopra & Chopra, 1957); and these uses are still common in the Himalay- as. Chopra (1933) and Chopra & Chopra (1957) outline some of the common modes of consumption of Cannabis in the Indian sub-continent: Masum: A special type of confection, for the preparation of 205 which bhang or ganja is either boiled in water or heated in clarified butter. The green resinous scum that appears on the surface is mixed with sugar and is then heated to form a paste, cut into small pieces. In some parts of the country, this prepa- ration is slightly different. HaAtwa: Bhang is boiled in a jaggery syrup. The filtrate is mixed with flour and clarified butter for making halwa. This preparation is common in the southern part of the Indian sub- continent. Curry: Bhang leaves are prepared in powdered form and then used for making curries, with the usual assortment of spices and vegetables. BHANG BEVERAGES: Bhang leaves are pounded and mixed with sugar, black pepper, and diluted with water. Almonds, resins, and even curds are mixed with pounded bhang leaves in the well-to-do strata of society. SMOKING OF CANNABIS: Ganja and charas are used for smok- ing in a wide variety of equipment in the sub-continent. ‘‘Chil- lam’’ or “‘hookah”’ (the water pipe) are the commonly em- ployed paraphernalia. Partly in view of the great significance given to Cannabis asa medicine in the Susrita (600 A.D.) — the Bible of the Indian system of medicine — there is a growing interest in revitaliza- tion of the indigenous Ayurvedic system of medicine in India. Government-supported clinics and research units exist where age-old treatments, making use of native herbs, are fully utilized and investigated. There is a tendency among the old societies to be influenced by modern civilization and technol- ogy. During this metamorphosis, their own cultural heritage is forgotten or denegated. Maintenance of fine cultural traits or the protection of the existing ones is a challenge, since they bespeak man’s achievements in his climbing the evolutionary scale. Nowhere does this fit more than in the case of Cannabis in the Himalayas. Such an investigation can discover new biody- namic principles associated with Cannabis. This paper is obvi- ously a step in that direction. The Himalayas are extensive, and Cannabis folklore there needs further research and attention. | observed the following modes of Cannabis utilization in the 206 Himalayas during my ecological investigations in that area. It is quite possible to encounter variants of these in other parts of the world. BHANG PaKoras: This preparation 1s made by mixing fresh or dried Cannabis leaves (Plate II) with chick-pea flour. A kind of electuary is prepared by adding water and the desired condi- ments (black or red pepper, ginger, and cumin seeds). The resulting dough is salted. Small balls of the dough are fried in mustard oil. These fried ‘pakoras’, as they are called, are eaten as snacks with tea or beer. The strength of this preparation is controlled by the amount of Cannabis leaves at the time of making the dough. On festive occasions, these ‘pakoras’ are sought with great felicity. BHANG PARANTHAS: For this extremely popular delicacy, crushed or powdered roasted seeds are stuffed in large, round wheat bread dough, either baked in an earthen oven or shallow-fried in a skillet. The resulting product is used like a regular bread by adults ina family. In some cases, Cannabis of desired quality is cultivated in private lots for this common preparation. BHANG BALLs: Fresh, tender leaves of Cannabis are finely powdered or crushed and mixed with a small amount of water. Small (1°’ diameter) balls are prepared from the mixture. Poppy seeds are added at times. These Cannabis balls are used as snacks with tea or coffee and sold freely in markets, especially during the summer months. The educated people seem to pay no attention to this product, however, since it is an accepted custom in the area (Plate III). BHANG AND Honey: A widespread belief holds that a concoc- tion of young Cannabis leaf powder and honey keeps youth, vitality, and virility. It is also used to maintain hair colour and texture. The preparation is made easily by the local inhabi- tants, since the ingredients are available on the spot. ‘*BHANG, HONEY. AND FULL Moon”: According to folklore, if bhang is collected from the plant during a full moon, it yields enormous amounts of resin of high quality, which is mixed with honey before consumption. Folklore thus becomes practical on occasions during the spring and early summer, when Can- aabis produces copious amounts of resin, a fairly well estab- 207 lished phenomenon in that part of the world. Perhaps the full moon may have a validity in scientific significance — a point that should be investigated. CANNABIS AND CosrRa: This folklore belief is accepted throughout this area: the cobra is killed and buried; Cannabis seeds grown on this site are presumed to yield extremely potent forms of marihuana, which is used for medicinal purposes, especially for tuberculosis. Such sites are naturally not com- mon, since such utilization would be limited to medicinal use. It demands, however, a scientific analysis of this folklore be- lief, since there may be other chemical substances of signifi- cance, in addition to tetrahydrocannabinol, which possess medicinal potentialities and which may be affected by this treatment of the soil. BHANG CHEWING: It is acommon practice among travellers — especially porters and caravans in the high Himalayan country — to chew bhang leaves during their journeys at high eleva- tions. I came across a lonely traveller at about 10,000 feet, riding his mule (Plate IV) and fully equipped with Cannabis preparations to be used during his arduous journeys. The chew- ing of Cannabis leaves in the Himalayas may well be compared to the use of coca leaves in the Andes. BHANG SMOKE AND CHILDBIRTH: In the mountains and in the plains of Panjab and adjoining Pakistan, smoke in the room or house at childbirth from burning bhang seeds is considered to be aritual that drives off bad spirits and thus ensures health and luck to the new-born. Seeds are burned in an earthenware pot at regular intervals in the mornings and evenings. Again it seems obvious that this practice may have significance in suggesting the use of Cannabis seeds as a vermicide. I visited a small, isolated village in the foothills of the Hima- layas: a village with the reputation of Cannabis of great narco- tic strength. An exotic, religious fair, well attended, is held annually in the nearby forested grove inhabited by a few reli- gious mendicants. Bhang products are collected just before the onset of the Monsoons — a month earlier to be precise — since it is at this time of the year when the highest quality bhang is produced. After the rains, according to reports, the strength is diluted. There may be some logic to this folklore: tetrahydro- 208 cannabinol resin exudes in abundance in dry, hot weather, which is so typical of this area before the rains arrive. The practice of collecting the resinous secretion before the rains is common throughout the entire area. Some of the other common uses of Cannabis in the Himalayan area are: oil extracted from the seeds is used for rheumatism; leaves boiled in milk and used for stomach upset; seeds mixed with barley are eaten by the common folks of these hills, perhaps an unconscious following of the hymn of Atharva-Veda; Cannabis mixed with Datura Stramonium seeds for intoxication. | During field investigations, | met several curious, helpful inhabitants of the hamlets. Practically everybody suggested a strong correlation between strong odour and narcotic potency of Cannabis. A biochemical analysis of various populations is planned to determine the validity of this folklore belief. Finally, the uses of Cannabis drugs in the Indian sub-conti- nent can be described under the following categories (Chopra & Chopra, 1957): 1) medical and quasi-medical use; 2) use in connection with religious and social customs; 3) euphoric pur- poses. Unfortunately, it is the use or abuse for euphoric pur- poses which causes public commotion, fear and indignation in the minds of many, sometimes obscuring the medicinal and religious significance of Cannabis in India. General belief holds that Cannabis is a means of escape from the realities of life. Dutt (1900) rightly summarizes the euphoric effect of Cannabis observed in Indian surroundings: Almost invariably the inebriation is of the most cheerful kind, causing the person to sing and dance, to eat food with great relish, and to seek aphrodisiac enjoyments. In persons of a quarrelsome disposition, it induces, as might be expected, an exasperation of their natural tendency. There is some truth to this statement, as evident in the Himalayan areas and other parts of the world. However, cau- tion and discretion must be exercised while trying to write a complete story of Cannabis, since its folklore may reveal new medicinal potentialities (as has been the case in the past with this and many other plants in the Himalayas). I hope, therefore, 209 that this account may stimulate further research before the tide of modern civilization sweeps away all folklore. Many more isolated Himalayan niches remain to be visited for data. A combination of folklore and modern tools of research in biol- ogy, biochemistry, and anthropology will surely result in a wealth of information. Investigation on Cannabis and biologi- cal problems associated with it have not yet received the sup- port and attention warranted by their immense potentialities for human welfare. International and national efforts must be made to understand basically this plant of such great signifi- cance. Further, the Himalayas are untapped, representing a relatively virgin field for folklore and scientific endeavors. ACKNOWLEDGMENTS I wish to thank Professor Richard Evans Schultes, Director, Botancial Museum, Harvard University, for his advice and encouragement during the course of my studies. Thanks are due the Government of India and the various State Govern- ments in the Himalayan area for permitting me to conduct research on Cannabis; inhabitants of the area studied for their generous help; Abha Kapila and Naveen Kapila for acting as guides. LITERATURE CITED Bouquet, R.J. 1950. Cannabis. Bulletin Narcotics 2:14-30. Chopra, R.N. 1933. Indigenous Drugs of India. The Art Press, Calcutta. Chopra, I.C. & R.N. Chopra. 1957. The use of Cannabis drugs in India. Bulletin Narcotics 9:4-29. Darmesteter, James. 1883. The Zend-Avesta. The Clarendon Press, Oxford, England. Dutt, Udoy C. 1900. The Materia Medica of the Hindus. Dwarkanath Mukherjee, Calcutta. Indian Hemp Drug Commission Report. 1893-94. Simla, India. 210 Majumdar, R.C. 1952. The Vedic Age. George Allen and Unwin Ltd., Lon- don. Nadkarni, K.M. 1954. Indian Materia Medica. Popular Book Depot, Bom- bay. Sharma, G.K. 1975. Altitudinal variation in leaf epidermal patterns of Can- nabis sativa. Bull Torrey Bot. Club 102:199-200. Watt, George. 1889. Dictionary of Economic Products of India. E.P. Dutton & Company, Calcutta. Whitney, William D. 1905. Atharva-Veda Samhita. Harvard Univesity Press, Cambridge, Massachusetts. 211 PLATE 41 i Plate.41. A Tibetan refugee near the India-Tibet border holding a Cannabis plant growing in the area. ZAz PLATE 42 ee | aterial being dried in the sun. Cannabis plant m Plate 42. nm —N PLATE 43 o- Plate 43. A shop in the area selling bhang balls. 214 PLATE 44 ; € ©, Plate 44. A mountain man chewing and smoking bhang during long mountainous journey. BoranicAL MuSEUM LEAFLETS VoL. 25, No. 7 CHEMICAL TEST FOR IDENTIFICATION OF COPROLITES ELIZABETH A. COUGHLIN Archaeological identification of coprolites has, for the most part, been based on visual and microscopic evaluations of gross morphology and of dietary botanical inclusions. These, combined with the absence of any major skeletal network, but possibly with inclusions of small dietary skeletal remains, have served as indicators of coprolites. Chemical analysis, however, of the suspected coprolite itself and of nearby associated material from the same stratum, level or feature offers further substantiation for identification. Significantly high levels of nitrates (which are the final and most highly oxidized form of nitrogen in the nitrogen cycle of biologic oxidation of organic nitrogen compounds) indicate the final stages of biological degradation and are thus suitable for indentification of coprolites. A colorimetric estimation of nitrates (NO3) can be made by measuring the yellow color produced by the reaction of nitrates with brucine at a wave-length of 410 millimicrons. PROCEDURE 1. PREPARATION OF STANDARDS A. Stock Nitrate Solution: dissolve 721.8 mg. anhydrous potassium nitrate, KNO3, and dilute to 1,000 ml. with distilled water. (This solution contains 100 mg /1 N.) B. Standard Nitrate Solution: dilute 10.00 ml. Stock Ni- trate Solution to 1,000 ml. with distilled water. Prepare immediately before using. C. Nitrate Standards: prepare standards in the range of 0.1 — 1mg/IN. by diluting 1.00, 2.00, 4.00, 7.00, and 10.00 217 to ml. Standard Nitrate Solution to 10.00 ml. with distilled water. (1.00 ml. of Standard Nitrate Solution = 1.00 mg. N.) PREPARATION OF SAMPLES A. D. Grind to a fine granular form a small amount of sus- pected coprolitic material. Add 10 ml. distilled water and shake thoroughly. Pass this mixture through a glass filter, separating out all loose particles. Collect the filtrate; this is the prepared sample. PREPARATION OF REAGENTS A. B. Sodium Chloride Solution: dissolve 300 g. sodium chloride and dilute to 1,000 ml. with distilled water. Sulfuric Acid Solution: add 500 ml. concentrated sulfur- ic acid to 125 ml. distilled water. Cool to room tempera- ture before using. . Brucine-Sulfanilic Acid Solution: dissolve | g. brucine sulfate and 0.1 g. sulfanilic acid in 70 ml. hot distilled water. Add 3 ml. concentrated hydrochloric acid, cool, and make up to 100 ml. with distilled water. PREPARATION OF BLANK A. Prepare reagent blank by measuring out 10 ml. distilled water. B. Run test procedure as in 5, including water baths and addition of reagents. TEsT! A. Place all standards, samples and blanks to be tested in test tubes in a cool water bath and add 2 ml. sodium chloride solution. 'This test eliminates interferences of nitrites ( NO>) and chlorides. Interferences due to excess organic material are removed by filtration through glass (See 2-C.). 218 B. Mix thoroughly by swirling by hand. Do not use mechanical or magnetic stirrers. Add 10 ml. sulfuric acid solution and swirl again. Replace in cool water bath. C. Add 0.5 ml. brucine-sulfanilic acid reagent; swirl by hand and place in a well stirred boiling water bath that maintains a temperature of not less than 95°C. D. After exactly 20 minutes, remove standards, samples and blanks and immerse in a cold water bath. E. When thermal equilibrium is reached at room tempera- ture, dry off the tubes, and read standards and samples against reagent blanks at 410 millimicrons in a spec- trophotometer. Run at least two standards and two blanks with each batch of samples. 6. CALCULATIONS A. Subtract reagent blanks from final absorbance readings of standards, and plot the resultant absorbance against mg. NO3-N/I1 (milligrams nitrate-nitrogen per liter). B.Correct the absorbance readings of the samples by sub- tracting their sample blank values from their final absor- bance values. C. Read the concentrations of NO3-N directly from the standard curve. D. mg. NO3-N mg./1 NO3 —N = ml. sample E. mg./1l NO3 = mg./1 NO3-N x 4.43 7. EVALUATION OF RESULTS A. Ahigh level of nitrate (NO3) in the suspected coprolitic material combined with comparatively lower levels of NO3 in associated accompanying material from the pA same level, stratum or feature would substantiate iden- tification of coprolites. This procedure was developed in the Ethnobotanical Labo- ratory of the Botanical Museum of Harvard University with the support and encouragement of Prof. Richard Evans Schultes, Director of the Museum. I thank Prof. Schultes and the museum staff for their contribution to this project. I also ex- press my appreciation to Miss Virginia Popper for initially suggesting to me the application of chemical techniques to ethnobotanical and archaeological identifications. No i) So BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS, OCTOBER 30, 1977 Vo-. 25, No. 8 STUDY OF PRE-CERAMIC MAIZE FROM HUARMEY, NORTH CENTRAL COAST OF PERU ALEXANDER GROBMAN, Duccio BONAVIA AND Davip H. KELLEY WITH PAUL C. MANGELSDORF AND JULIAN CAMARA-HERNANDEZ! ” In 1960, Harvard University received a collection of maize samples from the Valley of Huarmey in the north central coast of Peru. These samples represented prehistoric specimens and were sent by David H. Kelley, who at the time was associated with West Texas State University, and by Duccio Bonavia, who then was an associate research worker at the Institute of Anthropological Investigations of Lima. Although these samples were studied almost immediately, due to a series of circumstances the publication of the report has been delayed. We believe today, however, that this infor- mation is valuable for the knowledge of prehistoric cultigens, and itis, therefore, being published with slight modifications in the original manuscript, as a result of recent fundamental changes which have come about in our knowledge of the ar- chaeology of the Peruvian Coast. With reference to the subject, ‘Alexander Grobman, Associate Director General, Centro Internacional de Ag- ricultura Tropical, Cali, Colombia. Duccio Bonavia, Profesor Principal, Departamento de Biologia, Universidad Peruana Cayetano Heredia, Lima, Peru. David H. Kelley, Professor, Department of Archaeology, University of Calgary, Alberta, Canada. Paul C. Mangelsdorf, Fisher Professor of Natural History, Emeritus, Harvard University. Julian Camara-Hernandez, Professor Asociado, Facultad de Agronomia, Uni- versidad de Buenos Aires, Buenos Aires, Argentina. ?The section on description of maize remains was worked and written by Paul C. Mangelsdorf and Julian Camara-Hernandez. Published monthly except during July and August by the Botanical Museum, Harvard University, Cambridge, Massachusetts 02138. Printed by Harvard University Printing Office. Subscription: $20.00 a year, net, postpaid. Orders should be directed to Secretary of Publications at the above address. Application to mail at Second-Class Postage Rates is pending in Boston, Mass. only a few general notes were advanced (vide Kelley and Bonavia, 1963; Mangelsdorf and Camara-Hernandez, 1967), which are now being complemented with the present technical report on the botanical material. The present report, nevertheless, is a preliminary one. The work at Huarmey has not been concluded, and a final and more thorough study is in preparation. Up to this time, the site appears in the scientific literature with the denomination of Huarmey North 1, because a name was not known for that site. From now on we will call it Los Gavilanes, an old toponimic which the dwellers of the locality recall and its code will be PV35-1, according to Rowe’s nomen- clature. The locality of Huarmey is in the Valley of Huarmey, which is in the Department of Ancash, Province of Casma, District of Huarmey in the North Central Coast of Peru. The archaeologi- cal site which is being described is located in the lower part of the Valley, exactly at 78°10'21’’ longitude West and 10°02'45”’ latitude South. The site, which was originally located by Ed- ward Lanning, was a subject of interest of David H. Kelley, who made some test pits in the years 1957 and 1958. At the request of Paul C. Mangelsdorf, who at that time was Director of the Botanical Museum of Harvard University, in the year 1960 Duccio Bonavia excavated two stratigraphic cuts at the site and returned to it on numerous occasions in order to effect further observations. However, it is only in 1974 that Bonavia was able to make a more detailed and thorough work, still regarded as preliminary, on account of the size of the site and the complexity that it represents. This site is located in a sandy area, removed from the culti- vated area of the Valley, and a small range of hills separated it from the sea, while in its northern limit the remains of a fossil lagoon can be traced. This lagoon was formed on the bed of an ancient river, which is now dry and as a consequence of the conformation of ocean beach lines. The site is totally covered by a layer of eolic sand and on its surface can be observed patterns of a series of structures, apparently circular, whose function is not yet clear. No ceram- ics are present on the surface, while lithic material is easily one found. These are rough, simple tools, the majority with indefi- nite typology, made by percussion, and all of them on flakes. The only typical tools which characterize the site (and whose close counterparts are found only in the neighbouring Valley of Culebras and ina similar context), are some pebbles of various sizes, round or oval in shape, thin, and whose borders have been worked by unifacial percussion creating cutting edges. When Kelley made the first preliminary excavation in the year 1967, he found one maize cob, peanut shells, abundant samples of cotton, fragments of gourds, and residues of sea shell and bones. It is presumed that one sample of lima bean was also present, although it was never identified by a specialist. Later on, Kelley himself in 1958 continued search- ing for more botanical evidence in the surroundings of his preliminary excavation site, having been able to find 10 maize cobs and a corn tassel. This material was shown to Man- gelsdorf and because of its importance it was decided to go on with the search. When Bonavia excavated the site in the year 1960, he was able to find abundant residues of corn (consisting not only of cobs and ears, but also leaves and plants) besides lithic resi- dues (among which some disks which were mentioned before and one chipped stone point), textiles, mats and ropes. What called the attention most particularly was the finding of three pottery sherds during the excavations of the second stratig- raphic cut. Two of them were at the base of the eolic sand, that is on the surface of the site, at a depth of 15 cms., while the third appeared without any control, because of the collapse of one of the walls of the cut; it could be presumed that this one was associated with the two previous ones. The finding of these pottery sherds left some doubts about the possibility that the site could be pre-ceramic. However, in spite of many visits made by Bonavia to the site since 1960, he could never find on the surface other pottery fragments and during the excavations which took place in the season of 1974 nothing in the way of ceramic residues was found. Although the studies of the structures present a series of questions, in gen- eral terms all the cultural context which is associated with this corn belongs to the late pre-ceramic period of the Peruvian 223 Coast, what Lanning called Pre-ceramic VI (vide Lanning, 1967). Unfortunately, we are not in a position to analyze the three pottery fragments, a matter of controversy today, be- cause they were delivered to Lanning for their study and have been lost. Judging, however, on the basis of the recollections of Bonavia who found them, because of their paste characteristic, they do not correspond to early pottery as known on the Peruvian Coast. Bonavia suggests that these fragments could well have been brought by fishermen who had a permanent transit route over the site or else they could have been depo- sited in the pre-hispanic period, long after the site was aban- doned by its original dwellers. This statement is based on the fact that in the sandy areas surrounding the site, small surface sites have been found, which probably correspond to fisher- men who came down from the Valley and stayed there for a short period of time, and they belong in time from the Middle Horizon up to the Late Horizon. The pottery found at Los Gavilanes should correspond to these occupations. When the site was excavated in the year 1960 carbon samples were gathered in order to be utilized for dating through '4C. Due, however, to anerror, samples of corn were sent to labora- tories for determinations. The results obtained were totally inconsistent since the date fluctations were between 200 and 800 years before the present era, with a margin of error varying between 70 and 95 years. When this situation was investigated, we were informed that ‘‘corn cobs were particularly bad for use for '*C dating, be- cause growing corn had a most unusual rate of absorption of '4C, giving a high content and indicating spuriously late dates”’ (Gary Vescelius, personal communication, 1970). Further- more, there is a possibility of contamination of these samples, either through moisture in the deeper part of the site, in connec- tion with the fossil lagoon, which is nearby, and which up to now maintains some vegetation through its humidity. Also the corn was subjected to multiple manipulations before it went through the '*C process of dating. Up to a few years ago the existence of pre-ceramic corn in Peru was not only doubted, its acceptance was even forth- rightly refused by some archaeologists. Today the situation has 224 changed and pre-ceramic corn is accepted by a number of specialists (vide Lanning, 1967 and Moseley, 1975). The area of corn occurs through a coastal belt of approxi- mately 150 kms. which extends from the famous site of Las Haldas (in the vicinity of Casma) to Aspero (near Supe) (vide Moseley, p. 89). Most of this corn belongs to the Late Pre- ceramic that in terms of time means a period which extends from the year 2500 B.C. to the years 1800-1500 B.C. We think that in spite of existing doubts, the corn at Los Gavilanes is part of the same complex. New samples have been sent for new dating. At this time we have received two new datings, ob- tained on material excavated in the 1974 season, and which we believe form part of the same context to which plants exca- vated in 1960 belong. The first dating was made by the method of thermolumines- cence of burnt stones, at the Laboratoire de Cristallographie et de Physique Cristalline de la Faculté des Sciences of the Uni- versity of Bordeaux, France. The result obtained is 4800+ 500 years BP(BOR 20). On the other hand, a sample of burnt wood from the same context was dated by the '*C method at the Laboratoire du Radiocarbone du Comisariat al‘ Energie Atom- ique et du Centre National de la Recherche Scientifique de Gif-sur-Yvette, France. The date obtained is 3750+ 110 years BP (GIF 3564).' Also, in an indirect form, the results of the work of the Ayacucho Archaeological-Botanical Project, di- rected by Richard MacNeish (vide MacNeish, Nelken-Terner and Garcia Cook, 1970) in some respects provides a logical explanation to the problem. Though there are some questions, we believe that in the future these can be answered and that because of the dearth of botanical reports on the materials found at the pre-ceramic sites and because of the importance of the Huarmey corn, it Is desirable that we publish the result of our studies, pointing out again that they are only of a preliminary nature. Since this is a preliminary report, we have make no attempt to review the extensive literature that, in one way or another, 'Thanks are due to Claude Chauchat and Daniele Lavallée, for their help in securing the collaboration of this institution, and to Georgette Delibrias and Max Schvoerer of the two laboratories respectively, who made the actual datings. 22) has a bearing on our own findings. In a later, more complete report, we shall discuss some of the more recent contributions in South American archaeology that have come to our atten- tion, notably an article by Zevallos et al.2 THE MAIZE REMAINS The prehistoric remains of maize from the Huarmey site comprised a total of 238 specimens including all parts of the plant from the roots to anthers in the spikelets of the tassels. Their description follows: Coss: Cobs, 61 specimens in all, were found at all levels in the pits. With respect to their size the cobs reveal an evolution- ary sequence, those from the lower levels being generally shorter and more slender with fewer kernel rows and fewer spikelets per row. Since the majority of the cobs were not intact with respect to length and many of them were somewhat eroded, the only datum that could be obtained from all of the specimens was kernel-row number. The data for this charac- teristic are shown in TABLE 1, along with such other data as we TABLE | Kernel-row numbers of the cobs from five stratigraphic levels. Level 1 is most superficial: level 5 is deepest. Level Kernel-row numbers 4 8 10 12 14 l 2 2 2 4 8 l 3 | 1S 5 4 16 l 5 2 l “Carlos Zevallos M., Walton C. Galinat, Donald W. Lathrap, Earl R. Leng, Jorge G. Marcos, Kathleen M. Klumpp. 1977. **The San Pablo corn kernel and its friends’. Science, Vol. 196, pp. 385-389. 226 were able to obtain. The majority of the cobs from the three lowest levels were eight-rowed and none had more than ten rows. In contrast the majority of cobs in the two upper levels had more than eight rows and several had twelve or fourteen rows. One exceptional cob in level 3 (not included in Table 1) was distichous and had only four rows. Accompanying an increase in kernel-row number was an increase in the diameter of the rachis and the total number of spikelets. This conclusion is more a matter of observation than one of statistical averages since the number of intact cobs was limited. A trend can be illustrated, however, by comparing the intact cobs from levels 2, 3 and 4. The averages from these three levels are 133, 126 and 113 spikelets respectively. The early cobs from this site appear to represent a weak form of pod corn. The glumes are quite long in relation to the diame- ter of the rachis and they are soft and fleshy; not like the indurated lower glumes of corn’s relatives, teosinte or 7rip- sacum. With respect to known races of maize of Peru, some of these cobs could be assigned to the Peruvian popcorn race Confite Morocho described by Grobmanet al. (1961, Fig. 49). Like the slender cob illustrated by these authors the rachis is square in cross section and the cupules are shallow in outline. None of the cupules are as long as those of the unusually slender ear illustrated in Grobman et al. Cobs of this type are also quite fragile, breaking up easily into sections of one cupule each. Some of the smaller cobs have the stumps of terminal staminate spikes. The early cobs, which comprise the majority of the collec- tion may be related to the Peruvian race, Confite Morocho. This possibility is supported by cob morphology, number of kernel rows, which approaches an average of 8, kernel type, form and consistency of the glumes, etc. They bear also some resemblance to the Mexican race Chapalote, although they seem to be farther removed from it. Except for one cob, with four rows of spikelets and stiff lower glumes, which we would have classified as tripsacoid had we found it in the context of a Mexican ancient cob collection, all the others exhibited long, Zed semi-tunicate, soft glumes. Tripsacoid cobs, such as those we would ascribe to the result of hybridization of corn with either teosinte or Tripsacum, simply do not appear in the Huarmey collection. The Huarmey maize bears an interestingly close resem- blance to the Confite Morocho maize obtained by MacNeish et al. (1970 p. 38), from the Ayacucho, Peru caves that he explored, and which was dated as circa 4300 to 2800 B.C. In all respects it also coincides with segregants described by Grob- man ef al. (1962) as Confite Morocho, from the Los Cerrillos site in Ica. There are two exceptional cobs, one of which is illustrated in Plate 46A. This cob occurred in level 3. It is thick and tapering and has 14 kernel rows. It is quite similar to the predominating corn from the Huaca Prieta site described later. Plates 45 and 46 illustrate the variation in the cobs from the lower to the upper levels of this site. The range in size is by no means as great as it is in the prehistoric cobs from the several caves in Mexico and the southwestern United States. This may indicate that a shorter span of time is involved or it may suggest that in the absence of contamination by teosinte or Tripsacum the rate of evolution is less rapid. KERNELS: Only ten kernels were found. Several of these were immature and one was defective. The pericarp color in all kernels in which it could be determined was brown. The kernel shape, where it could be determined, was slightly pointed, (Fig. 1). Some ears of the Peruvian race Confite Morocho have kernels of this shape (Grobman et al., 1961, Fig. 48). Fig. 1. Enlarged diagram (2X) showing the shape of a kernel found in level 2. The shape is typical of the kernels of some ears of the Peruvian popcorn race Confite Morocho. All of the kernels found in this site have a brown pericarp color. ROOTS AND STALKS: Five pieces of stalks with attached roots show mainly that the root system of prehistoric maize is virtu- ally identical with that of modern maize. Diameters of the first internode above the root in the five specimens were 10.6, 4.3, 228 9.1,7.1, and 9.4 mm. These dimensions are much smaller than those of the modern race Confite Morocho which has an aver- age stalk diameter at the first internode of 16.0 mm. One of the five specimens had a split area on one side showing that it either had a tiller attached to it or was itself a tiller. This is the first archaeological evidence of tillering that we have encountered. (Plate 47) One of the pieces with roots attached had four internodes which, measured from the base upward, had lengths of 51, 93, 143, and 141 mm. respectively. There was an undeveloped ear attached to the third node above the ground and the remains of a husk system attached to the fourth node. An internode pat- tern plotted from the data (Fig. 2) is similar to that of the modern Confite Morocho (Grobman ef al., 1961). cm 16 Fig. 2. Internode pattern up to the i ear-bearing node of a stalk found in level 2. gh The pattern is almost identical with that of the Peruvian race Confite Morocho. cf. Grobman etal., 1961, Fig. 51. Ten pieces of stalk in addition to the five with attached roots were found. The diameter of these ranged from 5.6 to 12.8mm., the average being 8.0 mm. LEAF SHEATHS: Sixty-four specimens of leaf sheaths were identified. All of these were completely glabrous even lacking the weak pubescence usually found in modern varieties around the upper margins. In this respect the sheaths resemble those of the prehistoric corn from the caves in Tehuacan Valley. Some of the sheaths appear to have fitted rather loosely around the stalks; this is a characteristic which we have noted in some modern popcorn varieties and which have caused us to wonder whether it represents an adaptation to growth in regions of PB se) limited rainfall. We have often noticed that after a light shower in which the precipitation is not enough to wet the surface of the soil there is an area of moist soil surrounding the base of the stalks. What has happened is that rain falling on the leaves is directed toward the stalk moving from the upper leaves to the lower and finally to the soil at the base of the stalk where some of it probably reaches the roots. In plants with loose leaf sheaths, water funneled from the leaves is trapped in the spaces between the sheath and the stalk sometimes in substantial amounts so that puncturing the base of the sheaths causes a spurt of water. Can this trapped water be absorbed by the stalk? Plant physiologists to whom we have put this question have usually answered ‘“‘probably”’’ or words to that effect. However, a preliminary experiment that the senior author conducted with Galinat filling the sheath with water containing a vital dye showed no penetration of the dye into the tissues of the stalk. More refined experiments on this problem should be made. If water is not absorbed by the stalk it may be that at least it serves to reduce transpiration thus making the plant more efficient in its use of available moisture. LEAVES: Leaves and fragments of leaves comprised fifty-one specimens. One of those from level 4 is illustrated in Plate 49. The leaf had a width of 46 mm. and a venation index of 3.9 aGtuca=- _ eebipetet OP Fig. 3. Diagram show- ; Cee ing the anatomical features of 2 LARRY the lower epidermis of a leaf CS) found in level 3. The leaves of ~~ prehistoric corn have allofthe ©5 RS characteristics of those of QYRA rz ls aa modern corn: bulliform cells, long cells, silica cells, cork cells, and stomata. which is almost identical with the venation index 3.8 of modern Confite Morocho. Camara-Hernandez made a study of the lower epidermis of a leaf from level 3 and found it to have all of the characteristics described by others: bulliform cells, long cells, silica cells, cork cells, and stomata. His drawing showing these typical maize features is reproduced in Fig. 3. Husks: The husks of which there were 18 specimens are similar to those of modern maize except that they are shorter on the average and have more conspicuous parallel venation. There is a tendency for prominent veins to alternate with weaker ones. This is shown especially well in the photograph reproduced in Plate 48A. The anastomosing venation between two parallel veins is not completely lacking but it is less con- spicuous than that usually found in the husks of modern varie- ties. The husks from any one level are much longer than the longest cobs from the same level (Fig. 4) suggesting at first glance that the ears were well protected against damage from ear worms and other insects by husks extending far beyond the tip of the ear. An almost intact husk system from level 3, however, indicates that the ears may have been exposed at maturity. The shank of this particular specimen had four inter- Fig. 4. Diagram of a husk system found in level . 3. The uppermost internode of this specimen was mis- /\\ sing but a shred of its rind measuring 80 mm. remained | | | showing that the peduncle of the ear must have been at | | | least this long. The diagram of the cob is based on the | | longest cob found in this same level. The long peduncle ) | | suggests that the ear borne in such a husk system might | | | have been exposed at maturity and capable of dispers- \ J} | ing its seeds although enclosed and protected while young. The second internode from the base shows the [| scar of a branch which once must have been attached at | this point and which bore a second ear. The general | | structure of husk systems suchas these and the fact that | | the two outer husks are somewhat differentiated from \L the inner, raises the question whether in primitive | maize a husk system bearing two ears might have been | | the homolog of the staminate spikelets bearing two 4 florets. One-half actual size. 25) nodes measuring from the base upward: 11.5, 7.6, 9.3, 35.7 mm. respectively. The fifth internode was missing but a shred of its rind still remained and measured 79.5 mm. showing that the ear must have been borne on a long peduncle within the husk. This husk system with its fifth internode restored and terminated by the longest cob from the same level is shown diagramatically in Fig. 4. One of its noteworthy features is the scar of a branch on the second internode. Apparently husk systems of this type enclosed not a single ear but several, in this case probably two. A similar husk system from level | had the shred of a peduncle measuring 48 mm. in length. The original peduncle may have been considerably longer. There is a tendency in these prehistoric husk systems for the two outer husks to be somewhat differentiated from the re- maining inner ones in the thickness of the tissues. This is also true in modern varieties but the differentiation is by no means sharp, the transition from the outermost to the innermost being a gradual one. In the prehistoric husk systems the two outer husks are definitely thicker than the remaining inner ones which are almost tissue-paper like in their texture. The husk system showing the differences between the outer and inner husks is illustrated in Plate 48B. TASSELS, SPIKELETS, ANTHERS, AND POLLEN: Two almost complete tassels and 12 fragments were found. A diagram of one complete tassel is shown in Fig. 5. The tassels are similar to those of the Peruvian popcorn Confite Morocho (Grobman et al., 1961, Fig. 50) in the number of tassel branches and their arrangement. (Plate 50). The staminate spikelets of the prehistoric specimens are like those of modern maize, arranged in pairs, one member of each pair sessile the other pedicellate. Their glumes which are more rounded and less strongly keeled than those of many modern varieties, are beset with short hairs as are also the stems of the tassel branches. Spikelets from three tassel fragments in Level 4 had average glume lengths of 7.8 mm., from two tassels in Level 2, average lengths of 8.8 mm., and from 9 tassels in Level 1, average lengths of 7.5 mm. Although there is no evidence of an evolutionary series in the spikelets from these three levels, Zaz y Fig. 5. Diagram of a tassel in level 2. Except for the shorter length of its central spike and branches, this tassel is almost identical with that of the Peruvian race Confite Morocho. cf. Grobman et al., 1961, Fig. 50 One-fourth actual size. all are shorter than those of most varieties of modern corn. The spikelets from the three tassels in Level 4 are almost identical in their dimensions to the earliest spikelets found in the Tehuacan site in Mexico. Anthers containing pollen were found in three different tas- sels. Mounted in lactic acid the pollen grains swelled to the spherical, slightly oval shape characteristic of the pollen of modern maize. Pollen from two of the specimens included many empty grains, in one case 74 empty, 11 normal; in another 29 empty: 27 normal. The significance of these empty grains Is not apparent. The normal grains from these anthers measured 233 80.6 and 78.1 microns in diameter. Those froma third specimen had an average diameter of 86.6 microns. These dimensions are similar to those of the pollen grains of modern races such as Nal-Tel and Palomero Toluqueno of Mexico, which have pol- len diameters of 81.2 and 77.9 microns respectively (Galinat, 1961). PROPHYLLS: Several of the husks systems, four in all, had their prophylls still attached. These are identical in their characteristics to those of modern maize in having two distinct prominent keels. These are illustrated in Plate 48B. DISCUSSION The specimens from the Los Gavilanes site in Huarmey, including all parts of the plant, show that the prehistoric maize from this site is virtually identical to modern maize in all its characteristics, except size. The plant-to-plant variability is less than that found in contemporary races collected at any single location. With a few exceptions the specimens can be assigned to the Peruvian popcorn race Confite Morocho, as it is found today growing in the low to middle altitudes in the Central Andes of Peru, and as it was grown in the same loca- tions 6200 and 4700 years ago, as evidenced by MacNeish’s finds in Ayacucho, Peru caves (vide MacNeish et al. 1970). There seems to be little doubt, if any, that the Los Gavilanes maize is representative of corn grown on the Central North Coast of Peru, at a late pre-ceramic period, and that it was introduced to the Coast from the highlands of Peru at least about 5,000 years ago. Although there is some resemblance between the early Los Gavilanes maize and early popcorn of Mexico of the Chapalote race, itis premature to speculate on the possible significance of the similarities. It may be that early maize had similar overall morphological characteristics, and this is to be expected, as the forces of mutation, selection, and hybridization would not have had the time after domestication at various sites, to effect the profound qualitative and quantitative changes in morphol- ogy that are apparent in maize today. Yet, one thing is common 234 in the early Mexican and Peruvian corn, and that is the almost total absence of evidence of introgression with teosinte or Tripsacum. Tripsacoid corn is not alien to Peruvian archaeo- logical corn, but it appears at much later periods, which would indicate either influx into the Peruvian area of corn which hybridized with Tripsacum in either the lowlands east of the Andes, or in the Coast of the Choco area of Colombia, or of Central America or Mexico, or which hybridized with teosinte in Mexico or Central America. At any rate, this introduction would have occurred much later in history. SUMMARY 1. A collection of maize from a site named Los Gavilanes, located in the North Central Coast of Peru, in the Huarmey valley is analyzed. Judging from evidence available up to this time, the site corresponds to the Late Pre-ceramic of the cultural chronology of the Andean pre-hispanic Epoch. If it is true that certain problems concerning it still exist, there is concrete evidence that confirms the existence of this cultigen before the introduction of pottery on the Peruvian Coast. 2. The collection comprises 238 specimens including all parts of the plant, from the roots to the anthers and pollen grains. 3. The cobs reveal an evolutionary sequence from the lower to the higher levels. In their characteristics they resemble the Peruvian popcorn race Confite Morocho. 4. The stalks of the prehistoric corn are more slender than those of modern corn. The leaf sheaths are completely glabrous. The leaves have all the anatomical characteris- tics of those of modern corn. The husks are much longer than those of the longest cobs. This is regarded as evidence that asingle husk system may have enclosed more than one ear. 5. The spikelets of the tassel are similar to those of modern corn being in pairs, one member sessile, the other pedicel- led. Pollen grains are similar in size to those of modern 235 races of popcorn. The pattern of branching in the tassels is similar to that of the Peruvian popcorn race Confite Morocho. 6. The prehistoric maize from Huarmey is significant in dif- fering in its characteristics from two other Peruvian coast sites, Huaca Prieta and Ica. LITERATURE CITED GALINAT, Walton C. 1961. “‘Corn’s evolution and its significance for breeding’. Economic Botany, Vol. 15, pp. 320-325. GROBMAN, Alexander; SALHUANA, Wilfredo; SEVILLA, Ricardo; in collaboration with MANGELSDORF, Paul C. 1961. Races of maize in Peru. Nat. Academy of Sciences - Nat. Research Council. Publ. 915. Washington, D.C. KELLEY, David H.: BONAVIA, Duccio. 1963. **New evidence from pre- ceramic maize on the coast of Peru’’. Nawpa Pacha, 1, Institute of Andean Studies, Berkeley, California. pp. 39-41. LANNING, Edward P. 1967. Peru before the Incas. Prentice-Hall, Inc., Englewood Cliffs, New Jersey. MACNEISH, Richard Stockton: NELKEN-TERNER, Antoinette: GAR- CIA COOK, Angel. 1970. Second annual report of the Ayacucho Archaeological-Botanical Project. Robert S. Peabody Foundation for Archaeology, Phillips Academy, Andover, Massachusetts. MANGELSDORF, Paul C.;: CAMARA-HERNANDEZ, Julian. 1967. **Pre- historic maize from a site near Huarmey, Peru’’. Maize Genetics Cooperation Newsletter, Vol. 41, pp. 47-48. MOSELEY, Michael Edward. 1975. The maritime foundations of andean civilization. Cummings Publishing Company, Menlo Park, California. 236 PLATE 45 Plate 45. A, Upper right. Fragment of a cob from level 5. The long soft glumes indicate that this was a form of pod corn; upper center, two small cobs from level 4. Lower, typical cobs from level 4, actual size. B. enlargement x2.74, one of the small cobs from level 4 shown in A. C. Similar enlargement of the other small cob from level 4 shown in A. 2a] PLATE 46 Plate 46. A. Representative cobs from level 3. The center cob is not typical of this collection but resembles some of the cobs from the Huaca Prieta site. B. Representative cobs from level 2. Actual size. 238 PLATE 47 Plate 47. A. Stalks and roots from level 3. B. A root system from level 2. Actual size. 239 PLATE 48 Plate 48. A. The two outer husks of a husk system from level 4 illumi- nated from below to show the prominent parallel venation associated with a tendency for strong veins to alternate with weaker ones. Actual size. B. A prophyll and successive husks from an undeveloped second ear on a stalk from level 2. Note the prominent keels of the prophyll and the fact that the two outer husks are thicker than the inner ones suggesting a degree of differentiation. Actual size. 240 PLATE 49 Plate 49. A. Part of aleaf blade from level 4. A drawing illustrating the anatomy of the lower epidermis is reproduced in Fig. 3 and shows that the characteristics of the epidermis of prehistoric corn are similar to those of moderncorn. Actual size. B. A husk system from level 2, one-half actual size. C. A husk system from level 2 showing the shred of a broken peduncle indicating that the ear enclosed by this husk system was borne on a relatively long peduncle. One-half actual size. Another husk system of this kind is illustrated by the diagram in Fig. 4. 241 PLATE 50 Plate SO. A. Part of a tassel from level 4. Actual size. B. Fragment of a tassel from level 5 showing the relatively small spikelets. Actual size. C. A complete tassel from level 2. A central spike is present but is much less prominent than in the tassels of modern corn. One-half actual size. 242 BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY A NEW INFRAGENERIC CLASSIFICATION OF HEVEA RICHARD EVANS SCHULTES I. HISTORICAL CONSIDERATIONS ON INFRAGENERIC CLASSIFICATIONS The genus Hevea, now the source of 98% of the world’s natural rubber, was first known to botany in 1775, when Aublet accurately and thoroughly described Hevea guianensis from French Guiana. For half a century, this was the only species known to science. In 1824, Willdenow recognized a rubber-yielding tree col- lected near the mouth of the Amazon as a distinct species and, without actually describing it, he named it Siphonia brasilien- sis. A year later, Humboldt, Bonpland and Kunth described as Siphonia brasiliensis a plant from the Orinoco, where what we know as Hevea brasiliensis does not occur. Since Willdenow’s species is what we now recognize as true H. brasiliensis and since, in lieu of a description, he published diagnostic drawings of the critical parts of the plant, his “description” of H. brasiliensis is accepted on the basis of priority. The genus Siphonia was later shown to be congeneric with Hevea. Siphonia brasiliensis was transferred by Mueller Ar- goviensis in 1865 to H. brasiliensis. In 1854, five new species were described (under Siphonia) by Bentham, and the following year another was proposed by Spruce — all on the basis of the extensive collections sent from Brazil by the English plant explorer, Richard Spruce. $20.00 a year, net, postpaid. Orders should be directed to Secretary of Publications at the above address. Application to mail at Second-Class Postage Rates is pending in Boston, Mass. By this time, Bentham believed that some kind of in- frageneric classification could be significant in understanding the genus. He put what he then called Siphonia elastica, S. brasiliensis, S. discolor, S. Spruceana and S§. pauciflora into one Section, characterized by a sessile, divaricate-trilobed stigma; into another section, he placed §. /utea and S. rigidifolia, with a short, attenuate style. It is now recognized that this classification groups together species that are not closely related. Bentham pointed out, however, that, prior to Spruce’s field studies, little was known about floral structure in the genus: he wrote that, although the characters ‘have been verified in each instance in several, and often, in many flow- ers .. . it remains to be seen how far they may prove constant when we have specimens from a greater variety of sources’. A second attempt to an infrageneric classification was made in 1858 by Baillon in his Etude générale du groupe des Euphor- biacées. Using the epithet Siphonia, Baillon divided the species then known into two Sections: Hevea and Bisiphonia. Pointing out that there were, among the concepts then recog- nized, species which were intermediate, he placed what are now called Hevea guianensis and its variety lutea in Section Hevea; what are now known as H. brasiliensis ,H. pauciflora , H. Spruceana, H. Benthamiana and H. rigidifolia he included in Bisiphonia. The former Section had an isostemonous an- droecium and no disk, or, at best, an inconspicuous one; the latter, was characterized by having a diplostemonous an- droecium and a more or less well developed disk. Later, in 1864, Baillon merely enumerated seven species (unexplainably omitting H. guianensis) without making mention of an in- frageneric classification. Baillon’s infrageneric classification of Hevea into two groups, sections or series has come down to the present time, even though in recent years its naturalness has been questioned and its acceptance has been denied by most taxonomists who have worked on the genus during the past thirty years. In 1865, Mueller offered his classification of the Euphor- biaceae, placing the genus Hevea in Subtribe Heveeae — the only genus in this subtribe. He divided Hevea into two Sec- tions: Euhevea (equivalent to Baillon’s Section Hevea), made 244 up of one species, H. guianensis: and Section Bisiphonia (the same as Baillon’s Section Bisiphonia) comprising what are now knownas H. Benthamiana, H. brasiliensis ,H. guianensis var. lutea, H. pauciflora, H. rigidifolia and H. Spruceana. In 1866, he revised his family classification of the Euphorbiaceae but maintained his earlier infrageneric classification of Hevea with the same number of species. His only change in treatment consisted in a grouping together of the species of Section Bisiphonia, according to the shape of the staminate buds: H. Spruceana (with subovoid, obtuse buds with a calyx laciniate for only slightly more than half its length): H. pauciflora (with ovoid, obtuse buds, with a calyx divided for 2/3 and 3/4 its length); and H. rigidifolia, H. Benthamiana, H. brasiliensis and H. guianensis var. lutea (with oblong-conic-ovoid and acute buds, apically slightly twisted). Again, in 1874, Mueller divided the genus, as he had done previously, into Euhevea (H. euianensis) and Bisiphonia (all other species). At this time, he arranged the species in Bisiphonia into only two groups: those with staminate buds obtuse (H. Spruceana, H. pauciflora) and those with buds acuminate (H. rigidifolia, H. nitida,H. Benth- amiana. H. brasiliensis and H. guianensis var. lutea). No further attention was apparently given to infrageneric classification of Hevea until 1906, when Huber initiated his detailed studies of the genus. Huber, the first botanist ac- quainted with living trees in their native habitat to consider Hevea generically from a taxonomic viewpoint, followed the earlier division of Hevea into Euhevea and Bisiphonia. Under Bisiphonia, however, he made three Series, giving them tech- nical Latin designations. Into Series Luteae he put what he recognized as H. lutea, H. apiculata, H. cuneata, H. Benth- amiana, H. Duckei, H. paludosa and H. rigidifolia: with an- thers in two incomplete whorls and staminate buds acuminate: these specific concepts he separated into three groups on characters in the disk of the staminate flower. Series /nter- mediae comprised his H. minor, H. microphylla, H. Randiana and H. brasiliensis: with anthers in two complete whorls and acuminate buds: these species he divided into two groups, based on characters in the style. His Series Obtusiflorae in- cluded what he recognized as H. Spruceana, H. discolor, H. 245 similis, H. pauciflora, H. confusa, H. nitida, H. viridis, H. Kunthiana; he arranged these series into two groups based on characters of the disk of the pistillate flower, with the last three (incompletely known species) in a grouping which he called Incertae sedis. Huber maintained that Section Eu/hevea is ‘‘very natural and well characterized’’. While quite distinct from Euhevea, Bisiphonia is, he confessed, **not very homogeneous and does not have a rational subdivision’’ — for which reason he set up his three Series. As late as 1913, Huber still continued to maintain these two Sections and the three Series in Bisiphonia, believing that, in general, this treatment represented natural trends. He did state of Series Luteae, nonetheless, that ‘‘species in the Linnean sense seem almost non-existent in this group. ... With the present state of our understanding, all appear to be in move- ment and fluctuation, and we must be satisfied if we arrive at a rational grouping of small, provisional species.” In 1910, Pax used the division of Hevea into Sections Euhevea and Bisiphonia, separating the two solely on the basis of the number and placement of the anthers. Of the 17 species that he accepted, he grouped three in Euhevea (H. guianensis , AH. nigra, H. collina) and 14 in Bisiphonia (H. Benthamiana, H. Duckei, H. nitida, H. paludosa, H. brasiliensis, H. lutea, A. rigidifolia, H. spruceana ,H. similis ,H. discolor,H. minor. H. microphylla, H. pauciflora, H. membranacea). He pointed out that the flowers of Hevea exhibit few sharp characteristics of use in separating species and that the fruits and seeds, which might provide good differentiating characters, were not known for some species. He further pointed out that the differentiating character employed for Section Bisiphonia were not sharp, noting that he could find intermediates in the anthers of H. guianensis and H. lutea. It is now clear that Pax’s infrageneric classification, as well as those attempts that preceded his, were far from natural. Pax, a specialist in the Euphorbiaceae, was at a great disadvantage in not having seen Hevea growing in the natural state. In 1929, Ducke wrote that ‘‘the natural system of the Hevea is still to be made; the species are very difficult to group 246 because of their affinities that are too close’. It was not, however, until 1923 that he considered definitely the in- frageneric divisions of Hevea: ** While still awaiting fuller ma- terial of certain species, I have already been able to affirm that the sections Euhevea and Bisiphonia ... are not so well de- fined as one has thought: I have found, amongst the many specimens of H. guianensis, some that have the anther-whorl slightly irregular due to the insertion of one of the anthers a little too low. In this same species, in trees of one single locality, the staminate buds vary from wholly obtuse (almost globose) to rather distinctly acuminate’. This point of view he reiterated in 1935. Ducke’s silence on this matter in later publi- cations may be taken as an indication of abandonment of the whole system of grouping the species into subgeneric affinities. I know this to be true, for when we discussed this point in depth, he stated that he had no further use for the proposed infrageneric classifications that had been published. And our refusal to recognize these classifications was crystallized when, in 1945, we jointly reduced Hevea lutea (up to that time a typical member of Bisiphonia) to varietal status under H. guianensis (the only species of Euhevea). Ducke spent more than half a century studying wild Hevea in the Amazon, and he was, undoubtedly, the taxonomist most thoroughly acquainted with Hevea over most of its natural range. Ducke’s taxonomic outlook in Hevea underwent three distinct periods. In his earlier years, still under the influence of his teacher, Huber, Ducke often described minor variants as species (H. gracilis, H. Huberiana, H. humilior, H. mar- ginata). In what we may consider his intermediate stage, he reduced some of these *‘species”’ to varieties and forms and described a large number of additional infraspecific variants. Towards the end of his life, he recognized a limited number of species and fewer varieties and forms, reducing many of the concepts that he himself had previously described. In his papers on Hevea, Baldwin failed to discuss in- frageneric classification. That he did not consider the available treatments as natural, however, may be inferred from several of his statements. An example is the following opinion: **. Ducke found a tree which he considered to be inter- 247 mediate between H. guianensis and H. lutea, and for this and comparable reasons he and Schultes have recently made /utea a variety of H. guianensis. . . . One might with almost as much reason render the genus unispecific’’. In another context, he wrote that, while he preferred to recognize *‘nine — or fewer — species. ..., “‘in nature and in various localities entities so intergrade that if one wishes . . . he could. . . reduce the genus to one species and consider it in terms of trinomials with many forms appended’’. Nor did Seibert discuss in great detail Mueller’s two Sec- tions Euhevea and Bisiphonia, except to state that he had “arrived at the conclusion that exact number of anthers is of little taxonomic significance within the limits of certain tenden- cles’’, pointing out that the number of anthers ‘‘may vary within the species and between flowers on the same tree.’ Ever since 1945, when, jointly with Ducke, I reduced Hevea lutea to varietal rank under H. guianensis, 1 have considered the classical infragenetic grouping of species to be both un- workable and unnatural. Il. HISTORICAL NOTES ON HEVEA MICROPHYLLA A natural classification of species into infrageneric groupings should rest, whenever possible, preferably on several different characters — for example: both floral and fruit characters — in which little or no integradation is discernable. Extensive field work on wild Hevea in the Amazon and examination of thousands of specimens in the major herbaria have convinced me that such differences exist and that they may be used as the basis of an infrageneric classification which, in my opinion, is natural, showing two rather widely divergent trends in evolu- tionary development of the genus. In 1905, Ule, who had spent a long period studying Hevea in numerous areas of the Amazon, described a most interesting species: Hevea microphylla, which he had collected in fruit in 1902 on the Ilha Xibaru, slightly downstream from the mouth of the Rio Branco on the Rio Negro in Brazil. Later exploration has shown that this species is endemic to the Rio Negro, from the middle to the upper course of the river. 248 In 1910. Pax treated the species as comprising two varieties: var. typica and var. major, on the basis of differences in size of the leaflets. Unfortunately, Hevea microphylla, which only with the greatest difficulty and misunderstanding could be confounded with any other species, was, until recently, confused with Hemsley’s H. minor, now considered to be a synonym of H. pauciflora var. coriacea. In 1906, Huber suggested that Hevea microphylla might be synonymous with H. minor, pointing out several characters in which the two concepts, as described, seemed to agree. He admitted, nonetheless, that there ap- peared to be differences in other characters, so he chose ‘‘to consider H. microphylla a distinct species for the present’. Identifying erroneously a flowering collection of H. micro- phylla made by Ducke (Ducke 7027) in the lower-middle Rio Negro as H. minor, Huber published an extended description of H. minor. In describing the flowers of the Ducke specimens, Huber indicated still that the two species appeared to be close allies, although he believed that flowers were still unknown for H. microphylla. In 1913, he yet maintained H. minor and H. microphylla as distinct, including both in his Series /nter- mediae as he had done previously — but intimating that further studies might make it necessary to remove H. micro- phylla and H. minor from Series Intermedia and, together with H. rigidifolia, to form a new Series for them. Ducke apparently accepted Huber’'s identification of his flowering collection (Ducke 7027) as Hevea minor. He had collected topotypical material of H. microphylla (Ducke HJBR23750) which agreed in all characters with his earlier collection (Ducke 7027). Consequently, he reduced H. micro- phylla to synonymy under H. minor in 1935, maintaining this position in 1946. In 1947. Baldwin indicated apparent acceptance of Ducke’s treatment of Hevea microphylla. In the same year, I studied the type material of Hevea minor collected on the upper Rio Negro near the confluence with the Casiquiare and herbarium specimens of the Ule and the Ducke collections of H. microphylla. It became apparent immediately that the two concepts were completely distinct and not in any 249 way closely allied. I published the results of my studies, indic- ating that H. microphylla is, indeed, the most unique species in the genus and that, in addition to morphological characters easily to separate it from all other species, there are. likewise, strong ecological differences setting H. microphylla apart from H. minor: periodically and deeply flooded forested river banks in the former: scrub-forest in sandy, almost permanently flooded caatingas in the latter. Seibert accepted my treatment of H. microphylla as distinct: and, in 1949, Ducke (in litt.) likewise followed my interpretation, although, in publishing his acceptance of it in 1950, he stated that it was ‘‘lamentable. because it would have been better, for true scientific purpose, if that change could have been avoided.”’ For several years following my article in 1947, I was able to carry out intensive plant exploration in the Rio Negro basin of Brazil, Colombia and Venezuela, studying abundant stands of Hevea microphylla. These studies substantiated the unique- ness of this species and led, in 1952. toa paper on its range and variability and an extended description of the concept. At that time, I wrote: There are so many differentiating characters of the first magnitude to be found exclusively in H. microphylla that we are forced to regard the concept as standing entirely alone with no close allies in the genus’’. In 1967, whilst on the Alpha-Helix Amazon Expedition, I was fortunate again to meet with extensive stands of H. microphylla, not too distant from the type locality. These studies intensified my belief that we were concerned here with a species that had gone off on an evolutionary tangent of its own and that it, therefore, merited some special recognition in any treatment of infrageneric clas- sifications of the genus. II]. THE UNIQUENESS OF HEVEA MICROPHYLLA Hevea microphylla stands quite alone in the genus. It is unique in several basic characters — characters in both the flower and fruit and which are so distinct that there appear to be no intermediates. The pistillate flowers of Hevea microphylla differ markedly from those of all other species in having a greatly swollen torus 250 which is conspicuous not in the flowers but also at the base of the fruit. A torus is present, of course, in the pistillate flowers of all species but, except for Hevea pauciflora, it 1s so incon- spicuous as to be for all practical purposes of taxonomic use essentially non-existent. In H. pauciflora, it is sufficiently pro- nounced as to be easily visible, but it in no way approaches the size and conspicuousness of that of H. microphylla, nor is it obvious at the base of the ripened fruit. The fruit is unique in being pyramidal, triangular in cross section, conspicuously keeled and with a long-acute apex. The carpel walls are thin and leathery, made up of a thick-papery pericarp and an excessively thin, coriaceous endocarp. In all other species, the capsule is subglobose, ovoid or ellipsoid, trigastic or round in cross section, and emarginate, with a rounded or slightly mucronate tip. The carpel walls are thick and ligneous, made up of a more or less fleshy pericarp and a heavy, thick, woody endocarp. This unique structure of the capsule of Hevea microphylla is strongly reflected in the method of seed dissemination. In H. microphylla, the capsule dehisces slowly, not explosively, and the valves open gradually, twisting as they dry out, and adhere to the receptacle long after dehiscence. The seeds gently drop directly from the capsule and are not propelled violently a great distance fron the tree. In all other species, the capsule opens explosively, usually sending the seeds in several directions far beyond the area beneath the crown of the tree. The heavy, ligneous valves contort only slightly, if at all, and fall to the ground at the moment the capsule bursts open and frees the seeds. Only in Hevea Spruceana are the heavily ligneous val- ves persistent and, although the capsule does open explo- sively, the seeds are not propelled so far as in other species, primarily because of their greater size and weight. IV. INFRAGENERIC CLASSIFICATION Hevea Aublet, Hist. Pl. Guian. Frang. 2 (1775) 871. Subgenus Hevea Typus: Hevea guianensis Aublet Flos pistillatus toro valde inconspicuo, capsula matura haud 2a) manifesto. Capsula subglobosa, ovoidea vel ellipsoidea (nun- quam pyramidalis), transversaliter trigastra vel circularis, ecarinata, apice rotundata vel parum mucronata, eruptione dehiscens, semina ab arbore distante propullulans. Valvae crassae, pericarpio plus minusve carnoso atque epicarpio grosso, denso, lignoso, siccitate non convolutae, usualiter non perdurantes. The following species and varieties, as the genus is now understood, belong to this subgenus. Hevea Benthaminana Mueller Argoviensis in Linnaea 34 (1865) 204. Hevea brasiliensis (Willd. ex A. Juss.) Mueller Argoviensis loc. cit. 204. Hevea camporum Ducke in Arch. Jard. Bot. Rio Jan. 4 (1925) 111. Hevea guianensis Aublet loc. cit. 871. Hevea guianensis Aublet var. lutea (Spr. ex Benth.) Ducke et R.E.Schultes in Caldasia 3 (1945) 249. Hevea guianensis Aublet var. marginata (Ducke) Ducke loc. cit. 6 (1933) SI. Hevea nitida Martius ex Mueller Argoviensis in Martius FI. Bras. 11, pt. 2 (1874) 301. Hevea nitida Martius ex Mueller Argoviensis var. toxico- dendroides R.E Schultes et Vinton) R.E. Schultes in Bot. Mus. Leafl., Harvard Univ. 13 (1947) 11. Hevea pauciflora (Spr. ex Benth.) Mueller Argoviensis in Linnaea 34 (1865) 203. Hevea pauciflora (Spr. ex Benth.) Mueller Argoviensis var. coriacea (Ducke) Ducke in Arch. Inst. Biol. Veg. Rio Jan. ZXtos3) 200. Hevea rigidifolia (Spr. ex Benth.) Mueller Argoviensis loc. cit. 203. Hevea Spruceana (Benth.) Mueller Argoviensis loc. cit. 204. Subgenus Microphyllae R. E. Schultes subgen. nov. Typus: Hevea microphylla Ule Flos pistillatus toro manifeste incrassato capsulae maturae ad basim persistente, conspicuoque. Capsula pyramidalis, transversaliter triangularis, conspicue carinata, apice longe oe acuta, lente paulatimque (non eruptione) dehiscens, semina directe sub arbore cadens. Valvae tenues, coriaceae, pericar- pio crassipapyraceo atque endocarpio tenuissimo char- taceoque, siccitate valde convolutae, perdurantes. The only species in this subgenus is Hevea microphylla Ule in Engler Bot. Jahrb. 35 (1905) 669. EXPLANATION OF PLATE 51 1 and 2. habit. 3, leaf showing departure from normal shape. 4, valves of capsule showing mode of dehiscence. 5, seed. 6, pistillate bud, showing terminal spiralling. 7, staminate bud. 8, staminate flower with calyx removed. 9, pistillate flower with calyx removed, showing large torus. Drawn by ELMER W. SMITH 253 PLATE Sl] HEVEA Aylla - PHACTOP dle —N PLATE 52 H. microphylla H. Spruceana H. guianensis ) : 8 S ye y : 5S * z é g a =A oo re) : : E & ¢ oc H.nitida H. Benthamiana ee Plate 52. Comparison of the pistillate flowers of the nine known species of Hevea. Drawn by E. W. Smith. 255 ‘Sldodau FIVITaY WIFNZANAA — VIGWOTOD — TIZVUE OWSIN OLY 3JHL JO NISVd VTTAHEOUOIN VAAGH AONVaA NMONX aHL A@ Q3IVOIGNI SalLiTvo0T sos f ‘aqvw / ,-- N3aa AAVH SNOLLOAZTIOD / HOIHM WOUd SALLIIVDOIT e@ / ra ‘ALITVOOT AdAL OW Po / :Aau / st e Z [na Wr ppapy N PLATE 54 x e ( / —- ccm viatros } ( / , 08 ee = . paAoy Jo abuvry wate poaoH snuab oy3 fo sbuvy ——~ Sg OE + ? a Le 5 ; c , 4 / ; WNW INS wens | i votes Pas HONG Bh ees ) wee Fy } 3 _ / ? “ey Be paras woiwtse nz aNGABA VITAHdDOUDIW W3ASH omw WAASH ia? JO NOLLNSIYNLSIG IWUNLWN 4 257 BOTANICAL MUSEUM LEAFLETS VOLUME 25, No. 9 NOVEMBER 30, 1977 DE PLANTIS TOXICARIIS E MUNDO NOVO TROPICALE COMMENTATIONES XVII Virola as an oral hallucinogen among the Boras of Peru* RICHARD EVANS SCHULTES TONY SWAIN AND TIMOTHY C. PLOWMAN The hallucinogenic use of Virola bark was first reported from Amazonian Colombia in 1954 (Schultes, 1954), and preparation of a psychoactive snuff from a red, resin-like substance in the bark was described. Later investigations in Colombia and Brazil extended our knowledge of the use and methods of preparing this snuff in a variety of tribes (Biocca, 1966: Schultes & Holmstedt, 1968; Prance, 1970, 1972). In 1969, the custom of orally ingesting a paste made from the bark of Virola was first reported from information received from a Witoto Indian resident in the Leticia region of Colombia (Schultes, 1969). These natives no longer use the drug, but they remember its preparation by the older generations. Later, in April 1970, more intensive field studies in the Rio Karaparana area of Amazonian Colombia, where Witoto medicine men still employ Virola in witchcraft and medicine, clarified several points of uncertainty concerning the prepara- tion and use of the orally administered resin. The method followed by these Karaparana Witotos in making the thick paste from which small balls or pellets are shaped was fully described in 1976 (Schultes & Swain, 1976). It was pointed out that even in the ‘‘one limited area of the Witoto country along the Karaparana, the preparation of orally administered Virola resin varies appreciably from one Indian village to another and *The field work represented in this contribution was carried out as part of Phase VII of the Alpha Helix Amazon Expedition 1976-1977 and was funded by the National Science Foundation, Grant No. OCE 76-80874. 209 likewise in accord with the way in which the drug is to be utilized.” In April-May 1977, while on Phase VII of the Alpha-Helix Amazon Expedition, 1976-1977, we had an opportunity to carry out ethnotoxicological investigations in the region of Pebas in Amazonian Peru, where the numerically most impor- tant Indians belong to the Witoto and Bora tribes. Many of these Indians, originally from the Colombian region of the Rios Karaparana and Igaraparana, were transplanted to the Pebas region in the 1930’s. Substantial populations still live in the Karaparana-Igaraparana region under Colombian jurisdiction, and there is still some contact between the several groups. Many of the Witotos in the Pebas area are now so accultu- rated that even the older men — although some know the kinds of Virola once employed for the drug — no longer are familiar with the relatively simple methods of elaborating the pellets for hallucinogenic use. The Boras, on the other hand, are some- what less acculturated and conserve many of their older tribal customs, notwithstanding the inroads of Western religious and civil influences. We were able to witness on several occasions the prepara- tion of the Viro/a paste amongst a group of Boras living in Brillo Nuevo on the Rio Yaguasyacu, anaffluent of the Rio Ampiyacu which, in turn, empties into the Amazonas at the town of Pebas. These Boras no longer take Virola, or cumala, as it is commonly known in Peru, for hallucinogenic purposes of witchcraft, but older members of the group still remember how their elders prepared and used the drug. Knowledge of the methods of preparation of the product has been handed down even to the younger generation. What has apparently often been forgotten, we found, is which of the sundry species of Viro/a in the forests of the re- gion were chosen for their psychoactivity and which were es- chewed. Consequently, we had the Indians prepare paste from all of the species available and later sorted them out chemically in the laboratory: some containing the active tryptamines, others lacking these indolic compounds. A phytochemical summary of these analyses will be the subject of a later paper. We are here interested primarily in outlining the methods em- 260 ployed in the elaboration of the paste anda comparison of these methods with those followed by the Witoto Indians of the same original geographic area. The Karaparana Witotos select their trees by slashing a small strip of bark and tasting, sniffing and feeling the cambial layer, discarding trees which did not meet the right criteria: an ample cambial layer, bitter to taste and with a musty odour. That this custom is part of an old tradition was confirmed by an elderly Bora from the village of Tierra Firme on the Rio Ampiyacu who selected several Virolas and classified them according to their potency (which later chemical analysis proved to be correct): he had, in effect, chosen the species with the highest concen- tration of tryptamines as ascertained by his tasting and smelling the cut bark. It is obvious from our field observations in 1970 (Schultes & Swain, 1976) and 1977 that it is not the trees producing the most red resin that give the best preparation for inebriating effects or even contain the highest amount of tryptamines. Indeed, the Virola tree which produced the most copious amount of resin proved to contain no tryptamines either in the phloem or in the resin itself, although small amounts were found in the bark. Among the Boras, the first step consists in stripping bark from the trunk of the tree and carrying it back to the house for immediate processing. If the tree is standing in water (during the season of high water), the bark is stripped directly from the standing tree from a dugout canoe; if the tree grows on high land, the tree is usually felled for stripping. Strips of bark approximately two and a half feet long are cut from the tree with a machete, usually — and undoubtedly only for conveni- ence — from the lower four to eight feet of the trunk. If an unusually large amount of paste is to be prepared, bark is taken from other parts of the main trunk. This primary operation stands in sharp contrast to the Witoto method employed on the Karaparana where, in the forest, the shiny cambial layer left of the inner surface of the strips of bark and that are still adhering to the decorticated trunk is rasped off with the back of a machete, and the raspings are gathered carefully in a gourd for processing in the house (Schultes & Swain, 1976). 261 Upon returning, the Boras chip the hard, brittle outermost layer of bark from each strip, leaving only the thick, softer layer of the inner bark and phloem, called rem’-bee-ho-o in Bora. This cortical layer, now turned reddish brown with a congealed, oxidized “‘resin’’, is pounded on a log with a wooden mallet, until it is quite shredded. Cut into short pieces of convenient length, these shredded sections of bark are placed in a pot of a two-foot diameter with three or four inches of water at air temperature. They are allowed to soak for one half to three quarters of an hour with occasional kneading. The water soon becomes a chocolate brown colour. When the colour of the water is sufficiently deep, the pot Is slowly brought to a boil and, with the pieces of bark still in the liquid, it is boiled vigourously for about an hour, when the shredded bark is taken out, and, after the liquid is squeezed out back into the pot, it is discarded. The liquid is then boiled with almost constant stirring for another forty-five minutes to an hour, until a richly chocolate-coloured syrup remains. Stirring must be constant and careful towards the very end of the evaporation, so that a thick, sticky but homogeneous paste is left. The Boras call this paste ko’do. We would venture to assume that the Witoto technique is more efficient than that followed by the Boras. The nearly colourless liquid which rapidly turns reddish or brownish and which, for lack of a better term, we have called ‘‘resin’’. is present only in or near the cambial layer, not in the outer layers of bast or phloem. It is clear that the active principles them- selves — the tryptamines — occur mainly in the cambial sap and that boiling of the cambial tissue coagulates proteins and perhaps polysaccharides. When only this delicate cambial tis- sue is present, as in the Witoto method, it is obvious that such coagulation must proceed more efficiently than when, as in the Bora method, most of the material — consisting of shredded outer phloem — is relatively inert. The paste may be ingested directly without any further elab- oration. If the drug is to be kept for later use, however, the paste is made into small balls or pellets and rolled in a white or greyish powder referred to as ‘‘salt’’ or, in Bora, fi’-meh or 262 oo'me. This ‘‘salt’’ is the residue from the filtrate of ashes of one of two plants: one, known as pee-ve-ee-pa-a in Bora, an epiphytic species of the cyclanthaceous genus Carludovica , the leaves and stump of which are burned; the other a large palm, Scheelea sp., the leaves of which are reduced to ashes. The ashes of these plants are placed in a funnel made of a pliable piece of bark. Hot water is slowly poured into the funnel, passing through the ashes and dripping out into a recep- tacle placed under the funnel. The filtrate is then evaporated by heating, leaving the solid powder or “‘salt”’. Another contrast with the Karaparana Witoto preparation of Virola pellets lies in the much larger number of plants em- ployed by the Witotos as sources of the “‘salt”’ for coating the paste. It is of interest to note in passing that a ‘‘salt’’ from ashes of the leaves and stems of alow palm of the genus Chelyocarpus 1s similarly prepared by the Boras on this region for mixing with the thick syrup of tobacco, known in Bora as am-pl'-ree, applied to the tongue frequently during the use of the powdered coca, a narcotic characteristic of the region. The same tobacco preparation — with ‘‘salts’’ prepared from other plants — Is made by the Witotos of the Rios Karaparana and Igaraparana (Schultes, 1945). The Bora group at Brillo Nuevo recognized several species of Virola but pointed out as the “‘best’’ tree a species that occurs abundantly along the banks of the Rio Yaguasyacu. This ‘‘best’’ tree has been identified as Virola elongata (Benth.) Warburg (Plowman, Schultes et Tovar 7263). It is a stout, columnar tree up to 75 or 80 feet tall, with a diameter of two to two and a half feet, standing in deep water during most of the rainy season. The crown is not extensive for a tree of such height. The leaves are narrowly lanceolate-elliptic, greyish brown on the nether surface, bright green above: their consis- tency is firmly chartaceous. The bark ts hard, externally greyish black, light reddish brown within, about one quarter of an inch thick; when stripped from the trunk for use, an almost colourless resin-like liquid accumulates on the innermost sur- face and rapidly — within four or five minutes, some times even 263 sooner — becomes a rich brownish red. The local name of this tree in Spanish is cumala blanca; the Boras call it ko-de-ko, apparently a generic term for Virola. Several other species of Virola were indicated by these Boras, and paste was prepared from them. It was obvious that there was some confusion as a result of discontinuation of the native use of Virola in witchcraft. The species indicated were Virola surinamensis (Rol.) Warburg (cumala colorada) (Plowman, Schultes et Tovar 7260) and V. loretensis A. C. Smith (Plowman, Schultes et Tovar 7259). At the Bora town of Tierra Firme on the Rio Ampiyacu, the species indicated as the basis for the inebriating prepara- tion were: Virola Pavonis (DC.) A.C. Smith (Plowman, Schultes et Tovar 7091) and V. elongata (Plowman, Schultes et Tovar 7092). The very ‘‘strongest’’, these natives indicated, is represented by Plowman, Schultes et Tovar 7094, a sterile collection which we cannot identify and which may be an undescribed species of Virola. At Tierra Firme, Virola calo- phylloidea Markgraf (Plowman, Schultes et Tovar 7093) and Osteophloeum platyspermum (DC.) Warburg (Plowman, Schultes et Tovar 7095) were definitely indicated as ‘‘cumalas’”’ which are not employed in the elaboration of the narcotic paste. Amongst the Witotos, who live in the relatively new settle- ment of Puca Urquillo, near the mouth of the Rio Ampiyacu, on the other hand, a medicine man pointed out two species of Virola as possible sources of the hallucinogenic drug, even though he no longer knew the method of preparing the paste. These two species are: Virola elongata, of the forest well above flood level (Plowman, Tovar et Schultes 6920, 6595), and V. surinamensis, common along the deeply flooded banks of rivers and creeks as well as on higher level, known as cumala (Plowman, Tovar et Schultes 6688, 6920); the Witoto name for both kinds is vo0-koo’-na. We believe that our informant was confused, and, that actually, the Witotos formerly used only Virola elongata. Very significantly, these Witotos pointed out /ryanthera macrophylla (Benth.) Warburg (Plowman, Schultes et Tovar 69/9) as also a source of the narcotic paste. This report repre- 264 sents the first involving the related genus /ryanthera as an hallucinogenic group of plants. Voucher specimens of the Virolas cited above have been added to the Economic Herbarium of Oakes Ames in the Bo- tanical Museum of Harvard University, the herbarium in the Museo de Historica Natural ‘Javier Prado” in Lima, Peru, and the herbarium of the Instituto Nacional de Pesquizas da Amazonia in Manaos, Brazil. The material was determined by Prof. Richard Evans Schultes, Dr. Timothy C. Plowman and Dr. William Rodrigues. Species indicated in the foregoing discussion: Iryanthera macrophylla (Benth.) Warburg in Nov. Act. Nat. Cur. 68 (1897) 153. Osteophloem platyspermum (DC.) Warburg loc. cit. 162. Virola calophylloidea Markgraf in Fedde Repert. 19 (1923) 24. Virola elongata (Benth.) Warburg in Nov. Act. Nat. Cur. 68 (1897) 178. Virola loretensis A. C. Smith in Bull. Torrey Bot. Cl. 60 (1933) 95. Virola Pavonis (DC.) A.C. Smith in Brittonia 2 (1938) 504. Virola surinamensis (Rol.) Warburg in Nov. Act. Nat. Cur. 68 (1897) 208. REFERENCES Agurell, S., B. Holmstedt, J.-E. Lindgren and R. E. Schultes ** Alkaloids in certain species of Virola and other South American plants of ethnophar- macologic interest’’ in Acta Chem. Scand. 23 (1969) 903-916. Biocca, E. Viaggi tra gli indi alto Rio Negro — alto Orinoco 2(1966) 235-252. Consiglio Nacionale delle Ricerche, Rome. Holmstedt, B. and J.-E. Lindgren ‘‘Chemical constituents and pharmacology of South American snuffs”’ in D. Efron[Ed.] Ethnopharmacologic Search for Psychoactive Drugs, Public Health Serv. Publ. No. 1645 (1967) 339- 373. U.S. Gov't. Printing Off., Washington D.C. Koch-Griinberg, R. Vom Roraima zum Orinoco . . . 3 (1923) 368. Strecker und Schroder, Stuttgart. Prance, G. T. ‘‘Notes onthe use of plant hallucinogens in Amazonian Brazil” Econ. Bot. 24 (1970) 62-68. Prance, G. T. ‘Ethnobotanical notes from Amazonian Brazil’ Econ. Bot. 26 (1972) 221-237. Schultes, R. E. ‘‘El uso del tabaco entre los huitotos’’ in Agric. Trop. 1, No. 9 (1945) 19-22. Schultes, R. E. ‘‘A new narcotic snuff from the northwest Amazon” in Bot. Mus. Leafl. Harvard Univ. 16 (1954) 241-260. Schultes, R. E. **De plantis toxicariis e Mundo Novo tropicale commen- 265 tationes IV. Virola as an orally administered hallucinogen” in Bot. Mus. Leafl., Harvard Univ. 22 (1969) 133-164. Schultes, R. E. and A. Hofmann The Botany and Chemistry of Hallucino- gens. Charles C. Thomas, Springfield, Ill. (1973). Schultes, R. E. and B. Holmstedt **De plantis toxicariis e Mundo Novo tropicale commentationes II. The vegetal ingredients of the myristicace- ous snuffs of the northwest Amazon” in Rhodora 70 (1968) 113-160. Schultes, R. E. and T. Swain ** De plantis toxicariis e Mundo Novo tropicale commentationes XIII. Further notes on Virola as an orally administered hallucinogen” in Journ. Psyched. Drugs 8 (1976) 317-324. PLATE 35 AREA INHABITED BY THE 4 WITOTOS , BORAS “n° MUINANES wl 727% mabe 267 PLATE 56 Plate 56. Bora Indian stripping bark from trunk of Virola elongata for preparation of hallucinogenic paste. Brillo Nuevo, Rio Yaguasyacu, Loreto, Peru. Photograph: R. E. Schultes. 268 PLATE 57 Plate 57. Bora Indians pounding strips of Virola bark to separate the inner portion for boiling. Brillo Nuevo, Rio Yaguasyacu, Loreto, Peru. Photo- graph: R. E. Schultes. 269 PLATE 58 Plate S58. Inner part of Virola bark stripped from whole bark and macerated. ready for boiling. Brillo Nuevo, Rio Yaguasyacu, Loreto, Peru. Photograph: R. E. Schultes. 270 PLATE-39 Plate 59. Appearance of Virola bark after one hour of boiling and stirring in Bora method of preparation of hallucinogenic paste. Brillo Nuevo, Rio Yaguasyacu, Loreto, Peru. Photograph: R. E. Schultes. pa 8 PLATE 60 Plate 60. Apparatus used by Bora Indians in preparing the ‘‘salt’’ or filtrate of ashes for coating pills of Virola paste. Brillo, Nuevo, Rio Yaguasyacu, Loreto, Peru. Photograph: R. E. Schultes. ra BOTANICAL MUSEUM LEAFLETS HARVARD UNIVERSITY VoL. 25, No. 10 DE PLANTIS TOXICARIIS E MUNDO NOVO TROPICALE COMMENTATIONES XVIII Phytochemical examination of Spruce’s ethnobotanical collection of Anadenanthera peregrina. Richard Evans Schultes! Jan-Erik Lindgren? and Bo Holmstedt? Laurent Rivier? * One of the classical hallucinogens of the Americas is the snuff prepared from beans of the leguminous tree Anadenanth- era peregrina (L.)Speg., better known in the literature by its former name Piptadenia peregrina (L.)Benth. (1). Long known from the Orinoco River basin of Colombia and Venezuela, this psychoactive drug has been mentioned by virtually all of the early scientific explorers of the area. In 1916, it was identified by Safford as the source of the enigmatic cohoba, the narcotic snuff of the West Indies, the use and effects of which were seen among the Taino Indians of His- paniola by early Spanish explorers in 1496 (2). While the drug is no longer employed anywhere in the Carib- bean islands, the extent of the use of Anadenanthera peregrina has still not been clearly defined. It may be that, in isolated localities in the southern part of the Amazon Valley, the tree was until recently the source of a snuff. There is circumstantial evidence, too, that the very closely allied Anadenanthera co- ‘Botanical Museum of Harvard University, Cambridge, Massachusetts. 2Karolinska Institutet, Department of Toxicology, Swedish Medical Research Coun- cil, Stockholm, Sweden. 3Permanent address: Institute of Plant Biology and Physiology of the University, Lausanne, Switzerland. $20.00 a year, net, postpaid. Orders should be directed to Secretary of Publications at the above address. Application to mail at Second-Class Postage Rates is pending in Boston, Mass. lubrina was employed in preparing an intoxicating snuff known as céhil or huilca, used in former times in parts of northern Argentina, Paraguay and possibly in Bolivia and Peru (3). Our earliest botanical knowledge of niopo or yopo — as the snuff is called in the Orinoco — goes back to 1801, when von Humboldt and Bonpland encountered its use in Colombia and Venezuela (4). Kunth reported briefly on their observations: ‘*Ex seminibus tritis calci vivae admixtis fit tabacum nobile quo Indi Otomacos et Guajibos utuuntur.’” Humboldt iden- tified the source as Piptadenia Niopo, which he believed repre- sented the same species as Willdenow’s /nga Niopo. Hum- boldt, like the earlier explorer of the Orinoco, Padre Gumilla, erroneously believed that the intoxicating effects of the snuff could be attributed to the alkaline admixture and not to the seeds employed in elaborating the powder. The next major botanical encounter with the drug was that of Richard Spruce, who met with its use in June, 1854 amongst the Guahibos of the upper Orinoco (5). Spruce wrote that his ‘*specimens of the leaves, flowers and fruit agree so well with Kunth’s description of Acacia Niopo that | cannot doubt their being the same species; especially as I have traced the tree all the way from the Amazon to the Orinoco, and found it everywhere identical.’"’ An important point in Spruce’s meticulous observation of the preparation of the snuff, how- ever, is his statement that ‘‘there is no admixture of quicklime.”’ Spruce found ‘*. . .a wandering horde of Guahibo Indians .. encamped on the svannas of Maypures [on the Orinoco] and .. . an old man grinding Niopo seeds .. . . The seeds, being first roasted, are powdered on a wooden platter, nearly the shape of a watch-glass, but rather longer than broad (914 inches by 8 inches). It is held on the knee by a broad, thin handle, which is grasped in the left hand, while the fingers of the right hold a small spatula or pestle of the hard wood of the Palo de arco (Tecoma sp.) with which the seeds are crushed ... . For taking the snuff, they use an apparatus made of the leg-bones of herons .. . in the shape of the letter Y, or something like a tuning-fork, and the two upper tubes are tipped with small black perforated knobs (the endocarps of a 274 palm). The lower tube being inserted in the snuff-box and the knobs in the nostrils, the snuff is forcibly inhaled, with the effect of thoroughly narcotising a novice or indeed a practiced hand, if taken in sufficient quantity ....°> The apparatus which Spruce described and which he purchased at Maypures on the Orinoco may still be seen at the Royal Botanic Gardens at Kew. An illustration of this parephernalia is herewith pub- lished. There are also at Kew — in the Economic Botany Museum — specimens of the pods of Anadenanthera pereg- rina which Spruce collected in 1854 on the Colombo- Venezuelan border at the Savannahs of Maypures. These pods were purchased from an old Guahibo Indian who was grinding the seeds for preparation of the snuff. In our desire to analyze as many specimens of seeds of this species from as many localities as possible, we expressed to the authorities at Kew our interest in submitting some of Spruce’s 120-year old material to modern chemical examination. We owe a debt of gratitude to the former director, Dr. John Heslop-Harrison, the former Keeper, Dr. J. P. M. Brenan, Dr. Tony Swain, formerly director of the Biochemical Laboratory and Miss Rosemary Angel of the Economic Botany Museum for finding and making available to us the necessary pods and seeds. The analytic study is detailed below. We were encouraged to examine this important material collected by Spruce for several reasons (6,7). First: we wanted to compare its analysis with that of very recently collected material. Second: Spruce was far ahead of the customs of botanical explorers of his time in being willing to collect material of medicinal and narcotic plants for chemical analysis. Third: we had been successful in analyzing material of the hallucinogen Banisteriopsis Caapi of the Malpighiaceae, col- lected by Spruce on the Rio Uaupés of Brazil in 1852 (8). This material was examined in April 1968 in the Karolinska In- stitutet in Stockholm, 114 years after its collection. The yield of alkaloids was 0.4% as against 0.5% for a recently collected specimen of the same species. The alkaloid content of Spruce’s material consisted exclusively of harmine, as contrasted with aio harmine, harmaline and tetrahydroharmine, as well as two minor constituents in the modern material. As we did with the Spruce material of Banisteriopsis, the examination of the seeds of the collection Spruce 119 of Anadenanthera peregrina was compared with the analysis of similar freshly collected material. For several years, we have been studying a colony of beautiful trees of this species — obviously planted, perhaps some 40 years ago — in Barrio St. Just of Carolina, near San Juan, Puerto Rico. This colony grows ona hill immediately behind the El Comandante horse- racing track. Our most recent botanical studies on the colony were made in the month of December 1974, when the pods are still immature. Mature pods for the present analyses were also collected from the same colony of trees by Dr. Thomas Schubert and Mr. José Zambrana of the United States Depart- ment of Agriculture Forest Service, Institute of Tropical Forestry, Rio Piedras, Puerto Rico on March 13, 1975, when the pods had fully ripened. All of our collections from the Puerto Rican site are depo- sited in the Economic Herbarium of Oakes Ames in the Har- vard Botanical Museum. PUERTO RICO: La Carolina, Barrio St. Just, adjacent to Hipodromo. ‘‘Tree 60 feet tall. Pods brownish. Cojoba.’’ De- cember 8, 1970. R. E. Schultes 26091. — Same locality and date. *“‘Seedlings under tree of collection 2609/.' R. E. Schultes 26091 A. — Same locality. *‘Tree in grove on hillside. Height 70 feet. Pods green-brown, ripening black. Bark with large conical spines.”’ December 13, 1972. R. E. Schultes 26363. — Same locality and date. R. E. Schultes 26364. — Same locality. ““Tree 45 feet tall. Secondary forest. Cork black.’ December 12, 1974. R. E. Schultes, S. von R. Altschul et B. Holmstedt s.n. CHEMICAL ANALYSIS MATERIAL AND METHODs: The following botanical mate- rials will be referred to as below: seeds ‘‘December 1972’’— for immature seeds collected 276 in Puerto Rico. Voucher specimen: R. E. Schultes 26363, December 13, 1972; and seeds **March 1975°’— for mature seeds collected from the same colony of trees in Puerto Rico. Voucher speci- men: R. E. Schultes, S. von R. Altschul et B. Holmstedt s.n., December 12, 1974. REFERENCE SUBSTANCES: All reference substances have been previously described (9). ABBREVIATIONS: DMT N,N-dimethyltryptamine 5-MeO-DMT 5-methoxy-N,N-dimethyltryptamine 5-MeO-MMT 5-methoxy-N,N-monomethyltriptamine 5-OH-DMT 5-hydroxy-N,N-dimethyltryptamine or bufotenine MTHC 2-methyl-1 ,2 ,3,4-tetrahydro-S-carboline 6-MeO-THC 2-methyl-6-methoxy-1 ,2,3,4-tetrahy dro-6-carboline 6-MeO-DMTHC 1 ,2-dimethyl-6-methoxy-1 ,2,3,4-tetrahydro-f- -carboline ISOLATION OF THE ALKALOIpDS: The vegetal material was ground and extracted according to a procedure first used by Fish et al. (10). Gas CHROMATOGRAPHY (GC): The gas chromatographic analyses were performed with a Varian Model 2100 GC equip- ped with a flame ionization detector system. A 180 x 0.2 cm (i.d.) glass column was silanized and packed with 3% OV-I7 coated on Gas Chrom Q, 100-120 mesh (Applied Science Laboratories, State Coll., Pa.). The separations were obtained at a column temperature of 190°C with a nitrogen carrier gas flow rate of 30 ml per min. The vaporizer and the detector temperatures were 250°C and 300°C, respectively. The amounts of alkaloids were determined by peak heights using 5-hydroxy-N, N-dimethyltryptamine as a standard. GAS CHROMATOGRAPHY - MASS SPECTROMETRY (GC-MS): An LKB Model 9000 GC-MS (LKB-Produkter AB, Bromma, Sweden) was used to confirm the structure of the alkaloids. Separation was obtained ona 160 x 0.2 cm (i.d.) silanized glass column, packed with the same packing material as for the GC 277 analyses but maintained at 170°C. The flow rate of the helium carrier gas was 40 ml per min. The ionizing potential and the trap current were 70 eV and 60 wA, respectively. The ion source was kept at 250°C. Table 1. Gas chromatographic and mass spectrometric data for reference compounds. Compound® R, Mass spectrum® DMT oa 58 (base peak), 103, 105, 130, 143, 188 (M*) MTHC 4.3 78, 102, 115, 143 (base peak), 186 (M*) 5-MeO-DMT 5.8 58 (base peak), 103, 117. 160, 173, 218 (M*) 5-OH-DMT 8.4 58 (base peak), 103, 117, 146, 159, 204 (Mt) 4Ror abbreviations, see material and methods. bLKB 9000 GC-MS with helium as carrier gas on 3% OV-17 on Gas Chrom Qat 170 C. Ry= retention time in minutes. “Tonizing potential was 70 eV. m/e values of the major peaks are given. MASS FRAGMENTOGRAPHY (MF): In order to confirm the pres- ence or absence of minor alkaloids in the plant materials, the specific and sensitive method of mass fragmentography was used (11). The principles of the technique have already been described (12). The mass spectrometer was controlled by a PDP-12 computer system. The channels used were focussed carefully on the molecular ion of each compound of interest: m/e = 204 for 5S-OH-DMT; m/e = 186 for MTHC; m/e = 188 for DMT; and m/e = 218 for S-MeO-DMT. During another experi- ment m/e = 58 was chosen with two different sensitivities on two channels. 278 RESULTS The gas chromatographic trace of the chloroform-soluble bases from the seeds collected in 1854 by Spruce gave a single peak. Its mass spectrum was identical to that of S-OH-DMT. The mass fragmentogram of the same extract is in agreement with that result (Fig. 1). The extract of mature seeds freshly collected in Puerto Rico, seeds ‘‘ March 1975’’, showed several GC peaks. Beside 5-OH-DMT they have been identified by GC-MS and are DMT, MTHC and 5-MeO-DMT (Table 1). The mass fragmentographic recording confirms in a single run the presence of the four alkaloids (Fig. 2). The relative amount of the compounds in the plant material is given in Table 2. In the same table are listed the results of similar analyses of various plant parts of Anadenanthera peregrina, originating from the same colony of trees in Puerto Rico. Included in the table are also specimens of more or less well defined botanical or ethnological origin. DISCUSSION The finding of 5-OH-DMT as the only alkaloid in the Spruce material is significant for several reasons. First: it indicates that, with modern analytical tools, it is possible to detect and identify alkaloids in plant materials more than 100 years old. Second: identification of the botanical specimen is strengthened by the results of the chemical analyses, because the same compound has been found in both old and the freshly collected seeds. One previous analysis of seeds of Anadenanthera peregrina originating from Puerto Rico has shown the presence of 5-OH-DMT as the principal alkaloid (13). A sample of A. peregrina seeds collected in southern Venezuela contained 7.5% of SSOH-DMT (14, 15). Holmstedt and Lindgren (16) have reviewed the alkaloid composition of many specimens. DMT alone or together with 5-MeO-DMT has been isolated from Anadenanthera peregrina originating from Brazil (10). A similar composition was found in related species (9, 16). Table 2 illustrates the differences in alkaloid contents in various parts of Anadenanthera peregrina. The root contained 29 the highest amount of alkaloids. In this collection, 5-MeO- DMT was the predominant alkaloid in all plant parts, except the seeds, where DMT was found in the highest amount. How- ever, it should be mentioned here that the seeds were not fully ripe at the time of collection (seeds *‘December 1972”’). SAMPLE NO. : m WAITING PERIO SAMPLE PERIOD 1S ANADENANTHERA PEREGRINA PUERTO RICO CHANNEL MASS HEIGHT RET. TIME RATIO 1 204 869 8.409939 1 2 186 140 4.329992 - 161104 3 188 vee! 2-S3008 -818181 4 218 180 S.77000 - 207134 10@ 5-OH- DMT 80 _] ye DMT 60 _ a MTHC ees 20 _| 186 3 188 Z a re a 204 oO | \ ,218 0 T T T T ] 1 4 if 10 13 16 RETENTION TIME minutes Fig. 1. Mass fragmentogram (OV-17) of the alkaloidal fraction from the Anadenanthera peregrina seeds collected by Richard Spruce in 1854. 280 Since it may be assumed (in fact, examination of the seeds established) that the seeds collected by Spruce had matured, it was necessary for comparison to analyze similar material. Investigation of mature seeds (*‘March 1975”’) from the Puerto Rican locality, where immature seeds had been collected pre- SAMPLE NO. : 6 WAITING PERIOD 1 SAMPLE PERIOD 1S ANADENANTHERA PEREGRINA RICHARD SPRUCE CHANNEL MASS HEIGHT RET. TIME RATIO 1 204 430 8.47000 1 2 186 3 188 4 218 10@ 80 J 5-OH-DMT 60 40 _ 204 186 20 _— 188 =I na 218 ae Ts) Ww S T T T T 1 1 4+ 7 10 13 16 RETENTION TIME minutes Fig. 2. Mass fragmentogram (OV-17) of the alkaloidal fraction from the Anadenanthera peregrina seeds of Puerto Rico (seeds **March 1975"’) analysed in 1975). 281 Table 2. Distribution of the indole alkaloids. Voucher No. Species Part of Alkaloids Alkaloids %o the plant mg/100g dry plant R. Spruce A. peregrina Seeds 614 5-OH-DMT 100 119 o: Rio Negro, Brazil, 1854 R.E. Schultes, A. peregrina Seeds a) not 5-OH-DMT 80 S. von R. Altschul o: San Juan, Puerto Rico, determined DMT 19 and B. Holmstedt, 1975, “March 1975” (analysed 5-MeO-DMT 1 sin, num 1975) MTHC traces b) 3523 5-OH-DMT 100 (analysed 1977) R.E. Schultes A. peregrina Seeds 209 DMT 75 26363 o: San Juan, Puerto Rico, 5-MeO-DMT 19 1972, “March 1972” 5-OH-DMT 6 Seedlings 25 DMT 4 5-MeO-DMT 95 5-OH-DMT 1 Pods 13 DMT 8 without 5-MeO-DMT 91 seeds 5-OH-DMT 1 Leaves 107 DMT 12 5-MeO-DMT 88 Twigs 38 DMT 5 5-MeO-DMT 94 5-OH-DMT 1 Bark 410 DMT 5 5-MeO-DMT 95 Roots 699 DMT p. 5-MeO-DMT 97 5-OH-DMT 1 Anadenanthera presumably Seeds ] DMT 100 c: Biocca Cocco, 1963 o: Upper Orinoco El Platanal Machekototeri Anadenanthera Seeds 6 DMT 100 c: Bioccea Cocco, 1965 o: Upper Orinoco Rio Ocamo Anadenanthera Seeds 38 DMT 100 c: G. Seitz, 1965 Anadenanthera Seedlings 29) DMT 96 c: G. Seitz, 1965 5-MeO-DMT 4 R.E. Schultes A. peregrina Leaves 13 DMT 49 24625 o: Boa Vista, Brazil (9) 5-MeO-DMT 48 Bark 42 DMT 1 5-MeO-DMT 59 6-MeO-DMTHC 2 5-MeO-MMT 36 6-MeO-THC 2 Abbott Lab., 1948 Piptadenia peregrina Seeds 9 DMT 100 N2003C o: San Juan Puerto Rico, (16) Piptadenia peregrina Seedlings I DMT 100 c: J. Yde, 1964, H4685 Piptadenia, Tupari Seeds 13 DMT 15 c: Caspar, 1964 5-MeO-DMT 85 o: Guaporé, Brazil Schupfsnuff, Tupari Snuff 16 DMT 100 c: G. Baer, 1964 o: Brazil Yopo Snuff 16 DMT 100 c: L. Persson, 1966 o: R. Miriti-Parana Caqueta, Columbia collector origin c= viously, shows a different picture [mostly 5-OH-DMT and DMT, with less MTHC and 5-MeO-DMT (Fig. 2)]. Analysis of these seeds was done in August 1975, five months after collec- tion. No quantitation of alkaloid contents was performed at that time. We repeated the analysis of the same material two years later. At this time, the Puerto Rican seeds were no longer able to germinate. In the analysis of this material, the seeds (‘‘March 1975’) contained only 5-OH-DMT, with no trace of any of the other alkaloids found earlier. This fact might imply that the relative content of the various alkaloids upon storage follows with time a certain pattern. The seeds of the Puerto Rican material kept for two years and the 123 year old Spruce material thus contain the same alkaloid: 5-OH-DMT. Transformation of alkaloids during stor- age of botanical material is known to occur (17). CONCLUSIONS In yet another ethnobotanical collection made by Spruce more than a 100 years ago, it has been possible to identify alkaloidal material by the use of modern analytical techniques never dreamed of by this intrepid plant explorer. Of several alkaloids found in freshly collected reference material, only one remained in the Spruce collection: bufotenine (5-OH- DMT). Storage of freshly collected material for two years resulted in the disappearance of all alkaloids except 5-OH- DMT. This may raise speculation as to whether or not they were originally contained in the Spruce material. Our observa- tion stresses the importance of storage-time in addition to knowledge of plant part, soil, season and climatic conditions, when alkaloid analysis is carried out on seeds and on the snuffs prepared from them. ACKNOWLEDGMENTS Part of this work was supported by the Swedish Medical Research Council; the National Institutes of Health; the Bank of 283 Sweden Tercentenary Fund; the Wallenberg Foundation and by funds from the Karolinska Institutet. We wish to thank the Swiss National Research Foundation (Lausanne’s commis- sion) for a post-doctoral grant. Part of the research has been funded by the National Science Foundation. to 10. REFERENCES Altschul, S. von R. The genus Anadenanthera in Amerindian cultures. (1972). Botanical Museum, Harvard University, Cambridge, Mass. Safford, W. E. ‘Identity of cohoba, the narcotic snuff of ancient Haiti”’. Journ. Wash. Acad. Sci. 6 (1916) 548-562. Schultes, R. E. and A. Hofmann. The botany and chemistry of hal- lucinogens. (1973). Charles C. Thomas, Publisher, Springfield, III. von Humboldt, A. and A. Bonpland. Personal narrative of travels to the equinoctial regions of America. (Ed. and transl. T. Ross). (1852-53). Henry G. Bohn, London. Spruce, R. Notes of a botanist on the Amazon and Andes. (Ed. A. R. Wallace) 2 (1908) 426-630. Macmillan and Co., London. Schultes, R. E. ‘“‘Some impacts of Spruce’s Amazon explorations on modern phytochemical research’’. Rhodora 70 (1968) 313-339. Schultes, R. E. ‘‘The impact of Spruce’s Amazon explorations on modern phytochemical research’. Ciencia e Cultura 20 (1968) 37-49. Schultes, R. E., B. Holmstedt and J.-E. Lindgren. *‘De plantis to- xicariis e Mundo Novo tropicale commentationes III. Phytochemical examination of Spruce’s original collection of Banisteriopsis Caapi’’. Bot. Mus. Leafl., Harvard University, 22 (1969) 121-132. Agurell, S., B. Holmstedt, J.-E. Lindgren and R. E. Schultes. **Al- kaloids in certain species of Virola and other South American plants of ethnopharmacologic interest’. Acta Chemica Scand. 23. (1969) 903- 916. Fish, M. S., N. M. Johnson and E. C. Horning. **Piptadenia alkaloids. Indole bases of P. peregrina (L.) Benth. and related species”’. Journ. Am. Chem. Soc. 77 (1955) 5892-5895. Palmer, L. and B. Holmstedt. ‘‘Mass fragmentography: The use of the mass spectrometer as a selective and sensitive detector in gas chromatography’’. Science Tools, 22 (1975) 25-31; 35-39. Elkin, K., L. Pierrou, U. G. Ahlborg, B. Holmstedt and J.-E. Lindgren. ‘‘Computer-controlled mass fragmentography with digital signal proc- essing’’. Journ. Chromatogr. 81 (1973) 47-55. Stromberg, V. L. ‘‘The isolation of bufotenine from Piptadenia pereg- rina’’. Journ. Am. Chem. Soc. 76 (1954) 1707. Chagnon, N. A., P. Le Quesne and J. M. Cook. *‘Algunos aspectos de uso de drogas comercio y domesticacion de plantas entre los indigenas yanomamo de Venezuela y Brazil’’. Acta Cient. Venez. 21 (1970) 186- 193. 284 BF Chagnon, N. A., P. Le Quesne and J. M. Cook. ** Yanamam6o hallucino- gens: anthropological, botanical and chemical findings’’. Current An- thropology, 12 (1971) 72-74. Holmstedt, B. and J.-E. Lindgren. ‘“‘Chemical constituents and phar- macology of South American snuffs”’. In: Ethnopharmacologic search for psychoactive drugs. (Ed. D. Efron). U.S. Public Health Service Publ. No. 1645, (1967) 339-373. Waller, G. R. and E. K. Nowacki. Alkaloid biology and metabolism in plants. (1978). Plenum Publisher, New York. 285 PLATE 61 APPARATUS f Plate 61. Paraphernalia for preparing and taking yvopo snuff, collected on the Orinoco River by Richard Spruce. Photograph courtesy of the Royal Botanic Gardens, Kew. 286 PLATE 62 » li | ge ge ae I Mnyil iyi 4 8 5 FLY LA Wu a be Plate 62. Pods and beans of Anadenanthera peregrina collected on the Orinoco River in 1854 by Richard Spruce. Photograph courtesy of the Royal Botanic Gardens, Kew. 287 BotranicAL MuseEUM LEAFLETS vot. 25, No. 10 DecemBer, 1977 BRUNFELSIA IN ETHNOMEDICINE* Timothy Plowman** The genus Brunfelsia belongs to the alkaloid-rich family Solanaceae and usually is placed in the relatively advanced tribe Salpiglossideae. Brunfelsia is a medium-sized genus of about 42 species of small trees and shrubs: 22 species are confined to the West Indies; 20 species are found in tropical South America. Various species from South America have long been recog- nized by native peoples for their medicinal properties. Some of these plants are cultivated for use as household remedies with specific therapeutic effects. These effects have apparently been discovered independently by unrelated peoples in widely separated parts of the continent. Yet chemically and phar- macologically, species of Brunfelsia are still virtually unknown —a disturbing fact considering the many important drug plants in the Solanaceae which are commonly used in pharmacy to- day. At least five species of Brunfelsia are known to be of some medicinal importance. Other species of the genus are sus- pected of having pharmacological activity or of possessing alkaloids or other active constituents. My purpose in this ac- count is to review the literature on these plants and to present pertinent ethnobotanical data collected during my own field work and that of other workers, in order to rekindle the interest of chemists and medical researchers in this pharmacologically rich genus. 1. Brunfelsia uniflora (Pohl) D. Don The most important medicinal species of Brunfelsia is B. uniflora, the well known manacd root of the Brazilian phar- * Based in part upon ‘‘The South American Species of Brunfelsia (Solanaceae)’’, a doctoral dissertation presented at Harvard University, December, 1973. ** Present address: Botanical Museum of Harvard University, Oxford Street, Cam- bridge, Massachusetts 02138, U.S.A. 289 macopeia. This plant is often referred to, albeit incorrectly, by a later synonym, B. Hopeana (Hook.) Benth., particularly in horticultural and pharmaceutical literature. It has also been confused with two closely related species: B. australis Benth. and B. pilosa Plowman (Plowman, 1974). Brunfelsia uniflora is one of the most widely distributed species of the genus. It occurs throughout southeastern Brazil south to Sao Paulo, extending northward along the coast nearly to Belém do Para. Disjunct populations are found in the eastern Andes of southern Bolivia and northwestern Argentina, and in northern Venezuela. In its native Brazil, this plant was aboriginally known by a number of Tupi names. The most frequent is manacad (or its variant manacan), a word attributed to the most beautiful girl of the tribe and transferred to the most beautiful flower of the forest (von Marius, 1843). Throughout Brazil, the name /man- aca may be used for any species of Brunfelsia. However, in the pharmaceutical trade, the term ‘‘manaca root’ always refers to B. uniflora and will be used in this strict sense in this paper. Other vernacular names of Brunfelsia uniflora are related to its use in folk medicine. Cangamba and variants camgaba, cambamba, camganiba and cad-gamba mean the “‘tree of the gamba’’. Gamba is a species of opossum (Didelphis cancri- vora), known as mucura in Portuguese the odor of which the roots of B. uniflora are said to emit (Peckolt, 1909; Tastevin, 1922). Jeratacaca and variant jerataca loosely translate ‘snake bite remedy’, taken from the native name of the snake Mephitis suffocans (Tastevin, 1922). Umbura-puama 1s another name for manaca which means “medicine tree”’ (Peckolt, 1909). In addition to the indigenous Tup! names, smanacad also bears several Portuguese common names: mercurio vegetal (vegeta- ble mercury, referring to its antisyphilitic properties), mercurio dos pobres (poor man’s mercury), flor da Quaresma (Easter flower), flor de Natal, Santa Maria and boas noites (Peckolt, 1909), When the first Portuguese explorers arrived in Brazil, they found Brunfelsia uniflora in use by aboriginal Tupi payés or medicine men, employed both for healing and magical proper- 290 ties. An extract of the root was a constituent in arrow poisons (Peckolt, 1909). The first reference to manacad root appeared in the literature in 1648 in De Medicina Brasiliensi, an early materia medica written by Willem Piso. Piso was a Dutch physician who trav- eled in northeastern Brazil from 1637 to 1644 with the German physician Georg Marcgraf. Piso’s work provided a description of manacd root and its uses, as well as a line drawing, the first illustration of the genus Brunfelsia. He states that the scraped bark is a strong purgative, resembling scammony (Convolvulus scammoniae L.). Piso’s remarks and illustration were repeated by Marcgraf in his Historia Rerum Naturalium Brasiliae, pub- lished posthumously and bound together with the work of Piso. Manaca was first named scientifically Franciscea uniflora by Pohl in 1827. The correct combination Brunfelsia uniflora was made in 1829 by David Don, who recognized the similarity of Franciscea to the earlier genus Brunfelsia. In 1843, von Martius, an outstanding student of Brazilian medical botany, discussed at length the medicinal uses and pharmacological effects of manacd root. His observations were very thorough for the time and often copied by later authors. The substance of his account of manaca follows: ‘The whole plant, most of all the large root, stimulates the lymphatic system with great efficacy. It melts away the disease- producing parts and eliminates sweat and urine. It is very useful in syphilis and is called “vegetable mercury’ by some. The inner bark and all the herbaceous parts have a nauseating bitterness and are effective for fauces vellicantes. A small dose relaxes the body. A larger dose moves the bowels and the urine, produces abortion and expels the venom of snakebites. An excessive dose acts like a bitter poison... . « Among some tribes of Indians in the Amazon region, an extract of manacd is used in arrow poisons.” Further observations on the effects of manaca root appeared in 1871 in a work on toxic plants of Brazil by J.M. Caminhoa. Quoting several obscure authors, Caminhoa reported the fol- lowing effects not mentioned by von Martius: abundant saliva- tion, vertigo, general anesthesia, partial paralysis of the face, swollen tongue and turbid vision. He also mentions the great usefulness of the drug in treating rheumatism. 291 Baena, cited by Caminhoa, claimed that manacd was used by the Indians to produce “‘furious delirium and persistent insanity’ as well as “‘confusion of ideas, inconstant delirium and tremor’’. This is one of the few accounts reporting the use of manaca for narcotic or possibly hallucinogenic effects, in this case resembling belladonna intoxication. Another such account is found in a glossary of Tupi names of plants and animals (Tastevin, 1922): “One kind of manaca has the property of causing intoxication, blindness, and the retention of urine during the day; but after having drunk the infusion of the root or bark of this tree, a man is always happy in his hunting and fishing.” Unfortunately we do not know the specific identity of this kind of manaca. The roots of manaca used either fresh or dried and are considered to be the most effective part of the plant. All parts, however, are used medicinally in Brazil. The root is most often powdered or prepared as a fluid extract, of which a usual dose is 10 -30 minims (0.6 -1.8 cc.) three times daily. It is a powerful and energetic healing agent which has been used for many disorders. In recent times, its most general application has been against syphilis and rheumatism, and for its diuretic and diaphoretic properties. The known pharmacological effects and medicinal uses of manacad root are summarized as follows: SUMMARY OF PHARMACOLOGICAL EFFECTS OF MANACA ROOT (Brunfelsia uniflora) Diuretic (von Martius, 1843; Peckolt, 1909; Wren, 1956) Diaphoretic (von Martius, 1843: Brandt, 1895) Purgative (Piso, 1648; von Martius, 1843; Dragendorff, 1898; de Almeida Costa, 1935) Emetic (Dragendorff, 1898; Peckolt, 1909). . Alterative (von Martius, 1843; Peckolt, 1909; Wren, 1956) . Anesthetic (Caminhoa, 1871) . Abortifacient (von Martius, 1843; Brandl, 1895, de Almeida Costa, 1935). Emmenagogue (Peckolt, 1909; Le Cointe, 1947) Antirheumatic (Caminhoa, 1871; Dragendorff, 1898; Webb, 1948; Wren, 1956) 10. Antisyphilitic (von Martius, 1843; Kunkel, 1901; Peckolt, 1909; Webb, 1948) wn CoN AWS 292 11. Antiscrophular (Dragendorff, 1898) 12. Poisonous (von Martius, 1843; Brandl, 1895; Le Cointe, 1947; Webb, 1948) 13. Anti-inflammatory (Iyer et al., 1977) 14. Narcotic (Caminhoa, 1871) 15. Stimulates endocrine system (von Martius, 1843; Caminhoa, 1871; Brandl, 1895) 16. Stimulates lymphatic system (von Martius, 1843) 17. Lowers body temperature (Brandl, 1895) 18. Increases blood pressure and respiration (de Almeida Costa, 1935) 19. Produces parasthesia (Peckolt, 1909) 20. Produces muscular tremors and cramps (Brandl, 1895; de Almeida Costa, 1935) 21. Produces delirium, vertigo and clouded vision (Caminhoa, 1871) 22. Activates peristalsis (Brandl, 1895) The leaves of Brunfelsia uniflora are also employed medici- nally but only in the fresh state (Peckolt, 1909). They are considered to be less active pharmacologically than the roots. The leaves are most commonly used as an antidote for snake- bite. A tincture is prepared and given in frequent doses to the victim, and a poultice of the leaves is placed directly on the wound to ‘‘draw out the poison’’. Poultices are also employed for skin disorders such as eczema and syphilitic ulcers. The bark and young shoots of manacd are considered resolvent and, in high doses, emetic (Pereira, 1929). By means of ether, a perfume is extracted from the fragrant flowers (Correa, 1909). The first systematic investigations of manaca root began about 1880, when the drug stirred some interest among chemists and pharmacologists in Germany and in the United States. In 1880, J. L. Erwin attempted to determine the general classes of constituents in the root. He failed, however, to find any compound which could account for its potent effects. Brewer, in 1882, performed the first pharmacological studies with manacd by observing the effects of the fluid extract on cats and frogs and on himself. He concluded that manaca acts chiefly on the spinal chord by first stimulating, then abolishing the activity of the motor centers, with similar action in the respiratory center. All the glands were markedly stimulated, including salivary, gastric, intestinal, cutaneous, and the liver and kidneys. He found no effects on the brain or sense organs. Brewer’s self-experiment with manacd root furnishes us with arare, firsthand account of the effects on humans. Taking 293 the fluid extract on a full stomach, he experienced a feeling a restlessness followed by lassitude, a profuse sweating and an increase in amounts of saliva and urine. In 1884, Lenardson, a student of Dragendorff working in Dorpat (now Tartu in Estonia), discovered an alkaloid in man- acd root, the first to be isolated from the root. He named the compound manacine, which he characterized as an amor- phous, hygroscopic yellow powder with the empirical forumula C15H23Nq405 and a melting point of 115°. Manacine was sol- uble in water and alcohol but insoluble in benzene, ether and chloroform. It produced non-crystalline precipitates with sev- eral alkaloid precipitation agents. In addition to manacine, Lenardson found a fluorescent substance which he thought to be gelseminic acid. A second alkaloid franciscein was reported from manaca root in 1887 by Lascelles-Scott, but this was never substan- tiated by isolation and identification of the compound. The most complete study on manaca root during this period was conducted by Brandl in Germany (Brandl, 1895; Beckurts, 1895). He summarized Lenardson’s dissertation and presented the results of his own detailed chemical and pharmacological investigations. Brandl confirmed Lenardson’s discovery of manacine but claimed a different empirical formula C77H33N 70109 and melt- ing point of 125°. From the residue remaining after the alcoholic extraction of manacine, Brandl found an additional substance which he named manaceine. He characterized this constituent as an amorphous, white, highly refractive compound, soluble in water but insoluble in ether, chloroform and benzene. Brandl gave the empirical formula C,;5H25N 70g for man- aceine. When heated with water, manacine splits into man- aceine and a resinous, fluorescent substance which he consid- ered to be the aglycone esculetin (6, 7-dihydroxycoumarin). Although Brandl managed to isolate two alkaloids from the root, he never obtained a crystalline compound, nor was he able to characterize their structures. Schultes (1966) suggested that manacine is an ‘‘atropine- like’’ alkaloid. There is in fact no basis for this statement. 294 Manacine actually shows the opposite effects of atropine which inhibits rather than stimulates glandular secretions. At the present, the structural identity of manacine remains un- known. Brandl, following the work of Brewer, conducted phar- macological experiments on frogs, rabbits and guinea pigs, using the manacine and manaceine of his extractions. He found manacine to have an intense action, even in small doses. In mammals, it induced strong muscular tremors and epileptiform cramps, lowered temperature, followed by death due to re- spiratory paralysis. All the glands were strongly stimulated, as Brewer had observed earlier. Peristalsis was also increased. Glandular stimulation, but not peristalsis, was blocked by atropine. Frogs showed a general paralysis after an initial period of unrest, followed finally by heart arrest in diastole. With manaceine he observed a similar but less intense action. Peckolt (1909) repeated some of Brandl’s work on manaca. He found two products in the root, manacine and another alkaloid which he named brunfelsine, but he offered no further characterization of either compound. He stated that the seeds of another species, Brunfelsia brasiliensis (Spreng. ) Smith & Downs [reported as B. ramosissima (Pohl) Benth.], contain 1.14% brunfelsine but no manacine. Hoppe (1958) stated that the leaves and seeds of this species contain both manacine and brunfelsine. In its native Brazil, B. brasiliensis is considered poisonous according to data from herbarium specimens (V. Assis 142). Pammel (1911) listed two alkaloids for manaca root, man- acine and mandragorine. **Mandragorine”’ is a name given to an alkaloid isolated in 1889 from Mandragora officinarum L. This was later shown to be a mixture of I-hyoscyamine, 1-hyoscine (1-scopolamine) and a new alkaloid known as man- dragorine. This compound bears the empirical forumla C}5H}9O2N and forms a crystalline aurichloride with a melt- ing point of 124-126°. On hydrolysis, it yields tropic acid anda base resembling tropine (Henry, 1949; Manske & Holmes, 1950). There have, however, been no subsequent reports of tropane alkaloids in Brunfelsia species. More recently, the identity of the crystalline, blue-fluores- 29) cent substance in smanaca root has been identified (Mors & Ribeiro 1957). It had earlier been determined as gelseminic acid by Lenardson (1884) and as esculetin by Brandl (1895). This constituent is now known to be the aglycone scopoletin (6- methoxy-7-hydroxycoumarin) of the lactone glycoside scopo- lin. It is found in all parts of the plant and in several other species as well: Brunfelsia pauciflora (C. & S.) Benth. (re- ported as B. calycina var. macrantha (Lem.) Bailey & Raffill), B. brasiliensis (reported as B. ramosissima) and B. grandiflora D. Don. Machado de Campos (1964) found much smaller amounts of scopoletin in the seeds of B. uniflora and B. erandi- flora. This compound occurs in several other genera of Sol- anaceae including Atropa, Solanum and Lycopersicon. It may function as a regulator of growth processes in plants but is not known to be pharmacologically active in humans. With the exception of anatomical and pharmacognostic studies on manaca root (Hahmann, 1920; de Almeida Costa, 1935), only one recent paper has appeared on the nature and action of the drug. Iyer e¢ al. (1977) conducted hippocratic screening in rats of whole root and extracts of Brunfelsia uni- flora (reported as B. Hopeana). Administered intraperitoni- ally, the whole root showed the following dose-related symp- toms: decrease in spontaneous motor activitiy, irregular respi- ration, paralysis of hind and forelegs, analgesia, mixed convul- sions, hypersensitivity to sound, increase in pupil size and slight diuresis. These symptoms, indicating CNS depressant activity, were concentrated in a chloroform extract (**F’’), the lethal dose of which was about half that of the whole root. The authors also showed that this fraction has marked anti-inflam- matory activity when compared with phenylbutazone in reduc- ing Carrageenin-induced pedal edema in rats. These workers are Currently undertaking a detailed investigation of the com- ponents of the chloroform extract. Currently, manaca root is fully recognized in the Brazilian pharmacopeia and is considered a valuable remedy in that country where folk medicine is still very important in rural areas. Like many poorly known drugs of plant origin, manacd has been generally discredited in the United States for ‘‘lack of convincing evidence of its usefulness’’ (Osol and Farrar, 1955). 296 This constitutes an unjust appraisal of a potent drug with a possible future in treating rheumatism and arthritis. 2. Brunfelsia Mire Monachino In 1921, H.H. Rusby of Columbia University took part in the Mulford Biological Expedition to the Amazon. During this trip he collected a plant called miré in the Yungas region of Bolivia. The plant was sterile at the time of collection but Rusby noted its resemblance to manacad (Brunfelsia uniflora). He reported that the Indians of the region employed miré to expel cutaneous parasites and to ‘‘paralyze the voluntary muscles as in an alcoholic intoxication’’. They boiled the plant to extract the drug, a process which apparently does not injure the active constituents. Miré produced a profuse sweating capable of destroying all cutaneous parasites but with no disturbance of the senses or intellect (Rusby, 1924). Concurrently with Rusby’s initial report, T. S. Githens pub- lished his pharmacological studies on the effects of mire. He confirmed the drug’s paralyzing effect on the voluntary mus- cles through an action on the spinal chord. He also observed stimulation of the peripheral motor-apparatus indicated by muscular twitching. Both the sweat and salivary glands were stimulated in rabbits and frogs (Githens, 1924). Two years later, Githens (1926) published an article on the chemistry of miré. In analyzing extracts of the root and stem, he isolated three principles: 1. A strongly fluorescent body soluble in alcohol, ether and chloroform but insoluble in water. In mice, this portion caused paralysis, but without twitching of the muscles. 2. An alkaloid soluble in alcohol but precipitated from al- coholic solution by ether. He found 0.3% crude alkaloid which was pale yellow in freshly prepared solutions but soon became a reddish wine color on standing. This fraction was very active physiologically. 3. A second alkaloid was present which was soluble in al- cohol but not precipitated by the addition of ether. This body corresponded to 0.5% of the root and gave a permanently pale car yellow solution. Its action resembled that of the first alkaloidal portion. Mire was identified in 1925 as Brunfelsia hydrangeiformis (Pohl) Benth. by Youngken, who carried out a detailed phar- macognostic study of mire. Miré is now known to constitute a distinct species, B. Mire, described by Monachino in 1957. The plant is found in parts of Amazonian Peru and Brazil and is a close relative of B. hydrangeiformis which is restricted to southeastern Brazil. The findings of Rusby and Githens on miré recall the earlier studies on manacd root, especially the occurrence of two simi- lar alkaloidal fractions and a strongly fluorescent substance. In addition, the stimulation of the sweat and salivary glands are likewise known in manacad, suggesting that similar types of compounds may occur in unrelated species of the genus. One herbarium collection of miré (Cardenas 2813) bears the interesting comment that cattle die when they eat the leaves, a further indication of its toxic activity. Two other species, B. brasiliensis and B. grandiflora, are also reported to be poison- ous to cattle. B. Mire remains relatively unknown from chemi- cal and pharmacological standpoints and clearly merits additional phytochemical work. 3. Brunfelsia grandiflora D. Don Brunfelsia grandiflora is widely recognized in the upper Amazon for its potent drug effects. Yet its identity and uses have long been obscured in the literature by misidentifications and confused ethnobotanical reports. Specimens of B. grandi- flora in herbaria are consistently and erroneously determined as B. latifolia (Pohl) Benth., B. maritima Benth. or B. bono- dora (Vell.) Macbr. There have been numerous notes on her- barium labels that this species is medicinal, narcotic and/or poisonous, but reports in the literature are sparse and mislead- ing. Beginning in 1967 and in later publications, Schultes brought to the fore the question of the possible use of Brunfelsia as a hallucinogen, referring specifically to its widespread cultiva- 298 tion in the Colombian Putumayo (Schultes 1967, 1969, 1970a, 1970b, 1970c: Schultes & Hofmann, 1973). He suggested that this species (reported originally as B. maritima) may have been employed more extensively in the past and that, as native peoples have become acculturated, its use has died out. Brunfelsia grandiflora is distributed throughout western South America from Venezuela south to Bolivia and east to the Brazilian Amazon. It is also extensively cultivated in the American tropics as an ornamental. However, only in western South America are its curious medicinal properties recognized by native healers. B. grandiflora has one subspecies — subsp. Schultesii Plowman — recognized chiefly by its much smaller flowers and fruits. Both forms, however, seem to be used interchangeably in folk medicine. Subsp. Schultesii is more widespread in the lowlands and the form more likely to be employed. Many different Indian tribes in the Amazon region are ac- quainted with this species, and it is known by many vernacular names. The most widely used names are chiricaspi and chiric sanango, both Quechua words, found in southern Colombia and Amazonian Peru respectively. Chiricaspi means *‘cold tree’: chiric sanango signifies **cold medicine’. Both names incorporate the Quechua word chiric which means “cold”, in reference to the sensation of chills reputedly produced upon ingestion of the roots or bark. Chiric guayusa Is another variant found in lowland Ecuador (Pinkley, 1969). Brunfelsia grandiflora, like other plants which yield potent drugs, has multiple uses: as a medicine, as a narcotic and, in higher doses, as a poison. By no means exclusive of each other, these classes of usage frequently intergrade in everyday real- ity. BRUNFELSIA GRANDIFLORA AS A POISON Schultes (1967) mentions that this plant is considered potl- sonous to cattle near Leticia in the Colombian Amazon. Data from herbarium specimens collected in Bolivia (Steinbach 1805, 5487) also indicate that the plant is very poisonous, especially to cattle which occasionally eat the foliage. Another 299 collector (Heinrichs 496/) states that the roots are employed in Ecuador as a fish poison. BRUNFELSIA GRANDIFLORA AS A MEDICINE Collectors of the species frequently mention the effect of cold or chills produced when the ground bark or root is taken (Cuatrecasas 11275, Mexia 6444, Pinkley 43, 202, 444, 450, 457, Plowman 2019). Other reports indicate that the roots are employed against rheumatism (Mexia 6444, Plowman 2494, Woytkowski 6170). Still other workers have stated that the plant is used against fevers (Juajibioy 277; Pérez Arbelaez 688; Plowman 2040), against snakebite (Scolnik 1495) or simply that it is medicinal (Woytkowski 5525). My field work and that of others in the Amazon basin have served to substantiate these often vague claims of physiologi- cal activity. In the region around Iquitos, Peru, and probably throughout most of the Peruvian Amazon, Brunfelsia grandi- flora is one of the most important medicines against rheuma- tism and arthritis. One informant, a Kokama Indian from the Rio Ucayali, provided the following recipe for preparing the drug (Tina, 1969): “The root is scraped and placed in cold water or chicha de maiz. This is then taken in wineglassful doses. To increase the dose, the bark of other trees may be added, including remocaspi (Pithecelobium laetum Benth.), chuchuhuasi (Heisteria pallida Engl.) and Auacapurana (Campsiandra laurifolia Benth.). The root of chiric sanango may also be prepared with aguardiente (cane alcohol). About S50 grams of scraped root and bark are added to one liter of alcohol. A small glass is then drunk before meals until four liters have been consumed.”’ Pinkley, who worked extensively on the ethnobotany of the Kofan tribe of Ecuador and Colombia, found that the lowland Quechuas on the Rio Napo in Ecuador also utilize Brunfelsia grandiflora as a remedy for rheumatism: They take it if they have a burning in the lower part of their back. They place their hands in the area of the kidneys. Upon 300 making a drink from the leaves in hot water, they become ex- tremely chilled after drinking.” (Quoted in Schultes, 1966). Pinkley (1969) later added that the Kofans of the Putumayo also use the plant for high fevers and severe back pains. The leaves of this plant are also employed in the Iquitos region as a cure for bronchitis (Tina, 1969): ‘Twelve fresh leaves of chiric sanango are crushed up, then squeezed and the juice mixed with a little water. A spoonful is drunk twice a day, in the morning and at evening during three days.”’ In contrast to the root and bark, the leaves reputedly do not Cause nausea. The Siona Indians of the Colombian Putumayo similarly employ Brunfelsia grandiflora as an analgesic to alleviate pain. They say that it has a strong numbing effect, permitting one to walk long distances, even if the feet ache (Langdon, 1972). In an isolated report, Steward and Métraux (1948) state that the Chama Indians of Peru take the roots of Brunfelsia grandiflora as an aphrodisiac. The method of use is not given. BRUNFELSIA GRANDIFLORA AS A HALLUCINOGEN This species has long been suspected of possessing narcotic or hallucinogenic properties (Schultes, 1966, 1967; Schultes & Hofmann, 1973). Herbarium labels from Peru (Tessmann 3243) and southern Colombia (Bristol 1364, Pinkley 420) indicate that the plant is a borrachera, aterm which translates ‘‘intoxicant”’ and which is applied to the narcotic tree Daturas and other plants. Friedberg (1965) and Pinkley (1969) first disclosed that Brunfelsia species serve as admixtures to the hallucinogenic drink ayahuasca or yagé prepared from Banisteriopsis Caapi. In 1928, the French botanist Benoist published a new species, Brunfelsia Tastevinii, which he claimed was used as a hallucinogen in the Brazilian Amazon. Benoist named the plant for its discoverer P. Tastevin, a missionary and anthropologist who reported its use among the Kachinaua tribe of the Rio 301 Jordao, a tributary of the Rio Tarauaca. | have examined the type specimen of B. Tastevinii which is preserved at the Museum of Natural History in Paris. This plant is clearly B. grandiflora and Benoist’s B. Tastevinii must therefore be placed in synonymy with this species. The Kachinaua were said to cultivate the plant, which they called keva-honi, and to prepare a beverage from it. The effects of Keva-honi were described by Tastevin as follows: The juice of this plant plunges them [the Indians] into a kind of intoxication or stupefaction which lasts a littke more than a quarter of an hour and from which they acquire magical powers, enabling them to heal all sorts of diseases through incantations. While the effects of the drink act on their brains, they are unable to fall asleep. They believe they see all kinds of fantastic animals: dragons, tigers, wild boars, which attack them and tear them to bits, etc. This action of honi lasts four or five hours depending on the quantity ingested.” This account raises some points of skepticism. The descrip- tion of the effects of keva-honi is strikingly similar to that often given for Banisteriopsis intoxications (Rivier & Lindgren, 1972; Naranjo, 1973) and unlike any other reports of Brunfelsia intoxication. It is my contention that Tastevin confused these two plants. He himself stated earlier (Tastevin 1926) that the Kachinauas of the Upper Tarauaca knew and esteemed caapi (Banisteriopsis) for “‘learning the future, conversing with spirits, or dispelling bad luck’’. He further stated that the Panoan name for caapi is keya-honi or simply honi, *‘the liana’. It seems possible that he was shown a plant of Brunfel- sia used as an admixture to caapi and took it to be the main ingredient of the hallucinogenic mixture. Studies of hallucino- gen use among the Kachinaua in neighboring Peru (Der Mar- derosian et al. 1970; Kensinger, 1973) demonstrated the use of Banisteriopsis and Psychotria species to prepare the hallucin- ogenic drink called nixi pai. No mention is made of Brunfelsia admixtures, and it is possible that knowledge of the use of this drug is dying out, at least among certain segments of the tribe. Other reports of the use of Brunfelsia as an admixture to Banisteriopsis preparations are more substantial. In Iquitos, Peru, B. grandiflora is added to ayahuasca ‘“‘to give more 302 strength.’* It reputedly makes a sound “‘like rain in the ears” (Tina, 1969). A Witoto Indian living at Puca Urquillo on the Rio Ampiyaco (Peru) informed me that chiric sanango is taken to gain strength at the new moon. The bark is scraped and mixed with cold water to prepare the beverage. The Jivaro of Amazonian Ecuador and Peru are famous for their ritual use of Banisteriopsis Caapi, which they call natemd, to achieve trance-like states (Harner, 1968). A Cana- dian businessman, D. C. Webster, reported in a letter to R. E. Schultes the use of three plants in the preparation of the natema drink among the Jivaro, along with photographs of the plants. Besides natema (Banisteriopsis), they included chiricaspi (Brunfelsia grandiflora) and an unidentified liana called hiaji. To prepare the drink, the natemd is cut up into small pieces and boiled for at least 14 hours. Then the second two ingredients, also cut up, are added and the mixture boiled down to form a thick, muddy liquid (Webster, 1970). The supposed effects of this natema mixture, in addition to strong visual hallucinations, include the “melting away of dis- ease’’ with beneficial effects against arthritis, intestinal para- sites, tuberculosis, and poor vision. This is a good example of shamanistic healing methods, in which a strong hallucinogen provides the necessary spiritual contact with the forces of illness, while other potent herbal admixtures give specific therapeutic effects. In this preparation of natemd, the antiar- thritic effects may be attributable to the addition of Brunfelsia. Brunfelsia is known to play a part in shamanistic practices in still other tribes. Shamans often invoke the aid of a particular spirit helper in their healing ceremonies, which may take the form of a bird, snake, insect or plant. Shamans of the Lama tribe, who inhabit the region just west of Tarapoto in northern Peru, consider B. grandiflora a spiritual guide (Steward, 1948). Steward writes in the Handbook of South American Indians: ‘**The neophyte sorcerer dieted and took tobacco juice, cigars, avahuasca, and uniquely, Brunfelsia grandiflora and another liana. He acquired a general power from these plants but no internal ‘‘thorns’’. To cause illness, he impregnated a splinter with his power and cast it at his victim. To cure it, a shaman sucked out the splinter.’ (Steward & Metraux, 1948). 303 In southern Colombia, particularly the Putumayo region, Brunfelsia is also added to Banisteriopsis preparations among several tribes, including the Siona, Kofan and Inga. Among the Inga, several classes of chiricaspi are recognized. All of these are considered febrifuges and the term seems to be generic for plant medicines exhibiting this effect. Three kinds of chiricaspi are referable to B. grandiflora: picudo “*beaked” chiricaspi, salvaje “wild” chiricaspi, and chacruco *‘‘of cultivated ground” chiricaspi. Of these, picudo chiricaspi is considered to be the strongest variety. To counteract fevers, a small stem 30cm. long is used. The bark is scraped in cold water, let stand for two hours, then drunk. An Ingano curaca or healer at Mocoa told me that if three stems are used, one becomes intoxicated as with vagé and that it makes the whole body cold. Chacruco and salvaje chiricaspi do not differ morphologically from picudo chiricaspi but may represent chemical races of the plant. A fourth class of chiricaspi — calentura “‘fever” chiricaspi — is also known to the Inga. This shrub which grows in primary forest has been identified as Stephanopodium peruvianum P. & E. (Dichapetalaceae). The leaves are taken in cold water against fevers, as the vernacular names suggests. To date, no alkaloids have been encountered in this family (Raffauf, 1970). Another tribe of the Putumayo, the Siona, also employs Brunfelsia grandiflora which they designate generically as huha hai. Two classes of huha hai are recognized. Yai huha hai is not cultivated but collected wild in the forest. Jean Langdon, an anthropologist working with the Siona, supplied the follow- ing account of vai huha hai along with voucher specimens collected by an informant: “The plant is used as a drink to give visions as well as to alleviate pain. To drink it, they grate the stem and drink the juice that comes out. The leaves can also be used if they are mashed up. Nothing else is added to the preparation, but it is often taken in conjunction with vagé or voko (Paullinia Yoco). If taken with vage, it is drunk before drinking vagé.”’ The Siona describe the effect of vai huha hai as one of extreme coldness. It supposedly dulls all pains. The second 304 class of huha hai is cultivated and known as bi’a huha hai. It is cooked with yagé. Another Siona informant volunteered the following: ‘‘It makes you shiver when you drink yagé. It also makes your legs heavy and you feel like spines are sticking you. It is fresco, so it is good for curing sickness, as well as drinking with yagé.’’ (Langdon, 1970). The Kofan Indians of the Putumayo area are likewise famil- iar with the medicinal and intoxicating properties of Brunfelsia. Pinkley (1969, 1973) has stated that Brunfelsia, while not a common admixture to Banisteriopsis, among certain groups plays a role similar to that of Brugmansia in the magico-reli- gious ceremonies of the shaman. Brunfelsia grandiflora is sometimes taken by the Kofan shaman in order to diagnose disease (Pinkley, 1973). The Kofan, like other groups in the Putumayo, recognize three classes of Brunfelsia which are generically known as tsontinba’’k’o in the Kofan language. Two of these are refera- ble to Brunfelsia grandiflora and are distinguished primarily by where they grow. Soci (toucan) tsontinba’’k’o is cultivated in houseyards around their settlements. Chipiri ‘*small” tsontin- ba’’k’o grows wild in the surrounding secondary forest. 4. Brunfelsia chiricaspi Plowman The third kind of Brunfelsia known to the Kofan is called covi ‘‘tapir’’ tsontinba’’k’o. This plant is considered to be the strongest of the tsontinba’’k’o class and preferred for its potent drug effects. It belongs to a recently described species Brunfel- sia chiricaspi Plowman, known only from the Colombian Pu- tumayo and south to the Rio Coca in Ecuador. This species occurs only in primary forests and is not cultivated. While conducting ethnobotanical field work and general col- lecting in 1968, I visited the Kofan village of Santa Rosa on the Rio Guamués, in order to study firsthand the use of Brunfelsia and other medicinal plants. The Kofan settlements in the Col- ombian Putumayo are being subjected to rapid changes as a result of large scale oil drilling operations in their territory. This made the Indians especially suspicious of strangers and highly protective of their medicinal plant lore. 305 I made contact with an old man in the village, who was considered knowledgeable about plants and medicines. After some time, I asked him about the use of tsontinba’’k’o and yage. He informed me that hardly anyone uses this medicine anymore because it is considered very dangerous. He disa- vowed any knowledge of vagé or of taking tsontinba’'k’o for visions. Since it seemed no more information would be forth- coming, I asked the old Kofan to demonstrate the preparation of covi tsontinba’’k’o for me and my companion, a Kamsa Indian from the Sibundoy Valley. He reluctantly agreed to do it and we decided to drink the drug at his small hut. In view of the complete lack of firsthand experiments with this Brunfelsia, | include the following account of the effects of B. chiricaspi: December 3, 1968: Village of Santa Rosa, Rio Guamués, Com- isaria del Putumayo, Colombia. **My companion, Pedro, and I arrived at Santa Rosa from San Antonio well before sunset on the night our Kofan curaca had chosen to prepare covi tsontinba’’k’o. The curaca did not return to the village until nearly dark, carrying with him a handful of scraped bark which he had collected in the nearby forest. He said the plant is not common and he had had difficulty finding it. He extracted the juice of the greenish brown bark in a cup of cold water by wetting the bark and squeezing it repeatedly until the liquid became a murky light brown color. He then handed each of us half a cupful. The drink had a very bitter taste and pungent odor. We drank it quickly and sat down on the front porch of the curaca’s hut. “The effects of the drug appeared within about ten minutes. | first felt a tingling sensation in my lips, followed soon by the same sensation in my fingertips. This felt exactly like the feeling ex- perienced when your leg “‘falls asleep’’, when the blood rushes back. Along with the tingling, I felta pronounced vibrating in the affected parts. ‘The tingling soon spread into my mouth and upwards into my face; later into my hands and feet, tongue, elbows, and shoul- ders. After about an hour, the sensations were felt generally throughout my body, especially in my back, legs, face, lips and hands. There was a definite progression of the tingling from the base of my spine upwards towards the back of the neck, with ever-increasing intensity which centered at the base of my skull. ‘From the beginning | felt a strong urge to expectorate period- ically and later realized that I was actually frothing at the mouth. In spite of the extraordinary sensations running through my body, I remained mentally lucid. I felt quite agitated from the 306 strong tremors produced by the drug. I was also seized by periodic waves of cold, as frequently reported by the natives. “T next felt the tingling sensations and vibrations enter my head and scalp. Moving my head or hands increased the intensity of the sensations. The vibrations felt electric, penetrating my chest and back. Even then my thinking remained undisturbed though somewhat detached. ‘After another hour, I began feeling sharp stomach cramps and occasional spasms of nausea. | wanted to vomit but could not. We had fasted since the previous day according to the curaca’s instructions. The tingling remained very intense in my hands and feet and my head began to ache. A bitter taste de- veloped in my mouth and I felt vaguely cold. ‘*Sometime later — I was incapable of telling time at this point — I began to get used to the tingling sensations, which | found disquiting at first. I could not tell if | was becoming stronger or weaker. I became very dizzy with vertigo, which intensified precipitously. Everything started spinning to the right yet never seemed to move. My mind kept adjusting to the spin to set me right agin. There was a complete loss of muscular coordination at this point, and I could no longer walk or even stand up. I lay either prone or sat on the floor with my back against the wall for the rest of the night. The pains in my stomach became more acute. I continued to be nauseous and vertiginous and felt ex- tremely uncomfortable. ‘‘During the course of the evening, the curaca went to bed. We did not see him for the rest of the night except for a brief appearance upon our departure. Since we were feeling increas- ingly out of sorts in this strange place, we decided to return to the nearby village of San Antonio where we were staying in an abandoned jail cell. It was very difficult to move or stand up, but eventually we mustered enough strength to start for home. This took at least an hour with Pedro and I supporting each other, frequently stumbling and crawling along the dark trail through the forest. **When we arrived in San Antonio, we climbed into our ham- mocks to rest. I lay awake for a long time, still feeling the drug in my body, particularly the tingling. The stomach cramps and vertigo began to subside, and eventually I fell asleep, completely exhausted both mentally and physically. The next day, I felt extremely weak and nearly unable to move without great dis- comfort. I became very dizzy if I tried to stand up or walk. I could not eat anything and remained in my hammock. Only after two full days did I begin to recover and move around without becom- ing dizzy.” My companion later noted the following effects which he experienced with the drug: ‘‘swollen lips and heavy tongue, crazy in the head, cold sweat, stomach ache, nausea and weak 307 vomiting, urtication, inability to walk or move, and vertigo’. He also felt “‘the world was spinning around me like a great blue wheel. I felt that I was going to die’’. The effects described above correspond in some respects to accounts in the literature of the effects of other species of Brunfelsia, especially B. grandiflora. The notable tingling sen- sation, or parasthesia, apparently results from a peripheral vasal constriction of the capillaries. A well known drug with this effect is nicotine, and the same reaction is frequently reported in persons who have smoked their first cigarette. The increased stimulation of the salivary and sweat glands is highly reminiscent of the effects of manacd root discussed earlier. In view of the toxic effects which we experienced from Brunfelsia chiricaspi, it seems unlikely that anyone would knowingly ingest the plant in this dose except possibly in a desperate medical situation. More likely, smaller amounts are used which become further diluted when mixed with extracts of other plants such as yagé. We may also ask the question: why is Brunfelsia used in preparing yagé? Certainly its medicinal properties are impor- tant in this respect, for the use of yagé is grounded in its medicinal applications. Yagé itself is a strong purgative and may have vermicidal and bactericidal properties as well. Both these plants would then serve as strong medicinal agents irres- pective of their psychological or hallucinogenic effects. These are, however, of equal import to the shaman. It is well established that Banisteriopsis caapi, besides being a purgative, is a strong visual and perhaps auditory hallucino- gen. Sundry other plants may be added to preparations of B. caapi to vary and intensify the experience. Most notable among these are the leaves of Banisteriopsis Rusbyana and Psychotria viridis (Pinkley, 1969; Rivier & Lindgren, 1972). Both of these plants contain the potent hallucinogen N,N- dimethyltryptamine as their main active constituent, the ef- fects of which differ somewhat from the harmine derivatives found in B. Caapi (Agurell et al., 1968; Der Marderosian et al., 1968, 1970). Since N,N-dimethyltryptamine also produces strong visual effects, we can readily see why the shaman would include these plants in his brew. 308 Since Brunfelsia species appear to produce no striking visual hallucinations, we must look further for the rationale for in- cluding these plants in yagé preparations. Of the diverse physi- cal effects of Brunfelsia, I would single out the tingling sensa- tions as the most pronounced and bizarre and those which might best potentiate the hallucinatory yagé experience. By using smaller doses than I ingested, the Indians would be able to produce striking tactile hallucinations without the toxic side effects. The combination of these drugs may produce, perhaps synergistically, unique and other-worldy experiences and sen- sations. Because one often feels decreased sensitivity and numbness in the body with vagé alone, the addition of Brunfel- sia to the drink may also serve to create a greater physical awareness during the ceremony. Although each of these potent psychoactive plants is em- ployed individually for their specific effects, their use in com- bination was learned and perfected by native shamans through centuries of experimentation. Only these skilled practitioners fully understand the delicate questions of dosage and correct admixture to achieve specific physical and psychic effects. 5. Miscellaneous species Other species of Brunfelsia are known to contain active constituents such as alkaloids or to be used medicinally. Some of these suchas B. brasiliensis and B. pauciflora have already been mentioned. The fruits of B. australis, a species of Argen- tina and Paraguay, are added to food as a condiment by the Guarani Indians. Known as azucena or jazmin del monte, B. australis is noted on herbarium labels as being medicinal or poisonous. The root is used as B. uniflora as a remedy for syphilis (Woolston 57/), and the foliage is reputed to be harm- ful to horses (Pederson 10210). B. guianensis Benth., which grows in Amazonian Brazil as well as the Guianas, is also used like B. uniflora — as an antisyphilitic, antirheumatic, depura- tive and poison in high doses (Le Cointe, 1947). Brunfelsia species in the Caribbean are also known to be employed medicinally, though not nearly so commonly as in South America. Brunfelsia americana L., the most widespread 309 species in the Antilles, bears an astringent fruit which has been used as a tonic to cure chronic diarrhea and stomach problems (Descourtilz, 1833; Duss, 1897; Manfred, 1947). In the island of Dominica, it is called empoisonneur and is employed as a poison by the Island Caribs (Hodge & Taylor, 1957). Traces of cyanide have been found in the leaves and flowers as well as in the bark of the stem and root (Quisumbing, 1951). Chlorogenic acid is reported from the leaves (Politis, 1948). Scott and colleagues (1957) tested the leaf and stem of B. americana for alkaloids. They found two products: one a cry- stalline substance which melted at 125-130°; the other con- sisted of long needles which melted at 218-220°. They obtained positive alkaloid tests for both products but did no further work on the plant. Brunfelsia nitida Benth., a widely cultivated Cuban species, is used for herbal baths (Roig & Mesa, 1945). The fruits of this species were found to be strongly alkaloid-positive (Aleman Frias, 1972). Unnamed alkaloids have also been detected in the stems of B. undulata Sw., a Jamaican species (Willaman & Schubert, 1961); Willamam & Li, 1970), and in the leaves and fruits of B. Shaferi Britt. & Wils., a Cuban endemic (Aleman Frias, 1972). In conclusion, the need for modern detailed studies on the pharmacology and chemistry of Brunfelsia cannot be overem- phasized. I have outlined here what is known about the folk uses and pharmacology of the species known to be active. However, the entire genus merits intensive investigation to isolate and identify its alkaloidal and other constituents, which have eluded chemists for so long. The possible value of certain species in the treatment of arthritis and rheumatism is espe- cially important and worthy of detailed study using modern methods. ACKNOWLEDGMENTS Research reported in this paper was supported in part by the National Institutes of Health Training Grant (T T01 GM 00036-13) and by the National Science Foundation Evolution- ary Biology Training Grant (GB 7346, Reed Rollins, Principal 310 Investigator, Harvard University). I would also like to thank the following persons who kindly read the manuscript and offered useful comments: B. Holmstedt, J.E. Lindgren, R.E. Schultes, and P.G. Williams. Iam most grateful to H.V. Pink- ley and J. Langdon for sharing with me their ethnobotanical observations on Brunfelsia and to L.T. Bates for making the line drawing of B. mire. | would also like to thank the curators of the following herbaria for the loan of specimens pertinent to this study: BM, COL, ECON, F, G, GH, K, MO, NY, P, PH, PR, S, UC, US, W (abbreviations after P,.K. Holmgren & W. Keuken, 1974. Index Herbariorum, Part I, ed. 6.; Utrecht). LITERATURE CITED Agurell, S., B. Holmstedt, and J.E. Lindgren. 1968. Alkaloid Content of Banisteriopsis Rusbyana. American Journal of Pharmacy 140: 148-151. Aleman Frias, E. et al. 1972. Phytochemische Untersuchungen an Pflanzen der Kubanischen Flora. Die Kulturpflanzen 19: 417. de Almeida Costa, O. 1935. Estudo Farmacognostico de Manaca. Revista da Flora Medicinal. | (7): 345-360. Beckurts, H. 1889. Jahresbericht ber die Fortschritte der Pharmakognosie, Pharmacie und Toxicologie. Gottingen. 22: 162. Beckurts, H. 1895. Chemische und Pharmakologische Untersuchung der Manaca-Wurzel. Apotheker Zeitung 72: 622-623. Benoist, R. 1928. Une nouvelle espece de Brunfelsia (Solanacées), plante magique des Indiens du Haut-Amazone. Bulletin de la Societe Botanique de France 75: 295. Brandl, J. 1895. Chemisch-pharmakologische Untersuchung uber die Manaca-Wurzel. Zeitschrift fur Biologie 31: 251-292. Brewer, E.P. 1882. On the Physiological Action of Manaca. The Therapeutic Gazette, New series. 3(9): 326-330. Caminhoa, J.M. 1871. Das Plantas Toxicas do Brazil. Typographia Perse- veranca. Rio de Janeiro. pp. 143-144. Correa, M.P. 1909. Flora do Brasil. Typographia da Estatistica. Rio de Janeiro. p. 102. Der Marderosian, A.H., K.M. Kensinger, J. Chao, and F.J. Goldstein. 1970. The Use and Hallucinatory Principles of a Psychoactive Beverage of the Cashinahua Tribe (Amazon Basin). Drug Dependence 5: 7-14. Der Marderosian, A.H., H.V. Pinkley, and M.F. Dobbins, IV. 1968. Native Use and Occurrence of N,N-dimethyltryptamine in the Leaves of Banisteriopsis Rusbyana. American Journal of Pharmacy 140: 137-147. Descourtilz, M.E. 1833. Ed. 2. Flore Pittoresque et Medicale des Antilles. Paris. 2: 38-40. 311 Don, D. 1829. Observations on the Characters and Affinities of Darwinia, etc. Edinburgh New Philosophical Journal. July, 1829. p. 81. Dragendorff, G. 1898. Die Heilpflanzen der Verschiedenen Volker und Zeiten. Verlag von Ferdinand Enke. Stuttgart. p. 600. Duss, R.P. Le. 1897. Flore Phanerogamique des Antilles Frangaises. Protat Freres, Imprimeurs. Macon. p. 407. Erwin, J.L. 1880. Manaca — Proximate Properties of the Plant. Therapeutic Gazette. New series. | (7): 222-223. Friedberg, C. 1965. Des Banisteriopsis Utilisés comme Drogue en Amerique du Sud. Journal d’ Agriculture Tropicale et de Botanique Appliquée 12: 1-139, Githens, T.S. 1924. Preliminary Report on Studies on Mire. Journal of the American Pharmaceutical Association 13: 102. Githens, T.S. 1926. Additional Studies on Mire. Journal of the American Pharmaceutical Association 15: 1067-1068. Hahmann, C. 1920. Beitrage zur anatomischen Kenntnis der Brunfelsia Hopeana Benth., im besonderen deren Wurzel, Radix Manaca. Ange- wandte Botanik 2: 113-133, 179-191. Harner, M.J. 1968. The Sound of Rushing Water. Natural History Magazine 77(6): 28. Henry, T.A. 1949. Ed. 4. The Plant Alkaloids. The Blakiston Co. Philadel- phia. p. 83. Hodge, W.H. and D. Taylor. 1957. The Ethnobotany of the Island Caribs of Dominica. Webbia 12 (2): 603. Hoppe, H.A. 1958. Drogenkunde. Cram, De Gruyter & Co. Hamburg. pp. 396-397. Iyer, R.P., J.K. Brown, M.G. Chaubal and M.H. Malone. 1977. Brunfelsia hopeana |: Hippocratic Screening and Antiinflammatory Evaluation. Lloydia 40: 356-360. Kensinger, K.M. 1973. Banisteriopsis Usage Among the Peruvian Cashina- hua. In M.J. Harner, ed. Hallucinogens and Shamanism. Oxford Uni- versity Press. London. pp. 9-14. Kunkel, A.J. 1901. Handbuch der Toxikologie. Gustav Fischer Verlag. Jena. 2: 696. Langdon, J. 1972. Personal Communication. Lascelles-Scott, W. 1887. The Monthly Magazine. 46: 739, 47: 773. Le Cointe, P. 1947. Arvores e Plantas Utéis. Ed. 2. Amazonia Brasileira 3, ser. 5. Brasiliana 251: 279. Companhia Editora Nacional. Sao Paulo. Lenardson, R. 1884. Chemische Untersuchungen der rothen Manaca. Dis- sertation. University of Dorpat, Estonia. 37 pp. Machado de Campos, S. 1964. Scopoletin in Brunfelsia seeds. Anais da Academia Brasileira de Ciencias de Rio de Janeiro. 36 (1): SI1-513. Manfred, L. 1947. 7,000 Recetas Botanicas a base de Mil trescientos Plantas Medicinales. Editorial Kier. Buenos Aires. p. 152. Manske, R.H.F. and H.L. Holmes. 1950. The Alkaloids. Chemistry and Physiology. Academic Press. New York. 1: 313-314. 334. Marcgraf, G. 1648. Historia Rerum Naturalium Brasiliae. F. Hack. Leiden. p. 69. Martinez-Crovetto, R. 1968. La Alimentacion entre los Indios Guaranies de Misiones (Republica Argentina). Ethnobiologica 4: 24. S12 von Martius, C.F. P. 1843. Systema Materiae Medicae Vegetabilis Brasilien- sis. F. Fleischer. Leipzig. p. 67. Monachino, J.B. 1953. Miré, A New Species of Brunfelsia from Bolivia. Phytologia 4(5): 342-347. Mors, W.B. and O. Ribeiro. 1957. Occurrence of Scopoletin in the Genus Brunfelsia. Journal of Organic Chemistry 22: 978. Naranjo, C. 1973. Psychological Aspects of the Yagé Experience in an Experimental Setting. In M.J. Harner, ed. Hallucinogens and Shaman- ism. Oxford University Press. London. pp. 176-190. Osol, A. and G.E. Farrar. 1955. The Dispensatory of the United States of America. 25th Ed. J.P. Lippincott Co. Philadelphia. p. 1745. Pammel, J.H. 1911. A Manual of Poisonous Plants. The Torch Press. Cedar Rapids, Iowa. pp. 715, 853. Peckolt, T. 1909. Heil und Nutzpflanzen Brasiliens. Berichte der Deutschen Pharmazeutischen Gesellschaft 19: 307-315. Pereira, H. 1929. Diccionario das Plantas Uteéis do Estado do Sao Paulo. Typographia Brasil de Rothschild. Sao Paulo. Pinkley, H.V. 1969. Plant Admixtures to Ayahuasca, the South American Hallucinogenic Drink. Lloydia 32: 311-312. Pinkley, H.V. 1973. The Ethno-ecology of the Kofan Indians. Ph.D. Disser- tation. Harvard University. Cambridge, Mass. Piso, W. 1648. De Medicina Brasiliensi. F. Hack. Leiden, p. 85. Piso, W. 1659. De Indiae Utrisque Re Naturali et Medica. p. 184. Plowman, T. 1973. Four New Brunfelsias from Northeastern South America. Botanical Museum Leaflets, Harvard University 23(6): 245-272. Plowman, T. 1974. Two New Brazilian Species of Brunfelsia. Botanical Museum Leaflets, Harvard University 24(2): 37-48. Pohl, J.E. 1826. Plantarum Brasiliae Icones et Descriptiones. Vienna. pp. 1-8. Politis, J. 1948. Sur la Distribution de l’'acid Chlorogénique dans la famille des Solanacées et dans les organes de ces plantes. Comptes Rendus 226: 692-693. Quisumbing, E. 1951. Medicinal Plants of the Philippines. Technical Bulletin 16. Philippine Department of Agriculture and Natural Resources. Man- ila. Raffauf, R.F. 1970. A Handbook of Alkaloids and Alkaloid-containing Plants. John Wiley and Sons. New York. Rivier, L. and J.E. Lindgren. 1972. ‘‘Ayahuasca’’, the South American Hallucinogenic Drink: An Ethno-botanical and Chemical Investiga- tion. Economic Botany 26(2): 101-109. Roig & Mesa, J.T. 1945. Plantas Medicinales, Aromaticas 0 Venenosas de Cuba. Cultural S.A. La Habana. p. 325. Rusby, H.H. 1924. Miré. Journal of the American Pharmaceutical Associa- tion 13: 101-102. Schultes, R.E. 1966. The search for new natural hallucinogens. Lloydia 29: 302-304. Schultes, R.E. 1967. The Place of Ethnobotany in the Ethnopharmacologic Search for Psychotomimetic Drugs. In D.H. Efron, ed. The Ethno- pharmacologic Search for Psychoactive Drugs. U.S. Government Printing Office. Washington. pp. 44-46. S13 Schultes, R.E. 1969. The New World Indians and their Hallucinogenic Plants. Morris Arboretum Bulletin 21(1): 9. Schultes, R.E. 1970a. Notas etnotoxicologicas acerca de la flora amazonica de Colombia. In J. Idrobo, ed. II Simposio y Foro de la Biologia Tropical Amazonica. Editorial Pax. Bogota. p. 192. Schultes, R.E. 1970b. The Botanical and Chemical Distribution of Hal- lucinogens. Annual Review of Plant Physiology 21: 593. Schultes, R.E. 1970c. The Plant Kingdom and Hallucinogens. Bulletin on Narcotics 22(1): 43. Schultes, R.E. and A. Hofmann. 1973. The Botany and Chemistry of Hal- lucinogens. Charles C. Thomas. Springfield, Illinois. pp. 163-166. Scott, W.E., R.M. Ma, P.S. Schaffer and T.D. Fontaine. 1957. A survey of selected Solanaceae for alkaloids. Journal of the American Phar- maceutical Association 46 (5): 302-304. Steward, J.H. 1948. Tribes of the Montana: an Introduction. Handbook of South American Indians. Vol. 3: The Tropical Forest Tribes. Bureau of American Ethnology Bulletin 143. Washington, D.C. p. 531. Steward, J.H. and A. Metraux. 1948. Tribes of the Peruvian and Ecuadorian Montana. InJ.H. Steward, ed. Handbook of South American Indians. Vol. 3: The Tropical Forest Tribes. Bureau of American Ethnology. Bulletin 143. Washington, D.C. pp. 594, 605. Tastevin, C. 1922. Nomes de plantas e animaes em lingua Tupy. Revista do Museu Paulista 13: 688-763. Tastevin, C. 1926. The Upper Tarauaca. La Geographie. 45. Typed manu- script. Translated in: United States: Coordinator of Inter-American Affairs. The Middle Amazon. pp. 80-103. Tina, F. 1969. Personal communication. Iquitos, Peru. Webb, L.J. 1948. Guide to Medicinal and Poisonous Plants of Queensland. Council for Scientific and Industrial Research. Melbourne. p. 152. Webster, D.C. 1970. Letter to R.E. Schultes. February 5, 1970. Willaman, J.J. and B.G. Schubert. 1961. Alkaloid-bearing plants and their contained alkaloids. Technical Bulletin 1234. Agricultural Research Service. Washington, D.C. pp. 216-217. Willaman, J.J. and H.L. Li. 1970. Alkaloid-bearing plants and their contained alkaloids. 1957-1968. Lloydia (Supplement) 33 (3a): 203. Wren, R.C. 1956. Potter's New Cyclopedia of Botanical Drugs and Prepa- rations. Pitman and Sons, Ltd. London. p. 196. Youngken, H.W. 1925. The Anatomy and Botanical Position of Mire. Journal of the American Pharmaceutical Association 14 (3): 195-200. 314 PLATE 63 pe FACVLTATIBVS SIMPLICIVK Lis. IV. 85 Folia habet citriishaud diffimilia , paulo tamen longiora & molliora , & qualitate refri- gerandi & abftergendi pradita- Prater alias dotes, quibus excellunt,vulneribus atque ul- ceribus opitulantur, partefque vitiatas reparant.Cujus reiChi- rurgi noftri non ignari, ca in quotidianum ufum colligunt. Raro videas viatores iter fu- fcipere, nifi probe hoc reme- dio inftru&os. Succus oleofus denique ¢ filiquis expreffus , ad maturan- da apoftemata refervatur , & applicatur cum profpero fuc- ceffu. Car. XLII. DeManaca frutice,ejuf- que “bus Ocis umbrofis , maxime circa Aldeam Tapicirica lu- xuriat Frutex arborefcens 444- naca, cottice gryfco , ligno du- ro quidem, fed fragili , ex albo cinentio, foliis acuminatis. Flores fert exiguos, quo- rumaltercocruleo, alrer vero (quod ut rarum, ita maxime jucundum) lactco nitore nitefcit : Menfe Januario vegetus & valde fpeciofe florefcens confpicitur, atque integras filvas infigni fra- grantia Narciffi zmula implet. Flori fuccedit fruétus bacce Juniperi fimilis, fed inutilis. Radicem habct magnam, fo- lidam & albicantem, cujus me- dullofa fubftantia in pulverem re- daa, magna in Medicina poili- cetur & preftat. Incolz pene omnes,tam Lufi- tani quam Brafiliani, licet magni zftiment, tamen ob indomitas operationes haétenus in ufumad- mittere vix aufifuerunt. Quippe periculo non vacat hoc genus medicamenti , quod nimis vio- lenter corpus fuperne & interne moveat. Quamobrem tantum hominibus robuftiffimis exhiberi folet , idque additis corre@oriis, tum & juftaobfervata dofi, que Scammon¢i potius inferior quam fuperior efle debet : ad illud enim validum medicamenrum proxi- me accedit hc radix, verum non ita infipida eft , amarore enim & acore non plane deftiuitur. Wwe, ae a DS WN YF) Me [) “Ni L3 Cap. Plate 63. First Description and Illustration of Manaca Root (Brunfelsia uniflora) in Piso’s De Medicina Brasiliensi (1648). 315 PLATE 64 Plate 64. Illustration of Brunfelsia uniflora in von Martius’ Flora Brasilien- sis, Vol. 8 (1): plate 43. 1862, 316 PLATE 65 Plate 65. Medicinal Parts (Roots and Stem) of chiric sanango (Brunfelsia grandiflora subsp. Schultesii) collected on the Amazon River near Puerto Narino, Amazonas, Colombia (Plowman et al. 2407). 317 PLATE 66 at 7 '\ j “* x Sy Plate 66. Brunfelsia grandiflora D. Don. Colombia: Comisaria del Putumayo. Left: Shrub cultivated in Indian houseyard near Mocoa. Right: Wild plant found growing in secondary forest near Puerto Limon. 318 PLATE 6/7 BRUNFELSIA_ mire Monachino (i init 1a (ee i 4 4 A ly Y a, et Me Me Vt Yi: ATH y — NS IN Plate 67. Brunfelsia Mire Monachino. |. Habit. 2. Flowering Branch. 3. Fruit cluster. 319 PLATE 68 Plate 68. Brunfelsia americana L. From Burman’s Edition of Plumier’s Plantarum Americanarum (1756).