Journal of Ethnobiology VOLUME 14, NUMBER 1 SUMMER 1994 : Journal and Society Organization rich hes Deborah M. Pearsall, American ee Division, 107 Swallow Hall, Univer- sity of Missouri, Columbia, MO 65211 ASSOCIATE EDITOR (Spanish): Alejandro de Avila B., Department of Anthropology, University of California, Berkeley, CA 94720. NEWS & COMMENTS EDITOR: Gary J. Martin, 94 Blvd. Flandrin, 75116, Paris, France. FAX: 33 /1/ 45533001. BOOK REVIEW EDITOR: Carlos E.A. Coimbra, Jr., Escola Nacional de Saude Publica- FIOCRUZ, Fundacao Oswaldo Cruz, Nucleo de Doencas Endemicas, Rua Leopoldo Bulhoes-Manguinhos, 21.041 Rio de Janeiro-RJ-BRASIL. BOOK REVIEW EDITOR: Nancy J. Turner, Environmental Studies Program, P.O. Box 1700, University of Victoria, Victoria, B.;C. CANADA V8W 2Y2. PRESIDENT: Cecil H. Brown, Department of Anthropology, Northern Illinois University, DeKalb, Illinois 60115. PRESIDENT-ELECT: Catherine S. Fowler, Department of Anthropology, University of Nevada, Reno, Nevada 89557. SECRETARY/TREASURER: Brien A. Meilleur, Missouri Botanical Garden, Center for Plant Conservation, P.O. Box 299, — Louis, MO 63166 CONFERENCE COORDINATOR: Jan Timbrook, Department of Anthropology, Santa Bar- bara Museum of Natural History, oe Puesta Del Sol Road, Santa Barbara, CA 93105. BOARD OF TRUSTEES ROBERT A. BYE, JR., Universidad Nacional Aut6noma de México, MEXICO: ethnobotany, ethnoecology. TIMOTHY JOHNS, Macdonald College of McGill University, CANADA. Ex officio: Past Presidents Steven A. Weber, Amadeo M. Rea, Elizabeth S. Wing, and Paul Minnis; Permanent board member Steven D. Emslie; The Editor, President, President Elect, Secretary /Treasurer, and Conference Coordinator. EDITORIAL BOARD KAREN R. ADAMS, Crow Canyon Archaeological Center, USA; paleoethnobotany. EUGENE N. ANDERSON, University of California, Riverside, USA; focal BRENT BERLIN, University of California, Berkeley, U g , medi- cal ethnobotany. H. SORAYYA CARR, El Cerrito, CA, USA; zooarchaeology. GAYLE J. FRITZ, Washington University, St. Louis, USA; paleoethnobotany DAVID R. HARRIS, University College, London, ENGLAND; Sncerlens subsistence sys- tems, archaconet tany. TIMOTHY Bagot McGill University, CANADA; chemical ecology, ethnobotany. HARRIET V. KUHNLEIN, McGill University, CANADA; ethnonutrition, human nutrition GARY J. MARTIN, Grupo: de Apoyo al Desarrollo Etnico, Oaxaca, MEXICO; ethnobiological classification DARRELL A. POSEY, School of potheopology and Museum sg soca ae Oxford en _ sity, ENGLAND; natural resource management, ethnoecology, ethnoentomology, tropical cu tural ecology. _ AMADEO M. REA, San Diego Notes History Museum, USA; cultural ecology, zooarchae- ology, ethnotaxonomics. ELIZABETH J. REITZ, University of Georgia, USA; zooarchaeology. -MOLLIE S. TOLL, pean of New Mexico, USA; prehistoric and historic ethnobotany. _ Feature editors Carlos E.A. Coimbra and Nancy J. Turner (see above). Comment of Eehnablology i is pawn semi-annually. grease for publication, information for the “News and omments” and ©Society of Ethnobiology ISSN 0278-0771 Journal of Ethnobiology VOLUME 14, NUMBER 1 SUMMER 1994 Advertising Information Journal of Ethnobiology published by the Society of Ethnobiology Mailing Instructions. All initial advertising contracts and correspondence should be sent to: Secretary / Treasurer Society of Ethnobiology Brien Meilleur Missouri Botanical Garden Center for Plant Conservation P.O. Box 299 St. Louis, MO 63166 phone: (314) 577-9450 FAX (314) 577-9465 Insertion orders and camera ready copy should be sent to: Editor, Journal of Ethnobiology Dr. Deborah Pearsall American Archaeology Division 103 Swallow Hall University of Missouri Columbia, MO 65211 phone: (314) 882-3038 FAX (314) 882-9410 MISSOURI BOTANICAL NOV 17 1994 CONTENTS os GARDEN LIBRARY, BR MAE ea gl ha eat Cus gg ta ik we od Ye hag ds Bhan a She i TOWARD RECONSTRUCTING ANCIENT MAIZE: EXPERIMENTS IN PROCESSING AND CHARRING Susan Goette, Michele Williams, Sissel Johannessen, Christine A. Hastorf . .1 CHARACTERIZATION OF MESTIZO PLANT USE IN THE SIERRA DE MANANTLAN, JALISCO-COLIMA, MEXICO Bruce F. Benz, Francisco Santana M., Rosario Pineda L., Judith Cevallos E., Live Robles H., Dowitild de Ne he oo ew Gb Wk he oh Ke we 23 CHOICE OF FUEL FOR BAGACO STILLS HELPS MAINTAIN BIOLOGICAL DIVERSITY IN A TRADITIONAL PORTUGUESE AGRICULTURAL SYSTEM ee en oe ah ee eee 4 oe Oe 43 TRIBES, STATES, AND THE EXPLOITATION OF BIRDS: SOME COMPARISONS OF BORNEO AND NEW GUINEA er eT AIOE ners a aint bib arsine Saas oa Se eB 59 THE DISTRIBUTION AND ETHNOZOOLOGY OF REPTILES OF THE NORTHERN PORTION OF THE EGYPTIAN EASTERN DESERT Steven M. Goodman and Joseph J. Hobbs ................... 75 Ps Pee MMU I ng ay oa vas hee le ee ee RS A 101 BOOK REVIEWS Oat Bran, edited by Peter J. Wood eee NN ee le a wa aang gr day & a ge ee eee elas 21 Prosiding seminar dan lokakarya nasional etnobotani, Cisarua-Bogor, 19-20 Februari 1992, edited by Rusdy E. Nasution, Soedarsono Riswan, Prabowo Tjitropranoto, Eko Baroto Waluyo, Wahyono Martowikrido, Harini Roemantyo, and Salikin S. Wardoyo Tr I 2 0o: sae ded Ya gh Golan te Sosy wae an A Te a ae a2 Crops and Man (Second Edition), by Jack R. Harlan ee © Oe Oe ee a a ee SE he LA a ee ea 42 Histoire Illustreé du Caoutchouc, by Jean-Baptiste Serier, Antionette Diez, and Anne Van Dyk ei RO or I lO ee ee ea nl em 57 The Ethnobotany of the Chacobo Indians, Beni, Bolivia, by Brian M. Boom Wa Pr es ee iG PS es 58 iv CONTENTS Vol. 14, No. 1 The Tasaday Controversy: Assessing the Evidence, edited by Thomas N. Headland JOR EB LINE 6 oe a a IY eR AGS ONO eR OSS 74 Ayahuasca Visions: The Religious Iconography of a Peruvian Shaman, by Luis Eduardo Luna and Pablo Amaringo; Amazon Healer: The Life and Times of an Urban Shaman, by Marlene Dobkin de Rios; Guiado Pelo Lua: Xamanismo e Uso Ritual da Ayahuasca no Culto do Santo Daime, by Edward MacRae nn le rose SG. Oe a CLEAR eee 106 Before the Wilderness. Environmental Management by Native Californians, edited by Thomas C. Blackburn and Kat Anderson GRE REE a tA ee a We ole Als Aes aN 110 Altrove 1: Societa Italiana per lo Studio degli Stati de Coscienza, edited by Claudio Barberi, Antonio Bianchi, Gilberto Camilla, Francesco Festi, Marco Margnelli, Bruno Pochettino, and Giorgio Samiorini TE CR ea ahs a he ek oe es ke ag ee ee Ge ie Handbook of Edible Weeds, by James A. Duke NE TE SS Sig ik ote Ck a ee eke ee A eee eee 115 The Palaeoethnobotany of Franchthi Cave, by Julie M. Hansen TAMER le a ee Bl ae iG ae gh i es el ee Gi 117 Kava: The Pacific Drug, by Vincent Lebot, Mark Merlin, and Lamont Lindstrom Aromatic Plants and Essential Constituents, by Zhu Liangfeng, Li Younghua, Li Baoling, Lu Biyao, and Xia Nianhe michard Engus Selsiee .....0 2 Sek ee Be 2 BL ek avodart 121 Ethnobotany of the Waimiri Atroari Indians of Brazil, by William Milliken, Robert P. Miller, Sharon R. Pollard, and Elisa V. Wandelli Pe DO nb KR ae Se we ee ee ee a Oe 122 Funghetti, by Silvio Pagani AEE 5 oo oe ee ee oe OL Se an ee 124 New Directions in the Study of Plants and People: Research Contributions from the Institute of Economic Botany, edited by G. T. Prance and M. J. Balick CORE I COIN: 6. so ween cane ee eee pre eer aie at Pun iiea Ney a ae 125 50 Jahre LSD-Erfahrung: Eine Jubilaumsschrift, edited by Christian Ratsch AOR CRE pe eae 2. SO eS oe eee Oe i2Z7 ZauberPilze, edited by Ronald Rippchen RE CE ite ia oe i ce ek ee ee aes ee 129 Lista Anotada de las Plantas Medicinales de Uso Actual en el Estado de Quin- tana Roo, México, by Ma. T. Pulida Salas and L. Serralta Peraza ewer! Dye and Edetoe Linares... se ek es oS ee aes 131 Plants of the Gods: Their Sacred, Healing, and Hallucinogenic Powers (Reprint of 1979 Edition), by Richard Evans Schultes and Albert Hofmann Jonathan OF ok oh ac ea eS ee ee ee eee 132 Les Plantes des Dieux: Les Plantes Hallucinogénes, Botanique, et Etnologie (Translation of 1979 Edition), by Richard Evans Schultes and Albert Hofmann Jonathan OH 8 8 ove we ow 5 a ee 134 Summer 1994 JOURNAL OF ETHNOBIOLOGY Vv Rivers of Change: Essays on Early Agriculture in Eastern North America, by Bruce D. Smith ities cath inc SEAS ee ee ee ee 136 Persephone’s Quest: Entheogens and the Origins of Religion, by R. Gordon Wasson, Stella Kramrisch, Jonathan Ott, and Carl A. P. Ruck eo oe 8 Bee a eee hee ee 137 Wild Seasons: Gathering and Cooking Wild Plants of the Great Plains, by Kay Young WME OIR ERNE 5G Se ay eh bien ole cee a ee a oh 138 Lei | \e/ ere We PBs L - —— AA! r" WE OD . ts "74> . YOANCEked LER] > ZS EDITOR'S VIEWS This issue of the Journal takes us from the lab to the field and from Mexico to New Guinea. The geographical and topical diversity of these offerings are matched by the diversity of backgrounds of the authors—the five articles for this issue are one reflection of the breadth of on-going research in ethnobiology. We begin in the lab with experimental research on charring maize. It is note- worthy that this research, which will have wide application in paleoethnobotany, was conducted by undergraduates under the direction of Christine Hastorf and Sissel Johannessen. The next three articles each echo, in varying degrees, the theme of sustainable use of natural resources: what can be learned from peoples living in traditional relationships with the natural world? The first takes us to the Sierra de Manantlan, Mexico, and a report of long-term, collaborative research into mestizo plant use by Bruce Benz and colleagues. George Estabrook then presents the implications of fuel choice on maintaining biodiversity in Portugal. Christopher Healey’s paper comparing patterns of exploitation of birds in Borneo and New Guinea brings out the importance of understanding the wider social and political contexts of human-animal interrelationships. We end with a paper by Steven Goodman and Joseph Hobbs on the ethnozoology of reptiles in Egypt; this original research contributes much to knowledge of the distribution and classification of this often understudied group. To facilitate finding reviews of books in which you are interested, beginning with this issue we are listing book reviews by title, author, and reviewer (rather than just page number). Note also that there is a change in how back issues of the Journal are distributed: in a moment of weakness Cecil Brown volunteered to store and send these out. Write him at the Department of Anthropology, Northern Illinois University, DeKalb, IL 60115-2854 (Ph. 815-753-0246) to inquire about the availability and price of issues you may be missing. Finally, I would like to welcome H. Sorayya Carr and Gayle J. Fritz to the Editorial Board of the Journal. These new board members represent an expansion of the board, to insure that manuscript flow in any one research area does not overwhelm an individual board member. Let me take this opportunity to thank all Board Members for their hard work, and especially their efforts to speed up the manuscript review process. It still takes longer than we'd like to get manu- scripts through the complete review process; I thank our contributors for their patience—everyone involved with reviewing, editing, and compiling the Journal are volunteers who squeeze the Journal into already busy schedules. DMP res pa ne ial ib hy RHE ee EIS ee ec. YS ea, re eRe nae SP emits aes eee i J. Ethnobiol. 14(1):1-21 Summer 1994 TOWARD RECONSTRUCTING ANCIENT MAIZE: EXPERIMENTS IN PROCESSING AND CHARRING SUSAN GOETTE Anthropology Department University of Minnesota, Minneapolis MN MICHELE WILLIAMS Anthropology Department Washington University, St. Louis MO SISSEL JOHANNESSEN Army Corps of Engineers St. Paul MN CHRISTINE A. — Anthropology Departme University of California, ee CA ABSTRACT.—We report the results of two experiments designed to assess the effects of processing and charring on maize fragments, so as to allow improved interpretation of maize remains recovered from archaeological sites. In the first experiment, kernels of three varieties of modern Andean maize were processed by three methods—toasting, sprouting, and boiling with wood ash—and then charred. The three processing techniques produced diagnostic characteristics that survived charring. It was also found that dimensional changes with charring were greater in processed kernels than unprocessed kernels. In the second experiment, after establishing a set of charring conditions, ears of six varieties of Andean maize were fragmented and the kernels and cupules measured before and after charring to determine the direction, degree, and variability of distortion. RESUMEN.—Reportamos los resultados de dos experimentos disefiados para evaluar los efectos del procesamiento alimentario y la carbonizacion de frag- mentos de maiz, a fin de permitir una mejor interpretaci6n de los restos de maiz provenientes de sitios arqueoldgicos. En el primer experimento, se procesaron granos de tres variedades de maiz andino contemporaneo mediante tres métodos: fueron tostados, germinados, 0 hervidos con ceniza, y todos fueron después carbonizados. Las tres técnicas de procesamiento produjeron caracteristicas que perduraron después de la carbonizacién y que pueden servir como diagnéstico. Se encontré también que los cambios de dimensiones ocasionados por la car- bonizacion fueron mayores en granos procesados que en granos no procesados. En el segundo experimento, después de establecer ciertas condiciones de car- bonizaci6n, se fragmentaron mazorcas de seis variedades de maiz andino y se midieron los granos y las cipulas antes y después de la carbonizacién para determinar la direcci6n, grado y variabilidad de la deformaci6n. 2 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF Vol. 14, No. 1 RESUME.—Nous reportons les résultats de deux expériences destinées a évaluer les effects d’utilization culinaire et de carbonization sur les fragments de mais, de facon a améliorer l’interprétation des restes de mais provenant de contextes archaéologiques. Dans la premiére expérience, les graines de trois variétes de mais moderne des Andes ont été grillées, germinées, et bouilliees avec des cendres de bois, puis carbonizées. Ces trois techniques ont produit des charactéristiques diagnostiques qui ont survécu la carbonization. Entre autre, les changements dans les dimensions dus a la carbonization sont plus importants dans les graines préparées que nonpréparées. Dans la deuxieme expérience, apres avoir établi certaines conditions de carbonization, les épis de six variétés de mais andéens ont été fragmentés et les graines et cupules mesurées avant et apres la carbonization, de facon a préciser la direction, le degré et la variabilité des distortions. INTRODUCTION Domesticated maize (Zea mays subsp. mays) achieved perhaps the widest prehistoric distribution of any New World crop, spread by human agency from Mesoamerica north to the boreal forests of Canada and south to Argentina and Chile. In the process the maize ear underwent phenotypic and genotypic varia- tion into a myriad of colors, sizes, shapes, and textures. Hundreds of maize varieties were in use in the Americas at the time of European contact. These varieties, created and maintained by human groups each for its own particular purpose, were cultural artifacts. The recognition of these varietal differences in ancient maize remains recovered from archaeological sites is important for under- standing the long and complex interaction between people and maize. The difficulty inherent in distinguishing maize varieties on the basis of the morphologies of a few specimens (Bird 1970; Bird and Goodman 1978; Goodman and Paterniani 1969; King 1987) is increased by the fact that in many archaeologi- cal sites maize is preserved only if it was charred in antiquity. Further, maize ears generally have been fragmented through processing, charring, and other deposi- tional forces into loose kernels, kernel fragments, and cob fragments consisting of the hard cupules that held the kernels. Several researchers (Benz 1994; Bird and Bird 1980; Cutler 1956; Cutler and Blake 1973; Pearsall 1980; Johannessen et al. 1990; King 1987, 1994; Miksicek et al. 1981) have addressed the problem of develop- ing methods of measurement and statistical analysis that can be applied to classi- fying these charred fragmentary maize remains. If archaeological maize types are to be reconstructed, it is essential to know how accurately measurements taken on charred fragments reflect the attributes of the original ear. Heating and charring distort the size and shape of kernels and cupules. Also, various types of maize processing undoubtedly changed kernel characteristics. The effects of charring and processing can be assessed by experi- mental means, and here we report on two experiments toward this end. The results should be understood only as one piece of the complex puzzle of recon- structing ancient maize. The myriad varieties of maize as well as variables in processing and charring conditions make it unwise to use the results as formulas to be applied to every set of archaeological maize fragments. Several previous researchers have approached this same problem. Cutler (1956) and Cutler and Blake (1973) give the approximate effect of charring on cob Summer 1994 JOURNAL OF ETHNOBIOLOGY 3 parts and on kernels, but it is not clear whether their conclusions are based on observations of archaeological material or on experimental reconstructions. Pear- sall (1980) experimented with charring and parching modern maize kernels in order to arrive at an appropriate adjustment to reconstruct the original size and shape of a cache of charred archaeological kernels. However, she was not success- ful in replicating the condition of the archaeological kernels, which appeared to have little distortion, pericarp splitting, or extrusion of the endosperm. After heating for 1.5 hours in sand over a Bunsen burner, the modern charred kernels were extremely swollen and broken, which Pearsall attributes to too high heat or too rapid heating. Modern kernels that were parched but not charred showed less distortion, and Pearsall estimated that ancient charring produced size increases of 5% in length, 10% in width, and 50% in thickness, percentages midway between the changes in the experimental charred and parched kernels. More recently, King (1987, 1994) has experimented with the effect of different charring regimes, as well as differences with variation in processing techniques (boiling and alkali treatment) and endosperm types. She too was unable to reproduce by experiment the condition of archaeological kernels, her various charring regimes producing kernels that were either uncharred, or broken and fragile (neither of which would survive well in the archaeological record). She therefore also used parched rather than charred kernels to estimate change with charring. She found, with some variation, for 105 kernels of seven different cultivars the width and length increased about 3% but the thickness increased about 38% (King 1987:136). She also found that kernels previously boiled or made into hominy showed greater change after charring than did charred unprocessed kernels (King 1987:146-147). An experiment by Benz (1994), which used entire cobs rather than fragments, found that the charred cobs, although variably distorted, were still readily distin- guishable as to race in a multivariate comparison. These studies form a useful basis for understanding the effects of charring and processing, each pointing out directions for further work. First, a method must be found to char maize that replicates the appearance of archaeological kernels. Further, in any experiment the sample size should be substantial and include several varieties, the methods described in detail, and full range of varia- tion in the resultant data presented. In the present experiments we contribute to improved interpretation of ancient charred maize fragments by refining our assessments of the effects of charring and of various maize processing techniques. The experiments were designed to aid in analysis of charred fragmented maize recovered from sites in the Mantaro Valley (Johannessen and Hastorf 1989), an intermontane Andean valley in Peru, and the results should be applied with caution to other situations. In the first experiment, which assesses the effects of processing, the shelled kernels of three varieties of Andean maize were processed by three common and ancient Andean techniques (toasting, sprouting, and boiling with wood ash) and the processed kernels were then charred. Kernels were measured and their char- acteristics were noted at each stage. In the second experiment, dealing with the effects of charring, a method of charring that replicates the condition of archae- ological maize was developed. Then a sample of over 400 kernels and 200 cupules from ears of six modern Andean maize varieties were measured, charred, and GOETTE, WILLIAMS, JOHANNESSEN & HASTORF Vol. 14, No. 1 remeasured, and the changes produced by the charring were analyzed statis- tically. These two experiments allowed us to replicate archaeological maize char- acteristics and provided insights into signs of maize processing that may remain in the archaeological record. EXPERIMENT ONE: THE EFFECTS OF PROCESSING In this experiment we wanted to see if different processing techniques resulted in distinctive kernels whose characteristics might be expected to survive charring and be distinguishable in the archaeological record. We tested the effects of three common Andean maize processing techniques on traditional varieties of modern maize. For each of the processes, we noted the appearance of the kernels before processing, after processing, and after the processed kernels were charred. Mea- surements and photographs were taken at each stage. The three traditional processing techniques chosen were toasting, boiling with wood ash, and sprouting. Currently in the Peruvian Andes maize is com- monly processed for kancha (toasted maize), mote (boiled hominy), and chicha or ahka (beer). For kancha, kernels of soft or sweet maize are parched in a clay vessel over the fire until crunchy. For mote, kernels are boiled with wood ash until the pericarp is loosened. These are rubbed off by hand, and the resulting maiz pelado (peeled maize) is dried for storage. For chicha, the kernels are soaked for several days and then kept moist until they germinate. The wifiapo, or sprouted maize, is dried, milled, boiled with water, strained, and fermented to make the beer (Bird 1970; Cutler and Cardenas 1947; Gade 1975; Mejia Xesppe 1978). The antiquity of these processes is apparent in their descriptions in early ethnohistoric documents, and in old Quechua names such as moti, camcha, and wifiapo aque (chicha from sprouted maize) (Horkheimer 1973). Garcilaso de la Vega, born of an Inca mother in 1539, describes the traditional preparation of motis, camcha, and wifiapo as outlined above; “all this,” he says, “I saw with my own eyes, and I was nourished until I was nine or ten with ¢ara, which is maize” (Garcilaso de la Vega 1985 [1609-1617]:341). Prehistoric maize processing has also been discernable in some archaeological examples. Dried germinated maize, pre- sumably stored for chicha-making and dating to ca. A.D. 900-1400, has been recovered uncharred under very good conditions of preservation on the arid coast of Peru (Moore 1989). In wetter highland sites, however, where charring is necessary for preservation, the recognition of processed maize is more difficult. We wanted to see if these processing techniques resulted in characteristic kernels even after charring. Materials and methods—Gade (1975), Bird (1970), Cutler and Cardenas (1947), Nicholson (1960), and Rick and Anderson (1949) discuss the varieties of Andean maize traditionally preferred for each of these three processes. For the experi- ment, we selected a characteristic variety for each process: the sweet corn Chullpi for making kancha, the large-kernelled flour variety Cuzco for mote, and a flour variety Huilcaparu for making chicha. The maize types were obtained in 1989 as shelled kernels in Bolivian markets in Cochabamba and La Paz. We chose shelled Summer 1994 JOURNAL OF ETHNOBIOLOGY 5 < FIG. 1—Kernels of three maize types at stages of the three processing techniques. A: unprocessed Chullpi (note the shriveled endosperm of this sweet corn); B: Chullpi toasted for kancha (shows vertical split down back of kernel); C: charred kancha (vertical split down embryo); D: unprocessed Cuzco; E: pericarp and point of attachment removed by wood-ash processing for mote; F: charred mote; G: unpro- cessed Huilcaparu; H: sprouted wifiapo for chicha (note pericarp over embryo pushed away by hypocotyl and radicle); I: charred wifiapo with hypocotyl and radicle burned away. maize over whole ears because sources indicate that maize is traditionally graded and stored after shelling, and it is then this sample that is processed. The Chullpi used for kancha is an Andean sweet corn with very long kernels (Fig. 1a). The kancha process is very simple; dried Chullpi kernels are placed in a clay pot (we used a kancha pot from the central Andes) over high heat. A handful of kernels is toasted in three or four minutes while stirring constantly. The result- ing maize is toasted yellow with browned areas scattered across the swollen surface of the kernel. A distinctive crack in the pericarp occurs lengthwise either down the embryo area or the back of the kernel due to the puffing of the formerly shrunken “sugar” portion of the kernel (Fig. 1b). The radicle of the embryo pro- trudes upward through the cracked pericarp in many specimens. Mote, which is similar to North American hominy, is prepared in the Vil- canota Valley of Peru with the large floury kernels of Cuzco (Fig. 1d) (Gade 1975). We used approximately one cup of hardwood ashes in two quarts of water to process 50-100 kernels of Cuzco. The water and wood ash form a lye solution with 6 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF Vol. 14, No. 1 a pH of about 10. Once the ash and water mixture is boiling the kernels are added. The pericarps began to loosen after ten minutes of boiling over a medium-high flame. The kernels were then rinsed under running water while being rubbed together, removing any remaining pericarps and many of the points of attach- ment (Fig. le). Maiz pelado has a distinctive “hominy” smell and is light buttery yellow in color. Traditionally the peeled maize is added to soups or dried and stored for later use (Gade 1975). A second boiling in soup causes an enormous expansion of the kernels as they absorb water; the characteristic puffy appear- ance of mote or hominy results. Huilcaparu is a variety widely grown and commonly used to make chicha in the Cochabamba Valley (Cutler and Cardenas 1947:250) (Fig. 1g). The chicha- making process is long and complex, as illustrated by Cutler and Cardenas (1947). Freshly sprouted kernels (the malted grains introduce the enzyme diastase that changes sugars to alcohol through fermentation) are dried and then milled. The resulting flour is boiled, allowed to settle, and the supernatant is removed for fermentation. The fermenting process takes 3-5 days. The chicha processing technique we used was as follows: the maize was soaked overnight in water and a vermiculite mixture, and then sprouted for five days at 25°C in the moist vermiculite. When the majority of the kernels (15-20% of the kernels did not germinate) had radicles as long as the body of the seed, they were removed from the vermiculite and air-dried overnight. During ger- mination the expanding radicle and hypocotyl pushed away the pericarp cover- ing the embryo (Fig. 1h). The moist sprouted kernels were swollen to the limits of their pericarps, causing a puckered appearance across the tops of the kernels that was retained after drying. As the kernels dried, the radicle and hypocotyl became very delicate and broke off easily as did the pericarp covering the embryo. Nicholson (1960) states that in modern Peru the broken embryo parts are collected and saved for chicha production. Our processing sequence stopped here since in the next step the kernels are milled. Nicholson (1960) indicates that in the Andes maize is often sold or stored in the sprouted and dried state, and as we have seen, prehistoric examples of maize stored in this state have been found. e products of these three processing techniques were then charred by the method described in the next section of this paper, in sand over a Bunsen burner with intervals of cooling. Samples of the sprouted kernels were charred both in the wet and dried state. The toasted kancha kernels unexpectedly took the long- est time to char—up to 60 hours. The sprouted kernels took 24—50 hours, and the peeled and dried mote kernels took only 12 hours to char. The amount of endo- sperm extrusion (that is, the percentage of all charred kernels in which the endosperm expanded greatly with the heat and bubbled out through a split pericarp causing a fragile and greatly distorted kernel) ranged from 5-35% overall and correlated with the processing method. The mote kernels had the lowest percentage similar extrusion percentages of about 10-15%. The dry- charred chicha kernels had a very high extrusion percentage of 20-35%. This variation in the percentage of kernels that extrude with charring may be a reflec- tion of the relative ability of kernels processed by various methods to become part of the archaeological record, since extruded kernels are very fragile and unlikely to survive. Summer 1994 JOURNAL OF ETHNOBIOLOGY 7 Chullpi (n=150) N & WwW @ coo }He6 coom| |} 10000 ooo | [me eae | centimeters as eS kencha 8 ods ee om Th N 18 8 : ° 8 c 16 8 : Cuzco Sia ] fa : 8 (n=75) — '2 en es aoc = ° aa 1 ° ° ° +4 8 8 o - 8 e qj : 7 8 is rocess: Eee cae leone : mote ° Huilcaparu r n=150) o wmr{] fro | | ° 1a] 8 1.6 1.4 : 8 8 ” 3 ~ Bee 2 12 Bot gy or a ERS 5 : ° ‘Bos 3 =. process: eo ’ eae — Eg rT} chicha 4 7, © 24 —length— — width— —thickness— [__] unprocessed processed processed and charred FIG. 2.—Range, variation, and change in the kernel measurments with process- ing and charring. The five bars of all box-plots mark the 10th, 25th, 50th (mode), 75th, and 90th percentiles. To determine the metric changes resulting from the processing techniques, the kernels were measured at each stage; unprocessed, processed, and charred- processed. The measurements taken were length, width, thickness, and the angle of the two long sides (see below and Fig. 4 for details). One hundred and fifty kernels each of Chullpi and Huilcaparu and 75 kernels of Cuzco were measured at each stage. Results.—Results of the maize kernel measurements are illustrated in Fig. 2, which shows the range of variability and change in length, width, and thickness with each stage of the three techniques. In general we see the same directional changes in the charred processed kernels as we do in charred unprocessed kernels (see below and Table 2). The greatest change is in increased thickness, with a slight decrease in length, and little change in width. Fig. 2 shows that in most cases the change in shape takes place during charring rather than processing. Table 1 gives 8 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF _ Vol. 14, No.1 TABLE 1.—Average percentage change in kernel variables with processing and charring. Maize type Chullpi Cuzco Huilcaparu (process) (kancha) (mote) (chicha) Dimensions r P&C \y P&C P P&C length —2.0 -8.0 +12 +4.3 1 OF width +1.0 +4.5 —2.3 —4.2 +4.0 -1.6 thickness +29.9 +39.0 +4.1 +38.0 +13.1 +42.8 angle +31.0 +68.0 +15.6 +10.3 +5.8 +5.5 P: processed; P&C: processed and charred. Figures are percentage change in mean dimensions from the unprocessed kernels. the mean changes in percentage for each type after processing and again after the processed kernels are charred. We can see by a comparison with Table 2 that the processed kernels get much thicker with charring than do the unprocessed ker- nels (mean increase of 40% as opposed to 13%), although the other dimensions undergo much the same amount of change. This confirms King’s (1987) findings that processing does have a role in determining charred kernel shape, and further specifies that the major change is in the thickness. Perhaps more important is the appearance of the charred and processed ker- nels (Fig. 1c,f,i). The toasted Chullpi kancha kernels kept their characteristic ver- tically cracked embryo or kernel back after charring. Even the protruding radicles survived charring intact. The browned and puffed areas of the pericarp became fragile after charring but the pericarp retained its integrity. Charring increased the overall puffiness of the kancha kernels. The sprouted Huilcaparu lost the delicate hypocotyls and radicles with charring, leaving holes where they had emerged from the embryo. The pericarp covering the embryo was also lost during charring. Processed and charred sprouted Huilcaparu kernels have a vertical crack down the embryo, similar to that of the toasted Chullpi kernels, but are distinctive in that their embryos are depleted and sunken. Those kernels most resembling archaeological maize were the carbonized lye-treated mote kernels. The endosperm of the Cuzco, having lost its restricting pericarp in processing, expanded greatly with charring. The expansion left the embryo with a sunken appearance. Although sunken, the embryo was still persistent on most kernels even after the boiling and charring processes. Of the three processing techniques, the mote kernels were the quickest to char and were the most durable after charring, thereby making them the strongest candidates for preservation. In addition, they show the closest resemblance to much archaeological maize in lacking their pericarps, often their points of attach- ment, and occasionally their embryos. Processing maize with wood ash was a widespread practice in the Americas; the results of this process could make u much of the maize debris recovered from archaeological sites. King (1987:146) also reached this conclusion as a result of her experiments. Summer 1994 JOURNAL OF ETHNOBIOLOGY Conclusions.—We believe that the products of chicha, kancha, and mote produc- tion, as produced in our experiments, would be distinctive in the archaeological record. The remains of chicha production could be identified by the distinctive radicle/hypocotyl holes and the missing embryo pericarp. The chicha characteris- tics might occur in any sprouted maize so archaeological context must be consid- ered. Kancha kernels might be less distinctive because unprocessed kernels also puff during charring. However, the protruding radicle and the embryo crack would be good distinguishing characteristics for kernels that had been quickly parched over a hot fire. Mote kernels were the most distinctive products of the three processes in lacking the pericarp, often the point of attachment, and occa- sionally the embryo. EXPERIMENT TWO: THE EFFECTS OF CHARRING This experiment was designed, first of all, to devise a system of charring that would replicate the appearance of most charred archaeological maize. This would allow a more realistic estimate of the amount of distortion produced by such a charring method, and also provide insight into the kinds of conditions that may have preserved the maize we find archaeologically. Further, the experiment was intended to assess the effects of this charring on samples of kernels and cupules from a number of maize varieties. We concentrated in this case on loose kernels and cupules, rather than whole ears or cobs, since in our experience most charred archaeological maize is found in a fragmented state. Overall, 434 kernels and 221 cupules from six maize varieties were measured both before and after charring. Materials and methods.—Specimens of six cultivars of modern Andean maize were used for this experiment. The varieties were selected to give variation in size, shape, and endosperm type so that differences in the effect of charring could be assessed. The six varieties are (1) Confite puntiagudo, a popcorn with small pointed kernels, (2) Chullpi, a many-rowed sweet corn, (3) an unnamed variegated flour variety with imbricated yellow and red striped kernels, (4) San Geronimo, a white flour variety, (5) Morocho, a flint type with characteristic round kernels, and (6) Cuzco morado, a dark red, large-kernelled 8-row flour variety (Fig. 3). The four endosperm types represented have the following characteristics. Popcorn grains are composed mostly of a very hard vitreous endosperm with a small amount of soft starch in the center. Steam generated in the soft center causes it to explode with heating. Flint-type kernels also have a hard translucent endosperm with starch in the center, the proportions varying by variety. In flour varieties the endosperm consists of soft starch. In sweet corn much of the sugars are not converted into starches with maturation, and the kernels are translucent and shrivelled when dry (Sturtevant 1899; Purseglove 1972:303-304). Two ears of each variety were selected to provide the kernels and cupules for analysis. The two ears from each variety had the following row numbers: Confite, one 12-row and one 16-row; Chullpi, one 14-row and one 16-row; variegated, both 10-row; San Geronimo, both 10-row; Morocho, one 10-row and one with very irregu- lar rows that was counted as 9-row; and Cuzco morado, one 8-row and one 10-row. The sample of kernels and cupules from the two ears of each maize variety is of FIG. 3.—Six Andean maize types used in the charring experiment. A: Confite puntiagudo; B: Chullpi; C: Variegated; D: San Geronimo; E: Morocho; F: Cuzco morado. Scale is in centimeters. Summer 1994 JOURNAL OF ETHNOBIOLOGY 11 width — ' Cupule angie KERNEL width Ty length center length i. top view ) }) eaked Cross section of maize ear a 994 round square bea Cap types FIG. 4.—Kernel and cupule measurements used. course by no means considered representative of the morphological variation within the variety as a whole; the emphasis here is rather on the change and variation that comes about with charring. The ears were all collected from Andean markets or farmers 6-10 years ago, and have since been stored in a herbarium cabinet, and thus were thoroughly air-dry. Twenty percent of the kernels and cupules from each ear were measured. First, the length, center width, row number, and total number of kernels of each ear were recorded. All kernels were removed from each ear and 20% of the total were selected by picking blind-folded from a box. Selection of the cupule sample was more difficult, since cupules cannot readily be separated from an uncharred cob. Therefore, before charring, the cupule measurements had to be taken on the cob, and access to the cupules became the limiting factor in determining the sample. We experimented with sawing (Benz 1986), hammering, and hand- breaking to expose cob cross-sections. Because of the alternating arrangement of the cupule rows, sawing damaged the walls of the cupules, and hammering mashed the cupules. Hand-breaking best exposed a cross-section with intact cupules. Three breaks were made of each cob, one in the center and one toward each end. The cupules in the six exposed cross-sections were those used in the study, for roughly a 20% sample from each ear. Measurements of the kernels and cupules taken before and after charring were those that can readily be taken on archaeological specimens (Fig. 4). Kernel measurements were length, width, thickness, and angle of the two long sides. Kernel cap types were coded as round, square, or beaked. Cupule measurements were width, height, depth, center length, wing length, and angle. All measure- ments except angles and cupule depth were taken with sliding calipers to the 12 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF Vol. 14, No. 1 nearest 0.05 cm. Angles were measured to the nearest 5° by laying the kernel or cupule on a piece of laminated polar coordinate graph paper (delineating the 360° of a circle), lining one of the long sides on the 0° line and moving the kernel or cupule until the other long side was flush with a degree line. Cupule depths were measured from the front lip to the deepest part of the cupule pocket using a calibrated metal probe. Previous experiments in charring maize have resulted in extensively swollen, broken, extruded and fragile kernels (King 1987; Pearsall 1980). Not only are such kernels unlikely candidates for preservation, but both King and Pearsall note that archaeological maize remains often appear well-preserved with little apparent distortion, although often the kernel embryos are missing as well as much of the pericarp. The ancient conditions of charring that produced such maize remains have thus far been unduplicated experimentally. In this experiment we tried a number of charring techniques to find that which produced the least fragile and least distorted charred maize fragments. The most successful method was slow charring in a reducing atmosphere at relatively low temperatures, with periodic intervals of cooling. Test kernels and cupules were charred in sand over a Bunsen burner at a temperature of about 180—190° C. The fragments were heated for 1.5 hours, allowed to cool completely, heated again for 1.5 hours, cooled, and so on until completely charred through to the center. This method took an average of 16 hours burning time to char one sample of kernels. Kernels charred at the same temperature but without the cooling inter- vals showed more frequent endosperm extrusion; 25% of the kernels as compared to 17% with the cooling intervals. Cupules were charred on the cob using the same method, with a shorter charring time of 10-12 hours per cob. We found that kernels and cupules left to burn after they have been fully charred retain their integrity. They do not disintegrate, crumble, or become more fragile. The kernels and cupules from the six modern Andean maize varieties were then charred following this successful method, and the charred fragments were remeasured. Change in kernels and cupules.—The maize kernels generally became shorter, wider, and thicker with charring. The percentile box-plots in Fig. 5 show the range of variation and the change with charring in length, width, and thickness for each of the six maize types. The box-plots have the properties of (a) showing the central tendencies and full range of variation for the samples for each maize type and each variable, and (b) allowing comparison of sizes and shapes among varieties. Overall mean change for the kernels consisted of a 6% decrease in length, width was minimally affected with an increase of only 1%, and thickness increased most to an average of 13% (Table 2). These findings differ from previous experiments using parched maize (King 1987; Pearsall 1980), where it was reported that all three dimensions increased. The measured angles of the kernels undergo a slight average increase of about 6%. The kernel caps tended to become slightly more round; 8% of the square cap types and 6% of the beaked types became round after charring. Two indices useful in describing the shape of kernels from the front and the top, ratios of width/length and width/thickness respectively, also change with charring. Since width generally increases as length decreases, the width/ Summer 1994 JOURNAL OF ETHNOBIOLOGY TABLE 2.—Average percentage changes due to charring in six maize types. Figures show direction and percentage of change before and after charring. Dimensions Confite Morocho Chullpi Cuzco San Ger Variegated All Ears Kernels (n=94) (n=64) (n=85) (n=75) (n=72) (n=54) (n= 437) Length —14 “ae ie —4.1 —6.8 —8.8 =55 Width —0.4 +2.6 +20 0.0 +0.8 —0.6 +12 Thickness +10.4 +172 +131 +20.6 +10.0 +51 +125 Angle ae +6.7 +1.6 “8 +e +13.4 +55 Cupules (n= 47) (n=32) (n=42) (n=37) (n=36) (n=27) (n=221) Height +6.0 -14.9 91 = - =i2? “13 — Sa Width ae Nd wal 0 +63 “8G 174 —8.4 aed Depth 30.5 —18.2 “eS (=a —4.0 24 30.5 Center Length +9.6 +148 +448 +234 +267 +57.8 +26.4 Wing Length +7 +14.6 +44.2 +0.6 +25 +37.0 tinge ua Confite Sf = - Sendai a P - 2 ae i} s&s Cia ee ‘7 Fee hl ae. ss 7 8 Chullpi 14. * . Morocho tL 7 [-=-= ee s 1 ry e ° = 1 = = ° E ee er are 2 ea ss sa item ai ria? Mi - 3 Variegated if = z pone Pi: 7 Se tel ty: PP, i- pO Ps - o 6 . 7 — é length width thickness : length width thickness [_] uncharrea charred FIG. 5.—Range and variation (shown by percentile box-plots) in kernel measure- ments for each maize type before and after charring. 14 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF _ Vol. 14, No. 1 eee en See es Confite San ear one Geronimo centimeters nv centimeters an eo HL» ° CE oo 2 2 x ie ee oe | ==) o 2 SS Be , gigri Confite ° Morocho ear two centimeters a o +f} afJo centimeters o [He -+{Th oo ee ° ° Chullpi centimeters o ao ° (oh: centimeters on oe -_= {Th ° HE} ooo =| ° 3 a ° (ae aaa ae or 1@ 4 £9 ine -eg hd] “lee Pa7 4 ite afew ee eet A Laren height width depth fenter IP . Variegated 5. ae ae uncherred : charred 3 4 2 a | 21 3 4 fj ga CJ = A 7 height width depth tenga dion FIG. 6.—Range and variation in cupule measurements for each maize type before and after charring. Measurements for the two ears of Confite are shown sep- arately, because it was found after removal of the kernels that the cupules of the two ears were quite different. length ratio increases an average of 7%. Since thickness increases more than width, the width/thickness ratio decreases due to charring by an average of 12%. Fig. 5 and Table 2 show that although kernels of most varieties show the same general tendency for increase or decrease in each variable, the amount of change varies considerably. The amount of change does not correlate with endosperm type; the three flour varieties together show a wider range of variation than among the three other endosperm types (pop, flint, and sweet). For the cupules, we found that the cupule height, depth, and width decreased while the center and wing lengths increased with charring (Fig. 6). Mean overall change consists of a 8% decrease in height, a 30% decrease in depth, a 10% decrease in width, a 30% increase in center length, and a 16% increase in wing length. Table 2 shows that, again, while the direction of change was consistent in most varieties the amount of change varies greatly. Summer 1994 JOURNAL OF ETHNOBIOLOGY 15 Predicting row number from angle —The row number of a ear of corn is held to be among the more reliable indicators of its variety (Bird 1970; Cutler and Blake 1973; Goodman and Paterniani 1969). With fragmented archaeological maize where the row number can no longer be counted from the ear or cob, many researchers have used the angle measured on the two long sides of a kernel or cupule to approximate the original row number. This is based on the portion of the 360° occupied by the kernel or cupule; in an eight-row ear, for example, a kernel will occupy 45° (one-eighth of a circle) and a cupule 90° (each cupule bears two kernels). However, this angle can be affected by the fact that kernels are generally offset slightly so that their edges do not abut, and in some types the rows are irregularly arranged. Thus the generalization that row number = 360/ angle-of-kernel (or row number = angle-of-cupule X 2) does not necessarily reflect reality. Some attempts have been made to assess the accuracy of the angle method of determining row number. Pearsall (1980) measured the angles of 25 kernels from ears of known row number (8, 10, 12, and 14-row cobs), with limited success in predicting row number. She found that the 8- and 10-rowed measured as either 8 or 10 but couldn’t be further segregated; that kernels from 12-row ear measured as 8, 10, or 12-row; and that the 14-row measured fairly accurately. The measurements were made by the Cutler and Blake (1973) method of best-fit of the kernel to a number of pre-cut angles of 45° (8-row), 36° (10-row), and so on. Bohrer (1986), in an experiment measuring the kernels from two cobs (12- and 14-rowed), found that only 31% of the kernels from the 12-row cob and only 7% of the kernels from the 14-row cob gave measured angles that would have classified the kernels correctly. She does not state how the angles were measured. Kin (1987:128-129) found from measuring the angles of 160 kernels of eight varieties (measured by photo-copying the kernels and then drawing and measuring the angles), that 68% of the kernel angle measurements resulted in incorrect row number determinations. We tested the degree of accuracy of using angles to predict row number by plotting the measured angles against the angles calculated from the actual row number of the original ears. Fig. 7 shows the scatterplots, regression, and correla- tion for actual vs. expected angle for the kernels (uncharred and charred) and for the charred cupules (uncharred cupule angles could not be measured since they couldn’t be detached from the cob). The figure shows considerable variation in the measured angles of kernels from ears of the same row number, and overlap between and among row numbers. For uncharred kernels, only about 43% of the total variation in measured angles is attributable to the difference in row number. The predictive value of angles measured on the charred kernels was somewhat better (R? = .57). The angles measured on charred cupules were the best predictors of row number (R? = .64). This may be due to the fact that the expected angles of cupules are farther apart than those of kernels by a factor of two. In other words, 60° and 72° (cupule angles from 10- and 12-row cobs) are more easily discrimi- nated than 30° and 36° (kernel angles from 10- and 12-row cobs). An analysis of variance, however, does reveal significant differences (at 95%) in angle measurements between samples from most row numbers. The mean angle measurements of the kernels and cupules were very close to their expected angles (Fig. 8) in most cases. However, the kernels from the 14-row ears and the measured angle (°) measured angle (°) measured angle (°) GOETTE, WILLIAMS, JOHANNESSEN & HASTORF 16-row z ° a N y = 1.04x - 4.434, R-squared: 43 10-row 9-row* 8-row N o y = 1.166x - 6.914, R-squared: 566 35 40 expected angle (°) 16-row 14-row y = 1.036x - 4.231, R-squared: . 2-row 10-row 8-row eseysxasees nN > + Oo 4s so) OSS 70 #75 80 expected angle(°) uncharred kernels (n=437) charred kernels (n=434) charred cupules (n=221) Vol. 14, No. 1 FIG. 7.—Describing variance in measured angles. Scatterplots, regression, and correlation of the measured vs. expected angles assess the accuracy of using kernel and cupule angle measurement to reconstruct row number. Overlapping points are indicated by “sunflowers.” *One ear with uneven and varied rows was counted as a 9-row ear. Summer 1994 JOURNAL OF ETHNOBIOLOGY 17 o o ce | a coal pe L 1 L l | l 504 as ° 404 L ~ . CHARRED is ' | es " 30) > ian 254 © ie : ¥ 15 n=29 n=37 n=190 n=35 n=42 n=102 5 . 3 Ps Ps ° ° S So ° ° ROW S ° = C c = 1 NUMBER wo an ° N a © 100 @ on o i aa o 904 L 80 + oO ” . } é | 3 CHARRED > CUPULES . ) r (n=221) 0 } : ée ¥ zn ov net n=19 n=98 n=19 n=21 n=49 mean and 95% i ig intervals of measured angle © expected angle Gh difference significant at 95% sa difference not significant at 95% FIG. 8.—Analysis of variance in measured angles. The figure shows the mean angles and 95% confidence intervals of kernel and cupule samples from different row numbered ears, compares sample means to expected angles, and indicates the significance level of differences between groups (derived from analysis of variance). 18 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF _ Vol. 14, No. 1 cupules from the 9-row (one ear of Morocho had irregular and varied rows and was counted as being 9-row) had measured angles lower than expected, and thus the differences between the 14- and 16- row kernels and the 9- and 10-row and the 12- and 14-row cupules were not significant statistically. This suggests that measured angles from kernels and cupules often do reflect the actual row number on a statistical basis, although measurements on individual fragments have limited accuracy. A note on processing —An unexpected outcome of the experiment was the condi- tion of the charred kernels. Of over 400 kernels, none lost their pericarps or embryos during the burning process. Charred archaeological kernels, otherwise well-preserved, are often missing all or most of their pericarps and often their embryos, and we have tacitly assumed that these were lost during the charring process. However, this may not be so, since in this experiment every kernel without exception retained its pericarp and embryo intact, even after 20 hours of burning. It seems plausible that the loss of pericarp and embryo from archae- ological kernels with minimal distortion may have resulted from processing before they became charred, rather than from the charring itself. This bears out the findings from the experiment described above, from which we concluded that kernels processed with lye to remove their pericarps have the best chance of being preserved in the archaeological record. SUMMARY, DISCUSSION, AND CONCLUSIONS Our first experiment resulted in distinctive appearances for charred maize kernels subjected to three common Andean forms of processing. There is pres- ently enormous variation in the maize varieties preferred for mote, kancha, and chicha, and it is probably safe to assume that the prehistoric people of the Andes had as widely varied tastes as the modern residents. Since we know that different types of maize react somewhat differently to charring, we cannot make a direct comparison between the appearance of archaeological maize and our modern processed and charred maize. However, we believe that the distinguishing char- acteristics described in Experiment One are the results of the processing method and not just the maize variety. We cannot predict how hundreds of years in the soil affect charred maize. The grinding force of freezing and thawing could wear away at the persistent but fragile pericarps of kancha and chicha kernels leaving them naked like mote kernels. The embryos of the kernels could be preferred by animals or soil microbes, removing them before complete charring occurred. We can only sug- gest the many forces that could occur before, during, and after deposition, and thus cannot make direct comparisons between experimental processed and charred kernels produced in the laboratory and those recovered from archae- ological sites. But we do know that charred processed maize is quite distinct from unprocessed charred maize, and that the results of the three methods were dis- tinct from each other, and that these attributes may be preserved in ancient kernels. Finally, we gnize mote or hominy (pericarps removed by boiling with lye) as the process that results in maize most likely to be preserved and most resembling the condition of archaeological kernels. Summer 1994 JOURNAL OF ETHNOBIOLOGY 19 In the second of these experiments we found a method of charring maize successfully, and assessed the effects of charring on the size and shape of the kernels and cupules of six varieties of maize. Long, slow, intermittent heating ina reducing atmosphere produced charred kernels and cupules that were not exces- sively distorted or fragile; i.e., were good candidates for the kind of long-term preservation we see in archaeological maize fragments. These conditions may be similar to those that charred the ancient maize that we find—burial in soil or ash near hearths that were periodically kindled. Our research question concerned the degree to which measurements taken on charred maize fragments accurately reflect attributes of the original material. We found (a) that all measurements taken ona kernel or cupule are not affected to the same degree by charring, therefore the fragments change in shape as well as size; (b) kernels generally stay about the same width, and get slightly shorter and quite a bit thicker, while cupules decrease in all dimensions except length; (c) whereas the direction of change in a certain dimension was generally the same for the six varieties, we found considerable variation in the amount of change among vari- eties. Therefore no formula will entirely accurate in reconstructing the precharred attributes of maize kernels or cupules. Our results suggest generally that kernels stay about the same width, get about 5% shorter, and about 15% thicker, thus affecting the shape of the kernels, especially the width/thickness ratio. The height and width of cupules were found to decrease about 10%, and the length to increase roughly 20%, so that the most radical change is in the shape of the cupule from the top, becoming longer and narrower. We also found, as have previous researchers, that angle measurement of single kernels and cupules is not an accurate predictor of row number, having about 50% or less chance of accuracy with kernels and a somewhat better chance with cupules. However, the mean of a sample of measurements does often accurately reflect the row number. Our experiments have detailed the effects of only a few of many possible conditions of charring and processing. The variation in morphological changes in charred maize kernels and cupules, as well as the many unknown biases that occurred in the processes of deposition, preservation, and recovery, make the reconstruction of ancient maize types and their uses a daunting problem. We believe that the most fruitful avenue of research lies in a combination of multivari- ate statistical analysis of large systematic samples of ancient maize fragments, and continuing experimental work, not only on the conditions producing mor- phological change, but on the structure of variability within and between maize types (Johannessen et al. 1990; Johannessen and Hastorf 1989). All will agree that it is risky to draw conclusions about types and usage from a small sample of archaeological maize fragments. Nevertheless, standardized description and reporting of the raw data (so as to give the full range of vari- ability) from even small samples can eventually build up large data bases that in conjunction with results of experiments on modern maize can give a more confi- dent picture of local and regional patterns of ancient maize use. 20 GOETTE, WILLIAMS, JOHANNESSEN & HASTORF Vol. 14, No. 1 ACKNOWLEDGEMENTS We wish to thank Frances King and Deborah Pearsall for advice in designing the experiments. The research represents two Research Experiences for Undergraduates grants awarded under National Science Foundation grant BNS 84-51369. LITERATURE CITED BENZ, BRUCE F. 1986. Taxonomy and Evolution of Mexican Maize. Unpub- lished Ph. D. Dissertation, Department of Botany, University of Wisconsin, Madison. 994. Can prehistoric racial diver- sification be deciphered from burned corn cobs? Pp. 23-33 in Corn and Cul- ture in the Prehistoric New World. Sis- sel Johannessen and Christine A. Hastorf (editors) Westview Press, Boulder, Colorado. BIRD, ROBERT McK. 1970. Maize and its Natural and Cultural Environment in the Sierra of Hudnuco, Peru. Unpub- lished Ph. D. Dissertation, Department of Anthropology, University of Cali- fornia, Berkeley. and JUNIUS B. BIRD. 1980. Galli- nazo maize from the Chicama Valley, Peru. American Antiquity 45:325-332. BIRD, ROBERT McK. and MAJOR M. GOOD 1978. The races of maize: V. Grouping - maize races on the basis of ear morphology. Economic Botany 31:471-481. BOHRER, VORSILA L. 1986. Guideposts in ethnobotany. Journal of Ethnobiol- ogy 6:27-43. CUTLER, HUGH C. 1956. Plant remains. Pp. 174-183 in Paul S. Martin, John B. Rinaldo, Elaine A. Bluhm, and Hugh C. Cutler: Higgins Flat Pueblo, West- ern New Mexico. Fieldiana: Anthro- pology Series tgs 45, Chicago Natural History Muse and LEONARD BLAKE. 1973. Plants from archaeological sites east of the Rockies. Missouri Botanical Gar- den, St. Louis. CUTLER, HUGH C. and MARTIN CAR- DENAS. 1947. Chicha, a native South American beer. Harvard University, Botanical Museum Leaflets 13:33-60. GADE, DANIEL W. 1975. Plants, Man and the Land in the Vilcanota Valley of Peru. Dr. W. Junk B.V., The Hague GARCILASO DE LA VEGA, EI Inca. 1985 [1609-1617]. Comentarios reales de los Incas. Biblioteca clasicos del Peru No. 1., Editorial Andina, Lima. GOODMAN, M.M. and E. PATERNIANI. 1969. The races of maize: III. Choices of appropriate characters for racial class- ification. Economic Botany 23:265- 27d. HORKHEIMER, HANS. 1973. Alimenta- cién y obtencién de alimentos en el Peru prehispanico. Direccién Univer- sitaria de Biblioteca y Publicaciones de la Universidad Nacional Mayor de San Marcos, Lima. JOHANNESSEN, SISSEL, SUSAN GO- ETTE, and CHRISTINE A. HASTORF. 1990. Modern and ancient maize frag- ments: An experiment in variability. Journal of Quantitative Anthropology 2:179-200. ee ee SISSEL and CHRIS- . HASTORF. 1989. Corn and aaeae in central Andean prehistory. Science 244:690-692. KING, FRANCES B. 1987. Prehistoric Maize in Eastern North America: An Evolutionary Evaluation. Unpub- lished Ph. D. Dissertation, Department of Agronomy, University of Illinois, Variability in cob and kernel characteristics of North American maize cultivars. Pp. 35-54 in Corn and Culture in the Prehistoric New World. Sissel Johannessen and Christine A. Hastorf roan Westview Press, Boulder, Colora MEJIA XESPPE, TORIBIO. 1978. Kausay. Alimentacién de los Indios. Pp. 207- 225 in Tecnologia Andina. R. Ravines (editor). Instituto de Estudios Per- uanos, Lima. MIKSICEK, CHARLES H., ROBERT McK. ) PICKERSGILL, SARA DONAGHEY, JULLIETTE CART- WRIGHT, and NORMAN HAM- Summer 1994 MOND. 1981. Pre-Classic lowland maize from Cuello, Belize. Nature 289:56-59. MOORE, JERRY D. 1989. Pre- “Hispanic beer on the coast of Peru. American Anthropologist 91:682-695. NICHOLSON, G.E. 1960. Chicha maize types and chicha manufacture in Peru. omic Botany 14:290-299. JOURNAL OF ETHNOBIOLOGY 21 cache from Manabi Province, Ecuador. Economic Botany 34:344-351. Bus woah J.W. 1972. Tropical Crops: onocotyledons. Longman, London ene CHARLES M. and EDGAR AN- DERSON. 1949. On some uses of maize in the sierra of Ancash. Annals of the Missouri Botanical Garden 36:405-412. STURTEVANT, E.L. 1899. Varieties of corn. con PEARSALL, DEBORAH M. 1980. Anal- ysis of an archaeological maize kernel USDA Experiment Station, Bulletin 137. BOOK REVIEW Oat Bran. Peter J. Wood (editor). St. Paul, Minnesota: American Association of Cereal Chemists (3340 Pilot Knob Road, St. Paul, MN 55121-2097). 1993. Pp. 164. $90. No ISBN given. This compact volume of six chapters by nine recognised experts satisfies a need which is clearly set forth in the Foreword by the editor, Dr. PJ. Wood: “In 1989, the public appetite for oat bran was at its peak. Both the product itself and media reports describing miraculous health benefits were avidly consumed. . . . the American Association of Cereal Chemists. . . . suggested that a book be compiled that would attempt to describe the nature of oat bran, its means of manufacture and properties and what was known about its physiological effects.” This book fully satisfies the worthwhile attempt to set forth the actual facts and it does it with full coverage of the subject. The chapters describe: 1) Structure of Oat Bran and Distribution of Dietary Fiber components (R. Gary Fulcher and S. Shea Miller); 2) Current Practice and Novel Processes (D. Paton and M.K. Lenz); 3) Comparisons of Dietary Fiber and Selected Nutrient Compositions of Oat and Other Grain Fractions (J.A. Marlett); 4) Physiochemical Characteristics and Physiological Properties of Oat (1-3), (1-4)- B-D-Glutean (PJ. Wood); Physiological Responses to Dietary Oats in Animal Models (F.L. Schinnick and J.A. Marlett); 6) Hypocholesterolemic Effects of Oat Bran in Humans (J.A. Anderson and S.R. Bridges). Each chapter contains a com- prehensive bibliography of literature cited, and there follows a detailed index. The American Association of Cereal Chemists has published a number of outstanding books which I have reviewed. I consider this volume to be one of the finest, particularly from the point of view of coverage and presentation of the latest scientific data which corrects some of the misunderstandings and misinfor- mation that has been in circulation. Richard Evans Schultes Botanical Museum of Harvard University Cambridge, Massachusetts 02138 ye BOOK REVIEW Vol. 14, No. 1 BOOK REVIEW Prosiding seminar dan lokakarya nasional etnobotani, Cisarua-Bogor, 19-20 Februari 1992. Rusdy E. Nasution, Soedarsono Riswan, Prabowo Tijitro- pranoto, Eko Baroto Waluyo, Wahyono Martowikrido, Harini Roemantyo, and Salikin S. Wardoyo (editors). Jakarta: Departemen Pertanian dan Lem- baga Ilmu Pengetahuan Indonesia (LIPI) (Herbarium Bogoriense, Jalan Raya Juanda 22-24, Bogor, Indonesia), 1992. Pp. viii, 432. $11.25 U.S. (Rp 22500) (paperbound). ISBN 979-8006-71-2. Indonesia is biologically and culturally one of the most diverse countries in the world. An archipelago of over 17,000 islands covered largely by tropical rainforest, it contains upwards of 30,000 plant species and over a hundred distinct cultural units. Not surprizingly, there is a great deal of ethnobotanical knowledge present in the islands, much of which is being lost very rapidly. Some of this information has been collected by Indonesian ethnobotanists across the country. To date, the re- sults of these studies have been scattered across publications of a variety of disciplines, mostly Indonesian publications. In 1992, however, a conference was held in Bogor to attempt to bring together ethnobotanists from across the country. This book is a collection of 61 papers presented at the conference. All the papers are by Indonesian authors, and all are in Indonesian with English abstracts. Most of the papers are purely descriptive, some containing long tables listing names and uses of plants in one or another section of the country. Discussed also are plants used for food, fiber, medicine, contraception, and cosmetics, as well as in ethnoveterinary treatments and in traditional ceremonies. There is little of the theoretical bases pervasive in much European and North American ethnobotani- cal literature, such as ethnotaxonomy, chemical ecology, or optimization theory. Nevertheless, the book is remarkable for the breadth and diversity of topics covered, and for the wealth of valuable information contained in its pages. This book will prove to be immensely useful to anyone working on the eth- nobotany of insular Southeast Asia. Such progress deserves to be encouraged and fostered. Joseph E. Laferriére Arnold Arboretum of Harvard University 22 Divinity Ave. Cambridge MA 02138 USA J. Ethnobiol. 14(1):23-41 Summer 1994 CHARACTERIZATION OF MESTIZO PLANT USE IN THE SIERRA DE MANANTLAN, JALISCO-COLIMA, MEXICO BRUCE F. BENZ FRANCISCO SANTANA M. ROSARIO PINEDA L. JUDITH CEVALLOS E. LUIS ROBLES H. DOMITILA DE NIZ L. Instituto Manantldn de Ecologia y Conservacion de la Biodiversidad Universidad de Guadalajara Apartado Postal 1-3933 Guadalajara, Jalisco MEXICO, C.P. 44100 ABSTRACT.—Ethnobotanical research in the Sierra de Manantlan Biosphere Re- serve seeks to promote a local conservation ethic through acknowledgement, documentation, and application of existing indigenous knowledge and use of the local flora by the rural population. Use of and knowledge about the native plant species has been documented in nine rural communities over a three year period through interviews with more than 100 informants. Informants have been selected on the basis of their self-acknowledged experience and willingness to collaborate. More than half of the more than 650 plant species discussed in interviews have been reported to be employed for one or more pur Knowledge of a plant species’ use appears to be related to relative floristic abundance while various categories of use tend to focus on certain specific vege- tation types. The most frequently cited species are those which are either natu- rally widely distributed or respond positively to human disturbance. Information elicited from more than 100 informants suggests that a considerable amount of empirical knowledge is not shared among informants. For example, more than 20% of the species reported as useful are reported as such only by individual informants. This pattern appears to be independent of the rural community or general use category examined. Such idiosyncratic variability may stem from active experimentation by individuals or from local erosion of traditional knowl- edge through acculturation. RESUMEN.—I tigaci tnobotanicas en la Reserva de la Biosfera Sierra de Manantlan intentan promover una ética conservacionista local a través del reco- nocimiento, documentaci6n, y aplicacién del conocimiento existente y uso de la flora local por parte de la poblacién rural. El uso y conocimiento concerniente a las especies nativas de plantas se ha documentado por medio de entrevistas con mas de 100 informantes en nueve comunidades a través de tres afios. Se selec- cionaron los informantes en base de su propio conocimiento tematico y su dispo- sicién a colaborar. Mas de la mitad de las 650 especies de plantas utilizadas en entrevistas han sido reportadas como utiles para uno 0 mas propésitos. 24 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 Parece que el uso de las plantas depende de la abundancia relativa en la flora y varias categorias de uso parecen enforcarse en las especies de ciertos tipos de vegetacion. Las especies mas frecuentemente citadas como ttiles son aquellas que tienen una distribuci6n geografica amplia o responden positivamente a la pertur- bacion antropogénica. Informacién obtenida de informantes indica que una can- tidad considerable de conocimiento empirico no esta compartido entre ellos. Por ejemplo, mas de viente porciento de las especies reportadas como Utiles se reportan como tal solo por informantes individuales. Este patron parece ser inde- pendiente de la comunidad o categoria general de uso examinado. Tal variabi- lidad de idiosincratismo podria deberse a la experimentaci6n activa o de erosién de conocimiento tradicional impulsado por la aculturaci6n. RESUME.—La recherche ethnobotanique an sein de la Sierra de Manantlan cher- che a promouvoir une éthique de conservation locale en s’appuyant sur les con- naissances existantes et |’utilisation de la flaure locale par la population au- tochtone. Des interviews ont été réalisées avec plus de 100 informateurs dans neuf communautés et sur une période de trois ans afin de connaitre les espéces de plantes originaires et de savoir leur utilisation. Les informateurs ont été sélec- tionnés en fonction de leur connaissance thématique et de leur disposition a répondre. Plus de la moitié des plus de 650 espéces de plantes mentionnées dans les questionnaires sont utilisées pour une ou plusieurs fins. L'utilisation des espéces de plantes semble dépendre d’une abondance flauris- tique relative; et certains types d’utilisation semblent dépendre de certains types de végétation. Les espéces les plus fréquemment utilisées sont celles que l’on rencontre en abondance de facon naturelle, ou qui réagissent positivement a des perturbations d’origine humaine. Les renseignements obtenus des informateurs montrent qu’un nombre considérable de connaissance empirique nest apparem- ment pas divulgué entre les informateurs. Par exemple, 20% des espéces reportées comme étant utiles sont mentionnées par un seul et unique informateur. Ceci semble étre indépendant de la communauté ou du type d'utilisation examiné. Une telle variabilité idiosyncratique pourrait étre ralentie 4 travers une expérimen- tation active ou une érosion des connaissances traditionnelles par acculturation. INTRODUCTION The Sierra de Manantlan is situated along the border of Jalisco-Colima ap- proximately 50 km north of the port of Manzanillo and 20 km west of Volcan Colima (Fig. 1) in western Mexico. This small mountain range is situated at the confluence of three of Mexico’s major mountain systems: at the western margin of the Mexican Neo-volcanic axis, at the southern end of the Sierra Madre Occiden- tal, and at the northern-most extent of the Sierra Madre del Sur (Rzedowski 1978; Tamayo 1980). Recognition of the biological importance of this mountain range led to its being set aside to conserve its remarkable biodiversity (Jardel 1992). In fact, the present-day vegetation of this region, a mosaic of eight broadly defined types (Rzedowski 1978), contains a veritable wealth of plant and animal species, with more than 2500 species of vascular plants and 668 species of vertebrate fauna so far listed (Vazquez et al. 1990; Jardel 1992). The discovery of Zea diploperennis Iltis, Doebley, and Guzman, an endemic diploid perennial wild relative of maize (IItis et al. 1979; Iltis 1980) provided the initial impetus for its preservation and eventu- | % ' Po =. CHANTE =: | \I . / a 3 x Ce @ZENZONTLA “if e . f {/BARRANCA DE L a.cAcen ¢ NARANJERA Ue, ee > 19°30! — ’ 1 * sated ee A ee N EVADO da v 4 = + oo ili Vv. AND “arora ELCRUZ | . TERREROY bro ? il @ \ COLIMA al ) * \ er Xs a3 am a Yok tal | eae f ral # 3 S34 ia ia ase | i , Y / | | 8 a XQ i | -¢ ear | | & ee ; | y 02-5 10 20 | KMS FIG. 1—Geographic location of the Sierra de Manantlan Biosphere Reserve in western Mexico. Communities where informants were consulted are indicated by small circles. 26 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 ally for the federal decree establishing the Sierra de Manantlan as a Mexican Biosphere Reserve (139,000 ha; see Iltis 1980; Jardel 1992) and its eventual inclu- sion within UNESCO's Man and the Biosphere network of reserves. For millennia, the forested slopes of these mountains have provided many of the natural resources—agricultural soils, animal forage, and hunted and gath- ered products—nearby communities depend upon. Second, the forested slopes supply considerable quantities of runoff to three regionally important water- sheds, the Ayuquila-Armeria, the Marabasco, and the Purificacién, rivers that have been the basis for irrigation-based agriculture since before the arrival of the Spanish (Kelly 1945, 1949; Sauer 1948). Aside from the obvious economic motives for promoting a conservation and social development program in this mountainous region (Jardel 1992), the rich biological endowment of the Sierra de Manantlan Biosphere Reserve (SMBR) has proven to be exceedingly important for stimulating efforts to prevent local extinc- tion of many of the organisms that occur here and nowhere else. The objectives of the present study have been defined in the context of aims of the SMBR itself, which seek to integrate social with economic development and conservation to ensure that the local population adopts and/or maintains sustain- able practices of natural resource use and thus a sustainable environment. Goals of our ethnobotanical research are to describe existing patterns of plant utilization in and around the SMBR in pursuit of locally adapted and appropriate land use alternatives and to ascertain whether existing exploitation practices in any way threaten present or future natural resource availability. Our research focuses on describing the intensity of utilization of the species recognized as useful by the local inhabitants, and subsequently evaluating it to predict whether these utiliza- tion practices might conflict with the conservation objectives of this protected area. Our research also seeks to discern the structure of plant resource knowledge among the local inhabitants. Although our methodology initially sought to cor- roborate information provided by individual informants, the data obtained thus far suggest that such corroboration is relatively infrequent and variation between informants much more prevalent. In the following essay we evaluate plant use with respect to (1) the relative importance of plant families according to the abundance of utilized species, (2) the patterns of use with regard to vegetation type, (3) the intensity of use based upon the frequency of report of utilization, and (4) informant idiosyncracy in describing a species’ utility. THE AREA AND ITS PEOPLE The Sierra de Manantlan, like much of western Mexico, has been inhabited for at least the last 2000 years (Kelly 1945, 1949, 1981). At the time of Spanish contact, the population in the region was widely scattered with only the valley of Autlan supporting a nucleated population large enough to be referred to as a city (Laitner Benz 1992). While the region’s population at the time of Spanish contact consisted predominantly of Otomi speakers it also included people who spoke Nahua (Kelly 1945; Harvey 1972). In the Purificacién River valley, the population apparently spoke a large variety of languages, though it too had a Nahua overlay. Summer 1994 JOURNAL OF ETHNOBIOLOGY 27 The northeastern and southern slopes of the Sierra de Manantlan were appar- ently inhabited principally by Nahua speakers (Sauer 1948; Harvey 1972). Only a few indigenous Nahua speakers remain today, and they reside in the ejido of Ayotitlan in the south-central part of the Sierra de Manantlan. The current population in the Sierra de Manantlan is a mixed lot. While a few of the communities are inhabited by indigenous but Spanish-speaking people (e.g., Ayotitlan, Camichin, Cuzalapa, Tel Cruz), the inhabitants of many of the other communities are descendants of recent immigrants from outside the region. One community in particular, El Terrero, is inhabited by the descendants of immigrants from Michoacan who arrived with the timber boom in the 1940s (Jardel 1992). For the most part, the inhabitants of the Sierra de Manantlan live under very marginal socioeconomic conditions (see Jardel [1992] for details), While all of the communities studied can be reached by motorized vehicle, many of the roads are impassable during some or all of the rainy season, leaving these communities periodically cut off from surrounding areas except by foot or horse. At least half of these communities lack electricity, and five out of nine lack telephone, regular postal service, or transportation services. While water is carried or piped-in directly from nearby rivers or springs, its potability is seasonal. Illiteracy is rela- tively high (ranging from 15-40%) in these communities due to the lack of perma- nence of trained educators and the frequent truancy of students needed to tend the fields or livestock. The Reserve’s communities are primarily maize agri- cultural although the people now see cattle as an increasingly viable economic option; all raise a few chickens and pigs. El Terrero, which has an active timber industry, is the only community which has a nonagricultural economic base. The Sierra de Manantlan Biosphere Reserve protects a relatively large expanse of Cloud Forest (CF) although it comprises only a very small fraction of the total area of the Reserve (Jardel 1992). Tropical Deciduous Forest (TDF) comprises a very large percentage (25%) of the Reserve’s total area; the Reserve is apparently one of the few areas in the Neotropics where relatively undisturbed tracts of this formation have been set aside. The Reserve also protects large expanses of Pine (PF), Oak (OF), and Pine-Oak forests (POF), as well as Fir (Abies) (FF) and Tropical Subdeciduous Forest (TSF). The diversity of vegetation types provides habitat to more than 2,500 species of vascular plants (Vazquez et al. 1990; Santana M. unpub. data), including ca. 25 local and many more regional endemic species. METHODS The communities under study were initially selected in order to evaluate the local peoples’ awareness of the availability of the plant species present in this biosphere reserve. All nine communities have more or less ready access to six vegetation types—CE, OF, POF, TDF, TSF, and Gallery Forest—while only two of the communities have access to Fir Forest. As it turns out this suite of communities also provides a representative sample of the socioeconomic condi- tions prevailing in the region. Each community was visited periodically over the course of each year so that flowering and/or fruiting herbarium specimens of species that are locally available could be used to facilitate interviewing. Speci- 28 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ Vol. 14, No. 1 mens used in interviews were collected in relatively undisturbed vegetation and along paths located within two to three hours walk from the community. Her- barium specimens are collected in sets of five or more; at least two specimens are used in interviews assuring that three to five or more informants saw and commented on all of the species collected during a particular visit to any one community.! Plants were shown to informants in a freshly field-pressed state. Information was elicited about a species’ use by asking two questions. The first question is whether the informant recognizes and has a name for the plant, the second is whether the species is used for any purpose. If the informant provides a use for a particular species he/she is again asked whether it might have any additional use. Questioning continues in this way until the informant responds that he/she knows of no other use. We consulted numerous informants in each community in order to corroborate information provided by individual informants and to permit use of the frequency of informant response as a proxy measure for intensity of use. Individuals who were identified as knowledgeable in informal discussions with community officials and who expressed a willingness to endure our often lengthy interrogations partici- pated as informants. These primary informants have been repeatedly interviewed during the three years this research was underway. Other individuals have partic- ipated as well; these persons usually identified themselves as knowledgeable and either offered or agreed to be interviewed. Both male and female informants have been interviewed and we sought to include individuals of all age groups. The vast majority of these individuals are either natives or have spent a considerable part of their life in the community where they now reside. For the most part the interviews were conducted by persons who are also local residents; half of the interviewers were born and raised in the vicinity of the Sierra de Manantlan. Use of these resident locals (the authors FSM, JCE, and DDL) as interviewers has facilitated understanding of the information elicited from informants principally because many uses appear to be very local and the terminology used to describe such use often appears to be regionally, if not locally, unique. The information discussed here is based upon an analytical unit that has simplified the management and interpretation of the data obtained. This analyti- cal unit, one report of use, is the single mention of a part of one species for a particular use by one informant (cf. Alcorn 1984). For example, until 1990 gua- muchil (Pithecellobium dulce [Roxb.] Benth.) had been reported as useful by five different informants. One of these informants provides four reports of use: the “seed” (i.e., the aril) is edible, the leaf is medicinal, the trunk makes good fire- wood, and the wood is useful in house construction. Another informant indicated that the bark is used medicinally and that the seed is edible. A third informant recognized the root as medicinal. The fourth recognized the trunk as being suit- able for fence posts and for firewood. The fifth described the bark as medicinal, and like the fourth informant, reported that the trunk is useful for firewood and as fenceposts. In this example the total number of reports of use is 12. The data was computer-coded and manipulated using a variety of data man- agement and statistical programs. Nonparametric statistical tests (Sign, Chi- Summer 1994 JOURNAL OF ETHNOBIOLOGY 2 TABLE 1.—Floristic and ethnobotanical representation of the 11 most common families of vascular plants in the Sierra de Manantlan Biosphere Reserve! FLORISTIC INVENTORY2 ETHNOBOTANICAL INVENTORY Species? Reports of Use? Compositae (291) Leguminosae (37) Leguminosae (378) Leguminosae (213) Compositae = (20) Fagaceae (327) Gramineae (193) Euphorbiaceae (16) Verbenaceae _—— (118) Orchidaceae (126) Solanaceae (16) Solanaceae (118) Euphorbiaceae ( 62) Fagaceae (12) Moraceae (106) Solanaceae ( 51) Rubiaceae ( 9) Myrtaceae (100) Malvaceae ( 48) Moraceae ( 8) Compositae (100) Labiatae ( 45) Gramineae ( 8) Sterculiaceae (100) Rubiaceae ( 36) Malvaceae ( 7) Flacourtiaceae ( 83) Scrophulariaceae (33) Myrtaceae ( 6) Rosaceae ( 71) Fagaceae ( 31) Labiatae ( 6) Euphorbiaceae ( 61) 1 Species numbers in floristic inventory after Vazquez et al. 1990. 2 Numbers in parentheses are numbers of species 3 Numbers in parentheses are numbers of reports of use for all species. -square, calculation of Pearson’s correlation coefficients, and linear regression analyses) were obtained from these programs or calculated manually (Siegel 1956). RESULTS Is the useful flora a representative sample of the area's flora?—One of the questions posed initially was whether use of the flora is in any way related to floristic composition of the study area. Stated another way, is utilization of the flora determined by the relative abundances of certain taxonomic groups? There ap- pear to be two ways of examining this question: first, by comparing the relative numbers of species per family reported by the Reserve’s inhabitants with that of the area’s flora; second, by comparing the relative importance of each family based upon total number of reports of use and comparing it to the relative floristic importance of each family. Comparison was made using family rank (Table 1) based upon the number of species present in the flora and the number of species reported as useful by the Reserve's inhabitants. Only two of the 10 most speciose families in the Reserve’s flora—the Orchidaceae and Scrophulariaceae—do not provide a relatively large number of useful species (i.e., more than five species). While numerous species from both of these families have been employed in interviews, only three species of the Scrophulariaceae and a single species of orchid have been designated as useful. Comparing how families are ranked in the floristic and ethnobotanical inventories leads us to infer that little difference exists in the order of family importance using these measures. Eight of the 10 families with the largest number of species reported as useful are also among the 10 most speciose families in the Reserve's flora; in fact the order of relative importance of the 11 most speciose 30 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 families is not significantly different (Sign test; P < .2) from that of the Reserve's flora. Plant use in these nine communities of the SMBR thus appears to be related to relative floristic abundance. Comparing relative family order based upon fre- quency of report of use led to a similar conclusion, i.e., that no significant differ- ence in ranking existed (Sign test; P < .3). In this case five of the most speciose families of the Reserve's flora are in the top 10 most commonly reported families in the ethnobotanical inventory and two more are in the 15 most commonly reported (Table 1). Are all vegetation types subject to equal forms of use?—The specimens utilized in interviews were obtained from different types of vegetation. The aforementioned vegetation types are distinguished in part on physiognomy; for example, CF and TSF are similar in terms of tree diameters, heights, and shrub density, while TDF is quite distinct, with short, small-diameter trees the rule and much higher shrub densities (Rzedowski 1979; Benz unpub. data). Vegetation types are also distin- guished in part on floristic, phytogeographic, geographic, and climatic/phenolog- ical characteristics. Such differences in forest structure and phenology led us to question whether any one vegetation type might be characterized by a specific pattern of use. This interest stemmed from both a human foraging point of view, i.e., are there more edible products in any one type of vegetation?, or are the products available in one particular vegetation type more diverse than those from other vegetation types?, and from a conservation standpoint, i.e., is timber prefer- entially exploited from one or more types of vegetation? e specimens collected for use in interviews were obtained in nearly all 11 types of vegetation present in the Reserve but not all types of vegetation nor all categories of use are equally represented. Comparison of use and vegetation types thus is based upon only six vegetation types and eight of 14 types of use (Table 2). The null hypothesis is that no difference exists in the number of times a category of use is reported for all the species from each of the different vegetation types, that is, there is no a priori reason to expect that any one vegetation type is preferred over the others for any category of use. Acknowledging that a variable number of species were collected from each vegetation type and used in inter- views, that these species are for the most part represented in only one vegetation type, and that a variable number of informants were interviewed in each commu- nity, we suspect that certain types of vegetation might harbor species of similar habit or life form which, in turn might be subject to similar forms of use and, therefore, subject to characterization. We are willing to admit that similarities and differences of species’ uses across vegetation types might be attributed to the species present and their relative abundances in each vegetation type, or that the informants interviewed might have provided biased thematic knowledge; how- ever, for the moment, we focus on vegetation types as the source of this difference or similarity. Statistical comparison indicates that considerable difference exists with respect to the number of reports of use of the species from each of the different vegetation types (y2=200.5; 30 df; p < .001; Table 2). Oak Forest appears to be the principal vegetation type for obtaining species whose wood is utilized. Three of the five use categories—firewood, fenceposts, Summer 1994 TABLE 2.—Reports of use arranged according to ve JOURNAL OF ETHNOBIOLOGY use reported for the plant species by the local population. 31 getation type and type of TYPE OF VEGETATION TROPICAL OPICAL SUB- OAK GALLERY CLOUD PINE- DECIDUOUS DECIDUOUS Row TYPE OF USE FOREST FOREST FOREST OAK FOREST FOREST Total 82! 19 103 4 53 159 420 EDIBLE 96.2 22.2 74.0 20.8 59.4 1475 19.8% -1.6 -.8 4.1 -4.2 -1.0 1.3 95 30 36 12 33 85 291 FIREWOOD 66.6 15.4 913 14.4 41.1 102.2 13.7% 4.3 4] -2.5 -.7 -1.5 -2.3 65 9 23 5 21 67 190 FENCE POSTS 43.5 10.0 one 9.4 26.9 66.7 9.0% a9 -3 -2.1 -1.5 -1.3 0 71 1 45 13 11 64 205 CONSTRUCTION 470 10.8 36.1 10.1 29.0 72.0 9.7% 4.2 -3.2 i? 1.0 -3.8 -1.2 19 8 45 1 37 81 191 FORAGE 43.7 10.1 6 | 9.5 27.0 671 9.0% -4.5 -/ 2.3 -3.0 22 22 37 11 19 10 16 59 152 INSTRUMENTS 34.8 8.0 26.8 75 21.5 53.4 72% A 1.1 -1.7 1.0 “1.3 1.0 117 34 103 60 129 230 673 MEDICINAL 154.1 35.5 118.6 33.3 5.1 236.3 31.7% -4.1 m | -1.9 5.7 4.5 -.6 Column 486 122 374 105 300 745 2122 Total 22.9% 5.3% 17.6% 4.9% 14.1% 35.1% 100.0% ' The numbers in each cell from top to bottom refer to the observed frequency, (number of reports of use), the expected frequency, and the adjusted residual value. Adjusted residuals indicate the magni- tude and direction of the deviation of observed from expected standardized across all cells of the table. and construction—where wood is the forest product of interest show a higher than expected number of reports of use for OF than other types of vegetation (Table 2). This is probably due to frequent report of use of Quercus magnoliifolia Née, Q. gentryi C.H. Muller, and Q. elliptica Née. Reports of species’ use where OF appears to provide less than expected number of reports is where forage or medicinal uses are concerned. 8 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 Gallery Forest, a type of vegetation whose overstory is dominated by tall trees, appears to be subject to greater frequency of use than expected for firewood (i.e., Inga eriocarpa Benth., Salix humboldtiana Willd., Croton draco Schlecht., and Xylosma velutinum [Tulasne] Triana & Planchon) than for other vegetation types except OF. Contrary to expectation, species from Gallery Forest do not appear to be subject to use for construction purposes. Cloud Forest is one of the most diverse and highly endangered vegetation types in Mexico; its conservation is of high priority for the SMBR. The SMBR’s Cloud forest does provide a notable abundance of edible plant products (e.g., Prunus serotina Ehrh. ssp. capulli [Cav.] McVaugh, Rubus adenotrichos Schlecht., Smilax moranensis Mart. & Gal., and Crataegus pubescens [H.B.K.] Steud.). Pine-oak Forests cover a large part of the SMBR’s area. Species present in POF provide a relatively higher number of reports of medicinal use than species occur- ring in other vegetation types. Tropical Deciduous Forest does not appear to provide materials suitable for construction purposes. This is not surprising knowing that the arboreal species characteristic of this type of vegetation rarely exceed 7 m. TDF does, however, provide a relative abundance of species utilized for medicinal purposes (e.g., Vitex mollis H.B.K. f. iltisii Moldenke, Anoda cristata [L.] Schlecht., Plumbago scan- dens L., Guazuma ulmifolia Lam.). The focus of use in certain vegetation types is not totally unanticipated but may contradict the apparent taxonomic focus discussed earlier. In fact, it seems likely that focused use in these vegetation types might in fact be a reflection of relative taxonomic abundances, e.g., Oak Forest, dominated by three to five spe- cies of oaks, records uses focused on wood; Tropical Deciduous Forest with its abundance of Leguminosae, Euphorbiaceae, and Anacardiaceae provides a myr- iad of medicinal species. While floristic composition is undoubtedly a considera- tion in characterizing focus of use, very likely other factors should be considered in the future to fully understand why, for example, Cloud Forest provides an abundance of edible plant products (from a wide range of families) and Tropical Deciduous Forest is the focus of medicinal plant product extraction. Are important species subject to overexploitation? Focusing on the how, where, and what of plant resource use has been an over-riding concern of our research in the SMBR. This is due to the need to detect excessive use of plant species in order to identify which, if any, might require management alternatives to ensure that the species do not become endangered by overuse. Thus we sought a measure of relative importance or intensity of use to detect species whose importance might be adversely affected by human use. Relative ethnobotanical importance of plant species has been estimated for various reasons by a variety of methods. Prance et al. (1987) derived relative importance values of families by assigning weights (more important versus less important) to general use categories such as edible or construction, and combin- ing these weights with the number of times (i.e., different plant parts) a plant was cited as useful. Johns et al. (1990) calculated consensus values for medicinal species based on the number of informants who employed a given species in the treatment of the same illness and on the species’ relative abundance. While not all Summer 1994 JOURNAL OF ETHNOBIOLOGY 33 TABLE 3.—Twelve species with the greatest number of reports of use in the Sierra de Manantlan Biosphere Reserve's ethnobotanical inventory. Informants Communities Distribution Reports Reporting Reporting Types Parts se Species and Habitat! of Use U of Use Used Guazuma ulmifolia W,D,TF! 96 33 6 6 8 Quercus magnoliifolia W,N,OF 84 18 3 7 5 Quercus gentryi L,N,POF 82 Pte. 3 6 6 Vitex mollis W,C,TF 68 28 Yj Ne 8 Enterolobium cyclocarpum — W,D,TF 67 i7 i 9 8 Psidium guineense W,N,TF 53 16 3 Z 5 Byrsonima crassifolia wr? 50 an 5 Fs 7 Casearia corymbosa W,D,TF 46 24 5 5 5 Ficus insipida W,D,TSF 45 24 5 10 i Inga eriocarpa W,D,OF 45 20 6 9 z Quercus elliptica W,N,OF 43 10 2 6 4 Inga laurina W,N,TSF 36 12 7 4 1 Distribution and habitat: W = widespread, L = local; D = disturbed habitat, N = natural habitat, C = cultivated /disturbed ground; OF = Oak Forest, POF = Pine-Oak Forest, TF = Tropical Deciduous and Subdeciduous Forests, TSF = Tropical Subdeciduous Forest. species demonstrating high consensus values in their study were among the most frequently utilized, the majority of widely used species did have high consensus values. In this case consensus and frequency of use appear to be related. Turner (1988: 275-276, 278) calculated an index of cultural significance as a product of weights, each assigned according to the plant’s quality of “use” based on the plant’s cultural role in terms of its contribution to human survival, combined with an estimate of intensity of use and a scaled value of exclusivity of use. This index is a subjective but systematic attempt to measure relative importance of plant species. Phillips and Gentry (1993a, 1993b) developed an index, overall use value, based on the sum of the number of different uses reported for a species by an informant. This index is based on the number of times each informant saw a species and reported its use, summed over all informants, and divided by the total number of informants. These authors demonstrate that a species will have a high chance of being useful if it is large, a tree, has a high population density, is common, or grows fast (Phillips and Gentry 1993b). We employ a similar rational in assessing relative importance but separately list as indicators of importance the number of reports of use, the number of different parts utilized and distinct uses given each species, and the number of informants who employ a given species, as well as the number of different com- munities in which the species is recognized as useful (Table 3). As might be expected, in many cases the species most often cited as useful are the same as those for which the greatest variety of uses are reported; considering all taxa reported as useful, the number of reports and number of uses are correlated (r2 = 48, p < .001). Independent of this relationship, however, 12 species of the total 365 34 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 (see Table 3, Appendix A) present a significantly higher number of reports of use than the remaining 353, that is, their number of reports is greater than 2 standard deviation units from the mean (see Fig. 2, Appendix A). Frequency of report of use is probably related to abundance and availability (c.f. Johns et al. 1990; Phillips and Gentry 1993b). Hence it is not totally unexpected that five of these 12 species thrive in disturbed habitats (see Table 3) such as along paths in forests, that two are disturbed ground species that are frequently culti- vated, and that the five naturally occurring species are widely distributed in the Oak, Pine-Oak, or Tropical Forests of the SMBR, suggesting that tolerance to human disturbance and/or a wide habitat preferences might make certain species predisposed to human utilization (c.f. Bye and Linares 1983). How consistent are informants in reporting uses of plant species? Examination of the relative importance of plant species to the population of the SMBR also calls attention to the relatively large number of species that are considered useful by a single informant for a single purpose (Fig. 2). Considering all taxa designated as useful and all categories of use, 21% of these species (78 of 365) are cited as useful by a single informant. The percentage of species reported only once nearly dou- bles if we consider only those species used medicinally (85 of 221). This general trend has been noted at the level of community as well. In a typical visit to one of the nine communities, 55% of the species (64 of 116) employed in interviews were recognized as useful and 28% percent (18 of 64) of these were identified as useful by only one informant. Thus it would appear that at most 80% of the species cited as useful are subject to use by more than one individual. Neither the cultural or biological basis of this pattern, nor its significance, is currently understood, but we hypothesize that the apparently large proportion of idiosyncratic knowledge (more than 20%) existing among this population may be due either to experimen- tation or to the waning of traditional indigenous knowledge among the infor- mants of these mestizo communities (see Bernard et al. 1984). SUMMARY AND CONCLUSIONS Use of the plant resources in the SMBR appears to be a function of relative taxonomic abundances of the area’s flora. Floristically common plant families are represented by a greater number of species listed as useful. This is probably not uncommon in other areas of the world, though it has not, to our knowledge, been reported elsewhere in the ethnobotanical literature. The forms of use attributed to plant species in different types of vegetation are not uniform in the Sierra de Manantlan. While it might be expected that vegetation types that do contain woody or arboreal species are preferred sites for the collection of firewood or construction materials, the results discussed above suggest that differences exist in the use of species from five vegetation types: reports of use that focus on the wood of species from Oak and Gallery Forests are more numerous than from other vegetation types, Tropical Deciduous Forest and Pine Oak Forest species are more frequently identified as useful for medicinal purposes, and Cloud Forest appears to receive greater attention for its edible plant products than do the other vegetation types. Whether these tendencies NUMBER OF SPECIES . = | 40> Tt 20> @) { Hnlttets af 1.1 Paaataa? eB g_F g , 6) 20 40 60 80 100 NUMBER OF REPORTS OF USE FIG. 2.—Abundance diagram showing the number of reports of use of all useful species (1988-1990). Species exhibiting a significantly (Z > 2.1, P < .05) large number of reports of use are those numbering more than 36 (see Appendix A). This figure also depicts the relatively large number of species that are utilized for one purpose and by only one informant (Appendix A). 36 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 are due to a deliberate use of species found in these vegetation types, to the relative proximities of these vegetation types to habitation areas and the greater familiarity of informants with them, or to other sampling biases have not been tested. Plants that have significantly more reports of use are species with naturally widespread distributions or species that thrive in disturbed habitats. Humans might more frequently come into contact with such species, which would increase the possibility of experimentation. Once having been found suitable, the species would be included into the local ethnobotanical inventory and knowledge of its suitability widely disseminated. Widespread experimentation might then follow and lead to an even greater number of uses. While corroboration of a particular species’ use by more than one informant was hypothesized at the outset, the seemingly large proportion of species reported as useful by a single informant was an unanticipated result of our research. The large number of informants that we have interviewed could be one source of the seemingly large amount of idiosyncratic knowledge; that is, many informants might be expected to have a proportionately more varied knowledge of the local flora’s use than fewer informants. Alternatively, it is possible that the relatively large number of uniquely utilized species is due to identification errors by the informants. We have recorded such instances—where an informant refers to a specimen by a common name frequently applied to another species—but these seem rare and probably would not account for the 20% uniquely utilized species. Our informants appear to prefer to err on the conservative side by admitting not to know a plant or its use instead of incorrectly identifying it. One final considera- tion is also plausible: that a large proportion of idiosyncratic knowledge is typical (J. Alcorn, personal communication 1993). This possibility is supported by recog- nizing that each person has individual needs and that such individuality might require that only a small fraction of the total knowledge about a communities surroundings be shared among its inhabitants. These results lead us to suggest that conservation of biological diversity in the SMBR might provide context for continued experimentation and maintenance of traditional uses, hence, to the con- servation of traditional empirical knowledge. The manner in which knowledge about use of local plant resources is distributed suggests that programs to mod- ernize these communities that have homogenizing effects on information flow will displace opportunities for experimentation and for the transgenerational transmission of knowledge. Many informants appear to know much about a few species and a little about a large number of species. If we permit such moderniza- tion to occur without assuring opportunities to pass along this knowledge, or if we permit these forests and the wealth of species they contain to be destroyed, the rich lore and erudition possessed by these people will surely disappear. NOTE "Voucher specimens collected during this research are deposited in the herbarium of the Instituto Manantlan de Ecologia (ZEA) and the University of Wisconsin-Madison (WIS). Summer 1994 JOURNAL OF ETHNOBIOLOGY 37 ACKNOWLEDGEMENTS Financial support for this work was obtained from the University of Guadalajara. We thank H. Iltis, who suggested numerous improvements on an early draft, and him, T. Cochrane, and M. Wetter of WIS, who continue to provide us with identifications of our collections. Thanks also to L. Guzman for his assistance in curating our collections at ZEA. J. Alcorn, R. Bernard, and R. Bye made observations on the manuscript that improved clarity and data presentation for which we are grateful. Finally we want to thank all the informants in the Sierra de Manantlan for so willingly providing such a rich body of information about the flora and vegetation that surrounds them. Tanja Netscher provided the French translation. LITERATURE CITED ALCORN, JANIS B. 1984. Huastec Mayan. Ethnobotany. University of Texas Press, ustin. BERNARD, H. RUSSELL, PETER KILL- WORTH, DAVID KRONENFELD, and LEE SAILER. 1984. The problem of in- formant accuracy: The validity of ret- rospective data. Reape Review of Anthropology 3:495-517. BYE, ROBERT A. te EDELMIRA LI- NARES. 1983. The role of plants found in Mexican markets and their impor- tance in te eden a Jour- nal of Ethnobiology 3: HARVEY, HERBERT. 1972. ae Relaciones Geographicas, 1579-1586: Native Lan- guages. Pp. 279-323 in Handbook of the Middle American Indians, Vol. 12. Howard Cline and Robert Wauchope ee University of Texas Press, Aus ILTIS, HUGH. 1980. The 3rd University of Wisconsin-University of Guadalajara Teosinte Expedition to the Sierra de Manantlan, Jalisco, Mexico: December 28, 1979 to January 21, 1980. Back- corrections as Contributions from the University of Wisconsin Herbarium Vol. 1, No. 1, December 1983, Madison. , JOHN E. DOEBLEY, RAFAEL GUZMAN M., and BATIA PAZY. sc Zea diploperennis (Gramineae): A n teosinte from Mexico. Science 203:186- 188. JARDEL PELAEZ, ENRIQUE (Coordi- nador) 1992. Estrategia para la Conser- vacién de la Reserva de la Bidésfera Sierra de Manantlan. Laboratorio Nat- ural Las Joyas, Universidad de Gua- dalajara, ee Jalisco. JOHNS, TIMO . O. KOKWARO, and E Kenya: a quantitative crite- ria for sa us. Economic Botany :369-381. KELLY, ISABEL 1945. The Archaeology of the Autlan-Tuxcacuesco Area of Ja- lisco. I. The Autlan Zone. Ibero- americana 26, University of California Press, Berkeley. . 1949. Archaeology of the Autlan-Tuxcacuesco Valley. Vol. 2. The Tuxcacuesco-Zapotitlan zone. Ibero- americana 27, University of California Press, Berkeley. 981. Ceramic sequence in Colima: Capacha an Early Phase. An- thropological Papers of the University of Arizona No. 37, Tucson. LAITNER BENZ, KAREN. 1992. Organi- zaciOn Regional en el Area de Influen- cia de la Reserva de la Bidsfera Sierra de Manantlan, Jalisco-Colima, en el Siglo XVI. Pp. 319-337 in Origen y Desarrollo de la Civilizaci6n en el cee nite México. Brigitte Boehm e Lameiras and Phil C. Weigand (Coorinadores) El Colegio de Mic- hoacan, Zamor. PHILLIPS, OLIVER and ALWYN H. GENTRY. 1993a. The useful plants of Tambopata, Peru: I. Statistical hypoth- eses tests with a new quantitative technique. Economic Botany 47:15-32. 38 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 . 1993b. The useful plants of Tam- bopata, Peru: II. Additional hypo- thesis testing in quantitative eth- nobotany. Economic Botany 47:33-43. PRANCE, G. T., W. BALEE, B. BOOM, and R. CARNEIRO. 1987. Quantitative eth- nobotany and the case for conserva- tion = ae Conservation Biol- ogy 1:296-310. SAUER, CARL. 1948. Colima of New Spain in the Sixteenth Century. Ibero- americana 29, University of California Press, Berkeley. RZEDOWSKI, JERZY. 1978. La Vegetacién de México. Editorial Limusa, Mexico, DE SIEGEL, SIDNEY. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill, New York. TAMAYO, JORGE L. 1980. Geografia Mo- derna de México. 9th edition. Editorial imusa, Mexico, TURNER, NANCY J. 1988. “The impor- tance of a rose:’ " Evaluating the cul- tural significance of plants in Thomp- son and Lillooet Interior Salish. American Anthropologist 90:272-290. VAZQUEZ G., J. ANTONIO., RAMON CUEVAS G., THEODORE S. COCH- RANE, and HUGH H. ILTIS. 1990. Flora de la Reserva de la Biésfera Sierra de Manantlan, Jalisco, México. Universidad de Guadalajara, El Gru- llo, Jalisco. APPENDIX A REPORTS SPECIES Bunchosia mcvaughii Trichilia hirta Arachys hypogaea Euphorbia ariensis Hura polyandra Croton wilburi Euphorbia indivisa Oxalis hernandezii Penstemon roseus Pseudobombax ellipticum Calophyllum brasiliense 0 Eryngium nasturtiifolium uarea glabra Citrus limon Caesalpinia mexicana Cynoglossum pringlei Quercus castanea Piper amalago Asclepias angustifolia Porophyllum ruderale Heliotropium indicum Chusquea liebmannii Paspalum clavuliferum Digitaria horizontalis Rauvolfia canescens Tridax arr aiiie a sessei Baccharis icamintie a cc ce cee ce ce ee ee ee ee eee ee REPORTS SPE CIES Salix microphylla Tinantia longipedunculata Psacalium peltigerum Plumeria rubra Acacia angustissima tyrax sp. Chamaecrista punctulata oursetia mollis Spigelia scabrella Phoradendron reichenbachianum Rhytidostylis gracilis Antigonon flavescens Dalea obreniformis Lysiloma tergeminum Roripa nasturtium-aquaticum Raphanus raphanistrum Opuntia puberula Randia aculeata Tournefortia mutabilis Dyschoriste sesaaicort Salvia iodantha Arceuthobium olaboninn Acacia macilenta Anoda acerifolia Pavonia pleuranthera Malvaviscus arboreus Physalis nicandroides Hippocratea volubilis Ficus morazaniana Sida aggregata Senna occidentalis cl el eel ee ce ee ce cee ce ee ee ee ee ee Summer 1994 REPORTS NM NNN NNN NN NW NN NN NRE Bod 00 19 10 0-10.19 10 AD AD AD BD BS 0 Ds ee es JOURNAL OF ETHNOBIOLOGY SPECIES Conostegia volcanalis a Tournefortia densiflora Sambucus mexicana Iresine celosia Buddleia parviflora Salix bonplandiana Vigna lozanii Heimia salicifolia Hedyosmum mexicanum iper rosei Phoradendron amplifolium Dalea versicolor Fleischmannia arguta Leucocarpus perfoliatus Croton draco Citrus aurantium Martynia annua Senna foetidissima Hypoxis mexicana Trichilia americana Bursera grandifolia Scoparia dulcis Bursera fagaroides Acacia riparia Bursera bipinnata Paullinia tomentosa Senna fruticosa Picramnia antidesma Zanthoxylum arborescens Eugenia jambos Passiflora filipes 208 umbellata roton fragilis ever tha eglandulosa Solanum brachystachys Xanthosoma robustum Crusea longiflora Gnaphalium canescens Sapium pedicellatum Amaranthus spinosus Sonchus oleraceus Triumfetta gonophora Melochia adenodes REPORTS BWBWWWWWWWWWNNNNNNNNDND Pb PELE LEEPER WWWWWWWWWWWWWWWWww 39 SPECIES Chryosophylla nana Hamelia xorullensis Cestrum lanatum Ceiba aesculifolia Crataegus pubescens Curatella americana Cyperus hermaphroditus Rhus pachyrrhachis Cissampelos pareira Eleusine indica Iresine interrupta Sapium macrocarpum Calathea sp. Populus guzmanantlensis Rhychosia precatoria Cayaponia racemosa Muntingia calabura Cryptostegia grandiflora Passiflora podadenia Lycopersicon esculentum var. leptophyllum Licaria triandra Baccharis trinervis Aristolochia tequilana Xylosma velutinum Ixophorus unisetus Agonandra racemosa Allium glandulosum Struthanthus interruptus Euphorbia heterophylla Jatropha mcvaughii Cucumis anguria Citrullus vulgaris Verbesina greenmanii Karwinskia humboldtiana Crotalaria longirostrata Calliandra houstoniana Petiveria alliacea — glauca Cissus sicyoides Pithecellobium. lanceolatum Ipomoea bracteata Heteropterys laurifolia Machaerium salvadorense Cnidoscolus autlanensis 40 BENZ, SANTANA, PINEDA, CEVALLOS, ROBLES & NIZ_ Vol. 14, No. 1 APPENDIX A (continued) REPORTS DARADAKDADKDADUUIUUNUTTUTTGAATIAnGnaininnaaqana»i»rrR LP PEEKS HES PB SPECIES Achyranthes aspera Wigandia urens Calathea soconuscum Acacia farnesiana Crataeva palmeri Cordia spinescens Ficus cotinifolia Bauhinia divaricata Psidium sartorianum Caesalpinia pulcherrima Quercus glaucescens Calea urticifolia Annona reticulata Govenia superba Muhlenbergia speciosa Thevetia ovata Paullinia sessiliflora a atomaria Portulaca oleracea Stemmadenia tomentosa Parathesis villosa Dryopteris rosea Rhipidocladum racemiflorum Manihot intermedia Panicum hirticaule Phoebe pachypoda Tagetes lucida Witheringia stramonifolia Randia tetracantha Solanum lanceolatum Marrubium vulgare Pisonia aculeata Sommera grandis Cestrum aurantiacum Cenchrus ciliaris Oreopanax xalapensis Sida rhombifolia Melia azedarach Alvaradoa amorphoides Clethra hartwegii Jaltomata procumbens Vitis berlandieri Thouinia serrata Dendropanax arboreus Parthenium hysterophorus Combretum fruticosum Pteridium arachnoideum Guardiola tulocarpus REPORTS » OOO O00 WMH WWMWDMDOBDDDMDAHBMAWONINNNNNNNNNNNNNDDDADAAD SPECIES Alnus jorullensis Acacia cochliacantha Bursera simaruba Chamaedorea pochutlensis Hamelia patens Cladocolea loniceroides Xylosma flexuosum Maranta arundinacea Bromelia plumieri Bumelia cartilaginea Croton draco Argemone ochroleuca Fuchsia fulgens Lippia dulcis Pereskiopsis aquosa sap adstringens asianthaea ceanothifolia Tillandsia usneoides Begonia balmisiana ndia armata Nectandra ag ae a barclayi Peeve ai habeas uercus laeta Physalis philadelphica Juglans olanchana Ficus padifolia ptis albida Croton ciliato-glandulifera iccinus communis Verbena carolina Morisonia americana Spondias purpurea Hintonia latiflora Albizia tomentosa Buddleia sessiliflora Syngonium neglectum Vernonia capreifolia Agave maximiliana ple coccinea er aduncum Tithonia per eat hea Ilavea Quercus salicifolia Sideroxylon capiri Brosimum alicastrum Summer 1994 REPORTS 1 JOURNAL OF ETHNOBIOLOGY SPECIES Margaritaria nobilis Smilax moranensis Porophyllum punctatum Salix humboldtiana uercus rugosa Cercocarpus Lean a a orellana Pithecellobium | acatlense Aristolochia taliscana Crotalaria mollicula Ziziphus mexicana Celtis iguanaea ia polyandra Coccoloba barbadensis Coffea arabica Lysiloma microphyllum Celastrus pringlei Acacia jorge: acia hindsii Cileddabarat discolor Dorstenia drakena Rubus humistratus Vitex pyramidata Datura stramonium Rubus adenotrichos Solanum madrense Lepechinia caulescens Verbesina sphaerocephala Plumeria obtusa Pithecellobium dulce Miconia albicans Trichos um mexicanum Cochlospermum vitifolium Plumbago scandens Opuntia fuliginosa Annona purpurea Astianthus viminalis Ardisia revoluta Anoda cristata REPORTS 1 SPECIES Phytolacca icosandra Quercus peduncularis Cecropia obtusifolia Quercus obtusata Solanum candidum Ficus pertusa Calliandra laevis Symplocos prionophylla Casimiroa watsonii Quercus acutifolia Eugenia culminicola Ternstroemia lineata Lysiloma acapulcense Juglans major Solanum americanum Prunus serotina Acacia pennatula Inga laurina Quercus elliptica Ficus insipida Inga eriocarpa Casearia corym Byrsonima crassifolia Psidium guineense Enterolobium cyclocarpum itex mollis us gentryi Quercus magnoliifolia uazuma ulmifolia 42 BOOK REVIEW Vol. 14, No. 1 BOOK REVIEW Crops and Man (2nd edition). Jack R. Harlan. Madison, Wisconsin: American Society of Agronomy, Crop Science Society of America, 1992. Pp. xiii, 284. $34.00. ISBN 0-89118—032-X. It is a pleasure to see this second edition of Harlan’s excellent treatise on crops and their influence on human history. It is more than a text book, but it might serve in many courses in economic botany: it is a philosophical understanding of humans’ dependence on the Plant Kingdom. As expressed by the foreword writ- ten by two outstanding agronomists: “As Dr. Harlan taught us in the first edition . . and reinforces in this second edition, crops have shaped the evolution of human societies.” And as Harlan himself states: “In the second edition, I have tried to bring various features of crop evolution up-to-date, but something be- comes obsolete every day and every year. By the time this version is printed, some statements will be out of date and some views may be reversed. . . . There would be little fascination in science if it were static. For better or worse, I offer this version of Crops and Man.” The book is divided into 12 chapters which indicate the wide treatment of the story of plants and human affairs: Prologue—the Golden Age; Views on Agri- cultural Origin; What is a Crop?; What is a Weed?; Classification of Cultivated Plants; Dynamics of Domestication; Space, Time and Variation; the Near East; Indigenous African Agriculture; the Far East; the Americas; Epilogue—Who’s in Charge here? There is a bibliography of relevant titles for each chapter and a detailed subject index. The publication is in easily readable type, and the price is modest and within the range of student ability to purchase if it is used as a text book. With the increasing interdisciplinary approach to the study of plants and man, this well balanced contribution represents a welcome addition to the complex understand- ing of man’s dependence on the vegetation of the world. Richard Evans Schultes Botanical Museum of Harvard University Cambridge, Massachusetts 02138 J. Ethnobiol. 14(1):43-57 Summer 1994 CHOICE OF FUEL FOR BAGACO STILLS HELPS MAINTAIN BIOLOGICAL DIVERSITY IN A TRADITIONAL PORTUGUESE AGRICULTURAL SYSTEM GEORGE F. ESTABROOK The University of Michigan Herbarium and The Department of Biology Ann Arbor, MI 48109-1057 USA ABSTRACT.—The present vegetation on the shale hills of central interior Portugal is called mato. It consists of shrubs mostly in the heath (Ericaceae) and bean (Fabaceae) families. Farmers in this region harvest mato and, whenever they plant a crop, bury it in their cultivated plots to make the soil fertile. Farmers cut mato at ground level, leaving the woody root crown (caudex) just at ground level. Mato plants regrow from these caudices. They are harvested again every four years. In addition, the woody caudex of primarily one species, Erica arborea, is occasionally dug up and burned to distill a brandy-like liquid, called bagaco, from the mass of grape skins, seeds, and pulp that is left over after the fermented wine is drawn off. Interviewing and observing farmers revealed important uses of many different mato species. Sampling mato vegetation from areas regenerating from four to 30 years showed that E. arborea is competitively dominant and capable of reducing mato species diversity. The slow, cool burning qualities of any caudex would be adequate for a still fire, and farmers occasionally do use different species for this. However, the practice of occasionally removing the caudex of the competitive dominant for still fires serves to maintain the variety of useful species in the mato. The somewhat unfounded explanation that E. arborea caudex is best for still fires results every fall in appropriate and timely activity, and as such may serve better than would a more ecological, long-term explanation for the same practice. RESUMO.—O mato natural das formacoes xistosas do centro interior de Portugal € formado por arbustos principalmente das familias Ericaceae e Fabaceae. Os agricultores daquela regiao cortam o mato e, sempre que fazem uma nova cultura, enterram-no nas suas hortas para melhorar a fertilidade do solo. Os agricultores cortam o mato rente a superficie da terra, deixando assim as suas raizes lenhosas logo abaixo da superficie. O mato regenera-se a partir dessas raizes e é cortado de novo todos os quatro anos. Contudo, também as raizes lenhosas de Erica arborea sao por vézes arrancadas e queimadas na distilagao do bagaco. Foram identifi- cadas varias utilizagoes importantes de muitas espécies diferentes de mato. A amostragem da vegetacao do mato das areas em regeneracao durante quatro, oito, e trinta anos mostrou que a Erica arborea é a dominante competitiva e é capaz de reduzir a diversidade das espécies no mato. Embora para um fogo de destilacao sejam adequadas as caracteristicas de queima lenta e de baixa temperatura de qualquer raiz, a pratica existente de arrancar as raizes sO da dominante com- petetiva para queimar serve também para manter no mato a diversidade de especies titeis. A preferéncia de certa forma arbitraria pela queima da raiz da E. arborea em fogo lento proporciona faz cada outono uma lembranga temporal 44 ESTABROOK Vol. 14, No. 1 para uma atividade apropriada e desta forma pode servir melhor do que uma de longo térmo, mais ecolégoca. RESUME.—La végétation naturelle du mato des collines de schiste du centre intérieur du Portugal consiste d’arbustes, dont la plupart font partie des familles Ericaceae et Fabaceae. Les agriculteurs de cette région moissonnent le mato et, quand ils sement une nouvelle culture, ils en enterrent de grandes quantités pour engraisser le terrain. Les agriculteurs coupent le mato a quelques centimetres au dessus du sol, en laissant les racines épaisses et ligneuses juste en dessous du sol, desquelles poussent de minuscules racines qui s’enfoncent dans la pierre pour des dizaines de métres afin de faire monter des éléments nutrifs et de l’eau. Le mato repousse de ses racines épaisses et ligneuses et est moissoné de nouveau tous les quatre ans. Parfois, les racines d’une espéce, Erica arborea, sont arrachées et brailées pour la distillation du bagaco. Des entrevues avec des agriculteurs et des observations des agriculteurs ont révélé de divers usages importants de plusieurs espéces de mato. La végétation mato des locaux de quatre, huit, et trente ans de régénération a démontré que E. arborea est le dominateur compétitif et est capable d’abaisser la diversité d’espéces dans la végétation mato. Bien que les racines ligneuses de n’importe quelle espéce de mato pourraient servir a faire un feu de distillation parce qu’elles brilent lentement et pas trop fort, le fait qu’on n’arrache que les racines du E. arborea pour cet usage sert 4 maintentir la diversité d’espéces utiles du mato. INTRODUCTION An important reason to study a traditional agricultural system where it has supported a population for hundreds of years is to try to determine, from an ecological point of view, how various aspects of its technology contribute not only to the productivity but also to the sustainability of the system. Because traditional agricultural technology is usually developed empirically over generations, eco- logical explanations for some of the very specific, but seemingly arbitrary, prac- tices are not always apparent in the oral tradition of the contemporary popula- tion, especially when these practices are more related to long-term sustainability than to short-term productivity. It is remarkable how the persistent empiricism of human beings, struggling to make their living in nature, results in practices that make ecological sense, even though they may be codified in ritual or explained in ways that seem superficial or not compelling ecologically. Indeed, local practi- tioners may have concepts, equally justifiable but very different from those of academics, of what constitutes a useful explanation. This study of a traditional Portuguese agricultural system provides several examples, one of which is an ecological explanation for what initially seemed an arbitrary but nonetheless very specific fuel choice for the brief annual task of distilling a brandy-like liquid, called bagaco, from the mass of grape skins, seeds, and pulp that is left over after wine has been made. In rural villages in Portugal, grapes are harvested in the fall and made into wine. After the fermented wine has been drained from the fermenting vat and casked, alcohol is distilled from the leftover grape skins and pulp by heating them gently over a cool fire. The distillate, called bagaco in some regions, is about 40% ethanol and 60% water, plus traces of higher alcohols and impurities. A little of it Summer 1994 JOURNAL OF ETHNOBIOLOGY 45 is drunk, but traditionally most of it is used as a household chemical for treating minor injuries, sterilizing, and cleaning. The western and southern foothills of the Serra da Estrela (Fig. 1.), the highest range of mountains in Portugal, are made predominantly of Precambrian shale, with occasional quartzite intrusions. This shale easily erodes, producing very infertile clay soil of resedimented ilites high in iron and with low available water capacity (Azevedo and Ricardo 1973), and also producing very deep, steep-sided valleys that alternate with these quartzite peaks and ridges. Paths over the steep, crumbly rock offer poor footing. There is little or no rain in the summer months when temperatures often exceed 30° C. During the winter, temperatures are near 0° C at dawn, rising to near 15° C during the day. Frequent rains raise impassible torrents in the valley bottoms and erode from the hillsides what little soil may have accumulated during the past year. Human beings have been culturally and economically active in Portugal for thousands of years. However, low overall population densities before the six- teenth century, abundant nearby land that is more level and fertile, the harshness and infertility of these foothills, and the establishment there of Catholic church parishes not before the fourteenth and fifteenth centuries, together suggest that this area had remained largely unoccupied until the fourteenth or fifteenth centu- ries. Although a discussion of the biological, social, political, and economic fac- tors that may have motivated people to attempt to inhabit this region in the fourteenth century are beyond the scope of the work reported here, it seems plausible that many of the current practices directly observed in this study are based on techniques that have enabled people to inhabit successfully this harsh and infertile area since the fifteenth century. These techniques, and the self- sufficient village economies they supported, have largely disappeared from Por- tugal now. Refer to Pearson et al. (1987) for discussions of traditional Portuguese agricultural technology, and of the recent social, political, economic, and techno- logical changes that have contributed to its disappearance. All the details of the agricultural technology that enabled people to thrive in this marginal environment are beyond the scope of the work presented here, but a brief overview of the techniques used to create and maintain soil fertility is relevant. Cultivation of crops occurs on the steep hillsides in narrow terraces that are constructed of dry stone walls that hold the soil level. In winter, these terraces collect soil and water from above and help control water erosion. In the dry summer they facilitate irrigation by streams of water that trickle from slightly rising caves that have been dug above them about 10 m into the soft shale rock, where the shale is still wet from the rains of the past winter. To create fertility in the infertile clay soil in these terraces, large quantities of organic matter are collected, as brush from the hill tops, and mixed with the soil. Shrubs, mostly heaths and legumes, make up the scrubby vegetation type called mato, which occurs in central-interior Portugal on the tops and upper slopes of shale hills. The mato on any given place is cut near ground level every four years. Mato is cut from somewhere, two or three times a week, all year long, and removed to the village, where it is spread over the floor of indoor, ground level rooms that house goats. After two to four weeks, this old mato is removed and replaced with freshly cut mato. After its removal, the old cut mato is piled up, PORTUGAL Serra da Estréla Fundao @ MOROCCO Pampilhosa da Serra Rio Zezére Oleiros @ ' = ce Castelo Branco @ FIG. 1.—Map showing the location of the village of Ribeiros, at the headwaters of the Rio Pracais in the southwest foothills of the Serra da Estrela, Portugal. At the town of Pampilhosa da Serra, The Rio Pracais meets the Rio Unha, a tributary of Rio Zezere in the Rio Tejo drainage. The stippled area is an empoundment. Region of map is approximately the rectangle shown on the inserted outline of Iberia. Summer 1994 JOURNAL OF ETHNOBIOLOGY 47 TABLE 1.—Principle mato species of the region studied. Collector Scientific name Common name Family Number! Erica arborea L. mato negral Ericaceae 404 Ulex minor Roth tojo branco Fabaceae 407 Genista tricanthos Brot. tojo negro Fabaceae 406 Erica cinerea L. urze Ericaceae 403 Halimium ocymoidies (Lam.) unknown Cistaceae 408 Wilk. in Wilk & Lange Lithodora diffusa (Lag.) unknown Boraginaceae 401 .M.Johnson Caluna vulgaris (L.) Hull margarise Ericaceae 405 Erica umbellata L. negrela Ericaceae 402 Chamaespartum tridentatum carqueja Ericaceae 400 (L.) P. Gibbs ‘All voucher specimens were collected by G. F. Estabrook and are housed at MICH. and at planting time, buried in the soil of the cultivated terraces. Cut mato, enriched by goats, is the source of virtually all soil-borne plant nutrients, and much of the soil’s available water capacity. After a plant is cut, it regenerates from a woody root crown (caudex) just below the ground surface. These caudices ramify into an extensive system of fine roots, which penetrate for meters into the soft shale rock below. Although vir- tually all mato species regenerate in this way, the woody caudex of essentially only one, Erica arborea, is dug out and burned to distill bagaco. The caudex of E. arborea burns cool and slow, thus distilling the bagaco with a minimum of impuri- ties and water. Pine (Pinus pinaster), used inside the houses for cooking and warmth, would burn too hot, but any of the woody caudices of the mato species would burn cool and slow. Although the caudices of other mato species are occasionally used in conjunction with E. arborea, farmers clearly prefer E. arborea for still fires. Why principally just this one? They stated that it was used by their parents and grandparents, and that it is the best fuel for this task, but they never offered an explicit, functional or ecological explanation for their preference over other “roots.” Most of the principal mato species (Table 1) make distinct contributions, which this study will describe, to soil fertility and to other aspects of the local economy. Therefore, the maintenance of the species diversity of the mato is an important objective of this agricultural system. This study will also present spe- cies abundance data from plots of mato regenerating for differing numbers of years and subjected to different harvesting histories. These data show that Erica arborea, if not held in check, becomes the dominant species, and thus reduces species richness and diversity in the mato. The choice of the regeneration organ (the caudex) of E. arborea as a still fuel eliminates the domination of this species. Elimination of dominance maintains the species diversity in the mato, which contributes to the sustainability of this self-sufficient village economy. Grime 48 ESTABROOK Vol. 14, No. 1 (1979) discusses in more detail competitive dominance and disturbance-mediated co-existence in stress tolerant plants. It takes 9-10 ha of regenerating mato to supply enough organic matter to create fertility in 1 ha of cultivated terrace. It seems likely that the availability of mato may have begun to limit the amount of terrace under cultivation by the beginning of the nineteenth century or earlier. In this situation, all mato would have been managed for soil fertility, and thus cut every three or four years. Once the practice of removing a few Erica arborea caudices each fall was established, the potential for E. arborea to reduce or eliminate other valuable species would no longer be directly observed by the villagers. In the absence of these direct obser- vations, a reason to remove every fall a few E. arborea caudices to burn in the brief task of distilling bagaco would ensure that the practice happened every year, and thus might serve the local economy better than would a more objectively founded, ecological explanation that did not require a specific action at a specific time. MATERIALS AND METHODS The principle area studied is the village group of Ribeiros, located in the Freguesia de Cabril, Concelho de Pampilhosa da Serra, Distrito de Coimbra, Portugal, at about north 40° 06' by west 7° 54'. The village is located near the center of this region of eroded shale foothills, among the branching streamlets (called ribeiros, hence the name) at the headwaters of the Rio Pracais, a stream that 'runs down a deep, steep-sided gully to the Unha river in the Rio Tejo drainage, as shown in Fig. 1. The elevation of the village is 750 m, with the hill tops and ridges rising 100-300 m above the village. Ribeiros is the modern name of the coalescence of three original settlements (Sobralinho, Melho, and Sanguasuga, located about 1 km apart but separated by deep stream gullies), which, judging from church records, was probably established in the late six- teenth century. It continued to grow steadily, and thrived in the nineteenth and first half of the twentieth centuries, reaching a population peak of approx- imately 300 in 1940, when the first road capable of carrying a motorized vehicle was built into the area to construct an empoundment (Fig. 1) to generate electric power. By the late 1940s, Ribeiros had begun to lose population rapidly, and by 1988 at the conclusion of this study there were some 25 residents, mostly over 60 years old. Refer to Caldas (1981), Serrao (1982), and Brettell (1986) for a dis- cussion of possible reasons for the near universal demise of northern, interior Portuguese villages since the 1940s. In the 1980s, preindustrial agricultural technology was still practiced, if incompletely, by some of the residents of the villages of the Pracais valley, where I visited briefly in 1980, 1983, and 1984. The steep hillsides surrounding Ribeiros are covered with terraces, some of which may have been originally built over 400 years ago when residents and place names in Ribeiros are first mentioned in church birth records. By the 1980s approximately half of these terraces had been abandoned and about 40% had been planted to apple, fig, and olive within the last decade or so by largely absent owners. The remaining 10% were still in cultivation, mostly in corn, bean, potato, and some rye. In terraces closer to the Summer 1994 JOURNAL OF ETHNOBIOLOGY 49 village, vegetables and herbs are grown. The mato is harvested at about one tenth the rate that it was 50 years ago when, according to residents, all terraces were planted with seeds. Because agricultural practices in the village are in decline, much of the mato on the surrounding hilltops had not been harvested for varying lengths of time, up to 30 or more years. I lived in Ribeiros from August until December of 1987. Two married couples among the 21 permanent residents provided me with food and shelter, and intro- duced and endorsed me to the other residents. This endorsement was essential for any resident to speak freely with me. The residents consisted of eight married couples, no single men, and four to six women who were never married or were widows. Except for the wives of my hosts, the wives of the other six men were essentially not socially accessible to me. Three of the single women would talk readily and accept my help. During the first two months of my fieldwork, I observed people at work, took samples of soil and vegetation, and with the help of a tape recorder and inter- preter, learned the local dialect. During the last three months, I conducted infor- mal interviews on demography, agricultural technology, and economic activities. Typically I spent half of each day talking either repeatedly to the same 13 accessi- ble residents of Ribeiros, or to visitors to the village (nine occasions) or to resi- dents of a nearby village (12 occasions). I talked with people usually as long as they would give me their attention, from a few minutes to often an hour or more. I asked the same things in many different ways on different days of the same people and also of different people. I found that whenever different people talked about the same technical subject, their representations were mutually consistent, never contradictory. To determine and vouch mato species accurately, I collected plants in the mato near Ribeiros. These collections were identified at the herbarium of the Estacao Agronomica Nacional (LISE), in Oeiras near Lisboa, and named according to Nova Flora de Portugal (Franco 1971, 1984). Voucher specimens were deposited in this herbarium, and at The University of Michigan Herbarium (MICH). To calculate the diversity and abundance of the species of plants in the mato from areas subject to different harvesting regimes, I collected samples of vegeta- tion from four areas (referred to as Areas 1-4) in the mato-covered slopes to the north and to the east of Ribeiros. Area 1, located 130 m above, and 1 km from, Ribeiros, has been actively harvested for as long as residents can remember. Erica arborea caudex is still taken from here to distill bagaco. Area 2 is also still actively harvested, but E. arborea caudex has not been taken from it recently. This area is located 150 m above, and about 2 km from, Ribeiros. Areas 3 and 4 contain mato vegetation that has never been actively harvested. These areas, located on a nearly level hilltop shoulder, are about 250 m above, and 4 km distant from, Ribeiros. Rye, which grows without irrigation during the cold, wet winter, had been cultivated here but cultivation was abandoned about 30 years ago, largely because of the inconvenient distance of the fields from Ribeiros. Mato established spontaneously when this rye plot was abandoned. Area 3 is the eastern part of this shoulder, where above ground mato vegetation burned off 8 years ago and has since regenerated. It also contains young pines (Pinus pinaster), all less than 8 years old. This pine does not survive fires but grows readily from seed following 50 ESTABROOK Vol. 14, No. 1 fires or other disturbance. Area 4 is the western part that did not burn. Its pines and mato are approximately 30 years old. The residents’ description of the history of the vegetation in Areas 3 and 4 is corroborated by the age of the pines growing in these areas. Pines grow a swirl of branches from their trunk every year. For at least the first 20 and often up to 30 or 40 years, one can age pines by counting these swirls. Sometimes a few years’ swirls will be universally lost by wind or by a bud worm break out that inhibits the growth of swirls. This can be checked by counting growth rings. I cut downa 7-year-old sapling, whose rings and swirls matched. In Area 1 and Area 2, mato is cut on a four-year cycle and had been regenerat- ing for the past four years. Individual caudices regenerate growth 10-20 cm in circumference during four years. Areas 1 and 2 have only five or six abundant species. At the scale of a meter square, relative species abundance varies little throughout these areas. So, from an arbitrary one square meter plot in each of Area 1 and Area 2, all vegetation was cut at about 3 cm above the ground, the approximate height at which it is cut by residents when harvested for use. Plants were sorted by species into plastic bags, and removed the next day to Coimbra where the contents of each bag were dried and weighed. In Area 3, some plants have grown to three times the size of those in Area 1 or 2. Relative species abundance was quite variable at the scale of a square meter, but became more uniform for areas two or three times as large. For this reason, an arbitrary plot 2 m x 3 m was selected for harvesting. All vegetation was cut as described above and sorted by species. Because of the large amount of vegetation produced on this 6 m? plot, the quantity produced by each species was weighed wet in the field and approximately 0.5 kg was sealed wet in a plastic bag and removed the next day to Coimbra, where it was weighed and dried and re- weighed to determine percent dry weight. No vegetation samples were taken from Area 4, but the kinds, sizes, and relative abundances of these very large plants were recorded. To determine the potential of each mato species to enrich the soil with mineral nutrients, each plant species was analysed for levels of minerals, including nitro- gen and phosphorous, at the Laboratorio Agricola Quimica Ribelo da Silva in Lisboa. RESULTS An account of the history and technology of agriculture in this region, learned as a result of my interviews, archival research, and field observations, was presented above. The results presented here are of three kinds. First, what residents do with and say about the most common species in the mato establishes the conspicuous importance of maintaining the biological diversity of mato spe- cies. Second, the relative abundances of mato species measured from plots with different disturbance histories evidences that Erica arborea becomes dominant in plots where it is not periodically reduced. Third, the relative abundance among mato species of mineral nutrients essential for crops establishes the inconspicuous importance of maintaining the biological diversity of mato species for soil fertil- ization. Summer 1994 JOURNAL OF ETHNOBIOLOGY 51 TABLE 2.—Above ground accumulation in three areas of mato. Dry weight (gm) of accumulation per m? precedes average accumulation per year for each species. Areas Area 1 Area 2 Area 3 Size, age 1 m? 4 yr 1 m? 4 yr 2 yr Species Erica arborea 103 26 222 56 1,702 213 Ulex minor i 347 87 882 110 Genista tricanthos — 104 26 590 74 Erica cinerea _ 163 4] ao 4 Halimium ocymoides — iz 3 7 1 Lithodora difusa 5 1 — 3 0 Caluna vulgaris 30 8 241 60 63 8 Erica umbellata 256 64 _ — Chamaespartum tridentatum 816 204 198 50 —_ Erica arborea is called mato negral, which means grey or dark mato. Although this study reveals it to be the competitive dominant (Table 2), it is not considered a weed or otherwise undesirable by the village farmers. Its woody caudex is genu- inely valued as a fuel, and its foliage is also valued as goat forage and bedding. Like all the harvested mato, it is spread over the ground inside the goat houses, where it is enriched by goat urine and excrement before it is finally added to the soil. Ulex minor is called tojo branco, which means white tojo, even though it is covered with green leaves and prickers all year long. Genista tricanthos is called tojo negro, which means black tojo. It has green leaves in the winter that fall in the dry season, leaving a dark brown thorn scrub that not even goats will eat. Both tojos prick the hands of mato harvesters, making harvesting unpleasant and diffi- cult, but these plants are nonetheless harvested, included in goat bedding, and finally buried in the soil. Beyond repeating their preference for mato negral, resi- dents did not say why the regeneration organs of tojo branco and tojo negro are not dug up and burned in stills. Erica cinerea is called urze. It is valued for goat forage, although I rarely saw goats eating it, and highly valued for goat bedding. Except from the thorny tojos, there was nothing superficially apparent to set urze apart from the other cut mato spread on the floor in goat houses. Halimium ocymoides and Lithodora difusa are called mato plants by residents, but did not have more specific names that anyone remembered. Neither did residents describe specific uses for them. These species make up a very small percentage of the mato. Caluna vulgaris is called margarise by the residents of Ribeiros. This species, one of several known as heather, is common throughout northern Europe as well. In Ribeiros, its floral display in August and September is spectacular. Its prolific nectar production is recognized, and in order to increase the amount of honey i ESTABROOK Vol. 14, No. 1 TABLE 3.—Concentration (percent dry weight) of nitrogen (N), and phosphor- ous (P) in samples of mato species from study areas, and in a homogenized sample of old cut mato removed from the floor of a room housing goats. N r Erica arborea 0.86 0.055 Ulex minor 0.64 0.140 Genista tricanthos 1.24 0.101 Erica cinerea 0.64 0.160 Halimium ocymoides 0.59 0.080 Lithodora difusa 0.63 0.096 Caluna vulgaris 0.58 0.098 Erica umbellata 0.90 0.055 Chamaespartum tridentatum 0.48 0.100 Old mato, homogenized 1.44 0.377 collected by village bee keepers, mato is harvested less frequently during the flowering season of margarise. Erica umbellata is called negrela, a diminutive negral. Its caudex is not taken for fuel, and it is readily excluded by its more aggressive congener, E. arborea. Al- though urze is explicitly recognized, all three Ericas are valued as goat bedding. Chamaespartum tridentatum is called carqueja here and over most of northern Portugal. It is highly valued as goat forage. The stems are only slightly lignified, and the goats eat much more of it than of the other mato species. Like all har- vested mato, it becomes part of the goat pen bedding before being added to the soil. Table 2 presents the total and approximate annual above ground accumula- tion, in dry weight, of each species at each of Areas 1, 2, and 3. Area 4 was densely dominated by E. arborea, which had grown, true to its name, into gnarled trees, 2- 3 m high, overtopped with 30-year-old pines. Table 3 presents concentrations, in percent dry weight, of nitrogen and of phosphorous, in samples of mato species collected from the study areas, and in old cut mato removed from the floor of a room used to house goats. For both of these plant nutrients, the concentrations vary among the species by a factor of more than 2.5. Nutrient levels are clearly increased in mato that has been on the floor where goats are housed. DISCUSSION For the last 500 years or more, growth of plants of the mato has been essential for the maintenance of soil fertility in the Pracais valley and throughout central interior Portugal. The mato species play different and complementary roles to support the lives of the local people. Sugar production, fuel, goat forage, and soil fertility have been mentioned here. Thus, maintaining the species diversity of the mato is of genuine, immediate economic value to the residents. Some of their traditional practices can be understood by observing and interviewing the people Summer 1994 JOURNAL OF ETHNOBIOLOGY yi es) who employ them. Other dimensions of this understanding are suggested by evidence revealed by the decline or discontinuance of these practices, by experi- mentation, and by laboratory analysis. Mato negral is one of the many useful plants of the mato, but its competitive superiority would reduce species diversity if were not somehow controlled. The utility of its caudex (the regeneration organ) as a distilling fuel is the stated reason for digging up caudices every fall, even though the caudex of any mato plant would work well in a still fire. Over the past 100 years or more, people may have forgotten that this practice helps maintain mato species diversity because, when all the mato was being cut and properly managed to maintain both soil fertility and diversity, mato negral never had a chance to reveal its dominance. This more ecologically and observationally founded understanding of dominance reduction remains implicit in the traditional preference for mato negral caudex as the still fuel. It is much more important to the local economy to practice the appropriate activities at the right time than to explain them objectively as long term ecological phenomena. The residents of the Pracais valley often do explain their technology in very objective terms and with sound observational bases. Their explanation of irriga- tion technology and the factors that determine the height, width, and frequency of dry stone retaining walls for cultivated terraces, are two examples. Here, scientific explanations incorporate what people need to do to create and main- tain these structures. Thus to participants in this self-sufficient economy, an important part of the utility of an explanation is to help people remember what to do. Results of this study provide two other examples of local distinctions or explanations that seem to serve primarily to instruct people what to do or value, but that also have compelling scientific explanations of longer term effects. The naming of the three different species of Erica that occur in the mato provides one example. All three Ericas look similar enough to be considered congeners by taxonomists; indeed two of them, the large and small negral, are given similar names by residents even though they must be distinguished when it is time to distill bagaco. Residents give urze, Erica cinerea, separate folk generic status and value it highly as goat bedding, even though they give no compelling reasons. When the cut branches of urze are spread on the floor of the goat pen with those of other mato species, they cannot be readily distinguished and the goats do not seem to treat them differently. Residents do not know that urze branches, in comparison to those of the other mato species, have the highest phosphorous concentration. As Table 3 shows, urze has more than twice the phosphorous of most other mato species (only Ulex minor has comparable, but lower, levels), and three times the phosphorous of the other two Ericas. Because traditional practice is to use and value this species for goat bedding, which ultimately becomes soil enrichment, soil phosphorous levels are more effectively maintained. The ancient sedimentary ilite minerals in the shale-derived soils of central interior Portugal are especially poor in phosphorous and rich in iron (Azevedo and Ricardo 1973). Iron tends to chelate phosphorus so that it can not be taken up by plants. The release of phosphorous from decaying organic matter occurs at a slow enough rate that it can be taken up immediately by growing crop plants and not lost to the 54 ESTABROOK Vol. 14, No. 1 iron in the soil. Thus, organic matter has probably always been an essential source of phosphorous for this agricultural system. It is not surprising that hun- dreds of years of agricultural tradition in this region has distinguished the plant that is the best organic source of phosphorous from other members of its genus, and valued it as goat bedding, even though as goat bedding per se it has no special value. The second example of an explanation that seems not to have an observa- tional basis, but that is preferred by villagers because it helps people remember what to do, is provided by the concept of the goats’ bed. Like valuing urze for goat bedding and removing occasional Erica arborea caudices for still fires, the impor- tant consequence of spreading mato as goat bedding is not immediately apparent, and so an explanation that requires the appropriate activity is created. Residents spent about a quarter of their total economic effort cutting mato, hauling mato, spreading it out in goat pens to make a “bed” for goats, removing it from goat pens, piling it in heaps, and finally carrying it to cultivated terraces to dig into the soil before a new crop is planted. Although this effort is essential for the mainte- nance of soil fertility, it was always explained primarily as providing food or a bed for goats. Goats feed as foragers grazing at large during the day, and on weeds and thinnings pulled from the cultivated terraces and given to them, along with occasional rations of grain, when they return to their pens in the evening. Except for carqueja, the goats ate very little of the cut mato, all of which was spread out below them to make their “bed.” Even tojo negro, leafless, spiny wands of dense wood that goats won’t eat, is included in this “bed.” The goats stand up, rarely lying on their thorny bed of sticks and twigs. Why is cutting and carrying mato explained as a means of providing food and bedding for goats, when resi- dents are fully aware that their goats eat very little of it and rarely lie on it? Why is it not explained as a means of maintaining soil fertility, the need for which the residents are also fully aware? Spreading urze, tojo negro, and the rest of the cut mato in goat pens before adding it to the soil raises the ratio of nitrogen to carbon in cut mato (Table 3). When this old cut mato is buried in soil, the higher N/C ratio provides a microenvironment in which the balance of microbes is shifted towards more effective decay organisms that can decompose mato and release its nutrients during one growing season (Griffin 1972). In fact, no residual sticks or twigs were evident in the soil at the time of the corn harvest even though 7-10 metric tons dry weight per hectare of mato (mostly sticks and twigs) had been added at the time of corn planting. Effective, rapid decay of the dense woody branches of Genista tricanthos, which might otherwise decay more slowly than the fruticose twigs of some other mato species, is especially important because this species is highest in nitrogen, substantially higher than other species of mato (Table 3). Thus it is important for soil fertility to leave the caudices of the slow growing tojo negro in the ground, to include its dense, spiny branches in the harvest of cut mato, and especially to spread them out in the goat pens. Chemical analyses and microbial ecology are not evoked by residents to explain why they include inedible thorns in the “food” and “bed” of goats, but the consequences of their traditional agri- cultural practices are clear. Feeding and bedding goats is an explanation that reminds farmers what to do next, especially when the long term consequences of the activity are important, but not immediately apparent. This more proximal, but Summer 1994 JOURNAL OF ETHNOBIOLOGY ea. apparently less correct explanation, thus may serve a self-sufficient village econ- omy better than would a more ultimate, and apparently more correct, one. Other authors have discussed aspects of some of the ideas presented here. Brush (1986) documents the maintenance of the biological diversity of surround- ing areas by farmers practicing traditional methods. Brush (1992) also discusses the specific case of the persistent, deliberate maintenance within individual fields of high potato cultivar diversity by Andean farmers even following the introduc- tion and acceptance of new potato varieties bred by Green Revolution techniques. He lists some of the reasons why farmers might preserve this diversity: taste, interest, agronomic factors, economic opportunity, and prestige or social status, and observes that not a single ecological reason was given by farmers. Zimmer (1991) also discusses the maintenance within individual fields of high potato cultivar diversity by Andean farmers who have not accepted Green Revolution varieties, and describes prestige or social status as the most compelling proximal motive. Although neither author demonstrates, or even hypothesizes, a long term ecological effect of the maintenance of high potato cultivar diversity beneficial to these self-sufficient agricultural economies, attributing proximal prestige to those who maintain diversity would stimulate the practice and produce the ultimate benefit, if it did exist. There are many examples of the use of specific foods or medicines where the preventative or healing effects are known by practitioners who cannot explain, in scientific physiological or chemical terms, how they work. Kuhnlein (1981), Johns (1981), and Timbrook (1987) provide examples. Even though explanations for these practices may incorporate spiritual or magical concepts, by and large these food and medicinal practices are efficacious and people do understand the basic purpose for them, namely to maintain or restore health. These authors do not give examples of less relevant or somewhat artificial reasons, such as the examples of bedding goats or choosing still fuel discussed here, that maintain advantageous practices because they evoke appropriate activity. Concepts of utilitarian explanations and distinctions have been explicitly discussed by some authors. Alcorn (1981), in discussing Huastec perception of botanical resources, mentions invisible technology that not only enables plant use but also manages the plant resource, but gives no examples. Invisible technology may refer to parts of Huastec explanations with little or no observational basis that function to stimulate timely activity, with long term resource management effects not accounted for by the explanation. Hays (1982) suggests that distinctions among kinds of organisms made in self-sufficient agricultural economies may result in differential behavioral or atti- tudinal responses to the organisms distinguished with consequences that are useful or beneficial, even when the benefit cannot be described by those making the distinction. If the distinction is made, then the benefit is enjoyed, not because of the explanation but because of the behavior it elicits. The distinction of the phosphorous rich urze, whose name differs from negral and negrela, the other two Ericas, would seem to be an example of this phenomenon. The procedural, ritualized, unsubstantiated, or seemingly irrelevant explana- tions that elicit timely or appropriate behavior in self-sufficient farming commu- nities may describe practices that represent a deeper ecological or natural wis- 56 ESTABROOK Vol. 14, No. 1 dom. The wisdom of these practices (if not of their explanations) may transcend the short term, production orientation of modern agricultural technology, whose development has been in part motivated by the desire to convert natural resources to cash profits as fast as possible. It is becoming clear that many modern agri- cultural practices cannot be sustained without decimating the very natural re- sources on which productivity depends. Studying, recording, and understanding the human ecosystem in the Pracais valley, an ecosystem based on practices that for centuries have sustained agricultural production on poor soils, is especially relevant to the present challenge of developing technology for sustainable agri- culture to ensure the future well-being of people. Some aspects of this prein- dustrial technology were still available through the memory and activities of the aging residents of the Pracais valley. However, some access to the understanding of how things worked, and especially why things worked, is made available to us by studying the present breakdown of their traditional system. For these reasons, studies of preindustrial agricultural systems should be undertaken with any available evidence of how and why these past technologies were successful. ACKNOWLEDGEMENTS Dra. Graziana Goncalvez Engles, with her husband Prof. Horst Engles, first took me to the Pracais valley in 1980, where I saw that effective traditional agricultural technology was still practiced. I am grateful for the Engles’ help in 1987, when I made an intense study of this technology. I thank the residents of Ribeiros whose trust and interest made this work possible, and especially Joao de Santos who gave me shelter, and Maria Nunes who gave me food. I thank Pedro Nunes for his help with the local dialect. Dr. Jose Sovral Dias, director of the Laboratorio Quimica Agricola Ribelo da Silva, and his staff helped perform the chemical analyses of soils and vegetation. Personal support and encouragement by him and his family are gratefully acknowledged. The friendship and advice of Dr. Manual Bravo Lima, Chief of the Herbarium at the Estacao Agronomica Nacional in Oeiras, and the help of his staff member, Dra. Isabel Saraiva, with the identification of plants, were very valuable. The sabbatical program of the University of Michigan made possible my contin- uous presence in Portugal for six months in summer and fall of 1987. LITERATURE CITED ALCORN, JANIS B. 1981. Factors influenc- ing botanical resource perception among the Huastec: Suggestions for farming systems. Journal of Ethnobiol- ogy 6:151-167. 1992. Reconsidering the green revo- lution: Diversity and stability in cradle future ethnobotanical enquiry. Journal of Ethnobiology 1:221-230. AZEVEDO, ARIO LOBO and RUI PINTO RICARDO. 1973. Caracterizacao e Cons- tituicao do Solo. Terca edicao. Fun- dacao Gulbenkian, Lisboa. BRETTELL, CAROLINE B. 1986. Men Who Migrate, Women Who Wait: Population and History in a Portuguese Parish. Princeton University Press, Princeton, BRUSH, STEPHEN B. 1986. Genetic diver- sity and conservation in traditional areas of crop domestication. Human Ecology 20:145-167. CALDAS, JOAO CASTRO. 1981. Caseiros de Alto Minho: Adaptacao e declinio. A Pequena Agricultura em Portugal 7/8:203-216 FRANCO, JOAO DO AMARAL. 1971, 1984. Nova Flora de Portugal. Vol- umes I e II. Sociadade Astoria, Lisboa. GRIFFIN, D. F. 1972. Ecology of Soil Fungi. Syracuse University Press, Syr- acuse, NY. Summer 1994 GRIME, J. P. 1979. Plant Strategies and Vegetation Process. John Wiley and ns, Chichester, UK. HAYS, TERENCE E. 1982. Utilitarian/ adaptionist explanations of folk bio- logical classification: Some cautionary notes. Journal of Ethnobiology 2:89- JOHNS, TIMOTHY. 1981. The afiu and the JOURNAL OF ETHNOBIOLOGY 57 OTHY J. FINAN, TIMOTHY JOS- LING, MARK LANGWORTHY, ERIC MONKE, and STEFAN TANGER- MAN. (editors). 1987. Portuguese Agri- culture in Transition. Cornell Univer- sity Press, Ithaca, NY. SERRAO, JOEL. 1982. A Emigracao Por- tuguesa. = edicao. Livros Hori- zonte, Lisb TIMBROOK, TAN. 1987. Virtuous herbs: Plants in Chumash medicine. Journal of Ethnobiology 7:171-180. ZIMMER, KARL S. 1991. Managing diver- sity in potato and maize fields of the Peruvian Andes. Journal of Ethnobiol- ogy 11:23-49. maca. Journal of Ethnobiology 1:208- KUHNLEIN, HARRIET V. 1981. Dietary mineral ecology of the Hopi. Journal of Ethnobiology 1:84-94. PEARSON, SCOTT R., FRANCISCO AVI- LLEZ, JEFFERY W. BENTLEY, TIM- BOOK REVIEW Histoire [llustreé du Caoutchouc. Jean-Baptiste Serier, Antionette Diez, and Anne Van Dyk. France: Montpelier Cedex 1 (CIRAD-CP, BP5035, 34032), 1993. $27.00 U.S. (167 French Francs). (No ISBN found) It is almost impossible to “review” this extraordinary and unusual book because it depicts the story of rubber in 96 pages of illustrations. The 450 pictures record the history of rubber from the dinosaur age and that of early man through the use of the product in pre-conquest Mexican times to the “discovery” by Europeans, the early periods of the tapping of Hevea through to the beginnings of commercialisation of Amazonian rubber production to the introduction of Hevea brasiliensis to Asia and the establishment of the plantations in the Old World tropics. The book, extremely novel in its approach, should be useful in teaching economic botany courses and will certainly be of interest to general audiences. The artist-authors are to be congratulated for making available such an inter- esting detailed illustrated history of rubber, its uses, and commercial aspects of its development. Richard Evans Schultes Director Emeritus Botanical Museum of Harvard University Cambridge, Massachusetts 58 BOOK REVIEW Vol. 14, No. 1 BOOK REVIEW The Ethnobotany of the Chacobo Indians, Beni, Bolivia. Brian M. Boom. Ad- vances in Economic Botany, Volume 4. New York: New York Botanical Gar- den. 1987. Pp. 67. $15.00 (paperbound). ISBN 0-89327-312-0. This book is an important contribution to the ethnobotany of Amazonia, particularly of an area which is relatively underepresented in the literature. Its presentation makes it a useful tool for anthropologists and botanists alike. This highly concise work provides a brief introduction to the context and objectives of the study. It includes a description of the study area and an eth- nographic vignette of the Chacobo. Though largely historical it provides the reader some insight to the relative acculturation of the groups prior to 1983 when the author initiated his investigations. Field methods are described as including two basic approaches. These are the “artifact/interview” approach of anthropolo- gists and the “inventory/interview” technique in which informants are inter- viewed about names and uses of plants following the active collection of spec- imens. The largest section of the book contains an accounting of the 360 species collected within a one-hectare area south of the village of Alto Ivon. The collec- tion is distributed in at least 221 genera and 79 families. Where possible, each entry includes the Latin, Chacobo (with English translation), and Spanish names, local frequency, habit, habitat, geographic range of species, voucher citations and commentary on use. The remainder of the book considers the variable uses of plants in Chacobo culture. Of 305 species utilized, 102 are food plants, 75 gath- ered wild from the forest. The latter are of interest as they reinforce the author's suggestion that the Chacobo were traditionally hunter-gatherers as opposed to agriculturalists. Fuel plants include 22 species of trees, while species utilized in the context of construction and crafts include 68. Medicinal plants include 174 species, and are tabulated by ailment treated. Only five plants were considered useful due to their toxic properties. The most notable conclusion of Boom’s work is the fact that the Chacobo utilize 82% of all species and 95% of all trees, with utilization rather evenly distributed among medicinal, construction, and crafts and food categories. This book is well written, and contains excellent tables and illustrations and useful appendices on Chacobo names and nonvascular plants from the area. Beyond its obvious scientific value this book notes the significant implications of ethnoecological research in allocation of lands for self-sustaining Indian reserves. Mark G. Plew Department of Anthropology Boise State University Boise, ID 83725 J. Ethnobiol. 14(1):59-73 Summer 1994 TRIBES, STATES, AND THE EXPLOITATION OF BIRDS: SOME COMPARISONS OF BORNEO AND NEW GUINEA CHRISTOPHER HEALEY Anthropolo Northern Territory University arwin, Australia ABSTRACT.—Exploitation of birds for trade by tribalists of Borneo and New Guinea are compared. Traditionally, bird products in Borneo passed to overseas markets, but in New Guinea were mainly used locally. Contemporary exploitation is illustrated by case studies of a Punan village in Borneo and a Maring village in Papua New Guinea. The Punan are minimally involved in exploitation and con- trol of birds for markets, these roles having been assumed by outsiders backed by the state. Traditional exploitation of birds persists among the Maring, largely because bird products are unimportant in the national economy. The comparison indicates alternative consequences of the intrusion of the state for access to and control of biological resources by tribalists. Different forms of incorporation into the state have variable impact on tribalists’ ethnobiological systems, including use of resources in exchange. It is therefore appropriate for ethnobiologists to contextualize their studies by reference to contemporary polit- ical-economic systems. RESUMEN.—Se compara en este trabajo la explotacién de aves para el comercio por parte de grupos tribales de Borneo y Nueva Guinea. Tradicionalmente, los productos derivados de aves en Borneo pasaban al mercado externo, mientras que en Nueva Guinea eran destinados principalmente al uso local. La explotaci6én contemporanea es ilustrada mediante estudios de caso de una aldea Punan en Borneo y una aldea Maring en Paptia-Nueva Guinea. Los Penan estan invo- lucrados en forma minima en la explotacién y control de aves para el mercado, puesto que estas funciones han sido asumidas por personas externas a la comu- nidad, respaldadas por el estado. La explotacién tradicional de aves persiste entre los Maring, debido en buena medida a que los productos de aves no son impor- tantes en la economia nacional. La comparacién indica consecuencias alternativas de la intromisién del estado para el acceso a y control de recursos biolégicos por parte de los pueblos tribales. Las diferentes formas de incorporaci6n al estado tienen un impacto variable en los sistemas etnobiolégicos de las sociedades tribales, incluyendo el uso de recursos en el intercambio comercial. Es por ello apropiado que los etnobidlogos contextualicen sus estudios en referencia a sistemas politico-econdmicos contemporaneos. RESUME.—Les modes d’exploitation commerciale des oiseaux par les tribues de Bornéo et de Nouvelle Guinée sont comparés. Traditionellement, les produits oiselliers de Bornéo sont destinés aux marchés d’outre-mer, mais en Nouvelle Guinée, ils sont surtout utilisés localement. L’exploitation moderne de ces pro- duits est illustrée 4 l’aide d’examples tirés d’un village Punan de Bornéo et un 60 HEALEY Vol. 14, No. 1 village Maring de Papua Nouvelle Guinée. Les Punans s’occupent peu d’exploiter et de controller les oiseaux pour le marché, ce rdle ayant été assumé par des personnes de l’extérieur, appuyées par l'état. L’exploitation traditionelle des oiseaux persiste parmis les Marings, largement parceque les produits oiselliers ne forment pas une partie importante de 1’économie nationale. Les résultats de ces comparaisons indiquent des conséquences variables sui- vant l’intrusion de l’état concernant l’accés et le control des resources biologiques par les tribues. Les différents moyens d’incorporation au sein de l'état agissent de facon différente sur les systemes éthnobiologiques des tribues, y compris l’utiliza- tion des resources pour l’échange. I] est donc recommendé aux éthnobiologistes de contextualizer leurs recherches au sein des systémes politico-économiques contemporains. INTRODUCTION An important dimension of ethnobiological studies is the documentation of how biological resources are culturally utilized. Aspects of use include—but by no means exclusively so—processes of cultural appropriation from nature, and of the redistribution of these resources in exchange systems. One very important use to which biological resources have long been put is as objects of trade. It has been by trading local products that many pre-industrial communities linked them- selves into a larger economic and socio-political order. There has been a fruitful convergence of interest of ethnobiologists and human ecologists, particularly in relation to aspects of production and manage- ment of biological resources (e.g., Conklin 1957; Ellen 1983). A comparable conver- gence between ethnobiology and economic anthropology is yet to emerge. This is somewhat surprising, given the common interests of both economic and ecologi- cal anthropologists in systems of production and distribution, including indige- nous conceptions of the processes. The incorporation of tribal communities into encompassing political-economic systems has clearly had a profound impact upon the nature of systems of production and their sustaining systems of indige- nous knowledge In this paper I take up some of these issues through a comparison of two adjacent parts of the tribal world, and their different histories as suppliers of forest products, especially birds, to an international market. In particular, I am concerned with how the different trajectories of incorporation into a world- system resulted in different consequences for the continuing involvement of tribal communities in the exploitation and management of wild bird resources. Ethnobiology embraces a complex of knowledge and practice. It does not, however, constitute a “system” in itself, except as an analytic abstraction. Rather, ethnobiology is composed of elements of diverse aspects of socio-cultural sys- tems, including the ecological, economic, ideological, and cosmological. As such, any ethnobiological study should endeavor to indicate the relation of the particu- lar focus of study to the wider socio-cultural context within which ethnobiologi- cal data are embedd In this paper I am not concerned with the “content” and internal order of ethnobiological lore—the traditional focus of studies in ethnobiology, whether as knowledge of taxonomies, or practical, medicinal, magical, or ritual properties Summer 1994 JOURNAL OF ETHNOBIOLOGY 61 and uses of organisms. Rather, I focus on how biological resources come to assume a structural role in mediating the impact of the state on tribal, trading populations. Clearly, biological resources become transformed into tradeable commodities through the application of specific ethnobiological knowledge and practice. However, I am concerned here more with the consequences of this transformation than its organization and form. It is the intention of this paper, then, to develop some comments on how political-economic “world systems” might be conceived as impinging on tradi- tional systems of ethnobiological knowledge and practice. The emphasis is on the socio-cultural procurement practices and systems of redistribution of harvested biological resources. Ultimately, this discussion might lead to a theoretical conver- gence between anthropological studies of political economy and ethnobiology, as has already occurred in the fields of human ecology and ethnobiology. For millenia indigenous tribal people have sent the natural productions of their lands afar in trade and tribute. Subsistence economies of tribal communities were, to a greater or lesser extent, geared to the surplus production of commodi- ties that were in high demand by complex, state societies. Many of these goods were of little or no intrinsic value to those who produced them, other than for what useful or luxury items might be obtained in exchange. The range of prod- ucts of tribal lands in demand by settled societies is legion, but has included the fine furs and pelts of animals, feathers, ivory, aromatic gums, fine woods, per- fumes, spices, pigments, narcotics, and a bewildering variety of animal and plant products deemed to have medicinal or magical properties. In many parts of the globe, relations of mutual dependence rather than open exploitation characterized the political-economic relations between precapitalist trading states and the tribal communities on their peripheries. With the expan- sion of a capitalist world system, however, the relation of traditional communities to their natural resource base has become radically transformed. Many natural resources have lost their traditional value and been superceded or replaced by industrial manufactures, and, of course, by money. In many instances, those that retain value have become subject to commercial production processes that effec- tively deny the continuation of traditional forms of association with the resources and their management. It is conventional to regard the process of the incorporation of indigenous people into the global economy as involving the commoditization of resources— whether natural resources or human labor—which facilitates and encourages their exploitation and sale on an individualistic basis. In this process of commodi- tization, indigenous people have frequently lost socio- political somenny and privileged access to resources which formerly cl strategies (e.g., Nietschmann 1973). Two regions where these processes are relatively recent are the great equa- torial islands of Borneo and New Guinea. Both islands have long been the source of certain luxury goods, extracted from the forests by tribal communities, that have been important in international trade. In both areas, the influence of South- east Asian precapitalist states was weak, virtually nonexistent in most of New Guinea west of the Vogelkop (Birdshead) Peninsula. Similarly, effective control by European colonial powers came relatively late. 62 HEALEY Vol. 14, No. 1 aa meee 0 {i VIETNAM PB PHILIPPINES (yr a FIGURE 1.—Location of case studies. This paper, then, examines aspects of traditional exploitation of forest resources in New Guinea and Borneo, and certain recent transformations in the context of colonial and postcolonial developments. The emphasis is on the articulation of the production of jungle goods, especially derived from birds, with systems of ex- change, with particular attention to trade. The discussion is based primarily on my own research in New Guinea on the hunting of birds and trade in plumes (e.g., Healey 1980, 1990), and on the literature dealing with forest products in Borneo, but also includes some preliminary comparative remarks on a brief field trip to Indonesian Borneo. BORNEO AND NEW GUINEA The large tropical islands of New Guinea and Borneo (Fig. 1) lie either side of the Wallace Line, and exhibit a comparable diversity of flora and fauna (Beehler et al. 1986; Flannery 1990; Medway 1977; Smythies 1981). The richness of biological resources in both islands is paralleled by considerable ecological and cultural diversity that reaches its greatest elaboration in New Guinea. Forest resources are of material and ideological significance in the indigenous cultures of both islands, and have long been major items in extensive trading networks. The bulk of these goods are luxury items rather than staple foods. Trading networks developed to serve the demands of indigenous tribal groups for access to valuables that were deployed in local economies of competitive display or consumption. In many areas, however, trade networks filtered the products of the far interior to small coastal centers from where they entered the Summer 1994 JOURNAL OF ETHNOBIOLOGY 63 international passage of luxury and exotic goods serving the Far East, the Middle East, and Western Europe. The traditional societies of both islands can be broadly treated as small-scale tribal formations, mostly egalitarian and decentralized, although some central Borneo societies were stratified. The scale of socio-political units was highly variable, however, from a few score individuals to several thousands. Subsistence for the most part was based upon shifting cultivation—of dry rice in Borneo and tubers in New Guinea. Hunting and gathering were also important, to the extent that there were specialist forest dwellers in both islands practicing little or no agriculture. Most depended on sago palm for the carbohy- drate component of the diet, and were therefore principally confined to lower altitudes. One major difference between the two islands must be noted: centralized Malay states have been present in Borneo, mostly in coastal regions, since about the fourth century AD. It is through these states that interior tribes were linked to the outside world. Although there were no truly isolated, self-contained communities in New Guinea prior to the colonial era, contact with the outside world was at most tenuous, sporadic, and confined to the coastal belt in western and north-coastal New Guinea. Indeed, the Sultanate of Tidore in the Moluccas claimed suzerainty over the Western Papuan Isles and much of the Vogelkop Peninsula of what is now Irian Jaya, although its actual economic and political control was probably nominal (de Clercq 1889). In both islands, exploitation of forest resources served local demands for wild foods, building materials, magical and medicinal items, and valuable items of decoration on ceremonial occasions. Certain forest products were also traded widely. However, in each island, the ecological and social organization of trade was quite different, and this has led to a markedly different impact of the global economy in different areas. TRADITIONAL TRADE IN FOREST PRODUCTS The patterns of precolonial trade are quite different in the two islands!, and this is largely because of their respective connections to larger, international net- works of trade. Borneo was long an important source of jungle produce for international markets, especially in mainland Southeast Asia and China. Principal forest prod- ucts passed in this trade were edible birds’ nests, rattans, aromatic and decorative woods, camphor, gums, rhinoceros horn, “ivory” from the hornbill bird, and bezoar stones. The bulk of these goods were produced by tribal agriculturalists of the interior—the various so-called Dayak tribes—and specialist foragers, the forest nomads such as the Punan. Substantial proportions of these goods were exported to mainland Southeast Asia and China. Most of this export trade was controlled by elites of the various Malay sultanates dotted along the coastal belt at river mouths. Goods flowing in return for forest products were Indian and Indo- Chinese luxury items for consumption by coastal elites, as well as Indian textiles, Indo-Chinese brassware and porcelain, and salt for exchange with interior tribal 64 HEALEY Vol. 14, No. 1 people (e.g., Brown 1970; Dahlan 1975; Freeman 1970; Healey 1985b; King 1993; Metcalf 1982; Rousseau 1989). Unequal terms of trade operating between the Malay population of the coastal belt and tribal communities of the interior favored down-river commu- nities that could manipulate the supply of trade goods into the interior by control of river mouths. Coastal states were unstable polities, lacking structurally secure central authority. Their territories were generally poorly defined, and consisted of personal hereditary domains of the nobility interspersed with domains vested in the control of the sultan and his appointed officials. The decentralized structure of the state, coupled with the revenue-raising powers of domain-holders, was a critical source of instability of sultanates, with fractious noblemen and vassal states occasionally seeking to assert their indepen- dence from the sultan and establish themselves as rival, autonomous polities. Struggles for power within states led to escalating demands for jungle prod- ucts as a means of raising revenue to underwrite a sumptuous life-style, and to engage armed retainers (effectively pirates) to harrass the settlements and ship- ping of competitors. As a consequence, interior tribal producers of jungle prod- ucts were subject to periodic increasing demands for more produce and attempts to undermine their political and economic autonomy (Healey 1985b). It is important to note that interior tribal people appear to have had an awareness of the basic structure of the larger trade system in which they were embedded. This is indicated by population movements which were sometimes motivated by a desire to escape from disadvantageous trade relations with down- river agents of coastal states, or to gain easier access to other sources of exotic valuables. The ultimate result was the consolidation of structural instability of state systems, and of the mutually interdependent relation between tribes and states, that was historically reflected in the rise and fall of particular dynasties and states in the coastal belt, and in flurries of war, headhunting, and large-scale migrations of tribal people in the interior (Healey 1985b; Rousseau 1989). But ultimately, the position of states and tribal populations in a large system of political-economic relations ramifying out of the Far East and South Asia was crucially dependent upon the capacity of interior tribal groups to exploit forest resources. An integral aspect of this exploitation was the knowledge base itself. This was continually under potential threat through the tendency of coastal states to incor- porate autonomous tribes of shifting cultivators and hunter-gatherers into the state as dependent communities. This process tended to take the form of conver- sion to Islam and the adoption of sedentary agriculture. Similar processes continue today, under Indonesian government schemes encouraging re-settlement of communities at selected sites along major rivers. One might suggest that a consequence of such forms of incorporation into the state—premodern and modern—is an attenuation of the forms of attachment to, and exploitation of, forest resources by tribal groups, and an ultimate erosion of the traditional knowledge base upon which that exploitation rests. This is of minor consequence for the modern state, given that the primary sources of reve- nue in Borneo are oil, natural gas, coal, and timber products in capital-intensive industrial systems. Indeed, the current oil and timber boom in East Kalimantan Summer 1994 JOURNAL OF ETHNOBIOLOGY 65 makes it one of the richest of Indonesia’s 26 provinces (Pangestu 1989) although annual incomes of the great bulk of the population are extremely low. But while heavy extractive industry has supplanted the importance of jungle products as a source of state revenue, the collection of jungle produce is still an important source of income for more isolated rural dwellers (Jessup and Peluso 1986). The structure of trade in New Guinea was traditionally quite different from Borneo. In the precolonial period marine shells penetrated far into the interior, but for the most part, there was not the island-wide pattern of trade that may be discerned in Borneo. Compared to Borneo, the vast majority of trade goods were distributed over comparatively short distances, and the individual links between transactors were typically very short—at least on the mainland—often no more than 20 km at most (Healey 1980, 1990; Hughes 1977). But then, the known social and geographic world in the interior of New Guinea was itself typically very small. Nothing like the extensive journeyings that occurred in Borneo has been reported for mainland New Guinea. On the mainland a bewildering variety of artifacts, live domestic and wild animals, marine shells, bird plumes, mammal skins, crops, and other material objects, besides magical and ritual substances and knowledge, was passed in both barter and ceremonial exchange. The landscape was criss-crossed with com- plex, interlocking networks of exchange of material objects, but there were no large-scale, island-wide patterns of trade. Rather, there were a plethora of smaller, interlocking “systems” of trade, each with a rather different catalogue of goods involved. Unlike Borneo, New Guinea was never directly incorporated as a major sup- plier of forest products (or other goods) into major maritime trading spheres centered on precapitalist state systems. It is true that parts of western New Guinea were a major source of forest and marine products entering the Southeast Asian trading system, but the direct influence of traders and the agents of the Tidore sultanate seems to have been very limited (Hughes 1977). Nonetheless, the New Guinea mainland was the principal source of one forest product that excited Southeast Asian merchants and noblemen from early times: Bird of Paradise skins. The principal sources of these were concentrated in the Aru Islands, the Vogelkop, and nearby islands (Healey 1980). It is unlikely that extensive trade networks penetrated far inland prior to colonization. By the time Europeans became interested in the plume trade as a major commercial operation around the turn of the last century, they generally took over all aspects of production, becoming hunters themselves, or supplying local people with guns and ammunition, as well as managing the export of plumes (e.g., Doughty 1975; Gilliard 1969). Traditional patterns of production and supply were therefore only minimally incorporated into the large-scale commercial exploitation of the birds centered upon the European millinery industry. Nonetheless, bird plumes were probably the only item, other than marine shells, that ever enjoyed a widespread usage in the interior, and which were traded over an extensive area focussed on the central highland valleys of the eastern half of the island (modern Papua New Guinea) (Healey 1980). In essence, trade in plumes in this region converged on a comparatively small central area of densely populated highland valleys that “consume” large volumes of plumes as 66 HEALEY Vol. 14, No. 1 decorations and valuables, but which lacked direct access to adequate supplies because of extensive deforestation. Goods sent in exchange for plumes varied according to sector, but included marine shells, pigs, stone tools, salt, and various artifacts. Patterns of trade in such goods of localized provenance as stone tools, salt, and mineral pigments tended to radiate from the source areas, often on intersect- ing paths. But dominating patterns of trade in many inland regions were often defined by the flow of shells and plumes towards limited central areas of high consumption. In that sense, trade patterns within New Guinea, to the extent that we can reasonably refer to distinctive general orientations, tended to be directed towards an internal “sink,” rather than diverging towards overseas consumers, as in Borneo. To a large extent I think this can be attributed to the lack of a long- established demand for the products of New Guinea overseas, and the attendant minimal development of trading networks beyond the region. But, as in Borneo, trade in New Guinea continued into modern times. While some trade networks suffered substantial decline, especially in island regions (e.g., Harding 1967), others saw considerable growth in terms of the inclusion of new items of value, volumes of goods in circulation, velocities of passage of goods, and geographic scope of trade links (e.g., Healey 1990). Much trade that has persisted into the present remains essentially precapitalist in its organization. Although money is now widely used in trade, it has frequently been incorporated as a valuable rather than as a currency. Thus, the presence of money in trade does not amount to monetization, if the currency aspect of money is suppressed (Healey 1985a). Transformations that have occurred in trade systems in New Guinea are not simply the effect of the penetration of a money economy. Indeed, in New Guinea, some exchange systems have shown a capacity to absorb both cash and western commodities leading to an efflorescence of exchange (Gregory 1982; Healey 1990). One factor that insulated traditional trade from monetization was the fact that inland trade networks were not important in the delivery of specialist forest or other goods to international trade systems. New Guinea forest products were essentially of little commercial value to the outside world, except in small amounts for the curio market, and international trade never depended upon traditional trade for its supply. With a lack of pre-existing networks and infrastructure, the commercial, village-based exploitation of forest products has never significantly developed in New Guinea, nor has traditional trade in forest products served as a source of government revenue. By contrast, the exploitation of forest products has become a significant source of cash income for villagers and others in Borneo. In short, the history of commercialization of forest products and penetration of a cash econ- omy has been different in the two islands. CONTEMPORARY EXPLOITATION OF FOREST PRODUCTS Here I present two brief cases of village-based exploitation of forest resources. This exploitation and its significance for villagers must be considered in its legal context. In Papua New Guinea customary tenure is recognised in national law, Summer 1994 JOURNAL OF ETHNOBIOLOGY 67 together with customary rights of access to forest resources. This contrasts with the situation in Indonesia, where ownership rights to forest land and its products are vested in the national government, which ostensibly regulates trade in such valuable forest products as birds’ nests by registering traders and imposing a tax on sales. Jessup and Peluso (1986) have discussed the ecology of production of forest products in Indonesian East Kalimantan. They conclude that given the legal constraints on ownership of resources imposed by the state, the intrusion of outside collectors, poachers, and smugglers, and the poor regulation of the mar- ket by the state, local communities are unable to manage effectively their commu- nally-held forest resources. The high value of birds’ nests encourages state control as a source of revenue, and also raiding of nesting caves by outsiders, and has resulted in overharvesting in East Kalimantan (Jessup and Peluso 1986: 524). The impact of outsiders in the exploitation of minor forest products in Kali- mantan is illustrated by the case of the village of Long Sule, which in many respects encapsulates much of the past and contemporary progress of “develop- ment” in Indonesian Borneo. The village is located in Kecamatan (subdistrict) Kayan Hilir, Kabupaten (district) Bulungan, on the banks of the Kayan Iut River, a southern tributary of the upper reaches of the Kayan River. Because of the rugged terrain the area has so far escaped the ravages of the timber industry. Long Sule is a small village of about 300 people. It is one of three small villages clustered together in the middle of a virtually uninhabited stretch of hill- forest. Although it is some five days’ walk to the next nearest permanent habita- tion Long Sule is readily accessible to the outside world, as a small mission- maintained airstrip is adjacent to the village. Most of residents of Long Sule and its neighboring villages are ethnically Punan Aput, who were formerly forest nomads in the area. Sedentarization of nomads has a long history in Borneo (Sellato 1989) although the Indonesian government has hastened the process in recent decades in an attempt to consoli- date control over interior tribalists. In common with other forest nomads, the Punan Aput maintained, and indeed still do, important trading relations with stratified sedentary neighbors (Hoffman 1986; Sellato 1989), in this case the Kenyah. In particular, the Punan Aput are renowned locally for their fine rattan weaving. At Long Sule the Punan adopted Kenyah rice cultivation technology, as well as other cultural traits, nota- bly certain styles of dress, Hornbill-dance ritual, and various visual art motifs. The village, however, is not self-sufficient in rice, with the granaries exhausted after five or six months. Thereafter villagers depend upon subsidiary garden crops such as bananas, cassava, and taro. Traditional collection of wild sago and other food gathered from the forest, as well as hunting with blow-pipe and spear, and fishing remain important components of the subsistence economy. However, much rice is also purchased from the three local stores in the village. This rice is flown in by light aircraft, and derives from the surplus production of Kenyah agriculturalists elsewhere in Kalimantan. Outsiders as traders are thus an integral element of the village, providing both food and other commodities sold in small stores in the village to help meet Punan subsistence needs, and a source of limited income. 68 HEALEY Vol. 14, No. 1 Besides non-Punan (Bugis and Kenyah) store keepers, there are other out- siders resident in the village. These include several Kenyah men engaged in farming and collecting of forest products for the market, a pastor, and a school teacher and their families, and a group of Javanese alluvial gold workers. Long Sule is thus a small multiethnic community. Outsiders are mostly attracted by economic opportunities, and as providers of both goods and income are crucial to the very limited engagement of the Punan in the cash economy. There are no local cash crops, and the Punan are minimally involved in the collection of birds’ nests or aromatic garu woods. Both are found locally, though bird nesting caves are several days’ walk away. The Punan gain meager supplies of cash primarily through the sale of finely woven rattan bags and other handi- crafts to Kenyah and Bugis traders in the village, and by working as casual labor for Javanese gold prospectors. Despite the proximity of the airstrip, the Punan of Long Sule seldom travel, lacking the cause or financial means to do so. The airstrip is, however, a crucial factor in the current organization of the village, for it enables visits by officials of the government, the mission, and commercial interests such as geologists. It also facilitates the resupply of stores and the import of an impressive array of modern technology: generators, radio and television receivers controlled by store holders, as well as various items owned by Punan families such as a few outboard motors for canoes and children’s tricycles. Such possessions are evidence of the capacity of at least some households to amass quite considerable sums of cash despite the lack of local opportunities. Traditionally, the local Punan did not harvest birds’ nests or garu wood. Despite their value, and comparative ease of transporting these products to coastal markets, they still do not harvest them, leaving such exploitation entirely to Bugis and Kenyah traders. While Punan knowledge of the forest and skills as blowpipe hunters and gatherers remain an integral aspect of their subsistence utilization of the forest, they appear to have relinquished any possible collective control over, and management of, commercially valuable forest resources other than rattans for weaving. Arguably, it is the indirect intrusion of the state, which has encouraged and attempted to regulate trade in forest products, that has attracted outsiders to the village as commercial collectors, and inhibited the entry of the Punan into the trade and their potential role as managers of the resources. The scale and extent of the penetration of external commercial interests into this small village far from markets is striking. The situation in the Papua New Guinea village of Tsuwenkai is quite differ- ent. Though comparably isolated, the Kundagai Maring of Tsuwenkai have per- haps had more experience of the wider world. On the other hand, they retain customary control over their own lands, and outsiders—invariably mission per- sonnel—are rarely resident in the village. Tsuwenkai is located at about 1600 m above sea level on the flanks of the western Bismarck Range in the Jimi Valley, Western Highlands Province. One of over 20 Maring-speaking villages, Tsuwenkai is a community of about 300 people. Self-sufficient horticulture is the mainstay of the local economy. Most households earn modest income from smallholder production of coffee and occasional migrant Summer 1994 JOURNAL OF ETHNOBIOLOGY 69 labor. Money is now regarded as essential to meet demands of bridewealth and other ceremonial exchanges, and to purchase small luxuries and household com- forts, but it is not necessary for mere subsistence, as in Long Sule. Investment in items of industrial manufacture beyond simple hand tools, basic kitchen equip- ment, and small household luxuries like radios and tape recorders is virtually nonexistent. Kundagai territory in Tsuwenkai includes extensive tracts of high altitude primary forest, harboring several species of birds widely valued in much of the highlands, including various Birds of Paradise. The Kundagai have long been significant producers of plumes for trade towards central highlands consumers. Case history material indicates increasing export of plumes to the central highlands since just before first contact with the colonial administration in the mid-1950s. Until the 1980s little of this trade was directly with central highlanders, but rather over shorter links with trading partners in more nearby communities. In the two decades after initial contact the rate of trade in plumes increased considerably. This intensification of trade was sustained by an increase in the importation of plumes from more peripheral areas which were then passed on to ultimate consumers, rather than by increasing hunting locally. Local hunting, however, is important in augmenting the supply of plumes, and ensures that the Kundagai are able to export greater volumes of plumes than they import. Levels of hunting are regulated by a combination of social, technical, and ideological factors: (1) individual and collective property rights, by which only members of the local community may hunt plume-bearing birds, and by which individuals may lay claim to exclusive hunting rights at particular sites, such as Bird of Paradise display trees, hunting blinds, fruiting trees, and SO On; (2) explicit conservatory practices, in which hunters refrain from kill- ing all male Birds of Paradise visiting a communal display site, or fre- quenting a particular tract of forest, and voluntarily limit their hunting of female and immature birds; (3) beliefs that spirits are angered if hunters kill too much game. Just what amounts to “too much” is, however, equivocal, as the Kundagai also believe that unusual success is a sign of the favor of the spirits. Such beliefs are thus of dubious import in limiting hunting. However, dietary taboos may have some significafice. These apply to only some game— mostly mammals prized for their flesh—but make hunting generally less appealing to those subject to taboos. However, the eradication of warfare has resulted in most taboos assumed by warriors falling into abeyance, while Christianization has further eroded traditional taboos. (4) the restriction of hunting to simple technology and traditional techniques. Keen hunters display a detailed knowledge of the forest, much of it based upon personal experience. Their success in a hunting expedition is frequently limited by technical means of securing prey, rather than in their skills in locating it. The Kundagai own no shotguns, depending mainly on bows and arrows, traps, and improvised weapons. 70 HEALEY Vol. 14, No. 1 Although property rights, voluntary restraints, and technological factors limit hunting pressure with beneficial consequences for the conservation of game, it is not necessarily the case that the consequences are intended by the Kundagai. It is clear that on the basis of their extensive biological lore many Kundagai are conscious of the need to restrain hunting rates. It is another matter, however, to suggest that this appreciation alone, by individuals, is sufficient to have an effect upon levels of hunting by the community at large. Rather, I would suggest that what ultimately limits Kundagai hunting of plume-bearing birds is the cul- tural ideal of equivalence and egality in exchange. This ideal inhibits the emerg- ence of competetive and incremental exchange, and separates status from prowess in exchange. As such, trade is not a means for the accumulation of material goods or the generation of profit, and individual's participation in trade can be sustained adequately by only modest involvement in hunting. Should trade become commercialized, however, inherent restraints on hunt- ing pressure would be seriously modified, to the possible detriment of the capac- ity of the Kundagai to maintain a sustained harvest of plumes. In fact, by the mid-1980s, trade was increasingly monetized, and the traditional scale of custom- ary exchange rates for different goods had given way to vigorous bargaining over price between some traders. Contrary to the situation elsewhere in Papua New Guinea, this did not result in increased hunting of birds as a means of earning money. The reason lies in the fact that central highlanders’ demand for plumes from the Kundagai declined dramatically in the 1980s. Plumes have largely been replaced by money and other goods in ceremonial payments, and the occasions for traditional ceremonies requiring plumes as decorations greatly diminished. Central highlands traders no longer come as purchasers of plumes, but as sellers of pigs. Increasingly, they demand cash for pigs in their efforts to accumulate money for bridewealth and other payments. The Kundagai themselves were con- strained to acceed to these changing demands in trade in order to maintain their own high demand for pigs for re-deployment in life-crisis prestations. In essence, the declining demand for plumes in the central highlands and the growing emphasis on cash in exchange are symptomatic of the social and cultural transformations in the central highlands wrought by the intrusion and consolida- tion of the modern state. The political economy of the modern capitalist state in Papua New Guinea has rendered certain traditional objects, such as bird plumes, redundant to the social order. As such, it has ultimately undermined traditional trade where the objects of trade have no significant commercial value in national and international systems of exchange. This does not mean, of course, that hunt- ing is of no consequence in communities like Tsuwenkai. It is still a means of provisioning local demand for plumes and meat, and as a pleasurable pursuit in itself for the enthusiast. CONCLUSION The comparison of the exploitation of biological resources of the forest in Borneo and New Guinea indicate alternative consequences of the intrusion of the state. The incorporation of traditional, subsistence-oriented people as dependent communities within the wider political-economic structures of the modern na- Summer 1994 JOURNAL OF ETHNOBIOLOGY 71 tion-state does not lead to an inevitable commercialization and overexploitation of forest resources. This is a likely outcome where traditionally exploited forest products continue to have value in wider national or international systems of exchange, or acquire such value. This occurred in Borneo. On the other hand, in at least some parts of New Guinea, the intrusion of the state led to the modification, even collapse of traditional trade and associated systems of production. This has been to the detriment of some communities, such as the Kundagai, who are now enmeshed in an impoverished position as suppliers of cash to comparatively wealthy pig-providers of the central highlands who no longer desire Kundagai plumes. On the other hand, it has arguably meant that the traditionally exploited biological resource base has enjoyed a measure of protec- tion it might have otherwise lacked. In both examples of the impact of the state on the relation of subsistence- oriented communities to their forest resources we have seen significant changes in patterns of exploitation. The Punan of Long Sule abandoned a nomadic forag- ing lifestyle, adopting shifting agriculture and a more-or-less settled residence pattern. This amounts to a radical modification of their traditional relation with the forest though I do not know if this involved any substantial change in their ethnobiological knowledge base. Similarly, the collapse of the plume trade for the Kundagai has meant that hunting of plume-bearing birds is of little consequence as a specialist activity. In itself, this does not inhibit hunting, but it may well result in a shift in the focus of the hunt, for example, from plumes to meat. In both cases we can observe changing patterns in the exploitation of biolog- ical resources. I suggest that this is liable to have consequences for the eth- nobiological knowledge base itself, where that knowledge is significantly shaped y experience. For example, changing patterns of interactions with the environ- ment, as a consequence of the impact of the state, may lead to progressive loss of certain traditional skills, such as hunting, or a selective withering of the tradi- tional knowledge base?. So far this does not appear to have occurred among younger generations of the Kundagai. Incorporation into the modern state and a global economy have not yet resulted in a marked erosion of ethnobiological knowledge. However, a critical implication of the cases reviewed in this paper is that the particular forms taken by incorporation may have differential impact on systems of ethnobiological knowledge. These systems may be cognitively ordered in mod- ified forms. But we should not simply expect an inevitable impoverishment of the traditional ethnobiological systems. What I am therefore suggesting is that it is appropriate for ethnobiologists to contextualize their studies carefully by refer- ence to the political-economic constraints represented by the modern global econ- omy. In this way, the points of conflict and transformation between traditional systems of ethnobiological knowledge and intrusive alternative systems of knowl- edge may be more readily identified, rather than assumed. At the theoretical level this may lead to further efforts to strike some accommodation and convergence between ethnoscience, human ecology, and political economy within the holistic framework of anthropological discourse. 72 HEALEY Vol. 14, No. 1 NOTES 1What is meant by “precolonial” varies. Much of Borneo was not strictly colonized, although it fell under nominal control of Dutch and British protectorates from the mid nineteenth century. New Guinea was initially carved up among Dutch, German, and British interests in the late nineteenth century. Australia assumed control of the eastern half of the main island and its archipelagoes until the independence of Papua New Guinea in 1975. The former Dutch possessions in the west became the Indonesian Province of Irian Jaya in 1963. 2See Dwyer (1974) for an example of the loss of hunting skills and associated knowledge among younger generations of the Rofaifo of the New Guinea highlands. ACKNOWLEDGEMENTS Fieldwork in the Papua New Guinea highlands has been conducted at various times between 1972 and 1985 with the financial support of the University of Papua New Guinea, the PNG Department of Natural Resources, the Myer Foundation, the New York Zoologi- cal Society, and the Wenner-Gren Foundation for Anthropological Research. A preliminary field trip to Indonesian East Kalimantan in February 1990 was funded by the Northern Territory University. An earlier version of this paper was presented to the Second Interna- tional Congress of the International Society of Ethnobiology, Kunming, China, October 1990, and I thank participants for their helpful comments. LITERATURE CITED BEEHLER, BRUCE M., THANE K. PRATT, the New Guinea highlands. Oceania and DALE A. ZIMMERMAN. 1986. Birds of New Guinea. Princeton Uni- versity Press, Princeton. BROWN, D. E. 1970. Brunei: The Structure and History of a Bornean Malay Sulta- nate. Monograph of the Brunei Mu- seum, Brunei. CONKLIN, HAROLD C. 1957, Hanunoo Agriculture: A Re on an Integral System of Shifting Cultivation in the Philippines. Forestry Development Paper No. 12, FAO, Rome. DAHLAN, H. AHMAD. 1975. Dari Swapraja ke Kabupaten Kutai. Pe- merintah Daerah oo Kutai Kalimantan Timur, Samarinda. DE CLERCQ, S. S. A. 1889. Het gebied der Kalana fat of vier Radja’s in Westelijk Nieuw-Guinea. De Indische Gids 11:1297-1351. DOUGHTY, ROBIN W. 1975. Feather Fashions and Bird Preservation: A Study in Nature Protection. Univer- sity of California Press, Berkeley. DWYER, PETER. 1974. The price of pro- tein: Five hundred hours of hunting in 44:278-293 ELLEN, ROY. 1983. Environment, Subsis- tence, and System: The Ecology of Small-scale Social Formations. Cam- bridge University Press, Cambridge. FLANNERY, TIMOTHY. 1990. The Mam- mals of New Guinea. Robert Brown and Associates, Carina. FREEMAN, DEREK. 1970. Report on the Iban. Athlone Press, London. GILLIARD, C. THOMAS. 1969. Birds of Paradise and Bowerbirds. Weidefeld and Nicolson, London GREGORY, C. A. 1982. Gifts and Com- modities. Academic Press, London. HARDING, THOMAS G. 1967. Voyagers of the Vitiaz Strait: A Study of a New Guinea Trade System. Pytiuaiecs of Washington Press, Seat HEALEY, CHRISTOPHER J. 1980. Trade in ird plumes in the New Guinea region. beeeyra A of ne Anthropol- ogy M , Occasional Papers in Kuteopolony 10:249-275. . New Guinea inland trade: Transformation and resilience in the Summer 1994 context of eae penetration. Man- kind 15:127-144. . 1985b. Tribes and states in “pre- colonial” Borneo: Structural contra- dictions and the generation of piracy. Social Analysis 18:3-39. . 1990. Maring Hunters and Traders: Production and Exchange in the Papua New Guinea Highlands. Uni- versity of California Press, Berkeley. HOFFMAN, CARL. 1986. The Punan: Hunters and Gatherers of Borneo. University Microfilms International Research Press, Ann Ar No. Department of Prehistory, Research School of Pacific Studies, ade National University, Canberr. JESSUP, TIMOTHY C. and NANCY LEE PELUSO. 1986. Minor forest products in Proceedings of the Conference on Common Property Resource Manage- ment. National Academy Press, Wash- ington D.C. KING, VICTOR T. 1993. The Peoples of Borneo. Blackwell, Oxford. MEDWAY, LORD, 1977. sansa of Borneo. Monograph No. 7, Malaysian Branch of the Royal Asiatic Society, Kuala Lumpur. JOURNAL OF ETHNOBIOLOGY 73 METCALF, PETER. 1982. A Borneo Jour- ney into Death: Berawan Eschatology from its Rituals. University of Penn- sylvania Press, Philadelphia. ETSCHMANN, BERNARD 1973. Be- tween Land and Water: The Subsis- tence Ecology of the pia — astern Nicaragua. Sem Press, New York and London. PANGESTU, M. 1989. East Kalimantan: Beyond the timber and oil boom. Pp. 151-175 in Unity and Diversity: Regional Economic Development in Indonesia since 1970. H. Hill (editor). Oxford University Press, Singapore. ROUSSEAU, JEROME. 1989. Central Borneo: Ethnic a and Social Life in a Stratified Society. Oxford Univer- sity s, Oxford. SELLATO, BERNARD. 1989. Nomades et Ecole d eg études en sciences sociales, Par SMYTHIES, BERTRAM E. 1981. The Birds of Borneo. Third Edition. The Sabah Society/The Malayan Nature Society, Kota Kinabalu/Kuala Lumpur. 74 BOOK REVIEW Vol. 14, No. 1 BOOK REVIEW The Tasaday Controversy: Assessing the Evidence. Thomas N. Headland (Edi- tor), 1992. Washington: American Anthropological Association Scholarly Series Special Publication no. 28. Pp. 255. $19.95 (paperbound). ISBN 0-913167-51-7. The charge of “fraud” is one of the most devastating accusations which can be leveled against a scientist. Science as a whole, not just the individual scientist, can be adversely affected by the charge. Critics of science still use the example of the Piltdown skull to discredit the entire scientific community 80 years after this hoax was perpetrated. This book discusses the controversy about the recently “discovered” Tasaday “Stone Age” people of the island of Mindanao in the Philippines. It gives an evenly balanced presentation, containing articles arguing for the authenticity of the Tasaday, and others maintaining the discovery was a hoax perpetrated by members of the Marcos regime. Arguments on both sides are detailed, well- researched, and compellingly written. These are followed by several more papers by outside researchers attempting to reconstruct the facts objectively from the meager evidence available. In 1971, the Tasaday were described to the world as a Stone Age people with limited technology. Yet, even from the outset it was clear the Tasaday were no longer pristine. At best, they had already been affected by the people who brought them to the world’s attention; at worst, they were a set of local villagers paid to act “primitive.” No anthropologist today believes the early press reports of the Tasaday as holdovers living a life unchanged for millenia. The question now is how much of the early reports was accurate and how much was staged or simply exaggerated. The answer is probably somewhere between the two extreme viewpoints. The only thing certain about the controversy is that not enough is known about the facts. From the beginning of the saga, political and media involvement prevented adequate investigation by trained anthropologists. The entire truth will probably never be known. The number of books published on the Tasaday now outnumbers the Tasaday themselves. This book is, however, welcome in attempting to help settle the con- troversy. It is an issue in which we are all involved. Joseph E. Laferriére Arnold Arboretum of Harvard University 22 Divinity Ave. Cambridge MA 02138 USA J. Ethnobiol. 14(1):75-100 Summer 1994 THE DISTRIBUTION AND ETHNOZOOLOGY OF REPTILES F THE NORTHERN PORTION OF THE EGYPTIAN EASTERN DESERT STEVEN M. GOODMAN Field Museum of Natural History Roosevelt Road at Lake Shore Drive Chicago, Illinois 60605 JOSEPH J. HOBBS The Department of Geography University of Missouri Columbia, Missouri 65211 ABSTRACT.—In this paper we review the occurrence and distribution of rep- tiles known from the northern portion of the Egyptian Eastern Desert and the ethnozoology of these animals as viewed by a local Bedouin tribe, the Khush- maan Ma’aza. Particular emphasis is placed on reptile folklore, local names, taxonomy, use as medicine, and natural history as conceived by the Khush- maan; this information is contrasted with Western scientific thought. In most cases these two views are congruent with one another. The major exception is that the Bedouins consider several reptiles venomous which are not known to be so by herpetologists. RESUMEN.—En este trabajo resefiamos la presencia y distribucién de los rep- tiles conocidos de la porcién norte del Desierto Egipcio Oriental, y la etno- zoologia de estos animales segtin son vistos por una tribu local de beduinos, los Khushmaan Ma’aza. Ponemos énfasis particular en el folclor, nombres locales, usos como medicina e historia natural de los reptiles, tal y como son concebidos por los Kushmaan; esta informacién es contrastada con el pensamiento cien- tifico occidental. En la mayoria de los casos las dos visiones son reciprocamente congruentes. La principal excepci6n es el hecho de que los beduinos consideran venenosos a varios reptiles que no son considerados como tales por los herpetdlogos. RESUME.—Dans cet article, nous présentons une revue de la répartition géo- graphique des reptiles de la partie septentrionale du désert égyptien oriental, ainsi que l’ethnozoologie de ces espéces d’aprés la perception d’une tribu locale bédouine, la tribu Khushmaan Ma’aza. Nous discutons les aspects du folklore liés aux reptiles, les noms régionaux, la taxinomie, les utilisations médicinales et Vhistoire naturelle 4 travers la perception Khushmaan. Ces informations sont confrontées aux pensées scientifiques occidentales. Dans la plupart des cas, les deux perspectives ne sont pas opposées. La principale exception consiste dans le fait que les Khushmaan croient que certains reptiles sont venimeux, tandis que les herpétologistes réfutent cette croyance. 76 GOODMAN & HOBBS Vol. 14, No. 1 INTRODUCTION By the nature of their unusual locomotion, habits, and life-cycles, snakes and lizards are often the subject of intrigue and a unique folklore. In numerous cases these perceptions exemplify the secretive habits and calamitous mystique of rep- tiles, including aspects such as disease, poison, death, and the bizarre. Often times the initial basis for these notions and beliefs appears to be some astute knowledge of an animal’s habits, rather than the fantastic. Sometimes these natural history observations proceed through a series of cultural permutations that enrich the orig- inal information and become the fabric of indigenous environmental knowledge. Many groups of pastoral nomads of the North African deserts have a rich body of information about the natural world, including the local reptiles. This extensive knowledge reinforces the nearly universal observation by ethnozoolo- gists that various groups of people living in intimate contact with the natural world “know so much” about nature (Berlin 1992). To date only a small fraction of Bedouin ethnozoological knowledge had been researched and published. Other than cursory mention in several works (e.g., Bons 1959), little information on the ethnoherpetology of North African deserts has been published. Corkill (1935a, 1935b) discussed snake stories and snake traps from the Kordofan and Darfur provinces of the Sudan. Marinkelle (1959) reviewed the medicinal and nutritional uses of reptiles and amphibians found in the markets of Tunisia and Libya; he also mentioned some folk stories from the area. The folklore of Sudanese Nilotic people regarding a gecko was discussed by Cottam and Cottam (1923). In this paper we attempt to narrow the wide gap in the ethnoherpetology of North Africa with the knowledge possessed by the Khushmaan Ma’aza Bedouins of Egypt’s Eastern Desert. This presentation opens three subsequent oppor- tunities for analysis herein and in future ethnozoological research. First, by exam- ining Khushmaan nomenclature and perceptions of reptiles it is possible to learn how these people conceptualize some of the living things in their environment (Berlin 1992). This cultural information is important in its own right in filling existing gaps in knowledge about Bedouin peoples, and in allowing for poten- tially useful cross-cultural comparisons. Second, the environmental context of this cultural information may be quite instructive to Western science, particularly in disclosing the distribution, habitats, and habits of some Egyptian reptiles. Finally, the disparities between Khushmaan and Western scientific knowledge challenge the ethnoscientist with a puzzle: how can a people with such an inti- mate knowledge of nature be apparently so “wrong” about some major attributes of the animals they know? THE KHUSHMAAN The Khushmaan, a clan of the Ma‘aza, is comprised of some 250 households in Egypt, of which about half are based in the Eastern Desert between the Qift- Qusseir road to the south and the El Koriamat-Zafarana road to the north (Fig. 1). These Arabic-speaking tribesmen immigrated to Egypt from northwestern Arabia g about 200 years ago. They are primarily pastoral nomads, tending camels, sheep, and goats. There is also a hunting and gathering component of Kilometres T ak? FIG. 1—Topographical map of the northern portion of the Egyptian Eastern Desert. their economy. The nomads themselves consume some wild resources, such as the meat of Nubian ibex (Capra ibex nubiana), and sell others, including the seeds of Moringa trees and foliage of Artemisia plants, for cash to market buyers in the Nile Valley (Goodman and Hobbs 1988). The Bedouins also obtain necessary food- stuffs and clothing from sedentary populations (Hobbs 1986, 1989). 78 GOODMAN & HOBBS Vol. 14, No. 1 The Khushmaan classify all reptiles, with the possible exception of Uromastyx (see below) and the little-known marine turtles (which are regarded as fish), in the category duud, literally “worm.” All duud are believed to be egg-layers. Also in this taxonomic category are spiders, centipedes, ants, ticks, caterpillars, snails, beetles, and all other flightless, nonmammalian animals. Khushmaan folk medicine for snakebite relies principally on the intervention of a hawi (feminine, hawiyya), a kind of shaman whose only power is an ability to cure snake, spider, and scorpion bites and stings. Only certain persons can become a hawi or hawiyya. When he or she is an infant, the candidate is visited early in the morning on three successive days by a hawi or hawiyya who gives them a special drink and bestows his or her powers upon the candidate. The hawi or hawiyya does not administer medicine to snakebite victims, but rather breathes upon the bite, sometimes applying spittle to it, and recites special incantations. After five or six days, the patient usually recovers. Notably, the hawi or hawiyya is often supplemented by a “first aid” treatment, either cutting off the flesh around the bitten area with a knife; cauterizing the bite with a red-hot nail; or bleeding the bite by an incision, after blood has been brought to the skin surface by the vacuum action of a cup in which a match has been lit. An elderly Khush- maan man claimed that a piece of flesh from the rakhaam (Egyptian vulture, Neophron percnopterus) applied to the bite is sometimes an effective treatment. THE REPTILES OF THE NORTHERN EGYPTIAN EASTERN DESERT Several excellent works have been written on the reptiles of Egypt; however, the majority of these deal almost solely with the fauna of the Nile system (e.g., Anderson 1898; Flower 1933). In the past few decades some of the vast desert areas of Egypt have been surveyed zoologically and our knowledge of the local reptiles has increased many times over (e.g., Marx 1968; Capocaccia 1977). One area of the country where little information on the local reptile fauna is available is the Eastern Desert. Herein we restrict our discussion of this region to the northern portion, from the Nile Valley east to the Gulf of Suez and Red Sea, and from the Cairo-Suez Road south to the Idfu-Mersa el Alam Road (Fig. 1). This region is broader than the Khushmaan Ma’aza territory. Since 1980 we have been working on joint and independent research projects in remote portions of the Egyptian Eastern Desert, and have made observations and collections of the local fauna and flora. In this paper we summarize data on the reptiles of the northern portion of the Egyptian Eastern Desert, combining our own information with previously collected material housed in museums. To date, 30 reptile species have been recorded in the northern portion of the Egyptian Eastern Desert. These include: Gekkonidae—Hemidactylus turcicus, Ptyodactylus hasselquistii, Stenodactylus stenodactylus, and Tropiocolotes steudneri; Agamidae—Agama agama spinosa, Trapelus mutabilis, T. savignyi, Pseudotrapelus sinaita, Uromastyx aegyptius, and U. ocellatus; Lacertidae—Acanthodactylus bos- kianus, Mesalina guttulata, M. rubropunctata, and Ophisops elegans; Varanidae— Varanus griseus; Scincidae—Chalcides ocellatus and C. sepsoides; Colubridae— Coluber florulentus, C. rhodorhachis, C. rogersi, Lytorhynchus diadema, Malpolon moilensis, Psammophis schokari, P. aegyptius, and Spalerosophis diadema; Elapidae— Summer 1994 JOURNAL OF ETHNOBIOLOGY 79 Walterinnesia aegyptia; and Viperidae—Cerastes cerastes, C. vipera, Echis pyramidum, and E. coloratus. METHODS A considerable portion of the information presented here on the distribution of reptiles in the northern Eastern Desert is unpublished. Distributionally impor- tant specimen records for species not included in Figs. 2-10 are mentioned in Appendix 1. For documentary purposes we have cited the museum registration numbers of exceptional specimen records!. In order to distinguish the information gathered from the Khushmaan infor- mants from knowledge derived from our own work in the area, we have divided each “species” account into several headings. In most cases the information pre- sented under the heading “Distribution” and always under “Comments” is our own; while that under the balance of headings is strictly from the Khushmaan perspective and should not be viewed in the light of Western scientific thought. Any exceptions to this are explicitly noted. The systematic order and English common names generally follow Marx (1968) for reptiles, with the exception of the Agamidae which is after Moody (1980), and Tackholm (1974) for plants. The Khushmaan names for plants and their scientific counterparts are based on collections made by JJH in the Eastern Desert and deposited in the Herbarium of the National Research Center, Cairo, where they were kindly identified by Dr. Loutfy Boulos. The system used to transliterate Khushmaan Arabic words is based on Hobbs (1989). The coordinates of Egyptian localities mentioned in the text are presented in Table 1. THE REPTILES Family Gekkonidae brays; gecko Distribution.—Four species of geckos are known to inhabit the northern half of the Egyptian Eastern Desert: Hemidactylus turcicus (Turkish gecko), Ptyodactylus has- selquistii (fan-footed gecko), Stenodactylus stenodactylus (elegant gecko), and Tro- piocolotes steudneri (Steudner’s gecko) (Fig. 2; Appendix 1). Ptyodactylus is the most widely distributed gecko in the Eastern Desert. Bedouin taxonomy.—No distinction seems to be made by the Khushmaan among different types of geckos. Folklore—The brays is poisonous and people die from contact with it. The venom is in the spittle (riig) and is contracted from it via the animal's tongue and not by bite. The poison may be spread by the brays visiting camps at night and crawling over food utensils or water-carrying vessels. After coming in contact with con- taminated objects, the victim generally becomes extremely ill for about a week and then recovers. During that time the victim has no thirst, and may vomit after consuming liquids. 80 GOODMAN & HOBBS Vol. 14, No. 1 TABLE 1.—Gazetteer of Egyptian localities mentioned in the text. Locality Governorate N. Lat. E. Long. (°) () (°) (‘) Ain Sukhna Suez 29 5 ae 20 Beni Hassan Minya a 54 30 51 El Koriamat Giza 29 18 31 13 Gebel Abul Hassan Red Sea 26 a7 33 21 Gebel Galala el Qibyla Red Sea ca.28 50 32 30 Gebel Gharib Red Sea 28 07 32 54 Gebel Mogattam Cairo 30 02 31 17 Gebel Qattar Red Sea af 05 33 pe Gebel Shayib el Banat Red Sea 26 59 33 29 Gebel Suez Suez 29 55 32 20 Hurghada Red Sea a 14 33 50 Aswan 24 58 a2 52 Ismailiya Ismailiya 30 35 32 16 Katamiya Observatory Red Sea 4 56 31 49 Mersa el Alam Red Sea 25 04 34 54 Qift Qena 26 00 32 49 sseir Red Sea 26 06 34 17 Ras Gharib Red Sea 28 21 33 06 Ras Zafarana Red Sea 29 07 32 39 Suez Suez 29 58 32 33 Umm Diisi Red Sea 27 03 33 15 Wadi Abu Haadh Red Sea 28 18 32 48 Wadi at-Tarfa Red Sea ca.28 25 30 50 Wadi al-Maniih Red Sea 25 33 33 37 Wadi al-Radda Red Sea 27 08 33 20 Wadi Araba Red Sea ca.29 07 32 39 Wadi Arkas Red Sea 28 43 me 01 Wadi Askar Red Sea 29 01 32 04 Wadi el Asyuti Asyut 27 10 31 16 Wadi el Nasuri Suez 30 10 31 29 Wadi Gindali Suez 29 be 31 40 Wadi Hof Cairo 29 53 31 18 Wadi Iseili Suez 30 04 31 ao Wadi Qena Red Sea ca.26 12 32 44 Wadi Umm Haadh Red Sea 26 20 33 23 Wadi Umm Tinaydhab_ Red Sea 27 03 a 13 Wadi Umm Yasar Red Sea 27 03 33 11 Zafarana Red Sea 29 07 32 Ed Bedouin natural history.—Brays are known to eat jaru, the fruits of lasaf (Capparis cartilaginea). The snake as-sill al-argat (Coluber sp., see below) is a recognized predator of geckos. Comments.—No species of gecko is venomous. Similar beliefs on the dangerous nature of geckos are held by people residing in the Egyptian and Sudanese Nile Summer 1994 JOURNAL OF ETHNOBIOLOGY 81 31° 32° KS he = rd f | L - . : Taeceeres \ . f 's ° N ae i \ ) A \ oe hig \ a 1 Ce es a re ie St. Anthony's” Sinai Monastery oe, s Monastery 28° i 27°34 : 26° A aE a ge FD es ats ac RE OO SF I iene 25° bo = ' 1 T " i 32° 33° 34° 3s° FIG. 2.—The distribution of brays in the northern portion of the Egyptian Eastern Desert. Records include Ptyodactylus hasselquistii specimens (closed squares) and observations (open squares), Hemidactylus turcicus specimens (closed triangles), and Tropiocolotes steudneri specimens (open triangles). 82 GOODMAN & HOBBS Vol. 14, No. 1 Valley (Cottam and Cottam 1923; personal observations). It is plausible that geckos occasionally consume jaru fruits. Family Agamidae Abu sayha; Agama agama spinosa; Gray’s agama Distribution.—This species occurs in both the granitic and limestone mountainous regions of the Eastern Desert (Fig. 3). It occurs at high elevations, e.g., the summit of Gebel Shayib el Banat (2,187 m. above sea level). Bedouin etymology and taxonomy.—The word sayha, from which the name of this lizard was derived, means blue in Arabic. The term abu sayha is generally used for male Agama agama spinosa while the females are often put in the generic agamid category hibayna (see next entry). Bedouin natural history—This animal prefers rocky slopes. In the autumn (not in summer) the male abu sayha has red forelegs, like pants; the female is similar but lacks the prominent head spines. Six or seven eggs, very soft (the consistency of the skin on a person’s forefinger) are deposited in rocky clefts. It consumes the fruits of Capparis sp. and ants (Tregenza 1955). This lizard hibernates during the winter and, in this state, cannot move if picked up. hibayna; Agamidae lizards Distribution —Hibayna is the Khushmaan designation for several species of agamid This term lizards inhabiting the northern portion of the Eastern Desert (Fig. 4). This term generally denotes Pseudotrapelus sinaita (syn. Agama sinaita), the Sinai agama, but it is also used for Trapelus mutabilis (syn. Agama pallida and A. muta- bilis), the changeable agama; potentially Trapelus savignyi (syn. Agama savigny)), Savigny’s agama; and often female Agama a. spinosa, also known as abu sayha. T. savignyi is known only from the northern edge of the Eastern Desert (Appendix 1). In summary, any agamid other than male A. a. spinosa is classified by the Khushmaan as hibayna. Bedouin natural history.—Hibayna are known to eat lasaf fruits. They prefer rocky slopes or wadis with mixed sand and boulders. An important predator on these lizards is the snake as-sill al-argat (Coluber sp., see below). dhabb; Uromastyx spp. Distribution —Uromastyx aegyptius (syn. U. spinipes) (Egyptian dabb lizard) and U. ocellatus (eyed dabb lizard) inhabit the northern portion of the Eastern Desert and are known by the Khushmaan as dhabb. U. aegyptius is locally common from the Cairo-Suez road south to Wadi Qena (Fig. 5). It lives in dispersed colonies, generally in sandy or gravelly areas with sparse vegetation. U. ocellatus occupies the southern half of the Eastern Desert; most records are from south of the mountainous granitic area west of Hurghada (Fig. 5). This species lives solitarily Summer 1994 JOURNAL OF ETHNOBIOLOGY 83 31° 3 33° ua . 35° FIG. 3.—The distribution of abu sayha in the northern portion of the Egyptian Eastern Desert. Records include Agama agama spinosa specimens (closed triangles) and observations (open triangles). 84 GOODMAN & HOBBS Vol. 14, No. 1 33° 34° 35° ma 1 30°- | + 30° ) t 29° Tawa. ee + 29° St. “an Monastery Paul 4 alia 28° Lal 28° Ot bs al 27° ND " \ Wadi ay a oy Vv 2\ Gena af 26° sa 0 10 20 30 40 50 kK e ag ” Ki ! T ates qT | U 2 31° 32° 33° 34° al FIG. 4.—The distribution of hibayna in the northern portion of the Egyptian East- ern Desert. Records include Trapelus mutabilis specimens (closed squares) and Pseu- dotrapelus sinaita specimens (closed triangles) and observations (open triangles). Summer 1994 JOURNAL OF ETHNOBIOLOGY 85 in rocky or mountainous habitat with relatively dense vegetation. Marx (1968) reported a third species, U. acanthinurus (Bell’s dabb lizard), from the area; the specimen this record was based on has been examined (FMNH 164664), and it is a young U. aegyptius. A specimen obtained at Ismailiya (MHNP 1974.328) is refer- able to U. acanthinurus; however, some of the collecting details are not clear and this specimen should not be used as evidence for the occurrence of this species in the Eastern Desert. Bedouin taxonomy.—No linguistic distinction is made by the Khushmaan between U. aegyptius and U. ocellatus, although they recognize two types of dhabb: the large one (U. aegyptius) occurring in the habitat of gravel plains such as Wadi Qena and Wadi Araba, and the smaller one (U. ocellatus) in the mountainous regions typified by such locales as Gebel Qattar, Gebel Abul Hassan, Wadi Umm Yasar, and Umm Diisi. Some Khushmaan feel the dhabb belongs to the class known as hayawaan, the true ruminating animals, because it eats only plants. Others, however, point out that the dhabb is an egg-layer, unlike the other animals of the hayawaan category. The Khushmaan find significance in the resemblance between the hands of people or bani adam and the dhabb. Folklore —Some Khushmaan have eaten the flesh of this lizard, but this is haraam (forbidden). When the flesh of the dhabb is placed in a fire, it twitches and shakes. Once a Khushmaan threw a rock at a dhabb, hitting it on the head. The lizard put its hands to its head, like a person with a headache. The dhabb is much respected for saving the Prophet’s life. The Prophet Muhammad was fleeing from a person who wanted to kill him. After the Prophet entered a cave, a dhabb emerged and with his spine-covered tail erased the Prophet’s tracks in the sand, throwing off the pursuer. Bedouin natural history—The small dhabb is particularly fond of eating the flowers and seed pods of markh (Leptadenia pyrotechnica) and sayaal (Acacia raddiana). It also consumes kibaath (Launea spinosa), ‘awshiz (Lycium shawii), hurbith (Lotononis platycarpa), himaadh (Rumex vesicarius), and dharagrag (Trigonella stellata). The Khushmaan explain that the resemblance of the dhabb’s tail to the dhanaba dhabb plant (Blepharis ciliaris) accounts for the plant’s name, which means “tail of the dhabb.” The plant yahmiim dhabbaani (Trichodesma africana), “the dhabb’s yahmiim plant,” is named for the small dhabb’s fondness of eating it. The Khushmaan note that only four animals are active and feed at the hottest part of the day: the dhabb, dhabi (gazelle, Gazella dorcas), badan (ibex), and bill (camel). The dhabb goes into its hole in winter and does not surface for 40 days; there it eats its own dung to stay alive. The small dhabb lives under rocks, not in tunnels like its larger counterpart. Predators include abul-husayn (fox, Vulpes spp.), ihdayii and ‘ugaab (assorted hawks and eagles), and sagr (falcons). The Khushmaan have observed that if you give chase to a dhabb and beat the animal to its hole, it will “surrender” and allow you to pick it up. 86 GOODMAN & HOBBS Vol. 14, No. 1 29 F Porn : Monastery St e Paul's Monastery T ° x° 32° 33° 34° 35 FIG. 5.—The distribution of dhabb in the northern portion of the Egyptian East- ern Desert. Records of Uromastyx aegyptius include specimens (closed squares) and observations (open squares) and U. ocellatus specimens (closed triangles) and observations (open triangles). Summer 1994 JOURNAL OF ETHNOBIOLOGY 87 Family Lacertidae arabuuna; lizard Distribution.—Four species of lacertid lizards are known from the Khushmaan territory and all are collectively known as arabuuna. These include: Acanthodac- tylus boskianus, Bosc’s lizard; Mesalina guttulata (syn. Eremias guttulata), small- spotted lizard; Mesalina rubropunctata (syn. Eremias rubropunctata), red-spotted lizard; and Ophisops elegans, Ménétries lizard (Fig. 6; Appendix 1). Bedouin taxonomy.—The Khushmaan do not differentiate by name among any of the lacertid lizards living within their territory. Folklore —Arabuuna are respected by the Khushmaan and are not to be harmed. “Tf a little boy tries to catch it... an old man says to him, no, don’t kill the lizards, my son; they hold the keys to paradise.” The erebona [ = arabuuna] drinks not like bani adam and most animals, but with its tongue like a dog (Tregenza 1955). Bedouin natural history—tThese lizards tend to live in sandy wadi bottoms with sparse or essentially no vegetation. Known predators include the as-sill al-argat snake (Coluber sp., see below), the raahu (white stork, Ciconia ciconia), and the ghuraab (brown-necked raven, Corvus ruficollis). Family Varanidae warran or waral; Varanus griseus; desert monitor Distribution.—This species is rare and widely dispersed throughout the northern portion of the Eastern Desert (Fig. 7). Folklore—The warran is considered venomous. A Khushmaan informant related how the efficacy against snake bite of a plant called muliih (Reaumuria hirtella) was discovered. Long ago, a man saw a battle between the poisonous aaf snake (probably a cobra) and the warran. The monitor when fatigued and bitten ran periodically to a muliih bush and rubbed itself in it, returning repeatedly to do battle with the aaf. The man watching the fight uprooted the plant. The warran found the plant missing and could not “recharge,” and was quickly dispatched by the snake. Bedouin natural history—The warran is a voracious snake killer and uses its whip- like tail to dispatch prey. It sleeps under trees and does not dig a burrow. When the sun is high and there is no shade, the warran seeks shelter in rodent burrows and other holes, where there may be a hanash (Cerastes cerastes, see below). Comments.—The desert monitor is not venomous. The exceptionally strong and agile tail is used as a whip to stun and sometimes kill prey. The fine, sharp teeth readily cut through flesh. This species is known to excavate burrows (Vernet 1977). 88 GOODMAN & HOBBS Vol. 14, No. 1 s1* 32° i 34° a5” it | l 1 l j We l | l x» e j + 30° | / 7. eee : 29° 3 ’ h yo Monastery — 28°- 28° 27° T° 26° 5 ™ 0 10 20 30 40 50 Kilometres 25° Pl a° 32° | 33° 4° as FIG. 6.—The distribution of arabuuna in the northern portion of the Egyptian Eastern Desert. Records of Acanthodactylus boskianus include specimens (closed squares) and observations (open squares) and Mesalina guttulata specimens (closed triangles). Summer 1994 JOURNAL OF ETHNOBIOLOGY 89 31° Eg x” 34° 3s” \ | r 1 l l L 1 2° 0 (dC COO j - 39° ) we 0UC ee Paes ae - 29° St. Anthony's® Monastery St. Paul's Monastery 28° al 27° 7 gh aml s ra ants 6 Me Oren sila cre cle ele eitntiy Meme IRR) «ih A eR a 26° 25° =| i y | ' ie u , ° " ~ 31° 32° 33° «4 35 FIG. 7—The distribution of warran in the northern portion of the Egyptian Eastern Desert based on sight observations of Varanus griseus (closed triangles). 90 GOODMAN & HOBBS Vol. 14, No. 1 Family Scincidae mallaja; Chalcides spp.; skinks Distribution —Two species of scincids are known from the northern portion of the Eastern Desert: Chalcides ocellatus (eyed skink) and C. sepsoides (Audouin’s sand skink) (Appendix 1). Some records from the area may be of accidental introduc- tions by people; skinks regularly turn up in supplies and are occasionally carted between localities, perhaps most often from the Nile Valley to the Eastern Desert. For example, a specimen of C. ocellatus found by JJH in camp supplies in Wadi Umm Tinaydhab almost certainly was carried from Wadi al-Radda. Bedouin taxonomy.—No apparent distinction is made by the Khushmaan between these two species. Only C. ocellatus was captured in the company of our Bedouin informants. Another lizard, called lukaaz, described as similar to the mallaja and never observed by us, may well be C. sepsoides. Folklore —The mallaja is venomous and responsible for the death of many people, particularly the aged. The virulent spittle (riig) is passed via the tongue instead of fangs, in a similar fashion to the brays. If a person is “bitten” and then goes into sunlight, they will die immediately; even in the shade the chance of succumbing to the poison is great. Bedouin natural history—The Khushmaan consider the mallaja to be a rare animal in their territory. Many middle-aged Bedouin have never seen this animal. It is known to bury itself in guff, the accumulated needle-like leaves of the yasar tree (Moringa peregrina), or in sand. These skinks have the ability to disappear into and move quickly through sand. Comments.—No skink is known to be poisonous. Family Colubridae as-sill al-argat or sill abraq; Coluber spp. Distribution —Al of our records of this genus from the Khushmaan territory are of C. rhodorhachis, Jan’s desert racer (Fig. 8). However, specimens and records of C. florulen- tus (flowered snake) and C. rogersi (Roger’s snake) are known from the northern portion of the Eastern Desert (Appendix 1; Anderson 1898; Flower 1933; Marx 1968). Bedouin taxonomy.—This snake is classified as a type of aaf (probably a cobra). Only C. rhodorhachis was found by us in the company of our Bedouin compan- ions. Another snake, called dhaawi, about 50 cm long, black and white, and with a body shape like as-sill al-argat, may well be one or both of the other Coluber spp. known from the region. Folklore.—The as-sill al-argat is thought to be venomous. No deaths are known, but some bite-victims get sick. An infusion or poultice of the plant muliih is good treatment for the bite of this snake. Summer 1994 JOURNAL OF ETHNOBIOLOGY 91 31° 32° 33° 34° K - 1 1 l a | a L 1 1 on saiac eat eta ees . \ 4 30° Cairo A Suez p N \ f 30° + ) A ke y \ | id A ‘ae | : Aln f ae? Sukh bis ; EI Koriamat Fe ee Zafarana 29° - ae [ Gut } 29° St. Anthony's® \ of Sinai Monastery St. Paul's . Suez Monastery — Ras Gharib @ eo + 28° 28 a Sa os We | Hurghada} e\e : R e d Wy A“ : Be ers .¢ Sea 27° eal os a Mons i, ——_—d Claudianus i a } . tage, ff am la -~ % Wadi ” Qen s \ Poa al yO Guest 4 pees" jQena “i ae a ‘Gift Bie =’ %6 ee Beida j © 10 20 30 40 50 ———— Marsa Kilometres el Alam +25° = a \ T = T 1 | J ' 31° 32° cy 34° 35° FIG. 8.—The distribution of as-sill al-argat and hidhif in the northern portion of the E tian Eastern Desert. Records include specimens and photographs of Coluber rhodorhachis (closed squares), and specimens of Echis coloratus (closed triangles) and observations (open triangles). 92 GOODMAN & HOBBS Vol. 14, No. 1 Bedouin natural history—The as-sill al-argat is rare in the area. The snake’s loco- motion is different from the hanash (viper) in that it is not a side-winder but rather leaves an “S-shaped” track. The as-sill al-argat eats brays, arabuuna, small hibayna, and small abu shawk rodents (Acomys cahirinus). Comments.—No Coluber spp. is known to have fangs or to be poisonous. sill; Psammophis spp. Distribution —The systematic status of Psammophis in the area is unresolved. Marx (1958) described the species P. aegyptius (Saharan sand snake) from the Egyptian Western Desert and distinguished it from P. schokari (Schokari sand snake) by differences in coloration and scale counts. Since Marx’s paper the status of P. aegyptius has varied from a subspecies of P. schokari to full species status (e.g., Kramer and Schnurrenberger 1963; Marx 1968; Welch 1982). There are areas within Egypt where P. aegyptius and P. schokari are sympatric (Goodman et al. 1985), and both have been collected in the Egyptian Eastern Desert (Fig. 9). Bedouin natural history—The sill is often found under bushes such as gurdhy (Ochradenus baccutus), natash (Crotalaria aegyptiaca), and markh where they lie in wait for small birds attracted to the vegetation. They are known to take several types of birds including kalb is-hayl (bee-eater, Merops apiaster), fisaysi (war- blers, Family Sylviidae), and slaygaw (wheatears, Oenanthe spp.). This snake is aggressive and when handled will readily bite people, but is not venomous. It hibernates in winter. Family Elapidae aaf; cobra? Identification and distribution—We have not been able to capture or view any snakes referred to as the aaf in the company of our Bedouin informants. The identification of this animal with a single species of snake is problematical, in part because of the variation in the Bedouin’s descriptions of its appearence. It is reported by some Khushmaan as being a very long venomous snake, with a hood, and green to beige coloration. It is common in the riif (Nile Valley), but not in the desert. A very black aaf was once observed by a Bedouin in Wadi al-Maniih. Another Khushmaan description of the aaf is that it moves in a straight line like the sill, and is whitish grey with white spots (Tregenza 1958). The only elapid known from the Eastern Desert is the rare Walterinnesia aegyptia (Inne’s cobra), which occurs in the northern portion of the area (Appen- dix 1). It is completely black and the record from Wadi al-Maniih may well be of this species. Two other cobras, Naja naje (Egyptian cobra) and N. mossambica [ = N. nigricollis] (spitting cobra) are found in the Nile Valley, but to our knowledge have not been documented in the Eastern Desert very far from the valley. One Khush- maan mentioned that the aaf lives in the middle elevations (300-600 m above sea-level) of the basement-complex mountains, such as Gebel Qattar. Summer 1994 JOURNAL OF ETHNOBIOLOGY 93 = 2 33° 34° 35° ; ee ; : Lait L 1 l rp Caro | . 2° se, Pt ta St. ‘Anthony's ® Monastery St. Peute Monaster 28° : 27°- : Mee a ae ny ee a CO Ne Ne ‘: 25° a= 1 i. a8 - os = FIG. 9.—The distribution of sill in the northern portion of the Egyptian Eastern Desert. Records include specimens of Psammophis schokari (closed squares) and P. aegyptius (closed triangles) and observations of Psammophis spp. (open triangles) 94 GOODMAN & HOBBS Vol. 14, No. 1 Folklore-—In a battle between the warran and the aaf, people first learned about the medicinal qualities of a plant called muliih (see discussion of warran). Family Viperidae hanash; Cerastes cerastes; greater horned viper Distribution.—This species is widely distributed throughout the region (Fig. 10). It tends to occur in sandy wadi bottoms with sparse vegetation. A drainage west of Wadi Qena with a particularly high concentration of hanash was named by the Bedouins “Umm Duud,” “the mother of crawling creatures.” Folklore —The hanash is the only animal which the Khushmaan invariably kill when they encounter it. The virulent venom of the hanash poses a serious threat to these desert nomads, particularly children. Hanash are generally killed by direct blows. If the animal is hidden in a large bush such as gurdhy, the vegetation is set ablaze. Dispatched vipers are often buried in the ground, so that if a person or domestic animal steps on the bones they will not be envenomated. The ani- mal’s entire body is regarded as toxic. Thus, for example, if an ant has been feeding on the carcass of a viper and then crawls on your food or utensils, you may be indirectly poisoned. Bedouin natural history—The Khushmaan report that hanash come in different color phases and with or without horns. All hanash are side-winders. Some Bedouins believe that vipers with horns (abu guruun, the “father of horns”) are always males. It is reported that the ratio of hornless:horned hanash is about 6:1 or 7:1. For example, of seven vipers killed near Gebel Gharib in one week, only one had horns, and of six infant vipers found in one spot, only one possessed horns. There is some disagreement as to whether the fox eats this snake; some informants stated that fox do not consume vipers, while others said they readily do so. Fresh remains of a viper were found in a fox cache in Wadi Umm Haadh. Dogs (kalb) apparently eat hanash with no ill effects. The warran hunts the hanash by swishing its tail in rodent burrows and other holes where the snake resides. When it finds a sleeping viper, the warran strikes the snake with its whip- like tail, causing the snake to raise the front portion of the body. The warran then bites the snake’s head off; it always defeats the hanash. Vipers regularly take slaygaw and fisaysi. They often hunt for these birds from the ground below or from middle branches of markh bushes. When slaygaw (particularly the mourning wheatear, Oenanthe lugens) spot a viper they hover over it and give a distinctive alarm call. Wheatears performing this action are occasionally taken by the snake. In several cases Bedouin were tipped off about the presence of a hanash by the behavior of a slaygaw. This snake is also known to eat abu sayha (Tregenza 1958). Vipers are more common in summer than at other times of the year, and become more so after rain, when vegetation cover and rodent populations increase. They are regularly found under yasar and markh, the latter particularly in the summer. It is thought that during the winter, when vipers are in holes or Summer 1994 JOURNAL OF ETHNOBIOLOGY 31° 32° 33° 34° 3s° l : | l 1 | 1 it di, ae af \ 0° cairo 2 gy @ tie: N \ A hed q 2 a a” A \ ) ) \ (a ae q Ain cp Sukhn \ s a (per Koriamat B\ a i: Zafarana es en, a Gu P29" St Anthony's” gy, of Sinai Monastery St. Paul's * Suez Monastery Ras Ghar a 28° - a col el Sa Oo Ms r fh Y a x 3 > ae ae ge S = Ps — “7 O Hurghada \) R ° d 4 De Wa é \ ee e 27° - ¥ i Boh “ ee : s 8 a = 7