JOURNAL (of) ETHNOBIOLUGY

A.

_Matses Indian Rainforest Habitat

Classification and Mammalian Diversity in Amazonian Peru — Fleck and Harder

Ethnoecology of White Grubs (Coleoptera: Melolonthidae) among the Tzeltal Maya of Chiapas — Gomez et al.

identity and Current Ethnobotanic Knowledge of Francisco Hernandez’s “Cicimatic” — ochoterena-Booth

Management of Trees used in Mursik (Fermented Milk) Production in Trans-Nzoia Distric, Kenya — Mureithi et al.

Faunistic Resources Used as Medicines by Artisanal Fishermen from Siribinha Beach, State of Bahia, Brazil — costa-Neto and Marques

Folk Classification and Conservation of Bamboo in Xishuangbanna, Yunnan, Southwest China — Kanglin et al.

Volume 20, Number 1 Summer 2000

JOURNAL STAFF EDITOR: William Balée, Department of Anthropology, Tulane University, New Orleans, LA 70118

(wbalee@tulane.edu)

ASSOCIATE EDITOR — Alejandro de Avila B., A. P. 533, Oaxaca, Oaxaca C.P. 68000, MEXICO (serbo@anteque m)

copes meage TA ANT: Meredith Dudley, Department of Anthropology, Tulane University, New Orleans, LA

siren paige edu) BOOK REVIEW EDITOR: Michael K. Steinberg, Department of Geography-Anthropology, University of Southern e, Gorham, ME 04038 j du) SOCIETY OFFICERS PRESIDENT: phan M. Pearsall, American Archaeology Division, University of Missouri, Columbia, MO 65: PRESIDENT- ean Adams, Crow Canyon Archaeological Center, Cortez, CO SECRETAR RV/TREASURE R: rut Popper, UCLA Institute of Archaeology, Box 951519 Fowler A-210, Los Angeles, CA 90095-1510 CONFERENCE (¢ Mollie S. Toll, Museum of New Mexico, Office of Arct logical Studies, Box 2087,

Santa Fe, NM 87504

BOARD OF TRUSTEES Eugene N. Anderson, University of California, Riverside, CA Enrique Salm6n, Fort Lewis College, Durango, CO Gail Wagner, University of South Cai, Columbia, SC

Past eemicinst Steven A. Weber, Amadeo M. Rea, Pee S. Wing, Paul Minnis, Cecil Brown, Catherine S. 1, and Nancy J. Lsompige Permanent board member Steven D. Emslie. The editor, president, act esi ‘Z , and conference coordinator.

EDITORIAL BOARD Eugene N. Anderson, seeag Coce of California, Riverside, CA: ethnobotany, China, Maya. Scott Atran, CNRS, P. FRANC se eobniras tay beac donee Puen ‘enabsid Mae science, cc Brent Berlin, Gurion we Georgia, Athens Robert ye, Jr., Jardin Botanico, Liniwereidaul Nacional Artiecaia de iabce: México, D.E.,, ae ethno- lang. Mexico. H. Sorayya Carr, E] Cerrito, Spies zooarchaeolo,

logy. Nina Etkin, University of Hawaii, Honolulu » A: medical ethnobotany, the Pacific. Gayle J. Fritz, Washington Universit ty, St. Louis, MO: paleoethnobotany. Terence E. Hays, Rhode Island College, Providence RI: ethnobiology, P. ew Guin

apua N Chris 3 Healey, Northern ities University, Darwin, AUSTRALIA: ethnozoology, pe and New Guine Timothy Johns, Macdonald College of McGill Gndvensit Quebec, CANADA: chemical ecology, ata East

Africa Harriet V. Kubalein, McGill University, yee CANADA: ethno/human nutrition, First Nations of Canad

Brien A. Meilleur, Center for Plant Conservation, Missouri Botanical Garden, St. Louis, MO: es plant rvation, vison dili gar Naomi Miller, University of P. ine Philadelphia, PA: RES ethnobotany, Near Eastern Archaeology.

Gary Nabhan, Ait Su Desert Museum, Tucson, AZ: ethnobiology, Sonoran desert cultures.

Darrell A. Posey, Oxford Centre for the creating pene - —— Oxford University, Oxford, ENGLAND: natural resource management, ethn

Amadeo M. Rea, San Diego, CA iltural e ology,

Elizabeth J. Reitz, University of peseane Athens, GA: one.

Mollie S. Toll, Museum of New Mexico, Santa Fe, NM istoric ethnobiol

of

The Journal of Ethnobiology is published semi-annually. Manuscripts for publication and book review sections should be sent to the appropriate editors as listed on the inside back cover of this issue

© Society of Ethnobiology ISSN 0278-0771

COVER ILLUSTRATION: Figure of “Cicimatic” reproduced from Hernandez 1959, courtesy of Helga Ochoterena- Booth.

MISSOURI BOTANICAL Nov 1 6 2000 GARDEN LIBRARY

Journal of Ethnobiology

VOLUME 20, NUMBER 1 SUMMER 2000

CONTENTS

ETHNOBIOTICA

MATSES INDIAN RAINFOREST HABITAT CLASSIFICATION AND MAMMALIAN DIVERSITY IN AMAZONIAN PERU Fleck and Harder

1

ETHNOECOLOGY OF WHITE GRUBS (COLEOPTERA: MELOLONTHIDAE)

AMONG THE TZELTAL MAYA OF CHIAPAS Gomez et al.

IDENTITY AND CURRENT ETHNOBOTANIC KNOWLEDGE OF FRANCISCO HERNANDEZ’S “CICIMATIC” Ochoterena-Booth

MANAGEMENT OF TREES USED IN MURSIK (FERMENTED MILK) PRODUCTION IN TRANS-NZOIA DISTRIC, KENYA Mureithi et al

43

FAUNISTIC RESOURCES USED AS MEDICINES BY ARTISANAL FISHERMEN FROM SIRIBINHA BEACH, STATE OF BAHIA, BRAZIL Costa-Neto and Marques

FOLK CLASSIFICATION AND CONSERVATION OF BAMBOO IN XISHUANGBANNA, YUNNAN, SOUTHWEST CHINA

Kanglin et al.

113

BOOK REVIEWS OY, fas $40, ade

ETHNOBIOTICA

The Journal of Ethnobiology is its articles, book reviews, features, and other matter found between the covers of each issue, front to back. It is more, of course, than a paper product, though. The journal has for two ees served asa senses for seaterarepeai basic research, carried out from a variety of persp hic and tem- poral scale, into the relations among people, flora, and fauna. The authors of articles pub- lished in the journal have all had the advantage of seeing their work evaluated through a sometimes painful but mostly indispensable and excellent peer-review process. Now, the authors and their articles that have appeared in this journal, from the first issue of the first volume through the last issue of volume 19, are cross-listed in the fine index to articles prepared by Michael Thomas of the University of Florida. That index, inside its own cov- ers, accompanies this issue. It is the first index of the journal, and on behalf of the Board and myself, I thank Michael Thomas for his donation of time and expertise to this instru- mentally useful project.

The editorship of the Journal of Ethnobiology clearly cannot be a one-person respon- sibility, as my predecessors also have known quite well. It involves a team effort par excel- wage pideocapina! alae ie bodies se display in carrying out its scholarly mission must

e owed t tl idual t team. They are the authors, of course, whose names reside above their written outputs, as published herein. And the players include peer re- viewers of seas ai se — of reviews, translators of abatarss, and sone —e

matters. It is

playa who have donated their time, energy, and eepertibe ‘specifically to the publication

of volume 19 and this, the first issue of volume 20 Ms e., ‘ab issues published since I became editor in May 1999). First, I thank th f reviews, some of whom prefer anonymity. The others are: Karen Adams, Edward Anderson, Eugene Anderson, Alejandro d’Avila, Tim Bayliss-Smith, Brent Berlin, Cecil Brown, Paul Buell, Robert Bye, Javier Caballero, Alejandro Casas, Iris Engstrand, Nina Etkin, Gayle Fritz, Jaap Harden, David Harris, Terry Hays, Eugene Hunn, Timothy Johns, Leslie Main Johnson, Elaine Joyal, Heidi Lennstrom, W. Litcinger, Brien Meilleur, Madonna Moss, Gary Nabhan, Debby Pearsall, Darrell Posey, Amadeo Rea, Paul Richards, Wendy Townsend, Nancy Turner, and Yunxiang Yan.

In addition, I gratefully acknowledge Adeline Masquelier for translation into French and Monica Miranda for translation into Spanish of English abstracts. Appreciation is due Anne Totoraitis of Publications Services, University of Washington, for expert advice on production matters and to Meredith Dudley of Tulane University for tireless and efficient editorial assistance. And I am grateful to Dr. Teresa Soufas, Dean of the Liberal Arts and Sciences of Tulane University, for making available funds that have been used in support- ing aspects of the editorial and publication process here. I can verify that the team is mov- ing the journal down field. Together, as we get closer to the destination of being on sched- ule, I hope that you, faithful reader, can join me in recognizing that diverse, individual ef- forts constitute the essence of this journal.

Journal of Ethnobiology 20(1): 1-36 Summer 2000

MATSES INDIAN RAINFOREST HABITAT CLASSIFICATION AND MAMMALIAN DIVERSITY IN AMAZONIAN PERU

DAVID W. FLECK’ Department of Evolution, Ecology, and Organismal Biology The Ohio State University Columbus, Ohio 43210-1293

JOHN D. HARDER Department of Evolution, Ecology, and Organismal Biology The Ohio State University Columbus, Ohio 43210-1293

ABSTRACT.- The Matses Indians of northeastern Peru recognize 47 named rainforest habitat types within the Galvez River drainage basin. By combining named vegetative and geomorphological habitat designations, the Matses can distinguish 178 rainforest habitat types. The baie Sig basis of their habitat classification system was evaluated by d tics and mammalian species composition by plot sampling, trapping, and hunting in habitats near the Matses village of Nuevo San Juan. Highly significant (P<0.001) differences in measured vegetation structure eatin wee ss among 16 sampled Matses-recognized habitat types. H di tion of palm species (n=20) over the 16 sampled habitat types was reiccted. Captures of small mammals in 10 Matses-recognized habitats revealed a non-random distribution in species of marsupials (n=6) and small rodents (n=13). Mammal sightings and signs recorded while hunting with the Matses suggest that some species of mammals have a sufficiently strong preference for certain habitat types so as to make hunting more efficient by concentrating search effort for these species in specific habitat types. Differences in vegetation structure, palm species composition, and occurrence of small mammals demonstrate the ecological relevance of Matses-recognized habitat types.

Key words: Amazonia, habitat classification, mammals, Matses, rainforest.

RESUMEN.- Los nativos Matsés del nordeste del Peri reconocen 47 tipos de habitats de bosque lluvioso dentro de la cuenca del rio Galvez. Combinando sus designaciones vegetativas y geomorfolégicas de habitats, los Matsés pueden distinguir 178 tipos de bosque. La base biolégica de su sistema de clasificacién de habitats fue evalaada documentands caracteristicas ieeiaies y composicion

, atrapar con I trampas, y cazat en habitats reconocidos | om los Maines cane ‘de Sk comunidad nativa } d <U.UUL enl did 4metros de estructura d tacion entre | 8 16 Here. de caelsnaien d

: & 4 YW sdad hget teahyicia

(n= 20) en los 16 tipos de habitat fue rechazado. Capturas de mamiferos pequefios en aie habitats reconocidos por los Matsés reveld una oneeranre no-aleatoria Pp I yy peq 3). Mamiferos vistos

a q 2¢ . if 4 ry yy ee ae a Aan alanine

A = o

y FLECK & HARDER Vol. 20, No. 1

——- tienen preferencias para cierto ee os eS ——— por los

para qt ea mas eficiente cuando

tran esfuerzos de busca en ificos tipos de habitat Stecsiaes enla estructura de vegetacién, en la composicién de especies de palmeras, y en la

existencia de mamiferos pequefios demuestra la pertinencia ecolégica de estas

unidades.

RESUME.- Les matses, un indien du nord-est du Pérou, reconnaissent

a7 ype d‘habitats (pour lesquels ils si mocaa oe termes Bos de Ja forét . En combinant ces termes

Oo Sg a As s: rd + }

pour les différents habitats g g ie giq distinguer 178 types d'habitats pres biologique de |

a été évaluée en i de la végétation et des espéces ae me re avers heey ser de la végétation dans certaines parcelles-

een piege da

pres du village matses de Nice San Juan. Des différences trés ‘significatives (P<0.001) ont été trouvées abi - hercamrspid Bictirese mesurés pour la végétation dans les 16 types d’h L’hypothése d’une distribution homogéne des espéces de palmiers (n=20) pour les 16 types d’ habitats a été réfutée. Des captures de mammiferes de petite yaille dans dix

— matses arevent a

Pe an:

de sae sib (n=6) € et Lge petits JOT ENF eles: ee ie aaieaane et des signes

rencon

espéces ‘de mammiféres ont une préfé ffi t marquée pour certains types d’habitats reconnus par les matses pour que la chasse soit effectivement plus productive si les efforts Bont soncenties sur ces Dabuaks. _Des ciicrenices concernant la structure de la végétation et la composition des espéces

ainsi que l’existence de mammiféres de petite taille, dcnouice ee écologique des ces unités.

INTRODUCTION

The ongoing deforestation of the Amazon rainforest presents an urgent need to document its diversity and understand underlying ecological processes. Though it is widely recognized that high species richness in tropical rainforests is associ- ated with habitat heterogeneity, the patterns of habitat diversity within rainforest areas are poorly understood. Vegetation classifications of Brazilian Amazonia based primarily on flooding regimes, water quality, geographic location, and non-forest habitats within the Amazon basin (e.g., Pires 1973; Prance 1978, 1979; Braga 1979; Pires & Prance 1985). are useful for understanding variation on a large scale, but they are not suffi y detailed to describe habitat types present in a small local- ity. The classifications of Malleux (1982) and Encarnacién (1985, 1993), which are derived from the knowledge of foresters and local residents, respectively, are more detailed and thus more sensitive to variation within large habitat classes. How- ever, these classifications were designed for comparison of habitats throughout the Peruvian Amazon, and still lack detail, especially for terra firme habitats. De- scriptions of successional stages, initiated yearly by the deposition of sediments along large rivers (e.g., Salo et al. 1986; Lamotte 1990; Kalliola et al. 1991; Campbell et al. 1992), related well to habitat variation on a small scale, but these descriptions

Summer 2000 JOURNAL OF ETHNOBIOLOGY 3

are not applicable to upland rainforest, which covers the vast majority of Amazonia (approximately 85 percent [Prance 1978}).

Some indigenous peoples of Amazonia have extensive knowledge of rainforest communities. This knowledge is reflected in detailed habitat classifications (e.g., Carneiro 1983; Parker et al. 1983; Posey 1983; Alcorn 1984; Posey and Balée 1989; Balée 1994; Shepard et al. in press) which have potential for use in conjunction with scientific surveys, particularly in rapid assessment of rainforest communi- ties for conservation. A case in point is the rainforest habitat classification system of the Matses Indians of Northeastern Peru.

The Matses (also called Mayoruna; Panoan | Amazonian society consisting of about 1500 persons living along the Javari River and its tributaries in Peru and Brazil (Figure 1). In 1969 the Matses established peaceful contact with Summer Institute of Linguistics personnel (Vivar 1975), al- though they reportedly had intermittent contacts with rubber workers between 1920 and 1930 (Romanoff 1984), and it is possible that as early as the fifteenth century some of their ancestors may have been reduced in missions to the east of their present territory (Erikson, 1994). Prior to 1969, the Matses avoided contact by maintaining hostile relations with neighboring non-tribal Peruvians and Brazil- ians, and by staying far from navigable rivers in the area between the Javari and Ucayali Rivers, and to the east of the Javari (Romanoff 1984). In the 1980’s some groups moved away from the inland villages and settled on the banks of the Yaquerana (Upper Javari) and Galvez Rivers. Acculturation of the Matses to the national culture is proceeding rapidly, but because of their recent isolation, older individuals (>30 years of age) still possess undiminished traditional knowledge.

The Matses meet all their nutritional needs through traditional subsistence activities, including hunting, fishing, trapping, horticulture (primarily manioc, plantains, and corn), and collection of wild foods. They continue to procure the majority of their protein from hunting in upland forests for mammals and birds. The Matses use an elaborate system of rainforest habitat nomenclature and classi- fication to organize their knowledge of resource availability in order to conduct and discuss their subsistence activities more effectively. Their system allows them to identify as many as 104 types of primary rainforest and 74 types of secondary (successional) rainforest within the 8000-km? drainage basin of the Galvez River.

Such narrow definitions of habitat types in Amazonia have limitations, and local plant species composition might be better characterized by broad descrip- tions of soil and hydrology gradients (Kalliola et al. 1993). However, we present the Matses system as a complementary tool for describing Amazonian habitat di- versity, particularly in light of the utility of systems of categorization for establishing conservation policy. The Matses knowledge of rainforest habitats holds potential for description of ecological relationships as well as floristic diversity, considering that some Amazonian animals are known to be largely restricted to minor habitat types; for example, collared titi monkeys (Callicebus torquatus) are habitat special- ists in creekside forests (Peres 1993) and ichthyomyine rodents are almost never found away from bodies of water (Voss 1988).

This study was designed to provide preliminary biological descriptions of Matses habitat types and to investigate the extent to which Matses habitat desig- nations reflect quantifiable biological factors. To evaluate the ecological basis of

£ i] )areani dj

- FLECK & HARDER Vol. 20, No. 1

Matses habitat classification, we sampled 16 Matses-recognized habitat types that occurred within a 2 km radius from the Matses village of Nuevo San Juan in north- eastern Peru. The objectives of this study were: 1) to describe the Matses system of rainforest habitat identification and classification; 2) to evaluate if Matses-recog- nized habitat types exhibit distinctive vegetation characteristics with measurements of vegetation density, basal area, = bpest species composition; 3) to ascertain if Matses-recognized habitat typ it different small mammal ition and abundance with data obtained through systematic trapping; and 4) to see if mam- mals observed while hunting with the Matses exhibited differential use of Matses-recognized habitat types.

STUDY SITE

The study area was located along the Galvez River (a tributary of the Javari River) at the Matses village of Nuevo San Juan (73°9’50"W, 5°17’30"S, 150 m above sea level), in the district of Yaquerana, department of Loreto, in northeastern Peru

x : NY i8) SS. 7, ’ Oo : % O A a, © Iquitos eG . Pl PERU o Rivet = 5 java : “e Jenaro Herrera

Nueve

San BRAZIL

Juan

sateen tn Galvez River Buen Pe — ra) ive & Javari 0 100km ¥ ‘'s

FIGURE 1.— Location of Nuevo San Juan study site in northeastern Peru, showing the Galvez River drainage basin

Summer 2000 JOURNAL OF ETHNOBIOLOGY 5

(Figure 1). Average annual rainfall (2900 mm) and average annual temperature (25.9°C) were recorded at Jenaro Herrera, the location of the nearest weather sta- tion, 100 km west of Nuevo San Juan (Marengo 1983). The period of heavy precipitation extends from late December through mid April; July and August are the driest months. The Galvez is a blackwater river with a narrow floodplain that seldom extends more than 0.5 km on either side. Although the dry season is not exceptionally dry, the water level in the Galvez falls impressively, a total of 10 m from April to August.

The area around Nuevo San Juan is primary (“virgin”) rainforest except for gaps caused by windfalls and active and abandoned swiddens (0.5-2 ha horticul- tural plots) that have been cleared since the village was established in 1984. At the time that the Matses moved into the area, no villages had existed in the lower reaches of the Galvez for at least 25 years (Faura 1964). Woodroffe (1914), who visited the lower Galvez in 1905-6 reported that there was apparently no human habitation in the area except for a handful of rubber workers. None of the Matses who were interviewed recall there ever having been inhabitants i in the Galvez drain- age basin other than at the very headwaters. The Matses g rainforest areas within the Galvez basin that they identify as villages or swiddens of other tribes through the presence of pot shards, indicator plant species, such as the palm Elaeis oleifera, or distinctive tati , no such areas exist within a day’s walk from Nuevo San Juan, and so the Matses consider all areas surround- ing Nuevo San Juan that were not cleared by them or are visibly the result of a windfall to be primary forest. And we were not able to detect any areas around Nuevo San Juan that appeared to be advanced successional forest (but see Balée [1989] for the possible anthropogenic nature of apparently primary forests in Amazonia). The habitat classification system here includes only rainforest habi- tats, and so we did not consider beach vegetation or active Matses swiddens, which the Matses classify into at least three types based on the age and/or crop compo- sition of the swidden.

According to Matses informants, over the last 12 years abundance of some game animals has declined and densities of species adapted to secondary forest, such as agoutis (Dasyprocta fuliginosa; see Appendix C for mammal species au- thorities) and pacas (Agouti paca), have apparently increased; however, there is no evidence of extirpation of any species from the area.

METHODS

The data for this study were collected during two field seasons totaling 18 months from 1994 to 1996. From April to July 1994, twelve men from the Matses villages of Nuevo San Juan, Remoyacu, and Buen Pert were individually inter- viewed about the different habitat types that they recognized. An initial list of Matses names of rainforest habitat types was compiled from interview responses about the natural history of the local mammal fauna (Fleck 1997). Later, infor- mants were asked to list as many rainforest habitat names as they could and to describe them, and then to comment upon habitats listed by other Matses infor- mants. Subsequently, the informants were asked how they identified and classified these habitats and about the ecological relationships between mammals and these

6 FLECK & HARDER Vol. 20, No. 1

habitats. Only those habitat names that the Matses listed without my help were recorded in the initial listing, but when asked to describe habitats, they were also asked about habitat types mentioned by other informants. Interviews lasted from about 0.5 to 1.5 hours and were carried out without any other adults present. Trade items were exchanged for interviews, but these were given to informants prior to conducting the interview in order to make it clear that receiving the item did not depend upon the nature of their answers. While accompanying Matses on hunt- ing trips, they were asked to name habitats that we passed through and to explain what characteristics they used to recognized them. It is from these interviews and consultations that the final list of habitats was compiled. Habitat type names that were mentioned by only one informant or that were rejected as valid habitat types by more than half of the informants are not included in this paper.

Sixteen Matses-recognized rainforest habitat types (hereinafter, habitat types) that exist within a 2-km radius of Nuevo San Juan were selected for vegetation sampling. The goal of the habitat comparisons in this study was to determine if the basic classification units (named habitat types) of the Matses system were eco- logically relevant units. The purpose of our sampling design was not to provide a complete floristic or structural description of each habitat type, but rather to de- termine if Matses-recognized habitats could be distinguished one from another with data from limited sampling.

From April to July of 1996, eight 0.02-ha vegetation sampling plots (10 x 20 m) were established within each of the 16 habitat types by randomly selecting a start- ing point and a compass bearing for orientation of the plot. Two to four separate localities of each of the 16 habitat types were sampled; the number of plots per locality was related to the size of the habitat patch. At each plot, eight vegetation density estimates were conducted by using a 1 x 1 m density board marked witha 10 x 10 (10 cm) grid. The board was placed on the ground in a vertical position at a distance of 5 m from the observer and the number of squares more than 50 per- cent covered by vegetation were counted for the bottom half of the board and again for the top half. Diameter at breast height (DBH; 1.3 m) was measured for each tree within each plot; trees with stilt or buttress roots reaching above 1.3 m were measured just above the roots. From the DBH measurements, mean basal area per ha and mean number of trees >10 cm DBH per ha were calculated for each habitat type.

All identifiable palms (Palmae; palm nomenclature follows Henderson et al. 1995) taller than 1 m were identified and counted within each study plot. Palms were selected for study because they are salient, readily identified components of most Amazon rainforest habitats (Kahn et al. 1988), because Palmae is probably the most economically useful Neotropical plant family (Balick 1984), and because

alm fruits and seeds are also important resources for rainforest animals (Zona & Henderson 1989). Other plant species that the Matses indicated as important for identifying habitats were quantified at each plot: the number of Cecropia spp. (Moraceae) trees taller than 1 m, the number of Duroia hirsuta (Poeppig. & Endl.) Shumann (Rubiaceae) trees taller than 1 m, and the number of lianas >1 cm DBH. The following geomorphological data were also recorded at each plot: distance from the river (during highest water level), relative elevation (estimated elevation above lowest land within 50 m), perceived quality of drainage (during the dry

Summer 2000 JOURNAL OF ETHNOBIOLOGY 7

season), and water regime (maximum number of days a plot remains inundated). Duplicate sets of voucher specimens were deposited at the Instituto de Investigaciones de la Amazonia Peruana herbarium in Iquitos, Peru, at the her- barium at the Museo de Historia Natural de la Universidad Nacional Mayor de San Marcos in Lima, and at the New York Botanical Garden.

Two dichotomous keys were constructed for identification of Matses-defined habitats in the Nuevo San Juan area, one for geomorphologically-defined habitats and one for vegetatively-defined habitats (Appendices A and B). These keys were developed based on habitat characteristics used by the Matses in teaching DWF how to identify habitats.

Ten of the 16 selected habitat types and a Mat i idd d for small (<1 kg) mammals. The aim of this trapping was not to describe the entire mammalian composition of each habitat or to test for differences in composition with the surrounding habitat, but to determine if the sampled habitats exhibited detectable differences in small mammal composition. At each of the 10 habitats, 30 Sherman live traps, 10 Tomahawk live traps, 40 Victor rat traps, 10 Museum Spe- cial snap traps, five pitfall traps with drift fences, and five Matses-constructed deadfall traps (total of 100 traps) were set at each habitat type for 10 nights (total of 1000 trapnights). Traps were spaced evenly over a 1-ha area within each habitat type, at least 10 m away from the edge of the habitat, with half of the Sherman, Victor, and Museum Special traps set 0.5 to 2.5 m above the ground. Two of the eight 0.02 ha vegetation sampling plots were established within each trapping plot. Traps were baited every afternoon with ripe plantain (except for deadfall traps, which were baited with manioc) and checked in the morning. All animals were identified and removed; voucher specimens were prepared for all species and from all animals tion. Because the number of traps was limited, the 11 areas were not trapped simultaneously, but rather one or two at a time successively over a period of 90 d (20 April to 18 July 1996), a period that coincides with the end of the rainy season and the beginning of the dry season. Voucher specimens are deposited at the Museo de Historia Natural de la Universidad Nacional Mayor de San Marcos in Lima, and at the American Mu- seum of Natural History in New York.

From April 1995 to July 1996, DWF accompanied Matses on 108 hunting trips for a total of 583 h. The habitat types in which mammals were encountered during each hunt were recorded; two habitat names were recorded for each locality of observation, one for the vegetatively-defined habitat type and one for the geo- morphologically-defined habitat type. The location and habitat of signs of mammals, including tracks, dens, beds, scat, and scrapings were also recorded when the sign could be confidently identified to species with the help of Matses.

In order to determine if the frequency of observations of mammalian species per habitat was different than expected by random distribution, the amount of time spent by Matses hunting in each habitat type was estimated by pacing Matses hunting paths for a total distance of 10 km, recording the points at which habitat types began and ended. Pacing data were then mapped by application of a global positioning satellite receiver (Figure 2). Vegetatively- and geomorphologically- defined habitats were calculated separately. Although the sections of the Matses hunting paths that were paced might not represent a random sample of the habi-

8 FLECK & HARDER Vol. 20, No. 1

tats that the Matses cover on long hunting trips, the path-length estimates provide a rough estimate of the proportion of time that the Matses spend in each habitat type while hunting. Because Matses concentrate search time and hunting effort in selected habitat types, depending on the target species, sampling was not random during hunts. Therefore, these data were not subjected to statistical analysis. Also, this part of the study was not designed to test any mammal-habitat associations put forward by the Matses.

; «@ Matses houses at Nuevo San Juan Acte cuidi cue. small stream gallery forest —_

ens FIGURE 2.-- Geomorphologically-defined (shaded or patterned) and vegetatively-

defined (outlined in white) habitat types in the Nuevo San Juan area, showing overlap of the two classification systems.

Acte dada cue. large stream gallery forest

Summer 2000 JOURNAL OF ETHNOBIOLOGY 9

Vegetation densities, basal area and tree density were compared among the 16 sampled habitat types with one-way ANOVA tests and Tukey multiple compari- sons. A Pearson Chi square test was used to test homogeneity of palm species abundances over the sampled habitats. All data recorded at each plot were used to construct classification and regression trees (CART; Brieman et al. 1984). CART analyses were used to see if Matses-recognized habitat types could be predicted using the measured habitat parameters. Three classification and regression trees were constructed, one for the eight sampled vegetatively-defined habitat types, one for the eight sampled geomorphologically-defined habitat types, and one for all 16 sampled habitat types. The dichotomous habitat keys were then compared — the classification trees to see if the two methods produced similar results and

to | were important distinguishing fac- tors in both. Small mammal species diversity and abundance in the 10 trapped rainforest habitats and one swidden were analyzed using a chi-square test for ho- mogeneity of the distribution of animals (at three levels, family Didelphidae, family Echimyidae, and family Muridae) across the 11 habitats. Exact nonparametric con- ditional inference was used since the trapping data were sparse--there were many zero values for the number of animals of a species captured in a habitat, making large sample methods invalid.

if the same habitat

RESULTS AND DISCUSSION

The Matses recognize 40 named categories of primary rainforest habitats (of which 38 are terminal categories) and seven named categories of secondary rainforest habitats within the 8000-km? drainage basin of the Galvez River (Table 1). The Matses use different names for floodplain habitats while they are inun- dated during the rainy season, but these were not counted as different habitat types in this study (Figure 9). es Matses habitat classification system is divided into two separate (but physically t 1) geomorphologically- defined habitat types; and 2) vegetatively- -defined habitat types (Figures 2-4).

Vegetation density varied significantly among the 16 sampled habitat types. One-way ANOVA tests revealed highly significant (P<.001) differences for veg- etation density among 16 Matses-recognized habitat types, for both vegetation density below 0.5 m (F=58.90; Figure 5A) and from 0.5 m to 1 m (F=65.52; Figure 5B). Similarly, one-way ANOVA tests revealed highly significant (P<0.001) differ- ences for basal area (F=10.41; Figure 5C) and for tree density (F=9.06; Figure 5D) among the 16 sampled habitat types, though these characteristics were consider- ably less distinctive than was vegetation density. Significant differences among habitat types in the measured vegetation structure parameters indicate that habi- tat types are related to vegetation structure and, therefore, of interest for ecological investigation. Moreover, pairwise comparisons of each habitat with each of the other 15 habitats revealed significant differences (P<0.05 Tukey’s pairwise com- parisons) in at least one of the four measured vegetation structure parameters in all but three of 28 pairs of geomorphologically-defined habitats and in all but five of 28 pairs of vegetatively-defined habitats. For pairs of geomorphologically- and vegetatively-defined habitats, a higher proportion (12 of 64) did not differ signifi- cantly (at 95% C.I) in at least one vegetation structure parameter; however, some

10 FLECK & HARDER Vol. 20, No. 1

TABLE 1. Forty-seven named categories of rainforest habitat types recognized by the Matses within the Galvez River drainage basin in northeastern Peru. The Matses classify habitat types according to criteria of two majer ope 1) geomorphological features, and 2) vegetation characteristics. The most i in the Matses classification system is between floodplain rainforest along seasonally flooded rivers and upland rainforest that is not subjected to seasonal flooding. Matses also distinguish between primary rainforest and | habitats. Numbers refer to habitat types sampled in this study and are used in subsequent tables and figures. See Appendix D for linguistic description of Matses habitat terminology.

HABITAT TYPES DEFINED BY GEOMORPHOLOGICAL FEATURES Upland Rainforest

1 quiusudquid Non-flooding forest adjacent to a river 2 manan Hill crest (also called manan dadanquio) pieueiates Hill incline (also called macuésh potsen) tsimpiru Valley between upland hills acte dae cuéman Gallery forest along a large stream acte cuidi cuéman Gallery forest along a small stream 3 dépuen Ephemeral headwaters of a stream 4 mactac Poorly-drained muddy mineral lick tia* Upland palm swam anshantuc Permanently waterlogged swamp acte cuéman Floodplain 5 actiacho Low seasonally flooded forest (called acte mauan during flooding season 6 nacnédtsequid Levee flooded every year 7 mantses Levee flooded only on years of exceptionally high water levels (called mashcad during flooding season) chian cuéman Forest along a floodplain lake acte mactac Mineral lick in floodplain forest 8 itia dapa* Floodplain palm swamp (called itia mauan during flooding season) HABITAT TYPES CHARACTERIZED BY VEGETATION CHARACTERISTICS niméduc,* Primary Rainfores isanchoed Forest dominated by Oenocarpus bataua palms nistechoed Forest dominated by Iriartea deltoidea palms shuinte mapichoed Forest dominated by Attalea tessmanii palms budédchoed Forest dominated by A. butyracea palms 9 miochoed Forest with understory dominated by A. racemosa palms 10 budéd ushuchoed Forest with understory dominated by A. microcarpa palms 11 shubuchoed Forest with emer. dominated by Phytelephas macrocarpa 12 tanacchoed Forest with «id ooebi dominated by Lepidocaryum tenue palms dapaischoed Forest with understory and midstory dominated by A. phalerata palms cobisanchoed Swamp dominated by Euterpe precatoria palms tiantechoed Forest dominated by bamboo séntechoed Forest dominated by Cedrela sp. trees péncadchoed Forest dominated by pencad trees

manipadachoed Forest dominated by Musa sp. wild bananas

Summer 2000 JOURNAL OF ETHNOBIOLOGY 11

13 mayanén sebad Forest with open understory, dominated by Duroia hirsuta trees 14 isitodochoed Forest dense with many large lianas cuéte mampis Forest where only thin hardwood trees grow antinchoed Seasonally flooded forest dominated by A. maripa palms sinadchoed Seasonally flooded forest with understory dominated by Bactris cf. bifida palms shiuishchoed stoic flooded swamp forest dominated by Ficus spp. cana shétachoed y vegetation

SUCCESSIONAL HABITAT TYPES

15 tied shéni d dominated by Cecropia s hats and Marila spp. mayun tied Secondary forest f b d swidd villages > 50 yr old cuesbudaid recent blowdown characterized by creeping vines and no trees isitodo icsachoed Secondary forest thick with vines and young trees bucuchoed Secondary forest dominated by = spp. 16 sedquequid*** Secondary forest from blowd shift with many vines and few Cecropia spp. trees cuéteuidquio tabadquid —_ Secondary forest hard it t competing pi vegetation and vines * itia and itia dap included in both cl ti since they are — by permanently terlogged soil ll as being dominat ty sn pa “niméduc, int g ] fers to all Pp y in th ie ifi (niméd 9) it refers only to primary ae 41 2 ea aes | a | s £088 4 any of t ra} she a M4

ser nike is also used to refer wth a ee (Encarnacion 1993), primary { forest found on sandy soil here all trees are short and thin. The only chamizal in een far from Nuevo San Juan.

of these pairs were not expected to differ because they often overlap physically in nature.

Twenty species of palms were identified in the sampled plots of the 16 habitat types. Genera that could not be identified with confidence to species in the field (Geonoma and Bactris) were excluded from analyses. The null hypothesis of homo- geneity of the distribution of palm species over the habitats was rejected by the Pearson Chi-square analysis. In fact, some species of palms were present in 100 percent of the eight sample plots of some habitat types and absent in nearly all plots of other habitat types (Table 2; Figure 6). This is not surprising considering that Matses recognize and name many of their vegetatively-recognized habitats after palm species (see Appendix D for linguistic analysis of habitat names). Habi- tat types that had 100 percent frequency of occurrence of a palm species also had relatively high mean densities of that palm species.

Matses-perceived habitat types could be predicted with classification and re- gression trees using the measured variables. The classification and regression trees (CART) analysis of the eight geomorphologically-defined habitats correctly clas- sified all (N=64) of the sample plots into their Matses-recognized habitat type

12 FLECK & HARDER Vol. 20, No. 1

(Figure 7A). In the case of the eight vegetatively-defined habitat types, only four out of 64 sample plots (6.25%) were incorrectly classified (Figure 7B). When all 16 habitats were analyzed simultaneously, 12 of 128 (9.375%) were misclassified (i.e., the CART analysis categorized 12 plots differently than the Matses did). The CART analyses produced trees that were similar to the dichotomous identification keys developed using apie information (Appendices A and B), with many nodes at the same position

The 10 i habitats revealed differences in small mammal species com- position (Table 3), as well as species richness and abundance (Figure 8). The chi-square test for homogeneity of the distribution rejected the null hypothesis of homogeneity, indicating that the distribution of small mammals varies among the habitat types.

A

3 aes if — water level after rains quiusudquid——jmacuésh}—_mactac —-+_macuésh-+—-manan—+macuésh|}—dépuen— terra firme by river incline —_—smineral lick incline hillcrest incline , Stream

headwaters Habitat Number: 1 4 3

46 é pie ae

Pant

ap,

- alee s\t} I r ee cs “|

cami Ne ruaeZ : * Dry season tnacnedtséquid}———actiacho_—_+—__mantses_—__—itia dapa—_—+|

names low levee island dry floodable forest high levee palm swamp cay season }—acte mauan——+_-mashcad_—+_—itia mauan |

: flooded forest high levee island _ flooded palm swamp Habitat Number: 6 5 i / 8

FIGURE 3.-- Profiles of geomorphologically-defined habitat types: A) upland forest habitat types; B) floodplain forest habitat types, showing the annual range of water levels and dry season and rainy season names for the same habitat type.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 13

A large proportion of the observations of some species of mammals were in certain habitat types (Table 4). Many of these values for sightings or signs were more than one order of magnitude higher than would have been expected based on the estimated amount of time Matses spend hunting in each habitat. This sug- gests that despite the large sampling bias, the listed species might show an actual preference for those habitats.

The Matses system of habitat classification is different from other published rainforest classification systems (e.g., Pires 1973; Prance 1978, 1979; Braga 1979; Malleux 1982; Pires & Prance 1985; Encarnacién 1985, 1993) in that it recognizes an exceptionally large number of named habitats for a relatively small area and in that it uses two overlapping subsystems (geomomorphological and vegetative), rather than being strictly hierarchical. This study showed that these Matses-recog- nized habitat types can be recognized based on standard floristic and structural features (Figure 5). Moreover, these habitat types can be correctly predicted by CART analysis (Figure 7) and they can be identified with dichotomous keys (Ap- pendices A & B).

aerial

ye

serene m stemless palm forest stemless palm forest stemless palm forest treelet palm forest Habitat Number: 9 10 11

avait

hid

4 Th

: ALS : pease

L a a isitodochoed———+———tied shéni_ + sedquequid

‘demon’s swidden’ liana forest abandoned swidden secondary forest Habitat Number: 13 14 15 16

FIGURE 4.- Profiles of vegetatively-defined upland habitat types: A) habitat types dominated by understory palms; B) habitat types with other characteristic vegetation structures.

14 FLECK & HARDER Vol. 20, No. 1

Geomorphologically-defined habitats are identified by abiotic features includ- ing distance from a river, relative elevation, drainage quality, and water regime. Habitat types such as manan? ‘hill crest’, actiacho ‘seasonally flooded forest’, and quiusudquid ‘terra firme next to a river’ are identified using geomorphological features (Figure 3). All the rainforest in the Galvez River drainage basin is included in the geomorphologic classification system (Figure 2). Floristic composition and vegetation structure can be affected by water regime, drainage, topography, and

BOTTOM VEGETATION DENSITY TOP VEGETATION DENSITY

= 50 _ 0 = s B 0 o x 40 Z 40 ra a te 30 | 30 Ww Ww 6 5 © 20 © 20 e 8 <= 10 < 10 ae | 5 lial w” a ;

0 0 |

13459103 12118147 6121516 13459103 12118147 6121516 HABITAT HABITAT BASAL AREA (M7HA) TREE DENSITY

60 C 1000 + = et > 800 + Ww 2 40 7 o rad 3 600 A 30 - 593

<

& <= 400 - 2 20 + e 8 id all a

i :

4 1615141331279 610251811 4161514133127 9610251 811 HABITAT T:

FIGURE 5.— Mean (+SEM) vegetation density, basal area (m*/ha) and trees (>10 cm DBH) per ha for 16 Matses-recognized habitat types. Habitats in panel A are listed in order of increasing mean number of squares covered; habitats in panel B are listed in the same order as in panel A to illustrate differences in horizontal vegetation densi

between the lower level (0-0.5 m) and the higher level (0.5-1 m). Habitats in panel C are presented in order of increasing mean basal area; in panel D habitats are in the same order as in panel C to illustrate differences between basal areas and trees per ha in the same habitat types.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 15

TABLE 2.— Frequency of occurrence of 20 palm species in Matses-recognized habitat types. Values represent the number of 0.02-ha plots, out of eight sampled per habitat type, in which a species was recorded (values appear in bold type when the palm species is part of the name of the habitat type). Numbers preceding palm species (1-20) correspond to numbered drawings in Figure 6. Habitat type numbers (1-16) correspond to numbered habitats in Table 1 and Figures 2-4).

PALM SPECIES HABITAT TYPES bei 374 5 O° 7 8 8 1 11 ie is 14 15-36

1 Astrocaryum chambira 3 1 > s | ar ae: 2 Astrocaryum jauari 2

3. Astrocaryum murumuru 62 2 2 EE Sa ee 4,4.) 4 Attalea butyracea ee eS 2

5 Attalea insignis 2 3 6 Attalea microcarpa 2) 8 1 1 1 7 Attalea racemosa 5 8 aot a8 8 Attalea tessmantii 1 1 5 1 9 Euterpe precatoria i$ 6 8.8 8.6 Ye 3

10 Hyospathe elegans 3 1 1 nee

11 Iriartea deltoidea - 24 4 3 3 1

12 Iriartella stenocarpa Z 1

13 Lepidocaryum tenue 1 8

14 Mauritia flexuosa 8

15 Oenocarpus bataua 7. D0 3 S40 4°43 2 16 Oenocarpus mapora 1 Sie ae |

17 Pholidostachys synanthera 3 2

18 Phytelephas macrocarpa 3 3 8 2

19 Socratea exorrhiza 2.7. 6 3 2

20 Wettenia augusta ae 1

11 12 13 14 151617 18 19 20

FIGURE 6.— Drawings (to scale) of palms species identified in Matses-recognized habitats. Numbers correspond to numbered palm species in Table 2.

16 FLECK & HARDER

eee A distance from river <5m Renae PS x relative relative

elevation <7 m elevation >7 m

Mauntia Maunttia top top flexuosa flexuosa density density <3.5 >3.5 <25.875 >25.875

Actiacho Itiadapa Nacnéd. Mantses 5 7

Vol. 20, No. 1 | distance from river >5m | trees per trees per hectare <575 hectare >575 top top distance distance

density density fromriver from river <8.375 >8.375 <330m >330m

Mactac Dépuen Quiusud. Manan 3 1 2

ey | B top density top density <31.8125 >31.8125 aoe Wh me i ae | Attalea Attalea vines per vines per innleen i: <3.5 oe >3.5 plot <28.5 plot >28.5 | Attalea Attalea Budéd Cecropia Cecropia top top racemosa racemosa Ushuchoed _ spp. spp. density density <4.5 10 <22.5 >22.5 “a a 875 bottom bottom Miochoed Tanacchoed Tied Shéni_ Isitodo. Sedque* density density 9 14 16

Mayanén Shubuchoed* 13 10

* Habitat category contains 2 misclassified plots

FIGURE 7.— A) Classification and regression tree for eight geomorphologically-defined habitat types. Misclassification error rate was 0% for the 64 plots. Measured habitat characteristics used to construct the tree included distance of the plot from the Galvez river, relative elevation above lowest land within 50 m, number of trees per ha, number of Mauritia flexuosa palms in the plot, and horizontal vegetation density 0.5 to 1 m above the ground. B) Classification and regression tree for eight vegetatively-defined habitat types. Misclassification error rate was 6.25 percent (4 of 64 plots). Measured habitat characteristics used to construct the tree included horizontal vegetation density 0.5 to 1 m above the ground, number of Attalea microcarapa palms in the plot, number of vines, number of Attalea racemosa palms, number of Cecropia spp. trees, and horizontal

vegetation density below 0.5 m.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 17

distance from the river, and other physical factors (Duivenvoorden 1996), so geo- morphologically-defined habitats generally contain a circumscribed range of species and predictable structures.

Vegetatively-defined habitats are identified primarily by the presence of an obvious dominance by a plant species (e.g., certain palms) or plant life form (e.g., lianas) throughout the habitat. Miochoed ‘forest characterized by Attalea racemosa (an understory species of stemless palm)’ and isotodochoed ‘forest characterized by lianas’ are examples of habitat types defined this way (Figure 4). Named veg- etatively-defined habitats cover only 10-15 percent of the rainforest (Figure 2). The remainder of the area, called niméduc is not differentiated in the Matses classifica- tion.

Because the entire rainforest is divided into both geomorphologically defined habitats and vegetatively-defined habitats, the two must overlap (Figure 2). Some vegetatively-defined habitat types can occur in several different geomorphologi- cally-defined habitat types, so there is not a one-to-one correspondence between the two subsystems. By combining tl land the tative habi- tat classifications, the Matses can refer to any locality with more detail and efficiency (Table 5). This seems to be a very practical solution considering that it would re-

TABLE 3.— Captures of mammal species during 1000 trapnights in each of 10 Matses- recognized rainforest habitat types. (Common names in Appendix C.)

MAMMAL SPECIES TRAPPED HABITATS TOTAL OCCUR- CAP- o Sw oO ss 8 12.13 14. 15 RENCE -TURES

Didelphis marsupialis 1 2 Marmosa murina

Marmosops noctivagus

Metachirus nudicaudatus

Micoureus spp. 1 Philander mcilhennyi 1

Oecomys bicolor 1 Oecomys cf. slits 3 Oryzomys peren 1 Oryzomys ean cf. 1

Scolomys ucayalensis 1

Mesomys ferrugineus 1

Proechimys cuvieri 1 1 Proechimys sp. 1 2

Proechimys sp. 2 ee ee 2 Proechimys sp. 3 1

Proechimys sp. 4 3 S14 Proechimys sp. 5 Proechimys sp. indet.* Total Species (20) Total Captures . 6 7 6 16 18 13:2 6 % skulls hed by kill bar.

BP OWrRNR eB N eS — Set ee PSO Co 0) ae a

PNN WP RNR RP RRP RPNHP DS TIN W

i=) WWwWN W Dd pod

Ke)

N — oF N We

woe N a A Sarer oo oo

_ No oS

18 FLECK & HARDER Vol. 20, No. 1

SPECIES RICHNESS

oi oO |

A

marsupials

echimyids

NUMBER OF SPECIES OC]ANWARADNOO |

= P = murids

15119 13 8 7 1214 3 5 Swidden

HABITAT ABUNDANCE ” 50 B oO marsupials > 40 oa < CO 24 i. Oo echimyids ti ra) ag faa) a If H za ae E E oF murids

15 119 13 8 7 1214 3 5 Swidden HABITAT

FIGURE 8.— Species richness (A) and abundance (B) of marsupials, echimyid rodents, and murid rodents based on data from 1000 trapnights in 10 Matses-recognized habitat types (and in a Matses manioc swidden). Data are presented for three groups (taxa): marsupials (family Didelphidae); echimyid rodents (family Echimyidae); and murid rodents (family Muridae).

Summer 2000 JOURNAL OF ETHNOBIOLOGY 19

quire much repetition to include such detail in a strictly hierarchical system. It should be noted that lexemes for the two classes of habitat designations are not combined by forming compounds or lexicalized phrases, but rather may simply be mentioned in the same conversation to designate a more specific habitat type or to describe a particular locality.

The Matses system is al habitats | icular habitat type have several attributes in common (i.e., the categories are Sais: or logi- cally natural). Because traditional societies rely heavily upon the environment for subsistence, a habitat classification system that is useful for multiple subsistence activities (hunting, trapping, and gathering of food, medicines and construction material) would be useful and therefore more likely to be maintained in a culture.

TABLE 4.- Mammal species that were frequently detected in Matses-recognized habitat types. The percentages of time spent in habitat types while hunting were calculated based on paced trail lengths. Sightings include animals killed or observed while hunting. Signs include fresh tracks and new dens. Proportions of sightings/signs were calculated as the number of times a species was recorded in a habitat type, divided by the total number of times that species was recorded while hunting. Calculations are separate for geomorphologically-defined (1-8) and vegetatively-defined (9-16) habitat types.

HABITAT TYPE TIMEIN MAMMALSPECIES SIGHTINGS SIGNSIN HABITAT IN HABITAT HABITAT

1 quiusudquid 1-2% Saimiri sciureus 19% (5/26) 2 manan 30-40% Priodontes maximus 0% (0/1) 79%(31/39) 3 dépuen 3-5% Dasypus kappleri 70%(7/10) 65%(30/46) 4 mactac <1% Ateles chamek 43%(9/21) Tapirus terrestris 33%(2/6) 49%(23/47) Tayassu pecari 40%(6/15) 52%(22/42) 5 actiacho 5-10% Allouata seniculus 89%(16/18) Saimiri sciureus 62%(16/26) Isothrix bistriata 81%(18/22) 6 nacnédtsequid <1% Dasypus novemcinctus 50%(2/4) 53%(9/17) 7 mantses <1% Dasypus novemcinctus 25%(1/4) 29%(5/17) Agouti paca 20%(1/5) 23%(7 /30) 8 itiadapa 2-3% Cacajao calous 52%(11/2) Tapirus terrestris 33%(2/6) 26%(12/47) 9 miochoed 5-10% Dasypus kappleri 40%(4/10) 39%(18/46) 10 budédushuchoed 1-2% 11 shubuchoed 3-5% Mazama americana 20%(1/5) 27%(3/11) 12 tanacchoed 2-3% Pecari tajac 29% (2/7) 25%(15/59) Prsakenn rir tridactyla 33%(1/3) 28%(5/18)

13 mayanén sebad 2-3% : 14 isitodochoed 1-2% Choloepus hoffmanni 36%(4/11)

Cabassous unicinctus 0% (0/1) 48%(14/29) 15 tied shéni 5-10% Agouti paca 40%(2/5) 37%(11/30)

Dasyprocta fuliginosa — 60%(9/15) 44%(4/9) 16 sedquequid 2-3% Saguinus fuscicollis 38%(10/26)

Saguinus mystax 26%(8/31)

20 FLECK & HARDER Vol. 20, No. 1

TABLE 5.--Relationship between named rainforest habitat types, showing which vegetatively-characterized habitat types are found on which geomorphologically- defined habitat types. By using two names, the Matses can describe as many as 104 types of primary rainforest and 74 types of secondary rainforest.

VEGETATIVELY-DEFINED > GEOMORPHOLOGICALLY-DEFINED HABITAT TYPES.

HABITAT TYPES. Mananucquio Acte cuéman 2 er > 2: fae 8 8 qm oe ee RT Oe aoe wee aes 1 2G See ee Ret eae a echoes: t 8: @ebet topes bce, 84sted Sf 8 ee € tee f ek. 6 te se © 6 £6 @ oe. 2 4.8 u t ae'n ¢c nC a2 8 cm d hf. a4 fi Babee wv aa q uaa uo s cy u c i c é St. 8 i c q aa d ie u ee’ ee i m é d am aa n

Primary Forest

~~

wy

2

>

> = S

is) eh

~< ><

11 shubuchoe

ec eS PSPS we mS OS OS eR eS KK KK KKK KKK KK KOK x KKK XK ~~ *< x ~

rs ON OS Oe OR On

} 2 = as 2 i 2 ie) = =) is") a <>< >< >< > < >< >< >

cuéte mampis itia

antinchoed sinadchoed shiuishchoe siddattahoed itia dapa x

KK KK ~ ~K KK OX ~*~

Secondary Forest

~*~ ><

a

>

a

S

ie]

=

S

&

LW KK KK OK OK OX KK KK OK OK OX KKK KK KK KK KOK mK KK KK OK mK XK OK OX

~ Kx

x X< KKK KO KK OK OK OOS KK KK OK KKK KK

~K

cuéteuidquio tabad.

Summer 2000 JOURNAL OF ETHNOBIOLOGY Zi

For example, the Matses habitat shubuchoed ‘Phytelephas microcarpa stemless palm forest’ is notable to the Matses for containing palms for thatch (P. microcarpa), being located on good soil that is ideal for making swiddens, and having high densities of trees with edible fruits, which can be harvested seasonally and attract game.

Although the Matses habitat classification system is not entirely hierarchical, each of oe two subsystems) is. Geomorphologically-defined habitat types are clas- sified into two m aj upland rainforest that i by seasonal flooding of a river (Table 1; Pheu 3A), and acte cuéman, floodplain forest along a river that is subjected to seasonal flooding (Table 1; Figure 3B). Veg- etatively-defined habitats are arranged in a very shallow hierarchy and are placed into two general categories: niméduc ‘ primary rainforest’ and an unnamed cat- egory for secondary rainforest habitats (Table 1; Figure 4). Thus, niméduc has both a general and a specific definition (i.e., it is polysemous with referents at two taxo- nomic levels). In the general sense it means all primary rainforest (niméduc,), and in the specific sense it refers to all primary rainforest excluding the other named vegetatively-defined habitats (niméduc,). The Matses do not have a named cat- egory for “secondary forest”, but interview responses clearly show that they place successional forest habitat types into a satogory that is is ; separate from the named category for primary rainforest, niméduc, (i.e., “secondary forest” is a covert cat- egory [Berlin et al. 1968]).

Primary rainforest is characterized by high diversity and infrequent clump- ing of one plant species (Gentry 1992), so the Matses habitat types that are characterized by a dominant species of plant are the exception. The occurrence of vegetatively defined habitat types cannot be predicted by geomorphological fac- tors alone, but their distribution is probably related to some combination of edaphic, historical, and biological factors which favor dominance of some species. For ex- ample, higher densities of Lepidocaryum tenue, (the colonial treelet palm that dominates tanacchoed) were found in yellow ferralitic soil in higher densities than in poorly-drained gleyic soil (Kahn & Mejia 1987). The high densities of Duroia hirsuta and the dearth of understory vegetation in Devil’s gardens (called by Matses mayanén sebad ‘demon’s swidden’; Figure 4B) may be the result of a potentially allelopathic iridoid lactone (plumericin) produced by D. hirsuta (Page et al. 1994). The scarcity of pioneer species (e.g., Cecropia spp.) in sedquequid ‘natural second- ary forest’ (Figure 4B) compared to tied shéni ‘secondary forest from abandoned swiddens’ (Figure 4B) is likely due to advanced regeneration in natural treefall

aps from preexisting small trees in arrested growth stages that are not killed by treefalls (Uhl et al. 1988).

Miochoed ‘Attalea teml lm forest’ and budéd ushuchoed ‘Attalea microcarpa stemless palm forest’ (Figure 4A) have not been described as rainforest habitat types: in he iterature yeas Henderson [1994] noted that Attalea

colonies). Perhaps P. macrocarpa and L. tenue palm forest habitats are more likely to find their way into the literature because they are very important sources of thatch in the Peruvian Amazon, while Attalea spp. are not. Miochoed and budéd ushuchoed (and shubuchoed, ‘Phytelephas macrocarpa stemless palm forest’; aid 4A), however, are important to the aeons because great long-nosed a rmadillos, D. ,an important re found frequently in these habitats. Shubuchoed and miochoed had relatively high small

ye 4 FLECK & HARDER Vol. 20, No. 1

mammal abundance and species richness, especially for marsupials, compared to other trapped habitats. This may be due to large numbers of macroinvertebrates that thrive in the leaf litter collected in the bases of stemless palms (de Vasconcelos 1990) which may provide food for marsupials. Mayanén sebad ‘demon’s swidden’ is not an economically useful habitat for Matses, but these anomalous open zones in otherwise dense tropical forest are too obvious to go unrecognized.

Kahn (1987) found that in eastern Amazonia, differences in palm species com- position and abundance exist among hill plateaus, hill crests, hill slopes (inclines) and depressions between hills; these differences were attributed to differences in declivity (angle of slope) among the sites, which affected the drainage and canopy structure, thereby creating different abiotic and biotic conditions for palms. Simi- larly, relatively small variation in elevation (39 m) can affect rainforest tree species composition (Lieberman et al. 1985); hills in the Nuevo San Juan area can rise up to 60 m above adjacent gullies. These studies lend credibility to the Matses percep- tion that hill crests and hill inclines differ vegetatively.

Knowledge of rainforest habitats is important not only for describing floristic diversity, but also for understanding the ecology of animals in those areas. Capy- bara (Hydrochaeris hydrochaeris) in Amazonian Peru used beaches, Cecropia forests, and low levees more often than swamps, low flooded forests, and high levees (Soini & Soini 1992). Woolly monkeys (Lagothrix lagotricha) used colinas (inland hilly forest) and igap6 (seasonally flooded blackwater forest) more often than expected in Ama- zonian Colombia (Defler 1996). Squirrel monkeys (Saimiri sciureus) in Surinam showed a preference for liane forests and were found most often in that formation (Mittermeier & van Roosmalen 1981). Results of habitat-mammal associations re- corded while hunting with Matses reflect too much sampling bias to reliably determine habitat preferences by game mammals, but the high proportion of time that game species were found in certain habit to Matses hunters of recognizing many habitat types in order to hunt more 2 efficiently.

Very poor drainage and perhaps toxic levels of some minerals in the soil seem to inhibit growth of trees so that a conspicuously low basal area and tree density exist in mactac ‘muddy mineral lick’ Figure 3A). Mactac amen are Amportant to the Matses for hunting and they i ) of the high likelihood of finding game there. Tapirs (Tapirus terrestris), white-lipped peccaries (Tayassu pecari), and spider monkeys (Ateles chamek) were found very often in this habitat, as well as howler monkeys (Alouatta seniculus), collared peccaries (Pecari tajacu), and brocket deer (Mazama americana, M. gouazoupira).

The vegetation in dépuen ‘stream headwaters’ is neither conspicuously differ- ent from that in the surrounding habitats nor does this habitat type contain a high concentration of ly important plant species, but it is important for hunt- ing armadillos (Dasypus kappleri). D. kappleri make burrows in the eroded sides of dépuen gullies and the Matses have become quite skilled at detecting occupied burrows and flooding out the armadillos. The preferred location for searching for D. kappleri is in dépuen that overlaps miochoed, budéd ushuchoed, or shubuchoed. According to the Matses, armadillo paths are very common in these vegetatively- defined habitats because they contain good soil with large numbers of soil invertebrates. In fact, the preferred location for Matses to make swiddens is on shubuchoed, miochoed, or budéd ushuchoed, but not where these overlap dépuen,

make their pati IS

Summer 2000 JOURNAL OF ETHNOBIOLOGY 23

but rather where they occur on manan ‘hill crests’ and macuésh ‘hill incline’ or on quiusudquid ‘terra firme next to a river’. Thus, it can i seen that Matses subsis- tence activities and knowledge of natural history kn itive to habitat types that are not lexicalized, but that they can, nevertheless, refer to with preci- sion using a combination of names from the two habitat classification subsystems.

The Matses utilize their knowledge of habitat types to understand seasonal movements of animals. Folk natural history information from the Matses describes the movements of frugivores across several habitat types in response to habitat- specific seasonal availability of fruit and secondary foods (Harder & Fleck 1997). Many animals move between rainforest habitats during the course of the year (e.g., ungulates: Bodmer 1990), and utilize seasonally available resources in differ- ent habitat types within the upland and floodplain rainforests (e.g., primates: Peres 1994; Stevenson et al. 1994; Defler 1996).

Species richness of trapable small mammals was lower in actiacho ‘season- ally flooded forest’ than in any of the upland habitats, a trend similar to that found in upland rainforest and blackwater seasonally flooded rainforest habitats near the Ucayali River in Loreto, Peru (Fleck & Harder 1995). An important difference between two types of successional forest recognized by the Matses is that tied shéni had the highest abundance of small mammals of the 10 sampled rainforest habitats, while auxiliary trapping in sedquequid (350 trapnights) produced zero captures. Second to active swiddens, tied shéni is the Matses’ favorite habitat type for trapping Proechimys rats.

Use of local habitat classifications of indigenous people is not a substitute for extensive regional surveys as in Terborgh and Andresen (1998) or for broader de- scriptions based on gradients in soil types and hydrology. Nevertheless, there are several applications of indigenous cl systems for diversity inventories and management planning. For example, a researcher could consult locals about the habitat types they recognize and ask to be led to the different habitats, thus efficiently finding some habitat types that might contain fauna or flora that is rare elsewhere, and would otherwise be detected only by chance. One innovative ap- plication of folk classification systems is Shepard et al.’s (in press) utilization of the rainforest habitat classification system of the Matsigenka Indians of Amazo- nian Peru to interpret LANDSAT images. Another use of indigenous habitat classification and resource knowledge is in designing, implementing and manag- ing communal reserves, national parks and other natural protected areas with indigenous populations. A case in point is the use of Matsigenka ecological knowl- edge described by Shepard (in press) to form a baseline for implementing a recently-approved Conservation International project in the Vilcabamba Cordil- lera of Peru that engages the local indigenous groups as primary stewards of two communal reserves and as stakeholders in a proposed national park.

In order to develop effective conservation policy in Amazonian countries, it is essential to have an understanding of habitat heterogeneity in Amazonia, but un- fortunately at present there is not a habitat classification system for Amazonia available to scientists and policy makers that considers all minor habitat types such as those described in this paper. One way to develop a comprehensive habi- tat classification system for Amazonia would be to compile descriptions of habitat types recognized by locals and biologists throughout the Amazon basin, deter-

24 FLECK & HARDER Vol. 20, No. 1

mining which described habitats are similar enough to be considered a single habi- tat type, and determining whether habitats are geomorphologically or vegetatively-defined. The fact that Shepard et al. (in press) found a comparable classification system for the Matsigenka (more than 40 named habitat types in

land vegetation classifications) suggests that com- piling a classification of Amazonian habitats in this way would be practicable. The problem with this approach is time. Indigenous knowledge of habitat classifi- cation is passed in an oral natural history that depends upon active hunting in traditional ways. These ways are being threatened by the onslaught of western culture. As young men move to cities or adopt western methods of hunting with shot guns and flashlights, fewer will learn or become skilled in traditional ways that depend heavily upon the indigenous habitat classification. Thus, it is impor- tant to study native habitat classifications before they are lost to cultural change along with their potential value to ecology and conservation.

ACKNOWLEDGMENTS

We would like to thank Dr. Robert S. Voss of the American Museum of Natural History, and Pekka Soini and Kember Mas of the Instituto de Investigaciones de la Amazonia Peruana Wh hens ee aian and Dr. Andrew Henderson helped with identification of ‘Pints specimens. Maa Sarid oo Sergio Solari, and Elena Vivar helped with prelimina , and Dr. Robert M. Leighty and Shanggang Zhou provided statistical analyses. We a are a6 tect to Dr. Ralph E. J. Boerner who reviewed an earlier draft of this manuscript, and to the two anonymous reviewers who offered many insightful comments. This material is based upon work supported under a National Science Foundation Minority Graduate Fellowship, an Ohio State Dean’s Fellowship, and an Ohio State Osbourn Graduate Fellowship awarded to DWF. Travel expenses to Peru were provided by a Latin American Studies Program Tinker Foundation Foreign Field Research Grant. Most importantly, we are indebted to the Matses of Nueve en Juan, Remoyacu, and Buen Pera for sharing their knowledge and

their insight. \ ut their hospitality andr t tudy could I realized

NOTES

' The first author’s current address, to where correspondence should be sent, is: David W. Fleck

Department of Linguistics - MS 23

Rice University

P.O. Box 1892

Houston, Texas 77251-1892

* The orthography used here is the practical orthography developed by SIL personnel for Bible translation and pedagogical materials, which has become the official writing system for the Matses (Kneeland 1979). The alphabet is phonemically-based and modeled after Spanish orthography. To produce a pronunciation that approximates Matses, words writ- ten in this orthography may be pronounced as if reading Spanish, with the following ex- ceptions: @ is a high central unrounded vowel [ ]; c (spelled qu preceding e, é and i) is pronounced as a Loe stop word-finally and preceding consonants, and as [k] elsewhere; disr | flap [ ] between vowels, and as a [d] elsewhere; and ts should be read as an unvoiced veil affricate. Word-level stress is on even-numbered syllables (count- ing from left to right) unless otherwise marked.

Summer 2000

JOURNAL OF ETHNOBIOLOGY 25

LITERATURE CITED

ALCORN, JANICE B. 1984. Huastec Mayan peerage University of Texas Press,

Aus BALEE, ‘WILLIAM. 1989. The culture of

and Folk Strategies, Darrell A. Posey and William Balée aoe Advances in Economic Botany Vol. 7 New York Botanical Garden, Bronx, New York.

. 1994. Footprints of the Fore Ka’apor Ethnobotany - the Historical Ecology of Plant Utilization by a Amazonian People. Co tutta

University ew Y BALICK, MICHAEL J. 1984. Ethnobotany of ‘pa alms in the Neotropics. Pp. 9-23

aa

T. Prance and Jacquelyn. A, Kallunki Seah Advances in Economic Botany 1, The New York Botanical Garden,

DENNIS _ E.

1968. Covert categories and folk taxonomy. American Anthropologist 70:290-299. he tate ; . 1973. General principles of classification and biology. American Anthropologist 78: me 242 BODMER, RICHARDE. 1990. Responses « of

Amazon floodplain. Journal of Tropical Ecology 6:191-

BRAGA, PEDRO [VO SOARES 1979. Subdivisao fitogeografica, tipos de vegetacdo, conservacao e inventario floristico da floresta amaz6nica. Acta Amazonica 9:53-80.

BRIEMAN, LEO, JEROME H. FRIEDMAN, RICHARD A. OLSHEN and CHARLES J. STONE. ane Classification and Pa aia Trees. Wadsworth Advance

and Software, Bellmont, California,

CAMPBELL, DAVID G., JUDY L. STONE and ARITO ROSAS, JR. 1992. A comparison of the phytosociology and dynamics of three floodplain (vdrzea) forests of known ages, Rio Jurua, western Brazilian Amazon. Botanical Journal of the Linnean Society 108:213-237.

CARNEIRO, sec! L. ae The cultivation of manioc among the Kuikuru of the Upoer nee Pp. 65-111 in Adaptive Responses of Native Amazonians, Raymond B. Hames and William T. bigs (editors). Academic Press, New York.

CONKLIN, HAROLD Cr" 1962. Lexicographical treatment of folk taxonomies. International Journal of

Echimyidae) from the western Am of Peru. Proceedings of the Cpe gag Society of Washington 111:4

DEFLER, pect 1996. reed of the ranging patte oup of wil woolly oe (Lagotri lagothricha). American Journal of Primatology 38:289-302.

DE VASCONCELOS, HERALDO LUIS. 1990. Effects of litter collection by understory palms on the associated m

DUIVENVOORDEN, J. F. 1996. Patterns of f the

middle Caqueta area, Colombia, NW Amazonia. Biotropica 28:142-158.

EMMONS, LOUISE H. 1993. On the identity of Echimys didelphoides Desmarest, 1917 (Mammalia: Rodentia: Echimyidae). Proceedings of phe Biological Society of Washington 106:1-4.

and FRANCOIS FEER. 1997, Neotropical Rainforest Mammals: A Field Guide (2nd Edition). University of Chicago Press, Chicago.

ENCARNACION, FILOMENO. 1985. Introduccion a la flora y vegetacién de

ia d

formaciones vegetales en la llanura

amazonica. Candollea 40:237-252.

. El bosque y las formaciones vegetales en la llanura amazo6nica del Per. Alma Mater 6:95-114.

ERIKSON, PHILIPPE. 1994. Los Mayoru Pp. 3-127 in Guia Etnografica de la Alta Amazonia, Volumen II, Fernando Santos and Frederica agit (editors). Flacso, Quito, Ecuador.

26 FLECK & HARDER

FAURA GAIG, GUILLERMO F. 1964. Los Rios de la Amazonia Peruana. Colegio Militar Leoncio Prado, Callao, Peru.

FLECK, DAVID W. 1997. Mammalian Diversity in Rainforest Habitats as Recognized by Matses Indians in the Peruvian Amazon. M.S. — The Ohio State University, ee

D. HARDER. 1995. Ecology of _marsupials in two Amazonian tl t

Peru. Journal of Mammalogy 76:809- 818.

, ROBERT S. VOSS and JAMES L PATTON. 1999. Biological basis of saki Pasa folk ag recognized by the es Indians of Amazonia Per eases? areal of Primatology 20:1005-1027. GENTRY, ALWYN H. 1992. Tropical forest

their conservation significance. Oikos 63:19-28.

HARDER, JOHN D. and DAVID W. FLECK. 1997. Reproductive Saga 2 of New pials. P

rm and Lyn A. Hinds (editors). University of New South Wales Press, Sydney. HENDERSON, ANDREW. 1994. The Palms of the Amazon. Oxford University Press, New York. tvtime tet ARIA . GALEANO - and RODRIGO BERNAL. 1995. Field Guide to the Palms of the Americas. Princeton University Press, Princeton. HERSHKOVITZ, PHILIP. 1992. The South

(Marmosidae, — A taxonomic review with notes general morphology and relationships. Fieldiana Zoology, New

ries KAHN, F. 1987. The distribution of palms as a function of local topography in Amazonian _ terra-firme forests. oo 43:251-259. KEMBER MEJIA. 1987. Notes

tessmannii. Principes 31:1 erect and ALINE DE CASTRO. 1988. ‘Species richness and density of ms in terra firme forests of Amazonia. Biotropica 20:266-269.

Vol. 20, No. 1

KALLIOLA, RISTO, MAARIT PUHAKKA and WALTER DANJOY (editors). 1993

Jyvaskyla, Finland.

Sieistty , JUKKA SALO, HANNA TOUMISTO and KALLE RUOKOLAINEN. 1991. The dynamics, dictrihiition an

eraramn

vegetation in Peruvian Amazonia. Annales Botanici Fennici 28:225-239. KNEELAND, HARRIET. 1979. Lecciones

Verano, Documento de Trabajo Numero

LAMOTTE, SANDRINE. 1990. Fluvial d s and succession in the lower

RMAN, G. S. HART RN RODOLFO PERALTA. 1985. Small-scale altitudinal variation in lowland wet tropical forest vegetation. Journal of eee 73:505-516.

MALLEUX, JORGE. 1982. Inventarios Sees en Bosques_ seepage Universidad Nacional Agrari , Peru.

MARENGO ORSINI, JOSE. 1983. Estudio Agroclimatico en la Zona de Jenaro Herrera (Requena, Loreto) y Climatico en la Selva Baja Norte del Pert esis, Universidad Wacleedl Agraria La Molina, Lima, Peru.

MITTERMEIER, RUSSELL A. and MARC G. M ROOSMALEN. 1981. Preliminary observations on habitat utilization of diet in eight Surinam monkeys. Folia Primatologica 36:1-39.

PAGE, J. E., S. MADRINAN and G. H. N. TOWERS. 1994. Identification of a plant growth inhibiting iridoid lactone from

f the

‘Devil’s Garden’. Experientia 50:840-

PARKER, EUGENE, DARREL POSEY, JOHN FRECHIONE and LUIZ FRANCELINO DA SILVA. Resource exploitation in Amazonia:

Museum of Natural History 52:163-203.

Summer 2000

PATTON, JAMES L. and MARIA NAZARETH F. DA SILVA. 1997, oer oP ms gsinen of es four-

eyed ums (Didelphidae: Philander). ‘Souenal of xocheutiog: 78:90-102.

secsnlgeted and J. R. MALCOLM. 2000. Mammals ¢ of the Rio aioe and the evolutionar ological diversification of Bisel Bulletin of the Amer Ps an Museum of Natural

History 244:1-306.

PERES, CARLOS, A. 1993. Structure and spatial organization of an er mgr terra firme forest primate unity. Journal of Tropical Ecology 9: 259- 276.

1994, Pri e responses to phenological changes in an Amazonian

u Goeldi no Ano de Sesquicentenario; Pubicagdes Avulsas, Museu Paraense Emilio eng Belem 20:179-202. LEAN T. PRANCE. 1985. The vegetation types of the Brazilian Amazon. Pp. 109-145 in 0 Environments: Amazonia, Ghillea Prance and Thomas E. Lovejoy pave Pergamon Press, Oxford, England. POSEY, DARRELL A. 1983. Indigenous ecological

of the Amazon. Pp. 225-257 in The Dilemma of Amazonian Development, Emilio F. Moran (editor). Westview Press, iy Colorado

ee WILLIAM BALEE (editors). 1989. gore urce Management i

Strategies. Advances in Economic Botany Vol. 7, The New York Botanical Garden, Bronx, New York.

PRANCE, GHILLEAN TOLMIE. 1978. The origin and evolution of the Amazon flora. Interciencia 3:207-222.

HILLEAN. T. 1979. Notes on

ROMANOFF, STEVEN A. 1984. bates Adaptations in the Peruvian Amazon. Ph dissertation, Columbia University, New York.

JOURNAL OF ETHNOBIOLOGY 27

SALO, JUKKA, RISTO KALLIOLA, LMARI_ HAKKINEN, — YRJO MAKINEN, PEKKA NIEMELA, MAARIT PUHAKKA and PHYLLIS D. peseas 1986. River dynamics and the diversity of Amazon lowland forest. Nature 322:254-258.

SHEPARD, GLENN JR. in press. Resource

Vilcabamba. In A Rapid Biological Assessment of the Northern Cordillera Vilcabamba, Peru, Foret S Schulenberg (editor). RAP Workin

Papers 11. Conservation eal Washington, D

OUGLAS W. YU, BRUCE

MATEO Ethnobotanical ground-truthing and forest diversity in the Western Amazon. In Ethnobotany and Sara dean - Biological Diversity, Luisa

Carlson, and Eglée poser en feciitccs). Advances in Economic Botany, The New York Botanical Garden, Bronx, New York.

SOINL sap -. a SOINI. 1992. Ecolo del ronsoco o capibara ( Pa adoenis Coa en la reserva nacional Pacaya-Samiria, Per. Folia Amazonica 4:119-133.

STEVENSON, one R., MARCELA 1: QUINONES nd ORGE AHUMADA. 1994. Ecological peti:

(Lagothrix lagothricha) at bead a National Park, Colombia Primatology 32:123-140,

TERBORGH, JOHN and_ ELLEN ANDRESEN. 1998. The composition of A -§ £. 4 ‘4 1 1 and

regional scales. Journal of Tropical Ecology 14:64 :

UHL, CHRISTOPHER, KATHLEEN CLARK, NELDA DEZZEO and PEDRO MAQUIRINO. 1 Vegetation dynamics in Amazonian treefall gaps. Ecology 69:751-763.

VIVAR, JUDITH E. 1975. Los mayoruna: en la frontera Peru-Brasil. América Indigena, 35(2):329-347.

VOSS, ROBERT S. 1988. Systematics and ecology of ichthyomyine rodents (durciden): peters of morphological

A oe oe a American Museum of Natural History 188:259-493.

28

FLECK & HARDER Vol. 20, No. 1

ARRIN P. LUNDE AND WOODROFFE, JOSEPH F. 1914. The Upper

“NANCY. B. SIMMONS. in press. The Reaches of the Amazon. The MacMillan

mammals of Paracou, French Guiana: A Company, ore York.

Neotropical lowland rainforest fauna. ZONA, SCOTT and ANDREW

Part 2. Nonvolant species. Bulletin of the HENDERSON. 1989. A review of

American Museum of Natural History. animal-mediated seed dispersal of WILSON, DON E. & DEEANN M. palms. Selbyana 11:6-21.

Reference (2nd Edition). Smithsonian Institution Press, Washington D.C

APPENDIX A. Key for identification of Matses-named geomorphologically-defined rainforest habitat types found within 2 km of Nuevo San Juan

—

=

More than 3 m above normal peak river level, near or far from a river, never flooded by seasonal rise of a river Go to 2 Less than 3 m above normal peak river level, always near a river, subject to seasonal flooding of a river

Elevation rising 10-60 m above surrounding land Go to 3 In a valley between hills Go to 5 Adjacent to river, (up to 100 m from river)quiusudquid, terra firme adjacent to a river At least 100 m from river Go to 4 Top of hill with incline <15° manan, hill crest

Side of hill with incline >15°, 10 m above lowest point of valley

macuésh, hill incline

Along a stream Go to 6 Not along a stream Go to 8

Among several Sup gullies, gullies contain running water only during and im- mediately after rain épuen, stream headwaters Along a stream >1 m cK stream contains water all year

Along a stream >3 m wide, stream floods during heavy rains acte dada cuéman, gal- lery forest along large stream

Along a stream >1 m and <3 m wide, stream swells during rains, but does not over- flow banks during heavy rains .. acte cuidi cuéman, gallery forest along small stream Very poor drainage, ground always waterlogged or muddy........ mactac, mineral lick

Area between gallery forest and hill inclines, fair drainage, ground damp, but never waterlogged tsimpiruc, valley

“Island” elevated 7-13 m above surrounding land, does not flood during most of rainy season o to 10 Relatively flat land, floods during most of rainy season Go to 11

0-3 m below normal peak river level, flooded yearly, but only for a few weeks during highest water 1, low levee island 0-3 m above normal peak river level, only floods on years of exceptionably high river levels mantses, high levee island

Seman to a river or higher than land serene * it spies a river, drains well during dry seaso acho, seasonally flooded forest Never wad to a river, lower than land ting it f th , ground remains waterlogged during rainy season, dominated by Mauritia Pexuos palms sean nhiabiaineecnies

apa, palm swamp

Summer 2000 JOURNAL OF ETHNOBIOLOGY 29

APPENDIX B.- Key for identification of Matses-named vegetatively-defined rainforest habitat types found within 2 km of Nuevo San Juan

1

co

a

Primary rainforest vegetation, trees at least 40 m high, some trees with thick (DBH > 0.6 m) trunks o to2 Secondary rainforest vegetation, trees not reaching 40 m, no trees with thick trunks

Forest understory dominated by palms Go to 3 Forest understory not dominated by palms Go to 6 Forest understory dominated by treelet palms (Lepidocaryum tenue) ......... tanacchoed, Lepidocaryum tenue treelet palm forest

Forest understory dominated by stemless palms (Attalea or Phytelephas) ......... Go to 4 Forest understory dominated by Phytelephas sp. palms........... shubuchoed, Phytelephas macrocarpa dwarf palm forest

Forest understory dominated by Attalea sp. palms Go to 5

Forest understory dominated by Attalea racemosa palms .... miochoed, Attalea racemosa dwarf palm forest

Forest understory dominated by Attalea microcarpa palms ... budéd ushuchoed, Attalea microcarpa dwarf palm forest

Forest understory and midstory with low vegetation density and dominated by Duroia hirsuta (small dicot trees) ............... mayanén sebad, Duroia hirsuta ‘demon’s swidden’ Forest not dominated by Duroia hirsuta Go to 7

Forest dominated by numerous large and small lianas, high vegetation density isitodochoed, liana forest

Forest not dominated by any salient form of vegetation ... niméduc,, undifferentiated primary forest

Forest dominated by Cecropia spp. Marila spp. and other pioneer tree species, rela- tively few lianas, few primary forest species; from an abandoned Matses swidden tied shéni, abandoned swidden

Forest containing a wide variety of primary forest species that have sprouted vegeta- tively from stumps and roots of fallen trees, mixed with pioneer tree species, contains many small lianas and creeping vines; not from an abandoned swidden. sedquequid, natural secondary forest

Note: The keys in Appendices A and B can be used to describe any locality within 2 km of Nuevo San Juan, Peru using two habitat names, one geomorphologically-defined habitat name and one vegetatively-defined habitat name.

30

APPENDIX C.- List of 84 non-flying mammal species captured, observed (*), or reported

FLECK & HARDER

by Matses (**) in the Nuevo San Juan area in 1995-1996

Vol. 20, No. 1

LATIN NAME? ENGLISH NAME> MATSES NAME‘ DIDELPHIMORPHIA Caluromys lanatus western woolly opossum abuc checa Didelphis marsupialis common opossum piocos Gracilinanus kalinowskii! | Kalinowski’s gracile mouse opossum _— checampi Marmosa murina murine mouse opossum checampi Marmosops noctivagus White-bellied slender mouse opossum checampi j | Andean slender mouse opossu hecam Metachirus nudicaudatus _ brown 4-eyed opo checa déuisac Micoureus demerarae Long-furred woolly mouse opossum checampi Micoureus regina Short-furred woolly mouse opossum — checampi Monodelphis adusta Sepia short-tailed opossum ama Monodelphis emiliae Emilia’s short-tailed opossum ma

Philander mcilhennyi* XENARTHRA

Anderson’s gray four-eyed opossum

ya checa déuisac

Bradypus variegatus Brown-throated three-toed sloth méincanchush Choloepus cf. hoffmanni Southern two-toed slo shuinte Cabassous unicinctus* Southern naked-tailed armadillo mencudu Dasypus kapple Great long-nosed armadillo tsaues lasypus novemcinctus nine-banded armadillo sedudi Priodontes maxi giant armadil tae cea py aaa didactylus pygmy ant eater tsipud Myrmecophaga tridactyla t ant eater shaé Tamandua tetradactyla collared tamandua béui PRIMATES Callithrix pygmaea pygmy _— madun sipi Ps aguinus fuscicollis saddleback ta sipi cabédi Saguinus mystax Binck-chestedx hee ae tamarin sipi ésed Callimico goeldii* Goeldi’s e sipi chéshé

Alouatta seniculus

Saimiri sciureus CARNIVORA

y black spider monkey red uakari monkey

titi monke

white-fronted capuchin monkey brown capuchin monkey

common squirrel monkey

béchun ushu béchun chéshé

poshto béshuicquid anca

Atelocynus microtis**

Galictis vittata** Lontra longicaudis

short-eared dog bush dog

n southern river otter

mayanén opa achu camun bédi chéshé bédimpi

téstuc mauequid bédi

bédi piu batachoed bosen ushu bosen

Summer 2000

JOURNAL OF ETHNOBIOLOGY

Mustela africana** Amazon weasel opampi Pteronura brasiliensis gent river otter onina Bassaricyon ii ingo shéméin asua nasua South American coati tsise Potos flavus kinkajou cuichic Procyon cancrivorus** crab-eating raccoon tsise biecquid CETACEA Sotalia oma gray dolphin chishcan chéshé Inia geoffrens pink river dolphin chishcan piu PERISSODACTYLA Tapirus terrestris Brazilian tapir néishamé ARTIODACTYLA Pecari tajacu collared peccary shécten Tayassu pecari white-lipped peccary shéctenamé Mazama americana red brocket deer enad piu Mazama gouazoupira gray brocket deer senad tanun RODENTIA icrosciurus — Amazon dwarf squirrel capa cudu Sciurillus pus Neotropical pygmy squirrel cacsi Sciurus Fi olivian squirrel capampi Sciurus igniventris northern Amazon red squirrel capa Sciurus spadiceus southern Amazon red squirrel capa Nectomys apicalis? water rat ca tanun Oecomys b arboreal rice rat shubu pecquid Oecomys cf. trinitatis arboreal rice rat abuc macampi Oryzomys cf. macconnelli rice ra tacbid umu Oryzomys pe 3 ommon rice rat tacbid umu Oryzomys yunganus rice rat tacbid umu Scolomys ucayalensis gray spiny mouse tacbid umu Coendou prehensilis Brazilian porcupine isa Dinomys branic acarana tambis biecquid Hydrochaeris hydrochaeris capybara memupaid Dasyprocta fuligi#osa ack agouti mécueste Myoprocta pr ratti* pie acouchi tsatsin Agouti paca pac ambi Isothrix bistriata yellow-crowned aioe tailed tree rat abuc maca Makalata didelphoides? red-nosed tre uc maca Mesomys ferrugineu us* spiny tree oy abuc maca Proechimys brevicauda spiny rat tambisémpi oechimys cuvieri spiny rat tambisémpi Proechimys kulinae® spiny rat tambisémpi Proechimys simonst spiny rat a bisémpi Proechimys steerei spiny rat tambisémpi 4 Nomenclature follows Wilson & > Most common names from. Pome a ae (1997). x

© Only lexicalized terms a

the Matses (see Fleck et al. "1999 for ; paints overdifferentiation by the Matses), the Matses name given represents the non-terminal lexeme that as — bapsewie de to the scientific species. Also, —

Nuevo San Juan are presented here

1 Hershkovitz (1992) 2 Patton & da Silva (1997) 3 Patton et al. (2000) 4 Voss et al. (in press) 5 Emmons (1993) 6 da Silva (1998)

Ps

32 FLECK & HARDER Vol. 20, No. 1

APPENDIX D.- Linguistic analysis of habitat terminology.

Here all the habitat types listed in Table 1 are analyzed linguistically. Habitats are listed and discussed in three sections based on their analyzability. The first and second categories include terms that are not synchronically segmentable, and correspond to Conklin’s (1962:123) “unitary simple lexemes” and Berlin et al.’s (1973:217) Auitanialyzable) primary lexemes.” The first category includes lexemes thath g in Matses, and the second category includes polysemous lexemes. The third category includes names with more than one morpheme, cor- responding to Conklin’ s (1962: £23) “unitary complex lexemes” and Berlin et al.’s (1973:217) “ung ble) primary lexemes” None of the habitat names include morphemes that refer toa superordinate category, so there are no terms corresponding to Conklin’s (1962:123) “composite lexemes” or Berlin et al.’s (1973:123) “secondary lexemes.”

One notable trend in Matses habitat nomenclature is that all of the synchronically unanalyzable terms (category 1) are for geomorphologically-de- fined habitats, suggesting that these habitat names are older than those designating vegetatively-defined habitats. If indeed the geomorphological habitat classifica- tion subsystem is older, it is notable that plot it was this sub-system that was more easily classified by the CART analyses.

1) Non-polyse rphemic terms:

tsimpiduc ‘valley between hills’

anshantuc ‘permanently waterlogged swamp’

niméduc ‘primary forest/undifferentiated primary forest’ mananuc ‘upland forest’ (usually used with the emphatic -quio) manan ‘hill crest’

macuésh ‘hill incline’

mantses ‘high levee’

mashcad ‘levee island (flooding season term for mantses)’ actiacho ‘low seasonally flooded forest’

dépuen ‘stream headwaters’

Possible historical analyses.- The form uc appears to be a historical locative postposition. The only nouns in Matses that can appear in a locative phrase with- out a locative postposition are those ending in uc (these happen to all be habitat terms); so the term mananuc ‘upland forest’ is almost certainly historically de- rived from manan ‘hill crest’ and possibly once could be analyzed as ‘in the hills’. The term actiacho ‘low seasonally flooded forest’ obviously contains the word acte ‘water/river/stream’, but the form acho is not found elsewhere in the lan- guage (like cran- in English cranberry), so it is debatable whether this word is synchronically segmentable.

Matses has a productive but apparently very old process of noun incorpo- ration using abbreviated forms of body part terms prefixed to noun, verb, and adjective roots. The prefix provides a locative orientation in reference to an actual or metaphorical body part. The words listed above are no longer synchronically

Summer 2000 JOURNAL OF ETHNOBIOLOGY 33

segmentable, but the form ma in elevated topographical terms may be related to the prefix ma- ‘head.’ Similarly, the form tsi in tsipiruc ‘valley’may be the prefix tsi- ‘hips.’ And finally, the form dé in dépuen ‘stream headwaters’ may be the prefix dé- ‘nose’(cf. “upstream” is débiate-mi ‘nose-Directional.’

2) Polysemous hemic names mactac ‘mineral lick’ also means ‘mud’ itia ‘upland palm swamp’ also refers to the palm species, Mauritia flexuosa’

The reason for separating these terms from those in one is that it is not clear which of the meanings for these terms is the older one, making it questionable whether these are old lexemes or recent coining of new habitat names through metonymy. Note that niméduc is polysemous in the sense that it refers to catego- ries at two levels of habitat classification (‘primary forest’ and ‘undifferentiated primary forest’), but this polysemy does not bring into question whether this is a recently-introduced term for designating a habitat type.

3) Syncl ically analyzable names:

Geomorphologically-defined habitat terms are mostly nominalizations and loca- tive phrases, while vegetatively-defiened habitat terms, especially for primary — mostly. involve the noun Pp paraee enclitic —choed ‘characterized by,’ which is av that can be used to describe any animate or inatimate atiey besides rainforest habitats (e.g., the name for the tayra is batachoed ‘sweet- characterized.by’ because it eats fruits and steal papayas; or a man with a large belly may be teased by calling him chichanchoed ‘stomach.parasite- characterized.by’). However, all the terms listed below represent lexicalized terms (they are used consistently, they have restricted meanings, and they are treated differently grammatically from ad hoc descriptive phrases).

quiusud-quid ‘non-flooding forest next to a river’ rise.above-Agt.Nzr! (lit. ‘one that rises above’)

nacnéd-tsé-quid ‘low levee that is flooded every year’ stick.out-Dim-Agt.Nzr (lit. ‘one that sticks out a bit’)

sedque-quid ‘secondary forest from blowdown or river shine-Agt.Nzr shift with many vines and few Cecropia spp.

trees’ (lit. ‘one that shines/is bright [due to sun shining through the open canopy]’)

cuéte-uid-quio tabadquid ‘secondary forest where hardwood trees tree-

only-Emph stand-Agt.Nzr areout competing pioneer vegetation and vines’ (lit. ‘one where only dicot trees stand’)

cuesbud-aid ‘recent blowdown characterized by creeping fall.over-Pat.Nzr vines and no trees’ (lit. ‘fallen over’)

acte cuéma-n river edge-Loc

acte dada cuéma-n stream body edge-Loc

acte cuidi cuéma-n stream branch edge-Loc

chian cuéma-n lake edge-Loc

acte mauan river flooded.place

tied shéni

swidden old

mayu-n tied nonMatses-Gen swidden

mayan-n_ sebad demon-Gen swidden

cuéte mampis dicot.tree ?small

isan-choed Oenocarpus.mapora-char

niste-choed Tri artea daltnides -char

shuinte mapi-choed sloth head-char

budéd-choed Attalea.butyracea-char

mio-choed A tta ] eA racemosc a-char

budéd —_ushu-choed Attalea.butyracea white-char the name for A.microcarpa)

FLECK & HARDER

‘floodplain forest’ (lit. ‘beside a river’)

‘gallery forest along a large stream’ (lit. ‘beside the body of a stream’)

‘gallery forest along a small stream’ (lit. ‘beside the branch of a stream’)

‘forest along a floodplain lake’ (lit. ‘beside a lake’)

‘flooding season term for actiacho’ (lit. ‘flooded place by a river’)

‘secondary forest in abandoned Matses swiddens dominated by Cecropia spp. and Marila spp. trees’ (lit. ‘old swidden’) ‘secondary forest from abandoned swiddens or villages >50 yr old’ (lit. “non-Matses Indians’ swidden’)

‘forest with open understory, dominated by D. hirsuta trees’ (lit. ‘demon’s swidden’)

‘forest where only thin hardwood trees grow’ (lit. ‘small dicot trees’)

‘forest dominated by O. bataua palms’ (lit. one characterized by O. bataua palms’)

‘forest dominated by I. deltoidea palms’ ‘forest dominated by A. tessmanii palms’ (“sloth head” is the name for A. tessmanii) ‘forest dominated by A. butyracea palms’ ‘forest with understory dominated by A.racemosa palms’

‘forest with understory dominated by A.microcarpa palms’ (“white A.butyracea” is

Vol. 20, No. 1

Summer 2000

shubu-choed

Phytelephas macrocarpa-char

tanac-choed

Lepidocaryum.tenue-char

dapais-choed

Attalea ph alerata-char

cobisan-choed

Fi 1 terpe preca toria -char

tiante-choed bamboo-char

sénte-choed Cedrela-char

péncad-choed tree.species-char

mani pada-choed plantain flat-char isitodo-choed liana-char

antin-choed Attalea.maripa-char

sinad-choed Bactris-char

shiuish-choed Ficus-char

cana shéta-choed macaw beak-char isitodo icsa-choed

vine thicket-char

bucu-choed Cecropia-char

JOURNAL OF ETHNOBIOLOGY

‘forest with understory dominated by P. macrocarpa palms’

‘forest with understory dominated by L. tenue treelet palms’

‘forest with understory and midstory dominated by A. phalerata palms’

‘swamp dominated by E. precatoria palms’ ‘forest dominated by bamboo’

‘forest dominated by Cedrela sp. trees’ ‘forest dominated by pencad trees’

‘forest dominated by Musa wild banana plants’ (lit. ‘characterized by having flat [leaved] plantains

‘forest dense with many large lianas’ ‘seasonally flooded forest dominated by A. maripa palms’

‘seasonally flooded forest with understory dominated by Bactris cf. bifida palms’

‘seasonally flooded swamp forest dominanted by Ficus sp. trees’

‘low floodplain adjacent to the river with dense thorny vegetation’ (“macaw beak” is the name for a species of waterside shrub)

‘secondary forest thick with vines and young trees’ (lit. ‘characterized by vine thickets’)

‘secondary forest dominated by Cecropia sp. trees’

36 FLECK & HARDER Vol. 20, No. 1

acte mactac ‘mineral lick in floodplain forest’ river mineral.lick (lit. ‘mineral lick by a river’)

itia dapa ‘floodplain palm swamp’ palm.swamp large (lit. ‘big palm swamp’)

itia mauan ‘flooding season term for itia dapa’ palm.swamp flooded.place (lit. ‘flooded place in a palm swamp’)

The last three terms listed in seen section contain words that are Eee to other

habitat t ; itia are not ut rather

occur at the same taxonomic level (and therefore are not * ‘composite lexemes” / “secondary lexemes.”

1 Morpheme gloss abbreviations: Agt.Nzr ‘Agent — Gen

char — iby ne

Bon Pat.Nzr ‘Patient oe Emph Emphatic’

Summer 2000 JOURNAL OF ETHNOBIOLOGY oF

The Ecological Indian: Myth and Reality. Shepard Krech III. W. W. Norton, New York. 1999. Illus., bibliography, index. $27.95. ISBN 0-393047-555.

Shepard Krech’s book The Ecological Indian is the best of the works that de- bunk the concept of “the Indian” or “the indigenous/traditional person” as a natural environmentalist and conservationist, at one with nature. As he puts it: “For while this image may occasionally serve or have served useful polemical or political ends, images of noble and ignoble indigenousness, including the Eco- logical Indian, are ultimately dehumanizing (p. 26).” As this passage implies, he rejects also the image of the ignoble savage, “cannibalistic, bloodthirsty, inhuman (p. 16).” However, the present book attacks only the nobler image, although the second remains-alas-far more common in popular culture, recent Hollywood mov- ies notwithstanding.

Most of the classic tales of aia oue waste of resources are at least men- tioned in this book. S me he credits. The core consists of six detailed case studies that speak to the seas of Ne ative American resource management. These studies have significance beyond the bounds of the book itself.

The first concerns Paul Martin’s famous case for humans as the cause of the extinctions of Pleistocene fauna at the end of the last Ice Age (Martin and Wright 1967). Krech does not look with favor on Martin’s thesis of a sudden, extremely rapid populations nena, spate ag rapid oo (This thesis is now un- necessary, t pre nd settlement in the Americas is becoming extensive. Martin can argue, reasonably, ree a long slow process, rather than having to insist on a wave of hunters marching in lockstep down the conti- nent singing “Stout Hearted Men.”) Krech also notes a singular lack of evidence in the form of kill sites; the argumentum ex nihil seems defensible here, because of the sheer volume of archaeological work done in the Americas since Martin’s book. We still await a significant number of finds.

Krech also points out that “minifaunal as well as megafaunal animals van- ished (p. 38).” The lost include small birds, dwarf pronghorns, storks, and other unlikely game for Paleo-American hunters. Martin's attempts to explain these as all due to human agency are notably lame; some were giant scavengers depen- dent on megafauna, but most were not. Moreover, the climate of the Americas changed exceedingly rapidly at the time. Pleistocene habitats changed radically. Many have no analogues today. Harrington’s mountain-goat, a southwestern spe- cies (and a very unlikely candidate for hunter-driven extinction), lived in habitats that combined elements now scattered from mountain forest to desert scrub; it could not survive the wringout its plant community broke up. Areas rich in megafauna became hot and dry. Large animals dependent on water would have been wiped out with or without human agency. “Six outher extinction events marked the last ten million years in North America (p. 40).”. There had been others earlier. All followed hard on periods of rapid drying.

Krech refers to the inappropriateness of the analogy to recent island ex- tinctions, which took place when people with far more sophisticated technologies entered far more circumscribed, predator-poor environments. I have mentioned

38 BOOK REVIEWS Vol. 20, No. 1

this problem. The Polynesians and other island pioneers brought dogs, pigs, rats, and new pathogens with them; analogy with recent island extinctions makes it certain that these did a good deal of the exterminating. The human invaders can be blamed in some sense for causing the extinction, but it was not really their deliberate bad management that ae it; it was their symbionts and parasites.

In short, the Pleistocene xti s are almost certainly not a simple matter of overhunting. I find it impossible to believe that humans were the main factor, but equally difficult to believe that hunters were not one of the factors. As I see it,

and increasing human popuission were eceany ona colli- sion course, especially as drying land als around water holes (cf. Glover 1997). Ambush must have been easy. appear was diffi- cult; the hunters would have had a hard time working out a plan, because every year the game diminished from natural causes. No doubt the hunters overshot in more ways than one.

Krech also provided a solid and reasonable review of fire in traditional Native American management. Krech, however, paid too much attention to large-scale prairie fire and less to the more carefully controlled and targeted burning that is known in California, the Northwest Coast, and other western landscapes. He con- siders all the evidence carefully, but provides much more detail on the more destructive cases. He might have profited from comparison with Australia, where set fires probably helped exterminate large marsupials (Flannery 1995) but has proved essential to the survival of many small ones (Nowak 1999; Pyne 1991).

Much less defensible is Krech’s stand on buffalo hunting. As he points out, the Indians at least left 60,000,000 buffalo on the plains for the white men to eliminate in the late 19th century. But, he maintained, the Indians wasted bison. This they did largely through the “buffalo jump”: Driving herds over cliffs and ravine rims. It undeniably involved a great waste of bison. However, Krech almost certainly exag- gerates the extent of this. Reading his work, one cannot escape the conclusion that, every time an Indian wanted a light snack, he drove a million buffalo over a cliff. The truth is more complex. For one thing, before the horse and gun, there were very few Indians on the plains. For another, not every jump got a million buffalo at a time-most managed to get very few indeed. For another, driving a herd of buffalo is exceedingly difficult even with the horse and gun; for tiny roving bands operating on foot, it must have been almost impossible. It would have, at the very least, taken the whole group a great deal of time to organize it. One would expect buffalo jumps to be very rare. This is indeed the case. Krech cites the displays at the Head-Smashed- In Buffalo Jump in Alberta as a source, so he is surely aware that those displays point out that there was only about one successful jump per quarter century. Other sites have even fewer jumps. Many were used once only. Of course, once a herd starts over a cliff, it may stampede, leading to the deaths of many times more buf- falo than a group could possibly butcher and utilize. Evidence, however, suggests that such a mass kill was a rare event. (Was it viewed as a tragedy?) Much more often, only a few buffalo went down, and these were more or less thoroughly used. Of course, the Ecological Indian of Hollywood stereotype would probably not have used the jump method at all.

One main theme of Krech’s book is his contention that the concept that the

Summer 2000 JOURNAL OF ETHNOBIOLOGY 39

Native American concept of rebirth of animal souls allowed the Indians to kill without compunction. Virtually universal in native North America is the theory that animal souls are immortal; when one animal body dies, the soul goes to seek out another. This is taken by Krech as a license to kill. What Krech ignores is that in every well-documented case known to me-and I have worked with several Native North American groups and have read mythology or ethnographic data from all of them-this theory is used as a justification for good management. Usu- ally it is explicitly so used. The souls are believed to go elsewhere, or to refuse to reincarnate at all, if humans treat the animals with disrespect-and the worst form of disrespect is deliberately taking more than one needs. Weirs that block whole rivers, slaughter of whole herds of game, trapping out whole populations of bea- ver, and other such overdrafts on the resource base are prohibited; such offense makes the spirits go elsewhere. In addition, most groups have concepts of Mas- ters of the Game, or leader animals, or some suc turally powerful animals that watch over the animal beings and punish humans who overhunt. As a con- servation measure, this is hopelessly inadequate in an age of shotguns, but it is reasonably effective in a simpler, less heavily armed economy. I know this be- cause it is still very much alive and taken very seriously in the area where I now work, the Maya communities of the Yucatan Peninsula. It still operates to reduce hunting pressure very substantially. Overhunting goes on, but at least some game survives.

Krech’s next case is that of the beaver, and here he reaches still farther out- though, again, he is careful to assign the worst blame to European enterprise for trapping the beavers out, and to luring the Native Americans to do most of the dirty work. But he seems to maintain that unchecked waste of beaver was ab- original. This is difficult to believe. Traditional Native American ideology of beaver conservation is too well-documented and widely documented to deny; it is still a feature of life in Canada. However, Krech follows the very shaky arguments of Robert Brightman (1993), who indeed found this conservation ideology, but main- tained it was a result of teaching by early fur traders and other white outsiders. The obvious problem with this is that Brightman, and Krech, rely largely on the testimony of fur traders who were trying to explain to the home office why fur- bearer numbers were thinning out. Blaming the Indians, who were claimed to waste beaver in spite of all the traders’ diligent directions to the contrary, was an obviously rather self-serving story. Perhaps it was true, at least locally. But the conservation ideology of the Indians, as documented by Brightman and virtually all other ethnographers, is the same general belief system that one finds from the Koyukon of Alaska to the Maya of Quintana Roo. It is encoded in myth and ritual all over the continent. It did not come from the fur traders. Nor do I believe that many fur traders were seriously interested in conserving the resource. They were, indeed, rather more prone to forestall rival groups by deliberately trapping out all beaver over vast areas. Ina single expedition, Peter Skene Ogden led a party that exterminated the beaver from most of Oregon and northern California (Ogden 1987, orig. ca. 1827). The protestations of early writers, and their blame of Indians (whom they tended to regard as drunken and bloodthirsty savages), ring rather hollow. Conversely, to maintain his theory, Krech is forced to dismiss virtually all

40 BOOK REVIEWS Vol. 20, No. 1

41

, from Frank Speck to Harvey Feit, as hopeless romantics in the grip of the Ecological Indian stereotype. Never mind that Speck and many others wrote at a time when the stereotype of the Ignoble Savage was overwhelmingly domi- nant and the Ecological Indian lay in the dim future.

On the Northwest Coast, I found that the t t strategies and the real awareness of how to use the environment were found most strongly among the very traditional men and women, many of whom spoke little English. By contrast, the glib generalities of the “ecological Indian” sort were indeed found mainly among the young and English-educated. So the rhetoric may be learned, but the substance was old.! In Mayaland, there is no conservation rhetoric to copy; Mexican rural development strategy is still overwhelmingly focused on de- stroying nature and everything natural just as fast as aa Thus, it iB safe to ascribe -— Maya t teaching or behavior that 8 j y aware to genuine tradition.

Books that paint the Native Americans in a good light-from frankly Ecological Indian sources (e.g. Hughes 1983) to more scholarly work (Berkes 1999)-focus on the best of ideology: on myth, cosmology, and teaching. Certainly, it is undeniable that all Native American peoples were intensely conscious of their environments, and encoded incredible amounts of knowledge-both pragmatic and “religious”-in their myths, tales, and cosmological teachings. Yet, as Krech points out, such people often compare the best of Native ideology with the worst of Western practice. Conversely, books that slant toward an Ignoble Savage view, such as Krech, and those that take this view to an extreme, including such intemperate writings as those of Martin Lewis (1992), Rod Preece (1999) and Matthew Ridley (1996), focus on the worst of practice: on overhunting, over-irrigating, overburning. These lat- ter writers have varied political agendas. Lewis intends to defend moderate political positions against those who see a need to dismantle western civilization wholesale. I agree with him-in fact, that is the theme of the present book too-but he had no need to trash the indigenous peoples in the process. Preece, who writes with the equally worthy goal of redeeming western civilization from charges of being anti-environment, unfortunately ruins his case by comparing the best of Western ideology with the worst alleged indigenous practice.

Comparing best ideology with worst practice in 20th century America, one would see a gap between the writings of John Muir, Rachel Carson, and Stewart Udall, on the one hand, and on the other the area of wilderness paved, the number of species exterminated, and the acres of forests permanently destroyed.

What is the truth? Muir and toxic waste dumps were both a part of America in the 20th century, and there was a great deal in between. The Native Americans too have a diverse record.

These negative authors do not consider the superb management of resources that is extremely thoroughly documented for Pueblo agriculture, California plant management, Northwest Coast fisheries, and Maya swidden agriculture (except for the Late Classic overcut!). But, on their part, the Ecological Indian writers delicately write around the issue of human frailty: the fact that the spirit may be willing, but the flesh is weak. Moral standards are normally made to be impossi- bly strict, since moralists are sadly aware that people almost always fall short of

Summer 2000 JOURNAL OF ETHNOBIOLOGY 41

precept. If one prohibits overhunting, one can hope to reduce it, but not to elimi- nate it.

These matters are highly political. The Ecological Indian theory, if taken seri- ously, might lead to giving Native Americans unlimited power over their own resources. This might have most unfortunate consequences. In a few limited ar- eas, it is indeed having such consequences already (see Terborgh 1999; and I have encountered the problem in a few Third World cases).

Conversely, the Ignoble Savage theory was and still is a quite open and calcu- lated justification for depriving indigenous people of their lands and resources. Clearly, people who wander about on the land, burning forests and thoughtlessly exterminating game animals, are not exerting any “true” property rights; they should be driven off the land for their own good, since all they do if left in control is ruin the land for everyone else. This logic is still common in the United States and Canada-I have heard it countless times from ranchers, conservation biolo- gists, fishermen and many others-and is even more frequent in Third World countries, where I have found it in control of local policy from Malaysia to Mexico (cf. Ascher 1999).

It is interesting to contrast = works on both sides with the classic eth- nographies from the days wh ying to document the facts, as well as the statements of the Native Americans elves, Those ethnogra- phies revealed the Indians to be superbly aware of their environments, and good but not perfect resource managers. The extrem , in both directions, is anew thing for anthropology-but, alas, all too Welwcen a groove for politicians.

NOTES

1 Krech makes a common mistake among anti-Indians of maintaining that Chief Seattle’s famous environmentalist speech, originally made in the 1850s, was a fake. It was, in fact, real. However, it was heavily larded with fuzzily Christian rhetoric in a semi-fictionalized rewrite in 1970-71. The exact words of the speech have been lost, but early versions agree on included Chief Seattle’s militant defense of his land, supported by some concrete and spe- cific environmental details. The remake added some general, fuzzily-virtuous sentiments, but did not radically change the sense of these particular passages. See Kaiser (1987).

LITERATURE CITED

1999. Why

ASCHER, WILLIAM. GLOVER, JEFFREY. 1997. Human Impact amu Parte Gy eee ey 2 1 Resources tho Rinra and Moaonfaimna af A1ictralia

Johns Hopkins eporeaces Press, Baltimore. BERKES, FIKRET. 1999. Sacred Ecology. Taylor and Francis, Phildelpha. RIGHTMAN, ROBERT. 1993. Grateful Prey: Rock Cree Human-Animal Relationships. University of California

RY, TIM. 1995. TI a3. F t E ti . Braziller, New York.

and North America during the Late Pleistocene. Ms.

HUGHES, J. DONALD. 1983. American Indian Ecology. Texas Western Press, El

KAISER, RUDOLPH. 1987. Chief Seattle’s

Arnold Krupat (editors.). University of California Press, Berkeley.

42 BOOK REVIEWS

LEWIS, MARTIN. 1992. Green Delusions. D i Durham.

H. E. WRIGHT (editors.). 1967. Pleistocene Extinctions: The Search — Yale University Press, New Hav

NOWAK, Bic 1999, Walker’ s Mammals of the World. 6th edition. Johns Hopkins University Press, Baltimore.

OGDEN, PETER SKENE. 1987. First Over the Siskiyous. Jeff LaLande (editor). Oregon Historical Society Press, Portland.

Vol. 20, No. 1

PREECE, ROD. 1999. Cultural Myths, Cultural Realities. University of British Columbia Press, Vancouver.

PYNE, STEPHEN. 1991. Burning Bush: A Fire 3 of Australia. Henry Holt, New Yor

RIDLEY, scant 1996. The Origins of Virtue. Penguin, New York.

TERBORGH, JOHN. 1999. Requiem for Nature. Island Press, Washington, DC, and Covelo, CA.

Eugene Anderson Department of Anthropology University of California Riverside

Journal of Ethnobiology 20(1): 43-59 Summer 2000

ETHNOECOLOGY OF WHITE GRUBS (COLEOPTERA: MELOLONTHIDAE) AMONG THE TZELTAL MAYA OF CHIAPAS

BENIGNO GOMEZ EI Colegio de la Frontera Sur (ECOSUR) A.P. 36, Tapachula. Chiapas, México 30700

ADRIANA CASTRO El Colegio de la Frontera Sur (ECOSUR) A.P. 63, San Cristobal de Las Casas, Chiapas, México 29290

CHRISTIANE JUNGHANS EI Colegio de la Frontera Sur (ECOSUR) A.P. 36, Tapachula. Chiapas, México 30700

LORENA RUIZ MONTOYA El Colegio de la Frontera Sur (ECOSUR) A.P. 63, San Cristobal de Las Casas, Chiapas, México 29290

FRANCISCO J. VILLALOBOS Facultad de Ciencias Agropecuarias Universidad Auténoma del Estado de Morelos Av. Universidad 1001, Col Chamilpa, Cuernavaca, México 62210

ABSTRACT.- A participatory study of white grubs of the family Melolonthidae among the Tzeltal Maya recorded traditional knowledge of this pest, and also maize cultivation practices utilized for deliberately or not managing the populations. This group of farmers has an ample knowledge of the bioecology of Melolonthidae present in their community. They know major life stages, and also natural enemies of larvae and adults, as well as the host plants used by the latter. Recorded agricultural practices that can reduce the damage caused by grubs include preparation of the fields; sowing; and hilling up soil around the plant. We contrast the knowledge of this Tzeltal group with knowledge generated by bioscientific methods, to make it possible to integrate and render the Tzeltal methods useful in possible programs for sustainable pest management.

Key words: White grubs, Melolonthidae, corn, Tzeltal Maya, Chiapas, traditional knowledge

1 a ae

abiaenabonobe A través d t gia | f a Aala gallin ina cipo0a (Coleoptera:

Melolonthidae) y y las prcticas pie a del cultivo del maiz que ‘realiza para el plaga. Este grupo de productores tiene un amplio conocimiento ‘de la bioecologia de los Melolonthidae que se presentan en su

44 GOMEZ et al. Vol. 20, No. 1

comunidad; cen su est lid la de los enemigos naturales de larvas y adultos, asi como las aed aceite de estos ultimos. Las practicas agricolas registradas que tienen un efecto en el dano causado por gallina Gren son: Preparacion de los terrenos, la siembra y el aporque. A trasta el cgoeane gees ‘ges este EFupo =e -tzeltal bases CORCCIEMENIDS eeeaisaspeiaes por el étodo

ey etes yl 3 FS

de manejo Rceanabie de plagas.

RESUME.- Une étude participatoire sur le vers blanc de la famille Mélolonthidae parmi les Tzeltal Maya démontré une connaissance traditionnelle de ce parasite et des pratiques en matiére de culture de mais utilisées pour contrdéler ces populations. Ce groupe de fermiers a une ample connaissance de la bio-écologie des Melolonthidae dans leur communauté. Ils connaissent les étapes principales de la vie de l’insecte, les ennemis naturels des larves et des adultes, ainsi que les Plantes oboe pee les adultes. Les pees agricoles décrites ci-dessous qui

ges 5 q la préparation des champs; le nett d i herbes; l’ encemencement; et la formation de monts de terre autour de la plante. Nous contrastons le savoir de ce groupe Tzeltal avec les connaissances produites par des méthodes scientifiques, pour permettre l’intégration et l’utilisation des méthodes Tzeltal dans de futurs programmes de lutte antiparasitaire.

INTRODUCTION

The highlands of Chiapas have hosted, for more than 500 years, a rural popula- tion of Tzotzil and Tzeltal Maya (de Vos 1980). Their deep-rooted history has given casei wide experience and enowier ss of local resources. Part of this traditional

recorded in the eth 1 work of Berlin et al. (1974, 1990), and the ethnozoological work of Hunn (1977), on Tzeltal folk classification.

However, there is more to be said about perception of Melolonthiids among the Tzeltal. Farmers of the Chiapas highlands refer to these insects as k’olom (Tzeltal Maya), k’onom (Tzotzil Maya) or gallina ciega (Spanish; lit. “blind chicken”). These organisms are the principal cause of losses to grain, vegetable, fruit and flower culture in the area (Ramirez et al. 1999). There is evidence (Ramirez and Castro 1997) that the level of damage caused by Melolonthiids in this region is similar to that recorded elsewhere in Central America (Quezada 1980; Rios-Rosillo y Romero- Parra 1982; Rodriguez del Bosque 1988; Morén 1993). However, we do not know precisely which species are pests in the area, nor do we know enough about their biology to propose a pest management plan with optimal possibilities for success (Morn 1986, 1993). Most of the literature on Mexico’s agricultural pests says that the name covers some eight species, but actually there are over 560 soil-dwelling larvae—root-eating ly either—in the “gallina ciega” com- plex (Deloya 1993; :Morén. 1983; Moron et al. 1996).

ntil now, no one seems to have studied ethnozoological aspects of scarab beetles in Mexico (Moron et al. 1997), still less the gallina ciega. We have only a few notes on consuming them as food (e.g. Ramos and Pino 1989: 21). As noted above, we consider it important no conduct ethnoecological studies on rural and indig- enous knowledge of the N | Empirical | ht be ver y useful

Oo oO

Summer 2000 JOURNAL OF ETHNOBIOLOGY 45

in developing strategies of management for these pests, and for conservation of other species in the same fami

Ethnoecology, like many other disciplines, — a role i in community develop- ment. This di ditional wisdom i , and relates it to productive practice as well as to global economy and to the world of rural cultivation (Vasquez 1992). Bentley (1992) notes that technical collaboration with rural people should be based on what they know (or do not know), including what they might need to learn, teaching it in such form that it can be consistent with what they do know, and can learn in a manner that allows synthesizing the new information with the old. With this view, we carried out the present work, taking into account also the point that success in sustainable pest management is based on a broad technical knowledge and/or traditional knowledge of the agroecosystem. Our objective was to record and analyze knowledge of Melolonthidae, and of maize cultivation practices for managing white grub popu- lations thereof, among the Tzeltal of Balun Canal, Chiapas, Mexico. Additionally, we evaluated the possibility that this knowledge could be integrated with strate- gies for sustainable management of these grubs.

The Study Area.— This investigation took place in the community of Baltin Canal, in the municipio of Tenejapa, Chiapas (a municipio is roughly equivalent to a county or township). This locality lies 22 km east of San Cristdbal de Las Casas, at 16 46’ 49” north and 92 32’ 12” west. It is some 2240 m asl. It has a temperate, subhumid climate with summer rain (C[w,][w]), with temperatures around 14-16 C. The veg- etation of the zone includes cultivated areas (principally maize and beans), and fragmentary remnants of oak-pine and montane mesophyll forest (based on data from the Laboratorio de Informacién Geografica y Estadistica de El Colegio de la Frontera Sur, 1997).

The Tzeltal of Baltin Canal.— The population of the co ity is composed of Tzeltal Maya. As of 1990, there were 500 persons distributed in in 80 households; 47% were male, 53% female; 35% did not speak Spanish (INEGI 1990).

In the same year, 45% were listed as economically active. Most (92%) were involved in primary production, principally maize cultivation (INEGI 1990). The fields were slash-and-burn, used intensively for two years. They were prepared for sowing in late winter and early spring. Sowing was done before, or at the beginning of, the May rains. The maize was harvested in autumn or the first part of winter. Because of the scarcity of cultivable land, as well as low yields and prob- lems with erosion and pests, the Tzeltal megibrated seasonally to work for pay in the coffee plantations or cattle ranches of Chiapas (Robledo 1994).

The Melolonthid pests. The term gallina ciega— literally, “blind hen” or “blind chicken”— is the common Spanish term for larvae of beetles of the family Melolonthidae. It also includes some Scarabaeid larvae, and other subterranean pests. According to Morén et al. (1997), this name has no known origin, and no equivalent in other languages. They note it may have arisen during the first years of Spanish colonization. The only relationships between these larvae and the name “blind chicken” seem to be their lack of conspicuous eyes and the possibility of their being eaten by chickens.

46 GOMEZ et al. Vol. 20, No. 1

The Melolonthids go through a full transformation: egg, larva, pupa and adult. This cycle can take one to seven years depending on their geographic position. Most tropical species have cycles that are annual or biennial depending on envi- ronmental conditions (Villalobos 1995).

Larvae of Melolonthids are often associated with grasses, legumes, rosaceous plants, and plants of the nightshade family (Mordén 1984, 1986). In Mexico they have attacked roots of maize, beans, sorghum, wheat, potatoes, rice, sugarcane, strawberries, carrots, spinach, tomatoes and onions (Moroén 1984; Rodriguez del Bosque 1988). Damage can be light (15% or less of roots), moderate (up to 40%), or severe (over 40%), depending on cultivation, environmental conditions, and state of development of the insect at a particular time (Villalobos 1995). The third larval stage causes the worst damage (Morén 1984).

METHODS

Because of criticisms of ethnoecological investigation, especially its methods (Vasquez 1992), we decided to modify and/or enrich the investigation with new techniques for recording knowledge. Particiaptory investigation can be very use- ful in that it proposes strategies focused on participation of a larger number of agents involved in the process of investigation

The community of Baltin Canal was chosen due to the favorable disposition of the people and to earlier data on knowledge of the grubs that the people had pro- vided before the investigation. Most of the heads of families were involved in a society called “New Balun Canal,” an informal organization (not officially regis- tered). This made it easy to work with participatory methods.

Field work was done from January to September of 1997, through 35 visits to the community, each one lasting two days and involving a Tzeltal interpreter. We held participative workshops and used various data-gathering techniques, such as direct observation and group interviews. We interviewed key informants in the course of informal conversation, and also gave guided and open-ended but stan- dardized interviews. We will proceed to describe various techniques used for recording data.

Participatory workshops.— We held four bimonthly meetings with members of the “New Baltin Canal” society, which included 54 heads of families. At the first work- shop, with 32 persons (some being absent for wage labor), we explained the research. The intention of the meeting was to motivate those present to participate actively in the investigation. In a second workshop, with 40 persons, the theme was the cultivation of maize and the activities involved in it. Systems of maize phenology were laid out, as each pl tage, and activities during the agricultural cycle. The third meeting (49 persons) served to make known the principal pests af- fecting maize cultivation. We used pictures of maize plants; participants wrote down or indicated the organisms and the affected part of the plant. Individuals voted for pests considered most damaging. Thus we developed an ordered list of the five principal pests. In the fourth participatory workshop, 30 members of the New Baltin Canal

Summer 2000 JOURNAL OF ETHNOBIOLOGY 47

organization worked on knowledge of larvae and adults of the Melolonthidae specifically. They were presented with: larvae si pupas preserved i in alcohol, and adults mounted on pins, to elicit cl They were shown larvae of Melolonthidae of different ER (Dynastinae and Melolonthinae), Scarabaeids, and other soil-inhabiting insects (Coleoptera: Elateridae; Lepidoptera; Noctuidae; Diptera), to elicit grouping and differentiation. Also they were pre- sented with adult scarab beetles collected in the area, and also with other scarabs (Cetoniinae and Scarabaeidae) that could be confused with them. Once these or-

anisms were classified, we wrote up on a sheet of cardboard the ecological attributes belonging to these organisms: where and how they live, what they eat, what enemies they have. When the Tzeltal referred to natural enemies, they were shown pictures of the animals (Hunn 1977), to make a more precise identification of the species to which they refer.

Group interviews.— These interviews involved local people directly involved in maize cultivation. As part of these activities, we encouraged conversations and inter- change of ideas and information about the grubs. During these interviews, we worked with an average of 15 participants, to facilitate application of participa- tory techniques such as going over field inspections with them, elaboration of diagrams, and group discussions.

Interviews with key informants. With the purpose of getting different perspectives on the problem of the grubs, interviews with key informants were given individu- ally to 13 persons of different ages. These persons were selected in the workshops, in accord with their participation and knowledge. With them, we went deeper into their knowledge of the biology, ecology and classification of the grubs. We also took up aspects of maize cultivation practices that intentionally or uninten- tionally managed populations of the insects. The value of this type of personal interview is that it avoids bias that can occur in group interviews. Biases can occur in these groups when expression of real opinions is inhibited.

Direct observation. This technique consists of observing intensively and system- atically the management of the grubs in maize cultivation. Information obtained in the field was contrasted with information obtained in the workshops and inter- views. As part of this activity, visits lasting two to eight days per month were made. Most visits to field sites lasted four days. Data thus obtained was organized and presented in the form of an agricultural calendar of maize cultivation.

Collection of entomological specimens.— We collected 320 specimens of soil (288 in nine fields and 32 in woodland near the community). Larvae of the gallina ciega complex were collected in situ, in unit sample of soil specimens (monoliths of 15 cm diameter by 20 cm depth), during February and March. Adults were captured in March to June (their flying period), using a light trap. Additional adult beetle material was obtained during a check of trees and bushes known to be wild host species, during twilight and early night (19: 00-22: ~ hours). rca were made together with farmers, so as to collect more precise | data. Also, we collected specimens of plants used as perches or host plants. Natural enemies were also captured, as were other soil-inhabiting larvae encountered. All this material

48 GOMEZ et al. Vol. 20, No. 1

was sent to specialists for adequate taxonomic determination, and deposited in the collections ECO-TA-E, M. A. Moroén and B. Gomez.

Contrast and analysis of knowledge.— Contrast and analysis of traditional and scien- tific knowledge was done in a qualitative method, using comparative tables.

RESULTS AND DISCUSSION

Taxonomy.— One hundred percent of the Tzeltal farmers referred to the grubs as k’olom. We do not know whether this word has any other significance. The Tzeltal can distinguish Melolonthid larvae from other soil larvae such as Noctuid moths (wajchan) and Diptera (me’toyiw). However, they cannot differentiate Scarabaeid from Melolonthid larvae, possibly because of their morphological similarity. In Mexico in general, gallina ciega applies to Scarabaeids, which suggests the term is equivalent to k’olom.

In contrast, the Tzeltal differentiate the adults: Scarabaeids are kutuntza, Melolonthids chimol or umo’ (Table 1). This finding differs from Hunn’s (1977:295- 297); he found that the Tzeltal used the term kuhtum ca for certain Scarabaeids (Geotrupinae) and Melolonthids (Cetoniinae). Hunn suggests that cimol is the word for rhinoceros beetles (Melolonidae: Dynastinae; possibly the species Xylorictes thestalus). He notes that the word umoh is utilized for June beetles (we think the genus Phyllophaga is meant and other similar twilight-flying scarabs (pos- sibly Anomala). He mentions that “cimol, umoh or kuhtum ca” are the same (see Hunn 1977:297). They use the terms as synonyms for any adults of the gallina ciega group. Our differences from Hunn may be due to the fact that we worked in one community, while he worked in various Tzeltal communities. It is true that in other Tzeltal communities one can encounter other names; in El Madronal (Amatenango del Valle) the name xkumuk is used (Ramirez and Castro 2000). Also, we find that the older name was umo’ and today the term chimol is more often used, since approximately the 1950s. This change has come about because of mi- grants from other communities, Tzotzil as well as Tzeltal. The name chimol or umo’ includes at least the following species in the community studied: Phyllophaga obsoleta, Anomala sticticoptera, and two possible new species: Phyllophaga sp. 1 group Phytalus, and P. sp. 2 group Anodentata. However, it is possible that other species of adult Melolonthids have no special name. These are: Hoplia mexicana, Xylorictes thestalus, Cyclocephala alexi, Phyllophaga sp. 3 group Schizorhina, and Ancognatha sellata. The larval phases of these are possibly included in the gallina ciega com- plex. The species that possess a name in Tzeltal during their adult phase are those which were sometimes consumed by local people, and/or species that come in great numbers around houses, attracted by lights. However, some species that inhabit forested land have no name (Gémez et al. 1999a).

In this context, it must be emphasized that k’olom is the term used in most of the Tzeltal region for the larvae stage. Scarabs possess different names in the vari- ous communities of the Tzeltal region: chimol, umo’, xkumuk’. Berlin (1973) notes that in folk classification, when a name goes beyond its geographic limits and extends to a wider region, it is because the term has gained a large cultural signifi-

Summer 2000 JOURNAL OF ETHNOBIOLOGY 49

TABLE 1.— Designation in Tzeltal of Baltin Canal, Chiapas; Spanish; and English for scarabaeiform beetles and larvae.

Description Tzeltal Spanish English I and Scarabaeid beetles k’olom gallina ciega beetle grub, white grub

Adults of Melolonthid species mentioned chimol, umo’ ronrones, etles, escarabajos de scarabs, June Junio bugs

Adults of S baeids (Geotrupes sp., Copris sp.) kutuntza Escarabajos, beetles, scarabs, estiercoleros, dungbeetles “rodacacas”

cance. The wide use of the name k’olom can be attributed to the importance of the insect as a pest of maize.

Bioecology.— Development, Apparently, the Tzeltal ‘esmuameie studied ignore fun-

damental aspects of the p wever, 100% of the 13 persons interviewed know that chimol or umo’ adults are ei progenitors of the larvae (k’olom). They suppose that these proceed from eggs of the chimol or umo’. Only two of the 13 know that the larva goes through a pupal stage. They have observed that from these “little balls in the ground” (as they call them) issue adult scarabs. The few people who know the pupal stage and relate the different stages to the adults are persons of advanced age. It is possible that the above follows from the relative short duration of the pupal stage—30-45 days in Phyllophaga according to Morén (1986). One could also argue that the pupal phase occurs when the soil is resting and the Tzeltal are not active in the fields, and because of this cannot de- tect the pupae.

Life cycle. — It is evident that the Tzeltal of Balun Canal have knowledge of part of the life cycle of these scarabs. However, they cannot recognize the distinct larval instars or (usually) associate the pupae with the life cycle. Also, during the third workshop, 100% of the participants agreed that the duration of the life cycle is one year. Those interviewed mentioned that each year there are larvae as well as adults, and that they occur in similar abundance each year. The group mentioned that the larvae are present in the maize fields during nine months of May to February, and reported that the greatest abundance of larvae occurs in August. The farmers also noted that the adults fly from March to June. They have observed that adults fly during a period of one to three hours, starting around 19:30 to 20:00 p.m. The foregoing fits with results of scientific research in highland Chiapasm Mexico (Ramirez and Castro 1997) and other countries of Central America (Lastres de Rueda 1996; Méndez et al. 1996; Mendoza 1996), and can be compared with field obser- vations (Table 2).

Host trees and perch sites. — In the third workshops, we found that 100% of partici- pants knew that the chimol or umo’ eats leaves of the ajil or jnak (Alnus acuminata

50 GOMEZ et al. Vol. 20, No. 1

TABLE 2.— Comparison of information obtained in this study with that reported in biological research in the area or in similar areas

THIS STUDY OTHER INVESTIGATIONS Time of larvae May - Feb. 7 May - 10 Jan. (Ramirez and Castro 1997)

Time of major damage August July - August

(Ramirez and Castro 1997;

Lastres de Rueda 1996;

Mendoza 1996; Méndez et al. 1996) Density of larvae in roots 44 indiv/m2 38 indiv /m?2 (Gémez et al. 1999b) during critical period Time of flying of adults March - June April - May (Ramirez and

Castro 1997);

April - June (Gomez et al. 1999a) Hour of flight 19:30 - 22:30 19:30 - 21:00 (Ramirez and

Castro 1997);

19:00 - 23:00 (Gomez et al. 1999a) Spatial distribution of the patches patches (King and Saunders 1984) larvae during period of attacks

Life cycle 1 year Univoltine, 1 year (King 1996)

ssp. arguta, Betulaceae) and the chiquinib (Quercus crispipilis, Fagaceae). In accord with what is reported for other Melolonthidae, another host tree of P. obsoleta could be Erythrina americana (Fal ; Mor6n 1997). 70% of the Tzeltal mentioned plums, peaches and pears as possible hosts. However, we observe that these trees do not become defoliated and are used only as perch sites by the scarabs. Host plants identified in other work (Gomez et al. 1999a) that are not recorded as part of tradi- tional Tzeltal knowledge are siban (Cornus excelsa, Cornaceae—used in the community for firewood); tujkulum chix (Solanum myriacanthum, Solanaceae), whose spines can wound people in the woods; and Senecio sancristobalensis (Compositae), which lacks a local name, possibly because it lacks use or impor- tance for this ethnic group. These three species are probably not part of the Tzeltal’s store of scarab knowledge because they grow in the woodlands and are visited by the scarabs at night (Gomez et al. 1999a), outside the view of the Tzeltal.

Habitat. - The Tzeltal know that the k’olom occurs in different habitats. They have observed that these larvae live in agricultural or forest soils, and that on various occasions it is possible to find them in rotting treetrunks. Similarly, they note that the larvae eat roots of maize, beans, potatoes, and various trees, as well as rotten trunks and decomposing leaves in the soil of the woodlands. The scientific litera- ture, and observation during the present study, suggest that the habitats of the various species of gallina ciega are different. Phyllophaga obsoleta and Anomala sticticoptera are principally associated with agricultural soils (King 1996; King and Saunders 1984; Mor6én 1988; Morén et al. 1997; Ramirez and Castro 1997), while

Summer 2000 JOURNAL OF ETHNOBIOLOGY 51

the rest of the species appear to be confined to forest soils (Mor6n et al. 1997; Ratcliffe and Delgado 1990).

Natural enemies. - The workshops and interviews indicated that various enemies are recognized (Table 3). The Tzeltal have observed that the larvae are consumed by animals such as the skunk (pay), armadillo (majiltibal), pig (chitam), domestic fowl (me’mut), and various wild birds (tetikil mut), especially the great-tailed grackle (jojmut). The Tzeltal of Balin Canal recognize as predators of adult Melolonthidae the domestic dog (tz’i), the cat (xawin) and the wildcat (cis balan). This suggests that the combined effect of domestic and wild animals animals could contribute to the regulation of grub populations.

With data collected in the fields, we could determine other less conspicuous natural enemies, unknown to the Tzeltal. T] fungus, Beauveria bassiana (Deuteromycetes), and a wasp, Pelecinus polyturathor (Hymenoptera). The lack of Tzeltal knowledge of such small or microscopic natural enemies is explained by Bentley (1992). This author mentions that size can be a limiting factor in tradi- tional knowledge. Moreover, Hunn (1977) suggests that the low density of solitary wasps like P. polyturathor can explain why they have no names in the community.

Use of Melolonthids for human food.— During the larval phase, Melolonthids are not eaten by the Tzeltal, and have no use in the area studied. In earlier times, there was a custom of eating scarabs toasted on a comal (flat griddle). This habit contin- ued until about 30 years ago. Today, few eat these beetles;! the custom has gradually been lost. A possible explanation is that the Tzeltal, in constant contact now with industrial products, have changed their foodways. This phenomenon could be an indicator of change in the quality of life among the Tzeltal, due to the introduction of new types of foodstuffs, leading to a decrease in the value placed on natural foods obtained in the fields. We have observed that, in the community studied,

TABLE 3.— Principle Natural Enemies of the Melolonthids Observed by the Tzeltal of Balun Canal, Chiapas

Tzeltalname Spanishname English name Scientific name Phase eaten Pay zorillo skunk Mephitis macroura and larvae ?Conepatus mesoleucus Mayiltibal armadillo armadillo Dasypus novemcinctus larvae Chitam cerdo pig Sus scrofa larvae Me’mut gallina fowl, chicken Gallus gallus larvae Jojmut zanate great-tailed Cassidix mexicanus larvae Tt perro dog Canis familiaris adults Xawin gato cat Felis domesticus adults Cis balam gato de monte —_margay Felis wiedii adults

52 GOMEZ et al. Vol. 20, No. 1

consuming adult Melolonthids are looked down on—it indicates low status. We do not know what effect this change may have had on increases of the populations of gallinas ciegas. Ramos and Pino (1989) record the consumption of Phyllophaga by Nyahnyu (Otomi) and Nahua in some regions of Mexico, but as larvae or pupae. Hunn (1977) notes that the Tzeltal eat chimol, but does not record which species. The results of our workshops and interviews suggest that the species were Phyllophaga obsoleta, P. sp.(Phytalus) and P.sp.(Anodentata) (Gomez et al. 1999a).

Agroecology.— In this section we present and discuss findings on certain agroecological aspects of the cultivation of maize (ixim) relevant to consideration of the grubs

Agronomic importance of the gallina ciega.- Twenty-eight percent of 49 farmers interviewed in the second workshop stated that the grubs are the most important pest of maize. pnt ae 30% soe sat that first place belonged to the gopher (baj,

ill others had otl inions: 17% voted for corn borers (wajchan,

Lepidoptera), a for rats and mice (ch ‘o), and 8% for squirrels (chuch). These re- sults suggest that the k’olom is the principal invertebrate pest of maize. However, there is no special form of control, except cleanup by hand or with a hoe

According to those interviewed, symptoms of damage by grubs are yellowing and acame of the plants; acame is a local term for blowdown (falling over of plants due to root loss followed by wind). Damaged shoots can easily be pulled from the ground. Nearby soil is then examined; about three to five grubs can be found per plant. 100% of the Tzeltal mentioned that damage is present year after year, at the same level, in patches in the fields. The period of the most severe attacks is August (Figure 1). Symptoms of damage are similar to those recorded in field work else- where in Mexico and Central America (Table 2).

Calendar and agricultural practices— The Tzeltal of Balin Canal have a calendar for their c activities (Table 4 and Figure 1). This calendar is flexible, exact dates being determined by environmental conditions prevalent in a given year. Berlin et al. (1974) observed a calendar similar to that reported in the present study. In this calendar, the agricultural cycle is strictly correlated with a “Tzeltal native calen- dar” (Figure 1).

The farmers of the community carry out agricultural practices that can have positive or negative effects, deliberate or unintended, on the grubs. Four—prepa- ration of land, cleanup, sowing, and hilling up soil around stalk bases (the calzado or aporque)—can have a negative impact on grub damage.

Land preparation. — The farmers state that preparing the soil (see Table 4) lets them loosen the ground, eliminate weeds, and kill potentially damaging organisms. Thus this practice permits better drainage and exposes the larvae of Melolonthidae to attacks by their natural enemies, reducing their populations. The former was also observed by Musick and Petty (1974, cited in Carballo 1996), who mentioned the negative effect that soil preparation had on the grubs.

Sowing. — Sowing of maize (Table 4) follows the initial cultivation. One hundred percent of the farmers stated that the sowing consists of putting 4-5 seeds in each hole. Sowing follows a more or less definite pattern: holes are made a meter apart

Summer 2000 JOURNAL OF ETHNOBIOLOGY 53

(in each direction). If this is done well, it is not thought of as a control measure, but it reduces the problem of grubs. The strategy can reduce damage by increasing the biomass of the maize root system, The epee can sustain more wind action, reduc- ing acame. It is possible that the p for root-eating activity in comparison with isolated individual stalks. Indeed, some plants affected by

TABLE 4.— Agricultural activities; their stated purposes; their possible effects on management of grubs

Activity Description Objective Effects on gallina ciega

Preparation If the field is being cultivated This clearstheland Grubs are eliminated

of the land for the first time, or after a to create a biotic and as found, being killed long period of abandonment, abioticenvironment _ by foot or hoe, or slash-and-burn cultivation favorable for leaving them exposed is used. If the field has been cultivation. to birds and sun. used recently, it is cleared by

and or with hoes. Then the soil is moved, with manual elimination of pests.

Sowing 4-5 seeds are placed ina hill. Sowing several seeds They can then better This establishes strengthens the withstand the root- cultivation. support system of consuming activity of

the plants. grubs and other soil pests.

Cleanup Weeding, manually or with This reduces It possibly diminishes hoe. Only one person (of 10 competition by the densities of grubs interviewed) used herbicides. weeds, and improves

visibility in the field.

Hilling This involves hilling soil and It provides a more It is a form of organic material around the solid support for the __ fertilization that can base of the maize plant, using eer careers diminish the damage hands or hoe. blowdown. It also by grubs through

sco the incorporating more humidity of the soil available nutrients around the maize stalk. in the root zone.

Doubling Plants are bent over, at ca. This is done to avoid It has no known effect

over 1.3 m above ground. rotting of maize in on the grubs.

the ear through humidity due to rains. It also reduces blowdown and bird damage. Harvest Ears are collected and taken This allows It has no evident

to a secure place. Removing _ utilization of the effect on the grubs. grains from ears is done just product

before consumption or

utilization of the grain.

54 GOMEZ et al. Vol. 20, No. 1

blowdown because of late grub damage can still render normal ears of corn and can be harvested.

Hilling up. — This is an activity (Table 4) carried out by the Tzeltal to give more firmness to the plants in the face of wind and rain. With this practice, consciously or unconsciously, the farmers avoid blowdown provoked by weakening of the root system by soil pests such as the grubs. This activity is done just before the damage becomes evident (May-June).

The Tzeltal indicate that blowdown can cause loss of the product, through damage by rats, squirrels and other animals or through decay. The function of hilling up in diminishing damage can be related to better growing of roots in the hill, which may also have more organic material. An alternative explanation has been proposed by Cruz (1999), who has observed that in one locality of the Chiapas highlands the hilling up increases the number of larvae in the roots. This greater density—which does not necessarily increase damage—is probably due to the lar- vae eating the organic matter added around the plant bases. Villalobos et al. (1997) have demonstrated experimentally that the content of organic material has a nega- tive effect on root-eating activity by white of Costelytra zelandica. The foregoing

Occidental calendar bac'ul Jolal ti? inh 7uct) muk hu can winkil bacul (20/Kil-8/1) i (10/11-29/I1) (14/V-2M1) (3/V1-22/V1) = ad ~ ~ (20/X11-8/1) . sakil ha hul oI cum Native Tzeltal (9/1-28/)) (30/11-18/1V) po eran peak sen (30/XI-19/Kll) calendar ahil tak Cay kin h? ucetik mak winkil mis (29/-17/M) (19/V-23/IV) (13/VII-1/VIll) (1/K-20/K) (10/x1-29/X1) mak hok’en @haw huk winkil yas kin (18/II-9/IN1) (24/1V-13/V) (23/VI-12M1N) (21%-9/Kl) AAAI} RABARSSA SAABASA Agricultural Sidieseseseseiosdeesesee practices Fenology of domesticated com 2 ' awe White grubs” seasonality “a 3 ‘-) A “ie %e i] & 3

Land preparation; [RAY sowing; (RBBB first cleanup; Faq] second cleanup; Ee third cleanup, hilling up; dobling over and’

Figure 1. — Agricultural cycle (calendar, practices and fenalogy) of domesticated corn in Balun Canal, municipio of Tenejapa, Chiapas and its relationship to white grubs' seasonality.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 55

suggests that hilling up benefits cultivation by adding nutrition to the plant as well as by reducing damage by grubs.

Cleanup. — The farmers carry out 2-3 cleanups of weeds in the fields, to reduce competition of weeds with maize (Table 4). They state that this can reduce infesta- tions of grubs, in that they can kill grubs while weeding. Various authors have suggested that vegetation in the form of pastures (King 1985, 1996) and weeds (Carballo 1996 and references therein) permit the soil to host higher densities of grubs. We assume that this is due to higher survival rates of immature stages, greater chances of oviposition (associated with less compaction of soil), greater availability of food, favorable microclimate, and lower levels of parasitism and predation. However, we think that cleanup of the field also eliminates alternative foods, and centers the grubs’ attention on the roots of the cultivated plants.

The Tzeltal and Sustainable Management of Grubs.— Sustainabl t of agri- is proplems caused by grubs has been defined as a strategy that presents ly different from integrated pest t, and which

is based in ecological, economic and social principles (Villalobos 1995). This au- thor recommends saving traditional agricultural knowledge and practice as one of the principles to achieve this aim. Taking account of this, we advised studying and presenting to the community the effects that could follow from sowing and hilling up in managing the grubs. These activities had not been seen as related to the problem. The information produced by these studies could through light on conditions in which the effect could be better exploited.

The intervention of domestic animals during preparation of the land and cleanup of weeds could be developed to reduce the population of grubs. The re- newal of consumption of adults by members of the community could also contribute to regulation of insect populations.

The information possessed by the Tzeltal of Balun Canal about hosts and hours of flying of the adults could be relevant in campaigns of massive collection, as proposed by Cruz et al. (1998). These campaigns could combine with initiatives to use adults of noxious species as food for domestic animals such as fowl and pigs.

The incorporation of organic matter in cultivation, directly via hilling up or through other means (incorporating agricultural wastes and animal dung), could be helpful. Such improvements could improve soil fertility and help reduce dam- age by grubs, and even help any possible beneficial activities of these insects (Villalobos 1994).

Information on ecology and life cycle of the beneficial and noxious species of Melolonthidae will be fundamental for proposing strategies for a sustainable man- agement of white grubs that would be viable in the community. Ultimately, it will be necessary to evaluate the economic significance of the grub damage in maize cultivation in Balin Canal, to get a clearer diagnostic of the problem and confront it better.

56 GOMEZ et al. Vol. 20, No. 1

CONCLUSIONS

The principal conclusions of this investigation are:

The k’olom is a Tzeltal term used for the larvae of a species complex of Melolonthid beetles, of which Phyllophaga obsoleta and Anomal sticticoptera possibly cause agri- cultural damage.

Among the Tzeltal, Melolonthids and Scarabaeids are differentiated as adults, but not as larvae.

The Tzeltal of Baltin Canal use the terms chimol or umo’ for the species Phyllophaga obsoleta, P. sp. 1 group Phytalus, P. sp. 2 group Anodentata, and Anomala sticticoptera.

The group knows the duration of the life cycle, the larval stage, the adults, and the hour of flight of the latter.

Few know the entire cycle; a large majority is ignorant of the pupal stage, and none are aware of the different larval instars.

The farmers recognize various natural enemies, all vertebrates. They do not know of entomopathogenic microorganisms or invertebrates that participate in natural regulation.

We record for the first time the ption of adults of Phyllophaga by humans in Mexico.

The Tzeltal of Balin Canal have agricultural practices that reduce damage by the grubs. These practices include preparation of land, sowing, and hilling up soil around cornstalk bases.

We should consider as highly valuable th tk ledge, and also non- awareness, found among the farmers.

NOTES

‘México leads the world in insect-eating, with more than 200 species consumed (Defoliart 1997), so it is not unusual that the group studied here consumes insects. Ramos and Pino (1989) mention that it is surprising that, though scarabs constitute a significant part of the order Insecta, their consumption worldwide is not well known. Some examples of Melolonthids consumed in other parts of the world include: Adults of the genus Pyronota are consumed by the Maori of New Zealand (Miller 1974); Podischnus agenor is eaten by the Yukpa of northeast Colombia and Megaceras crassum by the Tukanoan peoples of southeast Colombia (Defoliar 1997).

ACKNOWLEDGEMENTS

We want to express our gratitude to all the people and institutions that allowed us to complete this work: to the Tzeltal f. that shared their knowledge with us; to El Colegio de la Frontera Sur (ECOSUR), Instituto de Ecologia, A.C., Consejo Nacional de Ciencia y Tecnologia (CONACYT), FORD charauecic ha and me bsoppesteat project CONACYT 1716P-B “Percepcion y j p s cultivos de Los Altos de Chiapas”; to Martha

Summer 2000

JOURNAL OF ETHNOBIOLOGY Sf

Meza and Manuel Girén for helping us with the field work and for translating; to Ma. Silvia Mendoza, Manuel Anzueto, Concepcién Ramirez and Jorge Cruz for providing their support during the fieldwork and for their criticism; to Ada Luz Chame; Miguel A. Moron,

Leonardo Delgado and William de la Rosa for doing the

and to

Dr. Gene Anderson for his excellent translation of this paper into English and providing

constructive criticism.

ITERATURE CITED

BENTLEY, J.W. 1992. El rol _s los agricultores en el MIP. Ceiba 33 (1): 257-

BERLIN, B. 1973. Folk systematic in relation to biological classification an nomeclature. Annual Review of Ecology and Systematics 4: cigars

sey Dis DLOVE and P.H. RAVEN. 1974. Principe of Tzeltal Plant Classification: An introduction to the botanical ethnogra 7 of a Mayan ey people of ar Chiapas.

cademic Press, New Yi

ee A. BERLIN, DE. BREEDLOVE, T. DUNCAN, V.M. JARA ASTORGA, R.M. LAUGHLIN, T. VELASCO. 1990. La herbolaria médica tzeltal-tzoltzil en Los Altos iapas, un ensayo preliminar sobre las cincuenta especies botanicas de uso mas frecuente. PROCOMITH. Gobierno del Estado de Chiapas (Serie Nuestros Pueblos 3).

gee nag V., M. 1996. Las practicas de

ulti n mai su efecto sobre Phylloohage spp. Pp. 119-125 en Biologia y Control de Phyllophaga spp. Phillip Shanon y Manuel Carballo (Editores). CATIE-PRIAG. Tuarrialba, Costa Rica.

CRUZ, a 1999. Alternativas de manejo de gal lina ciega (Co oleoptera:

iL:

del Valle, Chiapas. Tesis de licenciatura (Biologia). sage ed i Ciencias y Artes de Chiapas. Méxi

a, A. CASTRO pre C RAMIREZ. 1998. Ca ampana escolar contra el «ronron» (Coleoptera: Melolonthidae) en Amatenango del Valle, Chiapas. Pp.

en Memorias del XXXIII Congreso Nacional de Entomologia. Sociedad Mexicana de Entomologia. Acapulco, Gro. México.

DEFOLIART, G. R. 1997. An overview of the role of edible insects in preserving biodiversity. Ecology, Food and Nutrition: 36: 109-132.

DELOYA, C. 1993. El género Phyllophaga Aires en nineiaaea Morelos, Mé (Coleopte Melolonthidae: Se iniaatinans Pp. 39-54 en Diversidad y Manejo de Plagas Subterraneas.

iguel A. Moron (Coordinador). re te oe de la IV Mesa Redonda sobre Plagas Subterrdneas. ocieda Mexicana de Entomologia, México.

del Estado de Chiapas, México. ial ar B., FJ. VILLALOBOS, L. RUIZ y TRO. 1999a. Observaciones sn la bioecologia de meloléntidos (Coleoptera: Scarabaeoidea) en una localidad de Los Altos de Chiapas, México. Acta Zool. Mex. (n.s.) 78:173-

GOMEZ, B., F.J. VILLALOBOS, L. RUIZ, A. CASTRO y J. VALLE. 1999b. El complejo (Coleoptera:

Melolonthidae) en maiz en Los Altos de

tiempo de uso agricola y la materia organica ie ne Fol Entomol. Mex 107:1-20.

HUNN, E.S. 1977. Tzeltal Folk Zoology. The classication of discontinuities in nature.

cademic Press, N )

INEGI. 1990. XI Censo General de Poblacion y Vivienda, 1990. Chiapas: resultados definitivos por “eres (integracio6n territorial). Tomo I. Méx

G, A . 1985. Factats affecting infectacion by larvae of Phyllophaga spp. (Coleoptera: Scarabaeidae) in Costa Rica. Bulletin of Entomological Research 75: 417-427

Biologia, identificacién y distribucion de especies econdmicas de

0 spp. Phillip Shanon y Manuel C et CATIE-PRIAG. Tuarrialba, Costa

58 GOMEZ et al.

and J.L. SAUNDERS. 1984. Las

plagas de padi peor de los cultivos aes a yecopan s en América Central. ondon verseas te ee Satan CATIE. Turrialba, Costa Rica.

LASTRES DE RUEDA, _ Rae — de Phyllophaga spp. en Honduras. Pp. 15 en Biologia y one a Piviertas spp. Phillip Shanon y Manuel Carballo (Eds.). CATIE-PRIAG. Tuarrialba, Costa

Rica.

MENDEZ, E.R., H. RODRIGUEZ and F. TOUNDER 1996. Problematica de la gallina ciega (Phyllophaga sp.) en

n Biologia y Control

Manuel Carballo (Eds.). CATIE-PRIAG. Tuarrialba, Costa Rica

MENDOZA, P., H.E. 1996. 'Problematica de Phyllophaga spp. en El Salvador. Pp. 16- 19 en Biologia y Control de Phyllophaga spp. Phillip Shanon y Manuel Carballo igs CATIE-PRIAG. Tuarrialba, Costa

MORON, M. A. 1983. Introduccién a la pie anes Y eco ogia de los coleéntero if daficn] | México. Pp. 5-6 en Memorias de la II Mesa Redonda sobre Plagas del Suelo.

apingo-Sociedad Mexicana de Entomologia, Chapingo, México. . 1984. Escarabajos: 200 millones de anos de evoluci6n. pir scone 14. Instituto de Ecolo ogi DE. . 1986. El género Pinyllophaga en México. Publ. 20. Instituto de Ecologia, Ae. pee D.E ee ae de ha =e aan pris ee dae) co importancia agricola en ieee Pr 81- 102 en Memorias de la III Mesa Redonda jones a ie del suelo. Miguel A. Mor6n ied Mexicana de Sarees & ICI de México. Morelia, Michoacan

Vol. 20, No. 1

______ 1993. Las especies de Phyllophaga eee Melolonthidae) del Estado eracruz, México. Diversidad,

pA a Ria a Pp. 55-82 en

subterraneas, Miguel (Coordinador). Memorias de la IV Mesa Redonda sobre Plagas Subterrdneas. Sociedad Mexicana de Entomologia. México

HERNANDEZ and RAMIREZ. 1996. El complejo “gallin ciega” (Coleoptera: Melolonthidae) asociado con la cana de azuca Nayarit, México. Folia Entomol. Mex. 98:1-44

i, BL. RATCLIFFE. and €. DELOYA. 1997. Atlas de los Escarabajos de México. Coleoptera Lamellicornia.

ol. 1 Familia Melolonthidae. Comisién Nacional de la Biodiverisdad-Sociedad Mexicana de Entomologia, México.

MUSICK, K.J., H.B. PETTY. 1974. Insect control in conservation tillage. Pp. 47- 52 in Soil Conservation Society of America: A handbook for farmers.

QUEZADA, J.R. 1980. El manejo de plagas agricolas en Centroamérica. Folia Entomol. Mex. 45:16-27.

RAMIREZ, C. and A. E. CASTRO. 1997. El complejo ’ _ al _Mihllophage | y Anomala) en el cult del maiz Madronal, Mu vlictaic de Amatenango del be Pe se México. Pp. 6-7 e Mem de la Iil ann Lathipsciceioara de Escarabeidologia. Cuahutemoc Deloya eens ae de Ecologia, A. C.M

nenereee or oe 2000. El eauicie

“gallin ciega” (Coleoptera: Melolonthiae) - el cultivo del maiz en El Madron Municipio de Amatenango el Valle, Chiapas, México. Acta Zool. Mex.(n.s.) 79:17-41.

——_—=, 1. ARKEDONDO and CASTRO 1999. Biologia Comportamiento de la gallina ciega Phyllophaga (Phytalus) obsoleta; en la

Congreso Nacional de Entomologia. Soc. Mex. De _ Entomologia. Aguascalientes, México.

Summer 2000

RAMOS, J. and J.M. PINO. 1989. Los Insectos Comestibles en el México Antiguo. AGT Editor. México, D.F.

RATCLIFFE, B. and L. DELGADO. 1990. New species and notes of Cyclocephala from Mexico (Coleoptera: Scarabaeidae: Dynastinae). Folia Entomol. Mex. 80:41-

R{OS-ROSILLO, F. and $. ROMERO-

PARRA. 1982. Importancia de los danos

maiz por insectos del suelo en estado de ones — a as Folia Entomol. Mex.

ROBLEDO, G. 1994. Pueblos Baie de México. Tzotziles y Tzeltales. Instituto Nacional Indigenista - Secretaria de Desarrollo Social, México.

RODRIGUEZ DEL BOSQUE, L.A. 1988. Phyllophaga crinita (Burmeister) (Coleoptera una plaga del suelo (1855- 1988). Pp. 53- 80 en Memorias de la III Mesa Redonda sobre Plagas del Suelo. Miguel Angel Moron (Co ordinador).S Sociedad Mexicana de Entomologia, Morelia, Michoacan, México.

JOURNAL OF ETHNOBIOLOGY 59

Lets M.A. 1992. Etnoecologia para pre Le América Indigena 52(1 -2): 169-202.

iLL onos FJ. 1994. The contribution of melolonthid larvae to soil fertility. Pp. 129-143 en Resumen del 15 cn Mundial de la oo del Suelo Acapulco, Gro., Méxic

Sbihtasiess dr oe El aa sostenible de plagas del suelo: el caso de las larvas de season Pp. 69-89 en Control de Plagas peg pea oa se al

stin Aragoén (Ed.). Publicacién Secactal de la Sociedad Mexicana de Entomologia, Puebla, México.

ins? el. GOR, DJ, SAVILLE and R.B. CHAPMAN 1997. Interaction among soil organic matter, level of the indigenous Serratia entomophila in soil, peel disease and the feeding activity

A eae sie approach, Applied Soil Ecology: 5: 231-

Journal of Ethnobiology 20(1): 61-71 Summer 2000

IDENTITY AND CURRENT ETHNOBOTANICAL KNOWLEDGE OF FRANCISCO HERNANDEZ’S “CICIMATIC”

HELGA OCHOTERENA-BOOTH Instituto de Biologia, Universidad Nacional Autonoma de México Apdo. Postal 70-367 México DF 04510, MEXICO.

ABSTRACT. Francisco Hernandez’s “History of the Plants of New Spain”, written during the second half of the 16" century, is the main source of historical and 41 l 4 bs 11 j eee | 1 44] 1 4 Pa Baer 1 a . 1 smh oe

in Mexico. Despite the importance of this work, the lack of a universal system of

at ] 4 iT

J ilally Pp

in this volume. Currently, more than 2,000 plants remain unidentified and several previous identifications are questionable. Historical investigation of the uses of species belonging to the genus Ramirezella (Leguminosae) resulted in an identification proposed for Hernandez’s “Cicimatic” as Ramirezella strobilophora (Robinson) Rose, a conclusion in accord with Hernandez’s description, illustration and reported medicinal use. A Cicimatic was also mentioned in the work of Sahagtin (“General History about the Things of New Spain’), indicating that it was most likely a valuable plant during colonial times in Mexico. If Ramirezella strobilophora is the Cicimatic of Hernandez, the ethnobotanical traditions maintained over more than 400 years may indicate the potential pl} logical value of this species.

Key words: Ramirezella, Leguminosae, Mexico, Colonial times, Medicinal plants

RESUMEN.- La obra de Francisco Hernandez “Historia de las plantas de la Nueva Espafia”, escrita en la segunda mitad del siglo XVI, es la principal fuente de conocimiento histérico y etnobotdnico sobre las plantas que se conocian durante los tiempos de la Colonia en México. A pesar de la importancia de este trabajo, la falta de un sistema de nomenclatura universal en aquel tiempo hace dificil la identificacién de algunas de las plantas. Actualmente aun quedan mas de 2,000 plantas por identificar, ademas de varias con dudosas identificaciones, que requieren especial atenci6n de los taxénomos. Una investigacién sobre el posible uso de especies en el género Ramirezella (Leguminosae) durante los tiempos de la Colonia en México permite proponer una identificacion para el “Cicimatic” de Francisco Hernandez como Ramirezella strobilophora (Robinson) Rose. Esta conclusi6n se basa en la descripcién botanica, en la ilustraci6n y en uno de los usos medicinales reportados por Hernandez. Debido a que un Cicimatic también se menciona en el trabajo de Sahaguin (“Historia general de las cosas de la Nueva Espafia”), es posible suponer que se trataba de una planta valiosa durante los tiempos de la Colonia en México. Si R. strobilophora corresponde al Cicimatic de Hernandez, las tradiciones etnobotanicas que se han mantenido por mas de 400 afios podrian usarse como indicativas de un verdadero valor farmacoldgico de la especie.

62 OCHOTERENA-BOOTH Vol. 20, No. 1

RESUME- L’oeuvre de Francisco Hernandez “Histoire des plantes de la Nouvelle Espagne”, Scnte dans ie ese montié des XVI sénturi, c’est la principale source historique les plantes employer dans les Colonial temps au Mexique. En dépit de l’importance de c’est oeuvre, la manque d’un systéme de nomenclature universelle dans c’est temps difficulté la connaissance a l’identité des quelque plantes. Actuellement il y a plus que 2,000 plantes par identifier, en plus d’autre avec ciseatable saentite, ces nécessite de spéciale aberision par wig taxonomist. Une recherche sour le possible utilité “especes du genre R ) dans les Colonial temps au Mexique a permis de proposer une identité pour le “Cicimatic” de Francisco Hernandez comme Ramirezella strobilophora (Robinson) Rose. C’est conclusion est appui sur la description botanique, ]’illustration et Vutilisation médicinal rapporté par Hernandez. Puisque un Cicimatic est ti dans l’oeuvre du Sahagun (“Histoire générale de las choses de la Nouvelle Espagne”), il est possible du supposer qu'il éte un plante de valeur dans - ——— a au onan, “in ea i seihies aa cpreapancine avec le Cicimatic d’H Vil s’aveaux garder pour plus que 400 année ‘pouvoir indiquer une vrai valeur médicinale de c’est espéce.

INTRODUCTION

Francisco Hernandez’s History of the Plants of New Spain is the main source of historical and ethnobotanical knowledge about the plants that were known dur- ing early colonial times in Mexico. Herndndez’s work provides botanical descriptions for 3076 plants (Flores and Valdés 1979), together with their common names, uses and, in some cases, illustrations. The significance of Hernandez’s work is reflected in the various attempts to publish his entire contribution. Portions of it were published in four versions: in Mexico by Ximénez (1615); in Italy (1651); in Madrid (1790), and again in Mexico by the Instituto de Biologia, Universidad Nacional Autonoma de México (1942-1946). The only complete version of Hernandez’s work was published in seven volumes by the Universidad Nacional Autonoma de México between 1959 and 1984 as the result of a multidisciplinary effort that involved the participation of botanists, zoologists, linguists, geogra- phers, and historians.

Despite the unquestionable merit of Hernandez’s work, botanical writings of the era lacked a universal nomenclatural system. The use of common names that can refer to more than one species, be modified, or disappear with time, confounds the ability of modern workers to determine the identity of the species reported by Hernandez. Many researchers have provided significant contributions to the bet- ter understanding of the History of the Plants of New Spain (e.g., Sessé and Mocino 1887 a and b; Ramirez 1893; Altamirano 1896; Urbina 1897; Standley 1920-26; Batalla et al. 1942-1943; Miranda et al. 1946). However, the sheer volume of information and the more than 400 years that have passed since its creation still leave much interesting ethnobotanical information to be rescued.

To obtain a better understanding of the information for the more than three thousand plants mentioned by Hernandez, it is essential to know their taxonomic identity. Only about half of the plants mentioned by Hernandez have been stud-

Summer 2000 JOURNAL OF ETHNOBIOLOGY 63

ied (1,544). Based on the short but accurate botanical data provided by Hernandez as well as the occasional figures, identifications have been proposed for 98 names to the level of family, 249 to genera, and 667 to species (Flores and Valdés 1979). These names were compiled by Valdés and Flores (1984) in the seventh volume of The Complete Work of Francisco Hernandez. The remaining 530 of the studied names were not identified. The more than 2,062 names for which either no botanical com- ment has been given (Flores and Valdés 1979), or whose taxonomic identity is doubtful, stress the need for the participation of taxonomic specialists who could interpret Hernandez’s work.

In revising the genus Ramirezella (Leguminosae, Papilionoideae), it was noted that one species, R. strobilophora (Robinson) Rose, has many common names and traditional medicinal uses in Mexico (Ochoterena-Booth 1991). Distributed prima- rily along the Pacific slope of the Sierra Madre Occidental, from southern Sonora and Chihuahua to Nicaragua, R. strobilophora is known by 11 common names: Nowa (Chihuahua); Cuahuexutl, Ejote de Monte or Dichi-kuu (Guerrero); Frijolillo (Guerrero, Morelos and Oaxaca); Periquito Azul Grande (Morelos); Flor de Paloma or Ie-paloma, Gallinita (Oaxaca); Choreque (Chiapas) and Choncho (El Salvador). The root of R. strobilophora is used by indigenous groups in northwestern Mexico as a catalyst in the fermentation of Agave to prepare the beverage that the Warihios call batari (H.S. Gentry 2404, F, MEXU, US). The Raramuri (Taraumaras) use it for the same purpose during the fermentation of maize to produce tesgiiino (R. Bye 2847, COLO). In Oaxaca (Mexico) the bark of the liana is ground with water to treat fuegos (M. Sousa 7069 et al., MEXU), a kind of ulcer of the mouth (cold sores). In addition to this medicinal use, the boiled or toasted fruits are eaten locally (M. Sousa 7069 et al., MEXU; J. L. Viveros and A. Casas 332, MEXU). The present-day uses of Ramirezella strobilophora, as well as the great number of common names, se Ake ae a research on the possible uses of this species during colo- nial time

METHODOLOGY

Due to the morphological similarity between the genus Ramirezella and Phaseolus (the common bean), descriptions and illustrations of Hernandez that refer to beans (frijoles) were compared to species of Ramirezella. These were lo- cated using the indices of the History of the Plants of New Spain (Hernandez 1959). All descriptions that clearly did not correspond to Ramirezella were ruled out. The works of de la Cruz (1964, first edition 1552) and Sahaguin (1969, first edition 1590) were then consulted using Hernandez’s names in addition to beans (frijoles). The current ethnobotanical information of Ramirezella strobilophora was obtained from the notes on the labels of herbarium specimens.

RESULTS

Great resemblance was found between Ramirezella strobilophora and the illus- tration of the “Cicimatic” (Fig. 1A) from the History of the Plants of New Spain (Hernandez 1959). The similarity of the illustration was corroborated by the de-

64 OCHOTERENA-BOOTH Vol. 20, No. 1

scription of the “CICIMATIC or a plant similar to the cimatl” (Book 1, Chap. LVII), which can be translated as follows:

About the CICIMATIC or a plant similar to the cimatl, The root is like that of the turnip and fibrous; twining red stems are borne from it with three- foliolated leaves that are heart shaped and similar to those of the other beans, of which it is a species, and medium size legumes which are produced by purple flowers in cluster like groups. It has a cold and astringent tempera- ment. The root, when crushed and sprinkled, cures ulcers because it cleans them and favors healing; therefore many people call it palancapatli, which means medicine for ulcers. It relieves in an admirable manner the inflamed sick eyes, removes clouds and fleshy excrescence, stops discharges of the abdomen, cures cough and makes parturient [women in labor] stronger. The cooked root is good against dysentery. It grows in temperate or warm regions like the Mexican one.

The same name (Cicimatic) was found in the General History about the Things of New Spain from Sahagun (1969: Volume III, Book 11, Chap. 7, No. 232, Pg. 322) and the description also could be assigned to Ramirezella.

ba 7 ©

FIGURE 1A and B.— “Cicimatic” reproduced from Hernandez (1959).

Summer 2000 JOURNAL OF ETHNOBIOLOGY 65

DISCUSSION

About the name.— Santamaria (1942 and 1974) reported that “Senecio vulneraris” (pre- sumably Senecio vulneraria DC) is known as “Cicimate,” a name derived from “Cimatl.” Nevertheless, the description and illustration of Hernandez do not cor- respond to species of Senecio or any other Composite and therefore it is easy to discard this as a potential identification for Hernandez’s Cicimatic. Many of the common names reported by Hernandez are apparently not applied any more. This could be due to the lack of current ethnobotanical information about Mexican plants, or it could be that the names were lost after the more than 400 years that have passed since the work was written. The second case represents a likely pos- sibility for the Cicimatic since the name itself referred to another plant, as can be seen in the Hernandez’s translation: “a plant similar to the Cimatl. Nonetheless, the existence of another species known with that common name (Senecio vulneraria DC) allow us to still consider the first alternative. None of the 11 common names by which Ramirezella strobilophora is currently known is linguistically related to it. According to Martinez (1979), the name Cimatl refers to Phaseolus coccineus L., but this reference could have been obtained from a proposed identification of Hernandez’s Cimatl. According to Paso and aaewanee (1988) the word Cimatl “was applied to roots which nt, commonly perpendicular, and sometimes pivoting, whether they were edible or not... cimatl was equivalent to stump or underground axis.’

About the proposed identification.— Urbina (1897) proposed that the description and characteristics of the figure correspond with Canavalia villosa Benth. More recently, Baas: et al. kaso took up Urbina’ s identification (Valdés and Flores 1984).

of the genus Canavalia agree with Hernandez’s illustration, others do not. The stipules in this genus are small and deciduous, contrasting with the illustration of Hernandez (Fig. 1A), which shows very con- spicuous stipules. On the other hand, the inflorescence of Canavalia is a cluster in which the lower, more mature flowers are bigger than the upper ones, giving to it a conic aspect similar to Hernandez’s illustration. Although the flowers have the color mentioned by Hernandez, the morphology is different from the flowers of beans (Phaseolus) and can be easily distinguished. The fruits are comparatively larger than the legume of beans and have a rib along the side, which lends them a distinctive and characteristic aspect hard to cneeee witha bean.

Besides the lack of concl to interpret Hernandez’s Cicimatic as a Canavalia, there is no current common name known for any species in this genus that can be connected with Cicimatic or Palancapatli. Nor is there ethnobotanical support for this identification. Although the fruit of Canavalia is eaten in some regions (e.g. Guerrero, Mexico), there is no information about any medicinal use. In summary, although there are some similarities between the il- lustration and Canavalia, this identification is not supported.

About the description. Several species of the genus Ramirezella were described or considered at 5 some point as Phaseolus before that ‘genus was convincingly delim- ited. This refl imilarity of both genera. The description

oO oO

66 OCHOTERENA-BOOTH Vol. 20, No. 1

of the Cicimatic provided by Hernandez is consistent with the characteristics of the genus Ramirezella. Although the vegetative characteristics in most of the Phaseolinae species are very similar, the description of the root resembling “the one of the turnip and fibrous” fits the root morphology of the genus Ramirezella (Fig. 2C). The fruits of R. strobilophora (Fig. 2B) are bigger than those of wild beans (Phaseolus), reaching sizes between 11.5 and 17 cm. Therefore, they can be catego- rized as “medium size.” The flowers are grouped in a kind of inflorescence that is known as a pseudocluster (Fig. 2A). The color of the flowers of R. strobilophora varies from white to violet, so that the tonality corresponds to that mentioned by Hernandez.

About the uses.— Among the long list of medicinal uses reported by Hernandez for the “Cicimatic,” there is one in common with Ramirezella. Hernandez’s text de- scribes the capacity of Cicimatic to cure ulcers, a feature reflected in Hernandez’s translation of a second = name: _ pacmneapat or medicine for ulcers.” As mentioned previously, Rami is used in Oaxaca to treat cold sores,

which can be considered a kind of ulcer.

The properties of the root of Cicimatic were emphasized by Hernandez. It is interesting to note that the same part of the plant is currently used to elaborate fermented beverages. Gentry (1942) pointed out this employment of the root by the Warihios, but he unfortunately misidentified his collection number 2404, giv- ing to it the name Phaseolus caracalla L. instead of Ramirezella strobilophora. This error caused confusion in the subsequent use of the information, for in a later work, Gentry (1963) wrote:

The chopped pieces [of Agave’s stems] are put into large ollas [pots] of water, and as a catalyzer the root of a vine (nawo) (Phaseolus caracalla L.) is put in, which they say causes the water to ‘boil’. After a day or so the bub- bling stops and the batari is ripe for drinking. The older the brew becomes after this point, the weaker it grows and they speak of it depreciatively as ‘pasado’ [over-fermented]. If plenty is drunk, inebriation ensues. The drink has a sour astringent flavor.

In the above passage, Gentry did not cite any reference specimens, but it is clear that the information came from his 1942 publication because of the coinci- dent use, scientific name and common name, the latter, by the way, with nowd appearing as nawo

The fact that Ramirezella’s root affects the activity of bacteria and fungi may reflect the existence of some interesting compounds, as has been reported for Phaseolus (Litzinger 1983). Litzinger mentioned the presence of isoflavonoids, ste- roids, oxidative enzymes, non-proteic amino acids and indol alkaloids, which, according to Lappe and Ulloa (1989) take part in the catalytic conversion of the substrate during the elaboration of fermented beverages. Litzinger (1983) suggested that these substances could have medicinal applications or help in the treatment of intoxication.

Two interesting points regarding traditional uses of plants through history rise in this case. The first one is related to the loss of uses, the second to the acqui-

Summer 2000 JOURNAL OF ETHNOBIOLOGY

Bracts

Flower

FIGURE 2.— Morphological characteristics of Ramirezella: (A) inflorescence of R. " strobilophora (Robinson) Rose; (B) fruits (legumes) of R. strobilophora; (C) root of R. nitida iper.

68 OCHOTERENA-BOOTH Vol. 20, No. 1

sition of uses. Are there more medicinal uses for Ramirezella that we do not know? If not, why did the other medicinal uses reported by Hernandez get lost? On the other hand, were the roots of Ramirezella always used in the preparation of fer- mented drinks and Hernandez did not capture this information? If not, when and how was this used acquired? More ethnobotanical research is needed in order to try to understand these questions.

About the illustrations. As can be seen in the illustrations of the Cicimatic repro- duced from Hernandez’s work (Fig. 1 A and B), there are differences between the two plants illustrated. The plant of figure 1A has a shorter root than that of Figure 1B and the detail of the flower (bottom right Fig. 1B) does not correspond to one of the Phaseolinae group because of its radial symmetry. The plant of figure 1A can be identified as a kind of bean or a related group because it shows trifoliolated leaves with stipules and a legume similar to a green bean (bottom right).

Comparing the illustrations with the descriptions of Hernandez, figure 1B can be best assigned to AYECOCIMATL, which has the descriptive subtitle “a herb similar to the Cimatl” (Book 1, Chap. LV). The fact that the name Ayecocimatl also alludes to the Cimatl could be the cause of a mistake in the inclusion of this figure under the Cicimatic. In the description of the Ayecocimatl, Hernandez says that it has “...flowers at the end of the branches, scarlet and radiated as a star...,” just as it is represented in the detail of figure 1B. To assign this figure to the Ayecocimatl, which from the description was identified as Phaseolus coccineus L., opens the need for its reinterpretation, which in fact requires further study.

Figure 1A, on the other hand, corresponds to the description of the Cicimatic and Ramirezella, especially because of the inflorescences, which appears to be a many-flowered cluster, and has the general aspect of this genus (Fig. 2A). In the drawing of the inflorescences it is possible to distinguish structures that can be interpreted as buds protected by bracts (Fig. 1A). Relatively large and persistent bracts are characteristic of the genus Ramirezella. The fruit and vegetative charac- teristics can also be associated with this genus (Fig. 2 B and C).

About other sources of the XVI century—In the de la Cruz codex (1964), the first written work we know that refers to medicinal Mexican plants (originally published in 1552), also known as Badiano codex, there is no plant that can be related with the Cicimatic (Valdés et al.1992).

In the General History about the Things of New Spain (Volume III, Book 11, Chap. 7, No. 232, Pag. 322) Sahagtin (1969, first published in 1590) wrote:

There is another medicinal herb called cicimatic; it is a vine, with many very green leaves and wide growing in groups of three; it is like the beans; the green parts are not useful at all; the root has no flavor and is hard as a trunk, almost the size of the head of a person and large as an elbow; it has a thick bark, black outside and with thick red spots inside. Grounded, it is good for people with sick eyes that have a fleshy excrescence called ixnocapachiui; the ground-up root is covered with a cloth and squeezed over the eyes, after that, the fleshiness that covered the eyes is gone; it grows in all the mountains.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 69

Later in the same work (Volume III, Book 10, Chap. 28, No. 9, Pag. 170), Sahagun rote:

Against the sores outside the ears there are these remedies: take the leaf of coyolxochitl, grind it and mix it with océtzotl and put it over the sore, or grind it and mix it with the ajiya already mentioned and apply it on the sore, or take the herb called cicimatic in the [native] language, mix it with egg whites and apply it on the sore, or use all the other herbs that can be used to treat the rotten sores like the herb called chipilli and the stone of the avocado.

The woody and hard root of Ramirezella can reach up to 70 cm in length. A red resin is present in both the stem and the root (Ochoterena-Booth 1991), character- istics that coincide with Sahagtin’s description. The names mentioned by Sahagun are the same as Hernandez, which suggests that it was an important plant during prehispanic and colonial times in Mexico. Estrada Lugo (1989), probably follow- ing Urbina’s identification, suggested that the Sahagtin’s Cicimatic corresponds to Canavalia sp. The same arguments made in favor of the Ramirezella identifica- tion can be also applied here.

CONCLUSIONS

Due to the inherent problems interpreting a treatment greater than 400 years old, it would be incorrect to reject categorically any alternative identification for Hernandez’s plants. However, Hernandez’s description of Cicimatic as “...similar to those of the other beans, of which it is a species;” the evidence of stipules in the illustration (Fig. 1A); the characteristics of the inflorescence in the drawing, here interpreted as bracts (Figs. 1A and 2A), the kind of fruit (Figs. 1A and 2B); the uses for the plant, and the description of Sahagtin, more probably correspond with those of Ramirezella. If this is true, considering the distribution and morphological characteristics of its species, it appears to be R. strobilophora. With this new inter- pretation, interesting alternatives emerge for future research related with the ethnobotany and potential pharmacological value of the genus Ramirezella. This kind of research could reinforce the proposed identification for the Cicimatic and at the same time allow a better use of Mexican natural resources.

ACKNOWLEDGMENTS

I acknowledge the curators of the following Herbaria for the access to the herbarium specimens from the genus Ramirezella: AA, CAS, CHAPA, COLO, ENCB, F, FCME, GH, HUAA, LA’MAGATALL, LL, MEXU, MICH, MO, NY, TEX, UAT, US, WIS, XAL. I would like to thank Alfredo Pérez and Ismael Calzada for the phonographs of the inflorescence and fruits, respectively. I sincerely appreciate the comments on the content of the original manuscript from Alfonso Delgado, Robert Bye, Hilda Flores and Javier Valdés. I also appreciate the kind review of the English version by Debra Campbell, Gary Fick, David Bates, Donovan Bailey, Pamela White, and the comments of one anonymous reviewer and Iris H. W. Engstrand, which greatly improve the content of the paper.

70 OCHOTERENA-BOOTH

Vol. 20, No. 1

LITERATURE CITED

ALTAMIRANO, M. 1896. Historia natural anlicada dalnce + ; An

Inst. Med. Nac. Mex. 2: 261-272. .

BATALLA, M.A., D. RAMIREZ CANTU and I. RIBERA MORALES. 1942-43. In Historia de Las Plantas de Nueva Espana Vols. 1-2 (F. Hernandez). Imprenta Univ., Mexico D.F

DE LA CRUZ, M..-1964, Libellus de Medicinalibus Indorum Herbis

Reimpresion, IMSS, Mexico D.F.

ESTRADA LUGO, LIJ. 1989. El Cédice Florentino. nformacién Etnobotanica. Fortuna S.A-Colegio de Postgraduados de Chapingo, Mexico

FLORES OLVERA, M. and J. VALDES GUTIERREZ. 1979. La obra botanica del

Medicina Tradicional 2(6): 84-8

GENTRY, H. S. 1942. Rio Mayo ae A study of the flora tig vegetation of the valley of the Rio Mayo, Sonora. Carnegie tnstitation of Washington, Publication 527. Washington

______. 1963. The Warihio Indians of Sonora- Chihuahua: an ethnographic survey. Anthropological Papers No. 65. Smithsonian Institution. Washington,

Ds

Hernéndez, F. 1651. Nova Plantarum, Animalium Mineralium Mexicanorum Historia (Tesaurus). Roma.

. 1790. Opera cum edita, tum Inedita (Edicion Muted’, (ee Heredum, Madri

1942-46. Historia de Las Plantas de Nueva Espana. (Libros 1-24; Vol. II and Ill). Instituto de Biologia, UNAM, inne! D.F.

1959-84. Obras Completas (Vol. EVID. UNAM, Mexico D.F.

LAPPE, P. and M. ULLOA. 1989. Estudios Etnicos, Microbianos y Quimicos del Tesgiiino Tarahumara. Instituto de Biologia, UNAM, Mexico D.F.

LITZINGER, W. J. 1983. The Ethnobiology of Alcoholic Beverage ae by the Lacandon, Tarahumara, and other aboriginal Meanaieeican peo

esis, Doctor in Philosophy, University of Colorado.

MARTINEZ, M. 1979. Catalogo de Nombres Vulgares y Cientificos de Plantas Mexicanas. Fondo de Cultura Econémica. Mexico D.F.

MIRANDA, F., M.A. BATALLA, D. RAMIREZ CANTU and I. RIBERA MORALES. 1946. In Historia de las Plantas de Nueva Espana (F. Hernandez). Imprenta Univ., Mexico D.F.

sage int BOOTH, 191.

evisi6dn Taxonémica del Género Rypivien lla Rose (Fabaceae, Papilionoideae). Tesis Bidlogo, Facultad de Ciencias, UNAM, Mexico D.

PASO Y TRONCOSO, F. DEL. 1988. La Botanica Entre los Nahuas y Otros Sor aie Secretaria de Educacioén

Publica-Cienia de Mexico, Mexico D.F.

RAMIREZ, J. 1893. Sinoniuia vulgar y cientifica de las plantas de la Nueva Espana de M. Sessé y José ‘Mocifio. El Estudio 4(7):220-22

AHA 7 By DE. 1969. Historia General de las Cosas de Nueva Espana. Porrua,

Mexico D.F.

SANTAMARIA, F. J. 1942. Diccionario General de Americanismos. Editorial Pedro ae México, D. F.

ten oe ario de ete ‘Editorial Porrua S$. A. México, D.FE.

SESSE, M. and J.M. MOCINO. 1887a.

Naturaleza 2(1):

Appendix.

—______and___. 1887b. Flora mexicana. Naturaleza 2(2): Apéndice.

STANDLEY, P.C. 1920-26. Trees and shrubs of Seri Contr. U.S. Natl. Herb. 23(1- 5): 1

URBINA, ‘i 1897. Catalogo de ape mexicanas (Fanerogamas). Imp. M

al. Mex., Mexico D. VALDES, J. and H. FLORES. 1984.

n Obra Completas tomo VII (F. oe UNAM. Mexico D.F.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 71

XIMENEZ, F. 1615. Quatro libros de la naturaleza y virtudes de las plantas y

exten Oee and H. OCHOTERENA- BOOTH. 1992. La botdnica en el cédice

de la CRUZ. in Estudios Recientes Sobre el Libellus de Medicinallibus Indorum Herbis (J. Kumate, M.E. Pineda, C. Viesca, J. Sanfilipo, I. de la Pena Paez, J. Valdés Gutiérrez, H. Flores Olvera, H. Ochoterena-Booth and X. Lozoya). Secretaria de Salud, Mexico D.F.

animales que estén reunidos en el uso de la medicina de la Nueva Espana, y la método, y correcci6n, y preparacion que

viuda de Diego Lopez Dabalos, Mexico. Reimpr. 1888. Secretaria de Fomento, Mexico D.F.

72 BOOK REVIEWS Vol. 20, No. 1

Gentry’s Rio Mayo Plants: The Tropical Deciduous Forest & Environs of North- west Mexico. Paul Martin, David Yetman, Mark Fishbein, Phil Jenkins, Thomas Van Devender, and Rebecca Wilson, eds. The University of Arizona Press, Tuc- son. xvi and 558 pp., maps, diags., photos, notes, refs., and index. $75.00 cloth (ISBN 0-8165-1726-6).

Howard Scott Gentry is finally getting his due. This volume is a completely updated, re-edited version of Gentry’s original study of the Rio Mayo, in Sonora, Mexico. Here, in the moist canyons of the Rio Mayo, biologists continue to en- counter the juncture of Neo-Tropical and true Sonoran Desert plant species. The original 1942 release of Rio Mayo Plants was to be the first of many volumes pro- duced by one of the better field biologists of the century. Gentry did not consider himself to be a true botanist, and yet his contributions have endured, and his vol- ume on the Agaves of North America is still the most comprehensive treatise on the subject of these succulents.

This new version is a beautifully produced, almost intimidati of new information on the region and its flora. To be sure, portions of ‘Gentry’ s original prose are present, but only amount to about 40 pages in the new version. The editors of this tome have organized the volume into four sections. Part 1 consists of an overview of the Rio Mayo region, extant literature, and the contem- porary vegetation patterns. A list of localities cited in the study follows in Part 2, and part 3 is a portion of Gentry’s original work in describing the Rio Mayo. The bulk of this book is found in Part 4, which is a massive and useful annotated list of the Rio Mayo’s vascular plants. The editors canerully oon this information, and were cautious with their taxonomic ture; the region begs for further botanical and systematic work.

It is impossible to review a taxonomic list of new plant collections this vast. The contributors to the volume include 27 botanists and biologists, not including the large numbers of collectors involved in the revision of Rio Mayo Plants. Each individual plant citation provides the scientific name, common names in Spanish, as well as indigenous languages of the region (Guarijio, Mayo, and Pima). They also include habitat descriptions, the site of collection, and verbal descriptions of the plants as well as their past and current uses by locals. The first three sections of the book contain photos, tables, as well as illustrations completed by Paul Mirocha. Two maps are provided with the volume: one is a large foldout of the Rio Mayo region, while the smaller encapsulates the vegetation of the region. The annotated list, however, lacks any illustrations. The new volume is rather useless as a key in the field, yet the amount of information contained in the new list of plants justifies the cost of this volume, for true enthusiasts at least. The number of species described, or rather taxa, has more than doubled since the original volume appeared. In addition, the contributors explicate the Rio Mayo’s changing bioge- ography, and how current patterns of land-use are changing the vegetation composition of this area.

The writing is clear, pithy, and fairly uniform. This is an uncommonly good trait for an edited volume with a large number of contributors. The only flaw in the production of this updated Rio Mayo Plants volume is that the press adver-

Summer 2000 JOURNAL OF ETHNOBIOLOGY 73

tises the inclusion of a large color map: my copy was black and white. Under- graduates, graduate students, and professionals interested in biogeography, ethnobotany, and arid lands ecology will want (to use) this encyclopedic work. Most libraries, especially those located in the Southwestern U.S., might find it useful to buy two copies of this volume. Paul Martin and friends have given us a true gem, although as they note themselves, the mountains of southern Sonora beckon for more botanical work. Surely more treasures of the Sierra Madre re- main to be discovered. Had Gentry survived to see the release of this work, he would be pleased.

Eric Perramond Department of Geography & Environmental Science Stetson University, Deland, FL

as Le a4 ts 7 . E j ne ' > ur E ~ . E = Ss 2 proieaias on i - +51 h 2 vies i j ¥ f Jt ha } i * Tee - * _ Lt

Journal of Ethnobiology 20(1): 75-91 Summer 2000

MANAGEMENT OF TREES USED IN MURSIK (FERMENTED MILK) PRODUCTION IN TRANS-NZOIA DISTRICT, KENYA

WILLIAM MUREITHI Department of Forestry, Moi University, P.O. Box 1125, Eldoret Kenya

CHRISTOFFEL DEN BIGGELAAR Department of Interdisciplinary Studies 109-B East Hall Appalachian State University Boone, NC 28608 USA

EDWARD W. WESAKANIA and KURIA KAMAU Kaisagat Environmental Conservation Youth Group P.O. Box 119, Kipsaina via Kitale Kenya

CATHERINE GATUNDU Forest Action Network P.O. Box 21428, Nairobi Kenya

ABSTRACT.- Milk treatment using trees is an age-old practice of both sedentary and nomadic pastoral communities in Kenya. Due to economic, political and environmental pressures, many pastoralists have become settled farmers and turned to crop cultivation as their main means of survival. However, they have continued to keep some cows and to treat their milk using traditional practices, incorporating the desired tree species into their farming system. This paper presents information as to how species are identified and selected, how the trees are managed, management problems associated with the trees, and how farmers evaluate the results of continuing experimentation with trees used for murstk production.

Key words: Fermented milk, mursik, pastoralism, farming, Kenya

RESUMEN.-Tratamiento de leche usando arboles es una antigua practica de comunidades pastoriles sedentario y némadas en Kenya. Por la influencia econémico, politica y ambiental, muchos ganaderos se hacen agricultores y se mantienen con las cultivacién de comida cémo su superviviente. Ademas, ellos siguen manteniendo vacas y tratando la leche con practicos tradicionales, incorporando los arboles necesarios en su sistema de agricultura. Este papel va

76 MUREITHL et al. Vol. 20, No. 1

prrsenies PELE CION et cos los especies se identifican y escojan, c6mo los

as con los arboles, y como los granjeros evaluan los resultas de los experimentos continuos con los arboles en el uso del producci6n mursik.

RESUME.— Le fraitement du lait a - partir de certains — ‘auein la 1 sean lconaniies de

et nomadiques au ‘Kenya. A cause des pressions seed NN politiques et écologiques au cours des années, beaucoup de pastoralists sont devenus des cultivateurs comme moyen principal de survivre. Néanmoins, ils continuent d’élever des vaches laitiéres et d’utiliser leurs pratiques traditionnelles de traitement du lait en incorporant des espéces d’arbres utiliser pour ce but dans mae he = hl perettiens = informations ~ EASES Ee aNen et la

rencontrées, et les méthodes ae par les éleveurs pour évaluer les résultats des leurs expérimentations continuelles avec des espéces d’arbres utilisées pour la production de mursik.

INTRODUCTION

Tree diversity is generally low in farming systems compared to natural eco- systems such as forests. Even so, forest and tree resources provide many benefits and form an important part of the rural household economy. Trees are used in various ways for economic, social and cultural purposes. To a large extent, rural people themselves determine the tree species that grow on their farms and influ- ence each other in terms of what agroforestry practices to adopt or to reject. Extensive discussions with farmers in Trans-Nzoia District, Kenya, revealed that every tree growing on farmers’ land has a role to play in the household economy. However, as some common and widely used tree species are facing extinction in both their natural and human-modified habitats due to population pressures and increasing demand for cultivation land, efforts should be made to document the uses of tree resources so that good cultivation and conservation practices can be developed.

Such is the case with the tree species! used for the preservation of milk for the production of mursik, a traditional technology developed and widely used by various pastoral groups in Kenya. Mursik is the Kalenjin term for fermented milk, but the term is recognized and used by all ethnic groups in the research area. To date, little has been written about traditional milk preservation practices in Kenya. Articles found in most cases provide only very brief descriptions of the technol- ogy. Even the series of district socio-cultural profiles published in the mid-1980s devote only a few paragraphs to describing the technology (e.g., Were and Wanjala 1986; Wanjala and Nyamwaya 1986; and Were and Olenja 1986). Other articles summarize the process of sour or fermented milk production as an introduction to their main topic, the microbiological analysis of fermented milk products (e.g., Miyamoto et al. 1985; Ashenafi 1993; Feresu 1992; Isono et al. 1994; Kassaye et al. 1991; Mutukumira 1995; Mutukumira et al. 1995; and Nakamura et al. 1999). Shalo (1987) provides a generalized description of the pastoral methods of handling and

Summer 2000 JOURNAL OF ETHNOBIOLOGY 77

preserving milk practiced in Kenya, paying particular attention to the initiation and preparation of the milk storage gourds. The technology, however, is not re- stricted to Kenya, but is widely used in other Africa countries as well, for example the Sudan (Abdelgadir et al. 1998), Zimbabwe (Feresu 1992; Mutukumira et al. 1995; Mutukumira 1995), Tanzania (Isono et al. 1994) and Ethiopia (Ashenafi 1994; Kassaye et al. 1991).

More detailed descriptions of the process to produce iria ri matii, a fermented milk produced by the Meru, is provided by Kimonye and Robinson (1991), while the production of mursik is described by an anonymous author in Food Chain (Anon. 1994). The Meru people use charcoal from Olea europaea L. ssp africana (Mill.) P.Green to coat the inside of storage gourds; the attractive flavour/aroma of wood smoke is an essential characteristic of the product. The authors also describe the microbiology of the process involved. Wanjala and Nyamwaya (1986) reported about the production of murskik (sour milk) among the Tugen in Baringo District using charcoal obtained from Euclea divinorum Hiern. The purpose of doing so, according to these authors, is threefold: (1) it preserves the milk for a longer pe- riod; (2) sour milk is a strong and healthy meal in itself; and (3) it gives the milk colour and scent. The charcoal crushed into the gourd keeps it from wearing out fast, and it also erases the natural smell of a gourd when milk is drunk from it. Similar procedures are used by the Pokot and the Ilchamas (Wanjala and Nyamwaya 1986), but no mention is made of the species used for the charcoal to coat the milk storage gourds.

The Maasai in Kajiado District use Olea europaea L. for the treatment of their gourds for the preparation of osaroi (sour milk), which is believed to assist milk in fermentation and “gives it a pleasant flavour enjoyed by the Maasai people” (Were and Wanjala 1986). From the description given by Were and Wanjala (1986), it is unclear whether the Maasai coat the gourds with charcoal as is done by the Tugen, Pokot and Ilchamas, or whether the gourds are only smoked with a burning piece of wood. Miyamoto ef al. 1985 reported on the production of maztwa lala (Kiswahili for sour milk) by the Maasai in Nakuru, Narok and Kajiado Districts of the Rift Valley Province in Kenya. In a brief description on the preparation methods of maziwa lala, Miyamoto et al. state that a gourd is washed with hot water and rubbed with the burnt end of some chopped sticks from a tree known as mutamayio (this tree could not be identified in Beentje (1994) or in other sources). This is done for both flavouring and pasteurizing. An anonymous author writing in the jour- nal Food Chain (1994) about milk preservation by the Kalenjin in Baringo District provided a good description of the process of milk preservation using charcoal, as well as an explanation of how the technique works to preserve the milk. Like the other studies, the author identified only one tree used for this purpose ( ite), but did not provide a scientific name of this species. Ite could refer to Acacia mellifera which is spelled as Iti in Samburu and Oete or Eite in Maa according to Interna- tional Centre for Research in Agroforestry (ICRAF)(1992), or Oiti according to Beentje (1994). According to Ronoh (1987), the methods used to preserve milk by the Maasai, Kalenjins, Boranas, Turkanas, Pokot and Somalis are such that milk can be kept as long as three months.

Riley and Brokensha (1988) briefly described milk preservation practices among

78 MUREITH {et al. Vol. 20, No. 1

the Mbeere, who live on the semi-arid plains south of Mt. Kenya. The Mbeere use smoke to sterilize the gourds used for milk storage, contrary to farmers in the above mentioned studies who use charcoal. Nevertheless, the charcoal may still smoke when crushed in the gourds, thus indirectly providing this sterilizing ser- vice as reported in the article “Mursik - Fermented milk in Kenya” (Anon.,1994). Several tree species are used by the Mbeere to smoke their gourds, but they are different from those used by farmers in the other studies cited above, reflecting the different ecological and biogeographic conditions in each area. Kassaye et al. (1991) also reported on the use of smoke to prepare the storage gourds for the preparation of ititu (or concentrated, fermented milk) by pastoralists in Southern Ethiopia. Pastoralists there use wood from Acacia nilotica (L.) Willd. ex Delile to smoke the gourds. Another study from Ethiopia (Ashenafi 1994) also mentioned the smoking of fermenting vessels with Acacia nilotica wood by pastoralists in ru- ral areas. In the highlands, however, smoking of containers with olive wood is more common according to Ashenafi (1994). Smoking of milk gourds is also used by the Turkana as reported by Galvin (1985).

The above studies indicate that the use of charcoal and smoke from selected tree species is widespread among pastoral people in East Africa. However, the authors (with the exception of Riley and Brokensha 1988) make it appear is if each ethnic group only uses one specific species for the treatment of their milk, which may or may not be the case. In addition, none of these studies report on the tree species themselves, such as where pastoralists obtain the wood, whether the trees used in the process are actively planted and managed, or the specific problems farmers encounter using the technology (in particular related to the trees employed for the process) and ways they have tried to solve them. This article will address some of these gaps. It is based on a study of milk treatment by farmers in Trans Nzoia sian Kenya. Our intention is not to provide a detailed description of the milk ¢ process as such, but to provide more i about the trees associated with the technology and their management.

OBJECTIVES

The integration of woody species with crops and animals is an age-old prac- tice of people throughout the world. The formal study of what is now termed agroforestry, however, started only about 20 years ago. Despite heavy investment in research and extension in these two decades, agroforestry efforts have met mixed success, largely because nae le and ae ouabinea have not paid sufficient at-

seni (if any at all) to farmer. s designed and developed

through their own efforts (den Biggelaar 1996a), The realization that, indepen- dently of formal research and extension, farmers and y country carry out i tati on tree cultivation and t and

share findings with others, led to a major initiative by the Forest, Tree and People Programme (FTPP) to document these informal research and extension practices. A case study format was chosen by FTPP as the best way to study farmers’ experi- mental and information sharing practices and processes, enabling outsiders to understand their underlying rules and logic in different regions around the world.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 79

The objectives of the case studies were two-fold:

1. To document how selected farmers organize experiments and disseminate improved forest and tree management practices. The case studies were geared less at describing specific improved practices, but more at developing an un- derstanding of the “why” and “how” aspects behind the practices.

2. To define the current and potential role (if any) for outside institutions (e.g., NGOs, research, extension, donor-funded projects, universities) to support farmers in the above endeavors.

The study presented in this paper was one of four case studies undertaken in East Africa as part of the global FTPP initiative, and was guided by the above objectives. The specific topic was chosen after preliminary discussions with key informants in Trans-Nzoia District. The informants suggested milk treatment for this study, as it was an innovation developed by pastoral people themselves that is culturally important, widely spread across different ethnic communities in the research area, and still much in use to this day. For example, Kalenjin-speaking people (Kipsigis, Nandi and Tugen among others) believe that milk cannot be con- sumed fresh, but must undergo treatment before consumption. During the initial visit to the area, farmers mentioned several problems that could impede the fu- ture use and further development of the technology. Chief among these problems are the dwindling supplies of certain favoured tree species used for milk treat- ment in spite of having incorporated some of th pecies into the farming systems. However, a more thorough understanding of the technology and the exact nature of the problems would be necessary to determine how outside institutions could help farmers maintain and develop the technology further.

This study investigated the nature of the milk treatment procedures used by farmers in Trans Nzoia District, origins of the technology, and the selection, man- agement, and incorporation of trees used for this purpose within the farming system. We conclude the paper by identifying a number of areas in which research and development could be of assistance to improve and extend the use of this technology as a viable alternative to modern, expansive, capital-intensive milk processing plants.

METHODOLOGY

Data collection and analysis. The study was carried out in a three months period between August and October 1996, and consisted of three stages. Stage one in- volved a reconnaissance tour of the District by the researchers from Moi University together with the area agricultural extension officer and representatives of the Kaisagat Environmental Conservation Youth Group to identify the topic of the case study. During the tour, interviews were conducted with key informants (indi- vidual farmers and farmer groups). The majority of participating farmers were traditional pastoralists who had settled in the District and turned to farming as a survival strategy. They had a highly developed traditional knowledge on the use of tree resources for fodder, medicine (for livestock and human beings), food and milk preservation. Traditionally, these communities did not plant trees since natu-

80 MUREITHL et al. Vol. 20, No. 1

ral regeneration ensured a sufficient supply of trees for various uses. The demand for land for settlement and crop production, however, reduced forest cover and tree species diversity. Nowadays, farmers are actively planting and conserving tree species, especially those which address their cultural and economic needs such as species suitable for milk treatment.

Stage two involved collection of data from individual households. This stage involved direct observation of the milk preservation procedures used by farmers, and the use of structured interview schedules and informal interviews about the process, tree se — and their ar a wae! and problems and constraints

g method from division to location was adopted to ensure ro all ethnic ¢ groups in the two locations were included in the sample. Within the district, two divisions (Cherangani and Kwanza) were chosen randomly; in turn, one location was chosen at random in each of the two divisions. In each location, households of different ethnic groups were chosen based on their per- ceived knowledge and expertise of milk treatment based on information from key informants. Since women are the custodians of indigenous knowledge in the area of food preservation, the sampling procedure was directed towards them. How- ever, both the head of the household and his spouse (if applicable) were

inte d. A total of 60 farmers (36 from Cherangani and 24 from Kwanza) were

interviewed. Semi-structured interview schedules were used to guide the inter- views to assure that similar information was collected from each household. Nevertheless, based on the answers of respondents, additional questions were posed to seek clarifications and additional information. Questions in these inter- views included descriptions of the technology (often including a showing of milk gourds, utensils, prepared branches) to learn how it is milk treatment is done; perceptions of how and why the treatment is effective in preserving milk; choice of species used and their advantages and disadvantages; changes in species used from the past and/or from respondent area of origin, and the reasons for these changes; location(s) where f used in milk treatment; and prob- lems and constraints encountered with the technology in general and the desired species (esp. regarding their multiplication nd management) in particular. A stan- dard form designed by the researchers was used to collect information on each species used and cultivated for the purpose of milk preservation.

Stage three consisted of two community workshops in which the results of the surveys, observations and informal interviews were presented to the community for verification, discussion and further explanation where necessary. Small group discussions using a list of questions were used to further explore key issues re- lated to milk treatment (past, present and future).

Species identification.-In general, species were identified through their vernacular names used by the different ethnic groups in the area. The fact that two of the au- thors (Mr. Wesakania and Mr. Kamau) are also farmers in the study area, manage a small tree nursery for the youth group they lead, and are knowledgeable of the local vegetation greatly facilitated species identification. All but one of the species used for milk preservation are trees common throughout Kenya with which the Kenyan authors were familiar, and for which they knew both the local and scientific names. The exception was Lippia kituiensis (vernacular name is Mwokiot or Mwokyot in

Summer 2000 JOURNAL OF ETHNOBIOLOGY 81

Kipsigis), which was identified through Beentje (1994: 668). The identification of all species, however, was verified through consultations of Beentje (1994), ICRAF (1992), Gachathi (1989) and Teel (1984). Additional botanical information on the species was obtained from the Missouri Botanical Garden’s W? Tropicos VAST nomencla- tural database and authority files on the Internet (MOBOT, n.d.).

RESULTS AND DISCUSSION

Study area.-The study was conducted in Kwanza and Cherangani Divisions of Trans-Nzoia District in Western Kenya. The district covers 2,468 km? and has an elevation averaging 1800 m asl. Most of the rivers in the district are tributaries of the Nzoia River and flow throughout the year. The district has a highland equato- rial type of climate with a fairly well distributed ay aee annual precipitation of 1120 mm, and an average mean temperature of 18.6°C.

The District Development Plan (DDP) of 1994 estimated the district’s popula- tion at 462,748. Although no concrete data exist as to the exact numbers, there has been a steady in-flow of migrants from different parts of Kenya in the last 30 years (DDP 1994). This was confirmed by the fact that 98% of the farmers interviewed migrated to the district in the last 32 years. The main attraction for migrants into the area was the availability of land for settlement and a favorable climate for both cultivation and livestock.

The tenure system in the area has changed over the years. Initially, much of the land belonged to large individual farmers, mostly British settlers. After Kenya’s Independence in 1963, these settler farms were sold as group farms to individual small holders, or were expropriated by the government for settlement schemes and co-operative and corporate farms owned by the Agricultural Development Corporation and the Kenya Seed Company, among others. These large farms have in the recent past been subdivided and given out to individual (mainly small- holder) farmers, although a large number of these are not registered and farmers do not have title deeds to their land.

The total area under forest in the district is 50,292 hectares, but immigration and settlement have led to serious deforestation in many parts of the district. The incorporation of trees and shrubs in the farming systems has partly mitigated the loss of natural forests. Agroforestry has become one of the mua production activities in the district, supplying the bulk of the about 500,000 m° of fuel wood consumed annually (DDP 1994). Demand for fuel wood is bound to increase due to an increasing population. Through the combined efforts of the Forest Depart- ment and NGOs, over four million seedlings are produced and supplied to farmers annually to meet the growing demand for fuel wood and other tree products. The survival rate has, however, been very low, not exceeding 30% each year according to Forest Department and extension personnel. One of the reasons for the low survival rate is that seedlings provided are mostly exotics with no cultural impor- tance, leading to a lack of proper care to ensure survival.

Dairy farming is a major economic activity and constitutes a large Proportion of income for both the small and large-scale farmers. Farmers experience prob- lems in the marketing of their milk due to lack of storage facilities and a poor

82 MUREITH et al. Vol. 20, No. 1

transportation system. The dairy market is poorly developed in the district, with Kenya Cooperative Creameries (KCC) being the only commercial buyer of milk produced in the area; delayed payments for milk delivered to KCC exacerbated the problems experienced by farmers during much of 1995 and 1996. The need to treat and preserve milk at the farm level has, therefore, increased in importance despite the availability of modern milk processing facilities in Kenya.

Problems that led to uses of trees in milk treatment.-The use of trees in milk treatment is a common practice among farmers in Trans-Nzoia District. The technology has been practised for a long time by pastoral communities, and non-pastoralists (e.g., the Kikuyu) have adopted this practice from them. The farmers identified the fol- lowing problems and conditions leading to their experimentation with milk treatment:

¢ Milk is a major source of food for pastoral people, but many sipruispetigse = sen tose intolerant. Fermentation is, therefore, necessary to improve the of milk.

Z

The odour, taste, and flavour of fresh milk are not pleasing (farmers are of the opinion that fresh milk smells and tastes like cow urine) and need to be im- proved before it can be consumed. The charcoal helps to neutralize the undesirable odour and taste of the milk. Shalo (1987) reported that the finely divided charcoal inside the gourds has a wide surface area, and hence is very active absorbent of flavours in milk. The charcoal itself also imparts flavour to the milk (Shalo, 1987)

The white colour of milk is not acceptable to farmers, who accord a high aes- thetic value to the bluish-grey colour imparted to the milk by the charcoal. Similar findings were reported by Anon. (1994) and Shalo (1987).

Lack of markets and refrigeration, and the need to store milk for the dry sea- son (when milk production decreases due to a lack of pasture) required that excess milk be stored for longer time. There was, therefore, a need to preserve the milk and improve its shelf-life. For example, the Pokot developed chekha mwaka, a specially treated milk that could be stored for over one year with- out spoiling (findings from interviews for this study). This is much longer than the up to 3 month shelf-life reported by Ronoh (1987) earlier.

Gourds were and still are the best storage facilities to the farmers; they are cheap and easily available. However, gourds give milk stored in them a bitter taste. Moreover, gourds are hard to clean and produce a bad smell that affects the milk stored in them. Thus, treatment of milk was also necessary to neu- tralize the bad smell and bitter taste of the gourds.

Table 1 summarizes the reasons cited by respondents of why they treat their milk. Enhancement of taste and flavour was cited by 86% of the respondents as a reason for treating milk, although it was more important to farmers in Cherangani (92%) than to farmers in Kwanza (79%). The second reason cited was palatability (in the sense of being merely agreeable (WWWebster Dictionary, 2000) (75%), fol- lowed closely by the necessity of preserving the milk (cited by 73%). Again, there

Summer 2000 JOURNAL OF ETHNOBIOLOGY 83

are differences between the two Divisions, with preservation being more impor- tant to Cherangani farmers (83%) and palatability being more important to Kwanza farmers (79%). The differences between the two Divisions are largely due to the ethnic affiliations of the respondents. The medicinal value of treated milk was cited by 18% of respondents, but was more important in Kwanza (25%) than in Cherangani (14%).

The medicinal properties attributed to treated milk are imparted to it through the use of tree species that are known to cure diseases in animals or people. Me- dicinal tree species are the most frequently used species for milk treatment, as will be shown later. New species are often identified through already known medici- nal use. We also surmise that fermented milk may be beneficial to establishing a healthy intestinal flora, which is similar to the use of yoghurt for this purpose in Western cultures. Shalo (1987) mentioned that charcoal, being a material interme- diate between wood and ash, contains minerals that will benefit the nutritional value of milk which may, unintentionally, contribute to the overall health of its consumers.

Milk treatment procedure.— The first step in the treatment process involves the prepa- ration of the milk storage gourds. Women use the stalk (central nerve) of palm leaves to scrape the inside of the gourds to remove old fat and milk solids that cannot be removed by washing only. A few days before milk is to be treated, a small branch of a selected tree species, about the thickness of a thumb, is debarked and left to dry. One end of the dried piece of wood is then put in the fire to burn. When the end is completely burnt, it is gently crushed on the side and bottom of a cleaned milk storage gourd to crush it into charcoal dust. This procedure is re- peated several times until the gourd is completely coated on the inside. The excess dust is removed, and the gourd is then ready for milk storage. The quality of mursik (treated milk) obtained through this process is evaluated using the criteria of colour, smell and taste of the final product. Farmers claimed that their traditionally treated milk was superior to untreated milk or factory-processed milk from KCC.

A question in the community workshops about when the process was first used led to much lively discussion. Tracing oral histories and using age groups, participants agreed that the earliest reference would place its development about 300 years in the past. As there are no written documents to verify this, we ac-

TABLE 1.- Problems and conditions underlying the need for milk treatment. ie bl is Sot derives 2 cloacae ee ee % of respondents

Condition/problem Total (n=60) Cherangani (n=36) Kwanza (n=24) 92 ail

Taste / flavor 86

Palatability 75 83 58 Preservation 73 72 79 Smell 65 67 63 Color 55 14 25 Medicinal 11 50 63

NB: Multiple responses were allowed.

84 MUREITHI et al. Vol. 20, No. 1

cepted this as a reasonable date although the technology could in fact be much older. According to the farmers present at the discussion, little has changed in the treatment procedures over time, although the species used have changed espe- cially after people migrated to new areas where old, favoured trees could not be foun

Tree identification and selection procedure. - According to oral histories collected dur- ing the exercise, the id f the tree species and their parts to be used in milk treatment was a systematic exercise. It involved the participation of both men and women, but the role of men varied between communities. For example, among the Kikuyus (a traditional farming commu- nity), the men are involved in both identification of suitable species and the treatment of the milk. In traditional pastoral communities (for example, Kalenjin and Pokot), both men and women are involved in identification of suitable tree species, but only women would do the actual milk treatment. It should be noted, though, that over time the role of men in the exercise has gradually declined in these communities; women are now solely responsible for the development, imple- mentation and maintenance of the technology.

Not all tree species are suitable for milk treatment. The choice of species and the tree parts to be used are based on: (1) the availability of the species from natu- ral forests or tree patches, and (2) prior knowledge about the species, particularly their use for medicinal purposes for both livestock and human beings, or fodder for livestock. The selection process of potential milk treatment tree species involves smelling the leaves of the tree and/or the smoke produced when burning a branch of candidate species. It can also involve the chewing of specific tree parts such as leaves, stem, and bark. Species that produce a pleasant smell and have a good taste, are easy to burn and produce a porous charcoal to facilitate its crushing inside the gourds will be tried on milk on an experimental basis. Whether the tree would be adopted for the purpose of treating milk depended on the quality of the initial treatment trial; evaluation criteria to judge the results of the trial empha- sized the quality of the treated milk in terms of taste, colour, smell and shelf-life.

Presently, farmers use the following tree species for milk treatment (in order of preference obtained through a ranking procedure): Senna didymobotrya (Fresen.) Irwin&Barneby (syn. Cassia didymobotrya Fres.), Lippia kituiensis Vatke, Prunus africana (Hook.f.) Kalkm., Olea europaea L. ssp africana (Mill.) P.Green, Croton macrostachyus Del., Olea capensis L., and Rhus natalensis Krauss. Several other spe- cies are used as well, but by few farmers (Table 2); they include Euclea divinorum Hiern., Dombeya torrida (J.F.Gmel.) P.Bamps ssp torrida (syn. D. goetzenii K. Schum.), Bridelia micrantha (Hochst.) Baill., and Acacia gerrardi Benth. A few exotic species have been experimented on by women in the recent past (e.g. Acacia mearnsii DeWild. and Eucalyptus spp). A. mearnsii (black wattle) was found to have side effects on men, namely the blockage of the urinary tract, while Eucalyptus spp were found to be ineffective for the purpose and giving the milk a bad taste (which some farmers described to be similar to cold medicine). These species have there- fore been abandoned for use in milk treatment by the communities who tried them (e.g., the Kalenjin). Most of the above species are grown on the farm, but some can be found only in the dwindling natural forests of the district.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 85

Domestication and management of tree species used in milk treatment. —- Generally, where farmers have chosen to voluntarily grow trees on their farms, the species planted are carefully selected. In the study area, farmers’ choice of species varied accord- ing to gender, with men favouring species that provide income (from the sale of charcoal, poles, posts and timber) and women favouring species used in milk treat- ment or meeting fuelwood needs. It was found that among women, trees which could be used for milk treatment (e.g., Senna didymobotya) were preferred over species with a narrow fuelwood focus. However, women considered fuelwood of such species as a valuable secondary product. Because mursik has a special cul- tural meaning for women and is important to the welfare of the family, trees that were used to provide milk treatment ingredients were better taken care of by the women than trees providing other products. Species used for milk treatment are nowadays managed primarily by women, but all the members of the household will respect and take care of the trees. Men appreciate the important role of these species in the household economy and will (and can) not cut them for other pur- poses.

Amon ities where milk is a traditional and main source of food (such as the sedentary pastoralists) and where immediate storage and marketing facili- ties are lacking, trees used for milk treatment are not used for fuelwood or any other purpose. They are planted and managed solely for the purpose of milk treat-

TABLE 2.— Domestication of trees used for milk treatment.

Species Households Households Off-farm Other uses using (n) cultivating (n) source (n)

Senna didymobotrya (Fresen.)

Irwin & Barne

Cassia didymobotrya Fres.) 36 39 Fw, M Lippia kituiensis Vatke 34 34 Neighbor 1 Fw, M

Olea europaea L. ssp africana (Mill.)

P. Green 27 3 Forest 6 Fw Olea capensis L. 22 6 Forest - Fw Prunus africana (Hook.f.) Kalkm. 18 13 Fw, M, O, P Rhus natalensis Krauss 10 Forest 4 Fw, P Euclea divinorum Hiern. 3 11 Forest 4 Fw, P Dombeya torrida (J.F. Gmel.) P. Bamps ssp. torrida (syn. D. goetzenii K.Schum.) 9 5 Neighbor 1 Fw, O, P,R Acacia gerrardi Benth. 1 - Fw, P Acacia mearnsii DeWild. 5 6 2 Fw, P Eucalyptus spp 5 yg Fw, P, T Bridelia micrantha (Hochst.) Baill. 3 13 Fw, P 5 28 Fw, O, P

Croton macrostachyus Del.

Abbreviations: Uses: Fw=Fuelwood P= Poles, posts M = Medicinal R = Ropes O=Ornamentals T= Timber

86 MUREITHI et al. Vol. 20, No. 1

ment. In some communities, it is also taboo to cross-over or step on pieces of wood cut for the purpose of milk treatment.

The trees commonly used in milk treatment were found either scattered in the fields or growing around the homestead. There was a good relationship between the three most preferred species as identified in the preference ranking exercise and the presence of these species on the farm (Table 2). Olea spp. were widely used for milk treatment and demand for the species was high, but few households cul- tivated them because of their difficult r and t. On the other hand, species like Eucalyptus spp and Acacia mearnsii ranked very low for milk treatment, but were very common on the farm as they are easy to propagate and manage. The latter species were managed by men, who planted these fast-grow- ing species for income generation through the sale of fencing posts, small timber and charcoal.

In the two divisions, 44 percent of the farmers cultivated all the species they

use for milk treatment within their farms; 39 percent cultivated some but not all the species used. The remainder of the farmers (17%) did not grow any milk treat- ment species on their farms, but instead collected them from common property resources. Senna didymobotrya and eee meters were the most popular species used for milk treatment. Sixty percent of tl had planted Senna in their homesteads, while fifty-seven percent maintained Lippia kituiensis in thete fields. Prunus africana is another important species used for milk treatment across communities; over 20 percent of the farmers interviewed are cultivating it on their farms. Senna didymobotrya was planted primarily around the homesteads to enable women to: (1) protect it from browsing livestock; (2) lop some branches whenever they prepared mursik without having to walk long distances; and (3) minimize interference of the trees with other crops and farming activities including mecha- nization. The management techniques to maintain the species included protection against livestock, planting using seeds (for all species except Lippia kituiensis), and pollarding and thinning to control overcrowding and to stimulate the trees to grow many small branches for use in milk treatment.

Constraints.— Several traint identified regarding seed collection and stor- age, breakage of seed dormancy, and the management of trees. According to the farmers, many indigenous species do not produce seeds, making their propaga- tion difficult. Those that do produce seed may take a long time to reach maturity, as is the case with Olea capensis and O. europaea ssp. africana. Most farmers stated that milk treated with Olea apecies tasted the best but they had to look for alterna- tive species because few trees of ere remaining in the area. AGikuyu farmer explained how her husband had cecil a piece of olive wood from Kiambu District (over 600 km from Cherangani Division) and how she wisely used that piece for nearly one year in treating her milk. Farmers who do have a few olive trees left on their land carefully guard them and do not allow further cutting. The remaining specimens, often located on grazing fields or near river banks, are care- fully protected by the population.

A second problem is access to, and availability of, seedlings of useful and pre- ferred species as they are rarely found in village tree nurseries or the nurseries of

Summer 2000 JOURNAL OF ETHNOBIOLOGY 87

the Forest Department and NGOs. The farmers felt that many times their species of priority were not available in the local nurseries, leading them to plant trees about which they had very little local knowledge. This may be one of the reasons why the survival rate of seedlings was very low.

Farmers’ research activities on milk treatment.— At the time of the study, farmers stated that they were experimenting with various technologies to solve problems and constraints encountered with the cultivation of species used for milk treatment (as mentioned in the previous section). For example, some species well-liked for milk treatment such as Olea europaea ssp. africana, O. capensis and Rhus natalensis have become extinct on many farms (see Table 2 for the number of households growing these species) and have become rare in the neighbouring natural forest of Mount Elgon. Farmers continue their investigations of the multiplication of indigenous trees which are difficult to propagate due to lack of seeds and/or germination problems, for example Lippia kituiensis and Olea spp. Some farmers tried to plant cuttings and wildlings (i.e., seedlings collected from natural forests in the area) of Olea spp and Rhus natalensis, but without much success; however, they continue trying to augment the number of these species on their farms.

L. kituiensis is difficult to propagate | the seeds are very small and there- fore difficult to collect. Farmers have also experienced problems getting the seed to germinate; according to them, the species does not germinate and grow on crop land but only in pastures. It was surmised that this could be caused either by the breakage of seed dormancy when seeds pass through the digestive system of ani- mals, or by the seeds being buried too deeply in cultivated fields so that they would not germinate. Research attention to solve this problem would be much appreciated by farmers, as it was presently a much favoured species for milk treat- ment.

In addition, farmers are engaged in a continuous search for new medicinal plants both for human beings and livestock because the prices of drugs have esca- lated with a simultaneous decline in the quality of veterinary and health services. Eventually, some of these medicinal species may be tried and used for milk treat- ment, as many of the species presently used for milk treatment were identified through their prior use as medicinals.

Information sharing. Although farmers were (and some still are) actively engaged in experimentation on the use and management of trees, they are not sharing find- ings in any formal, organized methods of communicating information between or within different communities. These findings are similar to findings from research by den Biggelaar (1996) and Sperling et al. (1993) in Rwanda. Farmers and com- munities do learn from each other through informal channels such as observation of each other’s practices on the farm and discussions at social gatherings and cer- emonies. For example, farmers influence each other in the choice of tree species to plant through these informal ways of sharing information. Mostly, the informa- tion shared between communities concerned species with common economic uses such as fuel wood or milk treatment. Information on other, non-economic uses (for example, on medicinals), was rarely shared, as such uses were more tied up in cultural and social believe systems.

88 MUREITHL et al. Vol. 20, No. 1

CONCLUSIONS

Based on oral history recollections by farmers, the procedures for the treat- ment of milk have not changed much during the last 300 years, and, besides a change in species used over time, little innovation has taken place to improve them. Milk treatment remains largely a trial and error affair, in which the result cannot be obtained until the milk is ready for drinking, which is expensive in terms of time, money and milk (sometimes milk is lost because of adverse effects or unsuccessful treatment). The technology, therefore, appears to be ‘static’: There is either no need or opportunity for improving the process, or farmers depend on outside assistance for the further development of the technology (in the form of material, advice or ideas) which is not forthcoming. Innovations in milk treatment that did occur relate largely to the search for alternative species (which was some- times successful) to treat milk because favored trees became scarce in the environment, or because species encountered in Trans-Nzoia were different from those found in areas where farmers migrated from. The main conclusion from this case study (which was supported by the findings of the other studies in East Af- rica; e.g. den Biggelaar 1996a; Aluma et al. 1996; Njoka and Makenzie 1996) is that knowledge generation is the rule (each time milk is treated is different and a learn- ing experience from which new knowledge is obtained), but innovation is the exception. Networking with other pastoral people both within and outside Kenya (among communities not represented in Trans-Nzoia District such as the Maasai, Somali, Turkana and Samburu; Karamajong and Bayankole in Uganda; and Peul, Touareg, Fulani, Tutsi and Somali and other pastoralists in other parts of Africa) is recommended as one means to improve the technology. It may lead to new ideas, methods and species to be used for milk treatment by encouraging horizontal learn- ing and communication among communities around a theme of common interest.

The goal of FAO’s Farmer-initiated Research and Extension Practices initia- tive was to study and document local (agro)forestry knowledge and technologies with local people and communities in order to facilitate the identification, imple- mentation and evaluation of people’s own priorities for tree growing. Following findings of Scherr (1992, 1993) and van der Ploeg (1991), the initiative aimed at generating more reliable research and extension agendas than top-down ap- proaches, assuring that technologies that research does develop are more client-oriented, have greater local relevance, and are better grounded in local dy- namics of socioeconomic development. In view of the recommendations of Scherr and van der Ploeg, and in according to Objective 2, this study identified several opportunities for research and extension to collaborate with farmers in three areas which farmers themselves considered key to the continued use and further devel- opment of milk treatment: (1) collaboration to solve problems of propagation of favored tree species (i.e., 7 BEE dormancy, low — or =o sara aga avail- ability of seed); (2) i e to farmers to raise their g of these species to ensure a long-term, sustainable supply; and (3) investigations into possibilities for expanding the scale of mursik production and marketing (i.e., studies looking into the demand and supply situation, mursik quality and storage ability in rela- tion to the different species used to treat the milk, the potential markets for mursik

Summer 2000 JOURNAL OF ETHNOBIOLOGY 89

within and without Trans-Nzoia District, and consumer acceptance of milk not treated by someone from their own ethnic community). The collaboration between farmers and researchers i in solving these issues will not only increase the relevance of ongoing research and +}

it will empower, legitimize and enhance the existing endogenous capacities for identifying problems and developing solutions (den Biggelaar 1991).

One issue that is not resolved is the dissemination of innovations and tech- nologies. Farmer-to-farmer extension, even though highly touted in the literature as a promising and cheap(er) alternative to formal extension services, was foun to be all but absent in each of the four East Africa studies of the FTPP initiative (Aluma et al. 1996; den Biggelaar 1994, 1996a; Mureithi 1996; Njoka and Makenzie 1996). The absence of information and technology sharing was also observed by Sperling and Loevinsohn (1991) 3 in Rwanda, and den Biggelaar (1996b) in Kenya. In addition, formal extension activities in the Kwanza and Cherangani Divisions were virtually non-existent, while agricultural and livestock extension activities left a lot to be desired. More investigations and discussions on the role, function and form of both formal and informal dissemination practices is neces- sary to determine the best ways of informing people of new technologies and innovations, and to share research results.

NOTES

! Not all trees identified for use in milk treatment would be considered trees in the Western world view. In Bantu philosophy, however, there is ony a iburictions i trees (all plants that are not grasses) and grasses (all plants th 958). As the definition of Kagame better reflects farmers’ conception of reality, we ‘have cat it in

this study.

A previous version of this paper was presented at the 21st Annual Conference of the Soci- ety of Ethnobiology, Reno, Nevada, April 15-18, 1998

ACKNOWLEDGMENTS

> er operation of man

The success of this study depended on the efforts, people. Among them are the over sixty farmers who spared their most Valuable time to participate in the study and the community workshops. Thanks to them all. The study could not have been possible without the support and cooperation of Dr. James Kiyiapi, Head, Department of Forestry, Moi University. Special thanks go to the Forest Action Network, in particular Mr. Dominic Walubengo (its Director) for the facilitation and coordination of the “Farmer-initiated Research and Extension Practices” -initiative in East Africa. We are also very grateful for the financial support Ae oy aioe through its

lobal Farme

T

Dp

Forest, Tree and J initiated Research aed Extension Practices initiative.

90 MUREITH et al.

Vol. 20, No. 1

LITERATURE CITED

ABDELGADIR, W.S., T.K. AHMED and H.A. DIRAR. 1998. The traditional fermented milk products of the Sudan: a review. International Ee di of Food Microbiology 44 (1-2): 1

ALUMA, JOHN R.W., WOHN R.S. KABOGOZZA and SUSAN MUBBALA. 1996. A Case Study of TOKA Farmers Tree-related Innovations, Iganga District, Uganda. Forest, Trees and People Programme, Kampala, Uganda.

ANON. 1994. Mursik - fermented milk in Kenya. Food Chain No. 13: 9

ASHENAFI, M. 1994. Fate of Listeria monocytogenes during the souring of Ergo, a traditional ethiopian fermented milk. Journal of Dairy Science 77: 696- 702

BEENTIJE, H. 1994. Kenya Trees, Shrubs and Lianas. National Museums of Kenya, Nairobi, Kenya.

DEN BIGGELAAR, CHRISTOFFEL. 1996a. Farmer Experimentation an Innovation: A Case Study of Knowledge Generation Processes in Agroforestry Systems in Rwanda. Communit Forestry Case Study No. 12, Nancy Hart (editor). Food ig Agriculture oe Rom

1996b. iinianes Actors in the Agricultural / Agroforestry Knowledge System in Embu District: Developing a Collaborative Model for the Promotion of Agroforestry Technologies for Sustainable Development in Humid and

103. ane elaeio Center for Research in Agrofor ee 1994 ‘Farmer Knowledge and Experimentatio n with Trees and Tree Cultivation in oa Systems in Rwanda. Dissertation Department of See Michigan State University, East Lansing.

——_—— Farming systems development: synthesizing Se eeag and scientific knowledge system Agriculture and Human Values 8 (1- 2).

25-

DISTRICT DEVELOPMENT PLAN (DDP) 1994-1996, Trans-Nzoia. Ministry of Planning and National Development. Nairobi, Kenya.

FERESU, S. 1992. Fermented milk products in Zimbabwe. Pp. 80-85 in Applications of Biotechnology to Traditional Fermented Foods, S. Feresu. National Academy Press, Washington, ;

GACHATHI, EN. 1989. Kikuyu Botanical Dictionary of Plant Names and Uses. EN. Gachathi, Nairobi.

GALVIN, K. 1985. Food Procurement, Diet, Activities and Nutrition of Ngisonyoka Turkana Pastoralists in an Ecological and Social Context. Ph.D. Dissertation. Department of Anthropology, State University of New York, Binghamton.

INTERNATIONAL CENTER FOR RESEARCH IN AGROFORESTRY (ICRAF). 1992. A Selection of Useful bea and Shrubs for Kenya. ICRAF,

robi.

Boos Y., I. SHINGU and S. SHIMIZU. 1994. Identification and characteristics of lactic acid bacteria isolated from Masai fermented milk in Northern Tanzania. Bioscience, er ane and Biotechnology 58 (4): 660-664.

KAGAME, ALEXIS. 1958. ‘La philosophie bantu-rwandaise de |’étre. Classe des Sciences Morales et Politiques, Nouvelle série In-8 (VI) 1957-58. Académie Royale des Sciences Coloniales, Bruxelles.

KASSAYE, T., B.K. SIMPSON, J.P. SMITH and C.B. O’CONNOR. 1991. Chemical and microbiological characteristics of Ititu. Milchwissenschaft 46 (10): 649-652.

KIMONYE, J.M. and R.K. ROBINSON. 1991. Iria ri Matii - a traditional fermented milk from Kenya. Dairy Industries International 56 (2): 34-35.

MISSOURI BOTANICAL GARDENS (MOBOT). n.d. W*Tropicos, sero poe Internet URL: piss mobot.mobot.org/> (11/9/ ar

MIYAMOTO, K SIMON GG GICHURU, TORU AKIMOTO AND TOSHITAKA NAK 1985. Identification and Reon of lactic acid bacteria isolated from peerren fermented beverages in East Afric Japanese Journa eee Science 57 (3): 265-276.

Summer 2000

MUREITHI, WILLIAM. 1996. Milk

Trees and People Programme Kenya, Nairobi.

MUTUKUMIRA, A.N., J.A. NARVUS and R.K. ABRAHAMSEN. 1995. Review of traditionally-fermented milk in some sub-Saharan countries: focusing on Zimbabwe. Cultured Dairy products Journal 30 (1): 6-8, 10-11.

MUTUKUMIRA, A.N. 1995. Properties of amasi, a natural fermented milk

ae siete by

a milk producer mbabwe. Miele nseliate 50 (4): 301.205,

NAKAMURA, A.N.

OZAWA, H. ARIGA, rE. KOAZE, ae KITYUKIA, I. ARAT and T. URASHIMA. 1999. Sperm oe properties of Maziwa Lala, a Kenyan traditional fermented nil of Masai apres in Kenya. Milk Science 48 (1): 9-13. NJOKA, ELLIS and PAUL MAKENZIE. iy A Case of a pupae s

Kik apu, " Njoro Location, Naku Kenya. Forest Action Network - male Trees and People Programme Kenya,

obi. RILEY, BERNARD W. and DAVID BROKENSHA. 1988. The Mbeere in Kenya, Volume 1: Changing Rural Ecology. University Press of America,

Lanham.

RONOH, ALICEN. 1987. The role of women in milk production with special

asis on practical spplications.

Bulletin of the International Dairy Federation No. 221: 127-129.

SCHERR, SARA J. 1993. Meeting household needs: farmer tree-growing strategies in western Kenya. In Trees in Farmers’ Strategies: Responses to Agricultural Intensification, J.E.M. Arnold and P. Dewees (editors). Oxford University Press, Oxford..

JOURNAL OF ETHNOBIOLOGY 91

—_——_—. 1992. The role of extension in

agroforestry development: evidenc ern Kenya. Aaraterastess Systems 18: 47-68.

SHALO, PL. 1987. Pastoral method of handling and preserving milk. Bulletin of the International Dairy Federation No. 221: 110-112.

SPERLING, LOUISE and MICHAEL E. LOEVINSOHN. 1991. The Dynamics of Improved Bean Varieties Among Small Farmers in Rwanda. Unpublished mimeogra

d BEATRICE

- NTABOMVURA. 1993. Rethinking the

omer s role im sages breeding: in

Siege i vena aiierer Aye litite 29(4): 509-519.

TEEL, WAYNE. 1984. A Pocket ey of Trees and Seeds in Kenya. KENG Nairobi, Kenya.

VAN DER PLOEG, J.D. 1991. Landbouw als

NYAMWAYA (editors). 1986. Kenya Socio-Cultural Profiles: Baringo District.

African Studies, University of Nairobi,

airobi. WERE, GIDEON S. and J. OLENJA (editors). 1986. Narok District Socio- Cultural Profile. Nairobi: Ministry of

a

and The Institute of African Studies, aan of Nairobi.

eons HRIS L. WANJALA feos), re Kajiado District Socio- Cultural Profile. Ministry of Planning and National Development and The Institute of African Studies, University

of Nairobi, Nairobi WWWeebster Dictionary. 2000. Internet URL: <http://www.m-w.com/netdict.htm> (11/5/99)

© ee

Journal of Ethnobiology 20(1): 93-109 Summer 2000

FAUNISTIC RESOURCES USED AS MEDICINES BY ARTISANAL FISHERMEN FROM SIRIBINHA BEACH, STATE OF BAHIA, BRAZIL’

ERALDO M. COSTA-NETO AND JOSE GERALDO W. MARQUES epartamento de Ciéncias Bioldgicas Universidade Estadual de Feira de Santana, Km 3, BR 116, Campus Universitario, CEP 44031-460, Feira de Santana, Bahia, Brasil

ABSTRACT.- Artisanal fishermen from Siribinha Beach in the State of Bahia, Northeastern Brazil, have been using several marine/estuarine animal resources

fishes over other aquatic animals. Asthma, bronchitis, stroke, and wounds are the most usual illnesses treated by animal-based medicines. These results corroborate Marques’ zootherapeutic universality hypothesis. According to him, all human cultures that present a developed medical system do use animals as medicines. Further studies are requested in order to estimate the existence of bioactive compounds of pharmacological value in these bioresources.

Key words: Fishermen, marine resources, medicine, Bahia, Brazil

RESUMO.-Pescadores artesanais da Praia de Siribinha, estado da Bahia, Nordeste do Brasil, utilizam varios recursos bispsnpmn Seguane ae estuarinos como remédios populares . Registra , peixes, répteis e cetéceos. Observou-se 1 uma alta one ar de peixes sobre outros animais aquaticos. Asma, saree einen: e és sao as nteegeee mais usualmente tratadas

a hipdtese da éutica de Marques. De acordo com ele, toda cultura humana que apresenta um sistema médico desenvolvido utiliza-se de animais como remédios. Estudos posteriores sao necessarios a fim de avaliar a existéncia de compostos bioativos de valor farmacolégico nesses biorrecursos.

RESUME.- Les pécheurs seek de la plage de Siribinha dans l'état de Bahia, au nord-est du Brésil, utilisen ressources estuarines en tant que médecine nese Nous avons enregistré l’emploi de mollusques, de crustacés, d’échinodermes, de poissons, de reptiles et de cétacés. Nous avons noté une prédominance élevée de poissons par rapport aux autres animaux aquatiques. L’asthme, la bronchite, les attaques, et les blessures sont les maux les plus habituellement traités par les remédes 4 composante animale. Ces résultats corroborent I’hypothése d’universalité zoothérapeutique de Marques. Selon lui,

l’existence de composés bioactifs 4 valeur pharmacologique dans ces resources biologiques.

94 COSTA-NETO & MARQUES Vol. 20, No. 1

INTRODUCTION

“Naturally, fish fauna provides remedies, amulets, spurs, eyes, parts of the mandible, fins, fats, muscles, which take part in the current sea-shore folk medicine (...)”.

(Cascudo, 1972: 704)

Zootherapy is the healing of human diseases by using therapeutics that are obtained from animals, or ultimately are derived from them. As Marques (1994) states, “all human cultures that present a structured medical system utilize ani- mals as medicines.” Such a statement forms the basis of his ‘zootherapeutic universality hypothesis.’ Indeed, animals are therapeutic arsenals that have been playing significant roles in the healing processes, magic rituals, and religious prac- tices of peoples from the five continents.

The medicinal interaction between humans and animals has been shown both in indigenous and Western societies all over the world (Gudger 1925; Conconi and Pino 1988; Antonio 1994; Marques 1995; van Huis 1996; Costa-Neto 1996, 1999a, 1999b). As some authors have pointed out, animal-based medicines have been utilized since antiquity (Weiss 1947; Angeletti et al. 1992; Rosner 1992), where popu- lar remedies were elaborated from parts of the animal body, from products of its metabolism (corporal secretions and excrements), or from non-animal materials (nests and cocoons). An early record for animal-based medicine can be found in Tobias’ Book (Catholic Bible), in which Raphael the Angel would have prescribed the use of a fish’s liver content for the treatment of ophthalmic problems (Marques 1995).

The phenomenon of zootherapy has recently aroused the interest of many re- searchers from different branches of science, who have recorded this unusual cultural practice and sought for compounds with pharmacological action (Werner 1970; But et al. 1991; Bisset 1991; Faulkner 1992; Lazarus and Atilla 1993). This interest increases when it is considered that the annual global trade in animal- based medicinal products accounts for billions of dollars per year (Kunin and Lawton 1996). These authors have recorded that the investigation of folk medi- cines has proven a valuable tool in the developing art of bioprospecting for pharmaceutical compounds. Today from 252 essential chemicals that have been selected by the World Health Organization, 11.1% have plant origins, while 8.7% come from animals (Marques 1997).

In Brazil, an amazing number of about 300 animal species have been medici- nally used. These resources can be easily found as commercial items sold by herbalists and curers in market places all over the country (Marques, personal communication 1996). In relation to marine / estuarine animals used as medicines, Marques (1995) has already recorded a total of 66 fish species in the folk medicines of fishing communities from 13 Brazilian states. According to him, the medicinal use of fish seems to be a very usual pattern in fishermen communities. Begossi (1992) has found that fish resources valued as medicines are usually considered taboos as food, perhaps so that they may be available as folk medicines (drugstore hypothesis).

Summer 2000 JOURNAL OF ETHNOBIOLOGY 95

Unfortunately, many of the zootherapeutic resources include threatened spe- cies (IBAMA 1989). In fact, the diminishing number of fauna species, especially from neotropical areas, through hunting, depauperation of their ecosystems, and their varied uses has been enormous that most of them are becoming extinct even before they have been studied by science (Huxtable 1992). Hence, studies aimed towards traditional knowledge on animal use and its significance to men should be undertaken in order to lead to better ways of exploiting the natural resources, thus, their conservation, so that future generations may know and manage them.

This paper provides an overview of the phenomenon by illustrating 39 zootherapeutic species that are prescribed by artisanal fishermen from Siribinha Beach, which is in the city of Conde, in the state of Bahia, Northeastern Brazil. Subsequent research needs to be done not only to confirm the presence of sub- stances of medicinal value in these traditional remedies, but also to lead to a more ecologically sound exploration.

METHODOLOGY

Conde is a coastal city which is in the north region of the state of Bahia, north- eastern Brazil (Figure 1). This region presents a humid/sub-humid climate, a mean temperature of 24.5°C, a mean annual rainfall of 1412 mm, and vegetation that is characteristic of tropical coastal areas (CEI 1994). Siribinha fishing community was chosen as the study area due to its localization, relative isolation, biological and ecosystem diversity, the lack of bibliographical knowledge about it, as well as the degree of both social and environmental impact to which it is submitted.

Fieldwork was performed from March 1996 to March 1998. Cultural data on zootherapy were obtained through tape-recorded, open-ended interviews carried out with 54 informants, both male and female. The informants were questioned about zootherapeutic species, the raw materials used, modes of elaboration and administration of the folk remedies, as well as the diseases for which the folk rem- edies are prescribed. Native words were used in order to generate a confidence relationship between interviewer and interviewees. The interviewees were asked whether recording the conversations and taking photographs were permitted.

The medicinal use-value of each animal were estimated according to the fol- lowing equation:

UV =(2RM x C)/N,

UV represents the medicinal use value, RM refers to the total number of raw materials extracted from individual animals, C refers to the number of times which a particular animal has been cited, and N refers to the total number of informants questioned about zootherapy. Phillips and Gentry (1991, cited in Cotton 1993) de- veloped this quantitative method in order to calculate the relative usefulness of different plant species within a given community.

Medicinal raw materials were collected and then catalogued and deposited at Feira de Santana State University (UEFS) with other ethnobiology collections. Some specimens that were sent to specialists for taxonomic identification were also in the collection. Crustaceans were identified by Dr. Tereza Calado (Alagoas Federal University), echinoderms by Dr. Winston Leahy (Alagoas Federal University), fish

96 COSTA-NETO & MARQUES Vol. 20, No. 1

specimens by Dr. Paulo Duarte (Laboratory of Ichthyology at UEFS), and the re- maining animals were species known in this part of the country and were identified by the author using zoological references.

Borra ‘doltapicuru = = f Seribinha RSPLANADA/ i ‘ oA namife

~

1°46"

=

c a C~lRie on ° =

ACU OA TORRE \

FIGURE 1. —- Map showing location of the community of Siribinha where study was undertaken

Summer 2000 JOURNAL OF ETHNOBIOLOGY 97

RESULTS AND DISCUSSION

The medicinal use of marine/estuarine animals by the artisanal fishermen of Siribinha is one of the most consistent interactions that these fishermen perform with the local faunistic resources. Twenty-four fish species were recorded as hav- ing some therapeutic use when the fishermen were questioned about their folk medicine. Although interviews focused on fish-based medicines, fifteen other ani- mals with medicinal properties were also cited. This makes up a total of 39 resources, which are distributed in six scientific t ic categ , such as fish (62%), crustaceans (13%), reptiles (10%), echinoderms (8%), mollusks (5%), and mammals (2%). Zootherapeutic species, the raw materials utilized, and diseases for which they are prescribed for are found in the Appendix.

A total of 66 raw materials including scales, spur, shell, fat, skin, globe of the eye, tentacles, otolith are used in the elaboration of remedies to treat locally diag- nosed ailments. All of these resources, except the whale products, are relatively easy to obtain through hunting, fishing, or manual gathering. As can be seen in the Appendix, folk remedies are administered to the patients in the form of plas- ters, teas, smokes, and food. Teas, for example, are made by grinding the toasted or scraped parts of the body of the animals into powder or utilizing the whole toasted animal. Such is the case with echinoderms (starfish, sand dollar, and sea urchin) and syngnathids (seahorse). Fats, which are derived from 17 different ani- mals, are the most usual medicinal resources and Jed to treat a variety of diseases. As one fisherman observed, “the fat of a sea turtle sustains and heals any disease.” The great majority of the animals utilized in local medicine provide only one raw material, which is prescribed for the treatment of specific diseases. This is the case with squid, whose toasted internal shell is recommended for asthma.We found that queen triggerfish (Balistes vetula), sea turtles (Chelonia mydas, Eretmochelys imbricata, Caretta caretta, and Lepidochelys olivacea), and toadfish (Thalassophryne nattereri) were the most significant zootherapeuticals, with me- dicinal use-values (UV) of 0.92, 0.81, and 0.72, respectively. Although some zooth ti ies generated indices lower than 0.50, these were both medici- nally and culturally significant resources since a relatively high number of interviewees cited them. Such is the case with long-snout seahorse (Hippocampus reidi), with a UV value of 0.44, octopus (Octopus cf. variegatus), with UV value of 0.14, squid (Loligo sp.), with a UV value of 0.04, Atlantic tarpon (Tarpon atlanticus), witha UV value of 0.25, remora (Echeneis naucrates), with a UV value of 0.29, sharks, with a UV value of 0.29, and catfish (Bagre bagre, Sciadeichthys luniscutis, and Netuma barba), with a UV value of 0.12. The low rank of these species results from the fact that they do not represent multiple-use medicinal species.

Fishermen also use live fish, such as swamp eel (Synbranchus marmoratus) and cascarudo (Callichthys cf. callichthys). The former, with a UV value of 0.04, is pre- scribed for bronchitis and to make an infant child walk sooner. The latter, which presented a medicinal use-value of 0.02, is recommended for asthma and umbili- cal hernia. By spitting into their mouth and leaving them alive in the river, it is believed that these fish take bronchitis and asthma away; a child is thought to walk sooner by rubbing the fish over its legs; and cascarudos are recommended as food for the treatment of umbilical hernia. The procedure of applying live fish to

98 COSTA-NETO & MARQUES Vol. 20, No. 1

someone’s body or spitting into a fish’s mouth and then freeing it is a common practice both in Brazilian and other folk medicines (Branch and Silva 1983; Begossi and Braga 1992; Marques 1995). The placement of live fish under the soles of pa- tients suffering from jaundice or similar ailment was already cited by Shimshon Morpurgo in his book Responsa published in Venice in 1743 (Rosner 1992: 190).

Fishermen have a singular way to treat themselves from the injuries caused by poisonous fish, especially the toadfish (Thalassophryne nattereri). This fish has hollow opercular spines on its dorsal fin that are associated with bulky poison glands. Due to its bentonic habit, since it lives buried under the mud or sand in shallow waters, fishermen sometimes tread on it; the fish defends itself against the careless passer-by by injecting its poison in the person’s sole. In order to avoid pain and other more serious complications, fishermen take the fish’s globe of the eye (“goga do olho”) immediately after contact and rub it on the injured area. Guilherme Piso, a Dutch doctor who came to Brazil in the company of Prince Mauricio de Nassau in 1695, stated that “This fish is considered as having the remedy to its own poison in itself” (Piso 1957: 51). This kind of preventive medi- cine is performed with catfish (Bagre bagre, Sciadeichthys luniscutis, and Netuma barba) and the queen triggerfish (Balistes vetula). Another way of avoiding the pain caused by the toadfish sting is to eat the toadfish it without salt. In doing this, fishermen say they are immunized against its poison. According to the users’ tes- timony, it can be hypothesized that analgesic substances are indeed present in the bodies of these fish species. Further studies are requested in order to test this. According to Norse (1993), the chemical identities of the toxins of poisonous fish are still being determined.

We noted some similarities between the zootherapeutic healing practices cur- rently found in Siribinha Beach and in other fishing communities. For example, Begossi and Braga (1992) have recorded the medicinal use of ray by the fishermen from Tocantins River. Its sting is used for the treatment of asthma, cough, cold, and pneumonia. In the state of Alagoas, fishermen use the sting to treat them- selves for pneumonia (Marques 1995). In the Amazon folk medicine, the oil extracted from the liver of the ray Potomotrygon hystrix is used for the treatment of hernia and asthma (Branch and Silva 1983). In Siribinha, stingrays (Myliobates sp.; UV = 0.02) are also used in the folk veterinary medicine by employing the powder of a toasted sting to heal domestic animals’ wounds

A prescription involving toasted seahorses (Hippocampus spp.) is also both geographically and historically well disseminated. According to Botsaris (1995), the African slaves introduced the practice of using syngnathids as medicine in Brazil. As a result, these fish have been recommended in Afro-Brazilian traditional medicine as a tonic to treat physical debility, impotence, and asthma as well as rheumatism, bronchitis, and gastritis pe vl dbnale 1934; Marques 1995). The me- dicinal use of seal Brazil has bee fi d in at least eight states, Alagoas, Espirito Santo, Sao Paulo, Parana, Bahia, Santa Catarina, Piaui, and Rio de Janeiro (Marques 1995). In Siribinha Beach, toasted seahorses have been used in cases of asthma. Clinical and pharmacological research carried out on mice, with alcoholic extracts of Hippocampus, led to a weight increase of the uterus and ovaries and a prolongation of the estrogen period in females, while in males it caused an in-

Summer 2000 JOURNAL OF ETHNOBIOLOGY 99

crease in the weight of the prostate and testicles, and a prolongation of the time of erection (Botsaris 1995).

Other fish species are used medicinally, including the cod (Gadus cf. marhua; UV = 0.04), whose skin is put on furuncles, the curimata (Prochilodus sp.; UV = 0.02), whose fat is used as a plaster to treat boils, the electric ray (Narcine brasiliensis; UV = 0.05), whose fat is used for treatment of toothache, the grunt (Haemulon sp.; UV = 0.02), whose liver’s fat is put on swollen areas, the sheepshead porgy (Cala- mus pena?; UV = 0,02), whose toasted fin is used for curing asthma, the snook (Centropomus undecimalis; UV = 0.02), whose toasted fat is rubbed over swollen legs, the two-spot astyanax (Astyanax cf. bimaculatus; UV = 0.02), which is recom- mended for alcoholism, and the trahira (Hoplias malabaricus; UV = 0,10), whose fat is recommended for the treatment of asthma, bleeding, boils, wounds, snakebites, and conjunctivitis. Only one informant cited the medicinal use of the croak (Micropogonias furnieri; UV = 0.02), although was unable to identify the disease for which this fish is prescribed.

Siribinha fishermen use five folk species of crabs. These include the mangrove crab (Ucides cordatus; UV = 0.02), whose fat is recommended for treating women’s hemorrhage, the giant land crab (Cardisoma guanhumi; UV = 0.02), whose fat is used as a plaster for the cicatrization of wounds, the ghost crab (Ocypode quadrata; UV = 0.04), whose toasted whole body is used for asthma, the jellyfish crab (an uncollected specimen; UV = 0.04), which is also used for asthma, and the hermit crab (a pagurid; UV = 0.02), whose toasted whole body is recommended to treat women’s hemorrhage. Interestingly, the powder of crab shells in infusions has been reported in the state of Alagoas, northeastern Brazil, as an anti-asthmatic (Lages-Filho 1934). Indeed, pharmacological studies have shown the presence of anti-inflammatory, antibiotic, and anti-tumor substances in the bodies of crabs (Croft 1986).

The value of animal-based medicines. Although zootherapy is considered by some to be a weird, even absurd practice, its pertinence should be emphasized. As many researches point out, the significance of traditional medicines cannot be denied since they have become sources of drugs within modern medical science (Launet 1993; Lazarus and Attila 1993; Ferreira 1993; Marques 1997). Worldwide, a num- ber of pharmaceutical companies have been supporting research on marine animal-derived compounds to be used directly as medicines and as new chemical structures that could be turned into remedies (Norse 1993; Fusetani 1996). Regarding fish, several compounds have been extracted and are employed as remedies in standardized medicines (Hamada and Nagai 1995; Salte et al. 1996). Finkl (1984), for example, refers to Eptatretus stoutii, Dasyatis sabina, and Taricha sp. as sources of cardiac stimulants, antitumors, and analgesic, respectively. Oily fish, like cod, herring, salmon, and turbot, have a great medicinal value to human be- ings due to a polyunsaturated compound known as OMEGA-3. This substance helps with the prevention of arthritis (Adeodato 1997). The presence of an antico- agulant system in the plasma of Atlantic salmon (Salmo salar L.) and rainbow trout (Oncorhynchus mykiss Walbaun) has been confirmed, which supports similarities with the protein C anticoagulant system in mammals (Salte et al. 1996). Tetrodot- oxin (TTX), a water-soluble guanidinium derivative, is an example of a bioactive

100 COSTA-NETO & MARQUES Vol. 20, No. 1

compound produced by marine organisms such as puffer fish “that resembles procaine in its ability to inhibit transmission of nerve cells” (Colwell 1997). When diluted it acts as an extraordinary narcotic and analgesic (Bisset 1991). Maybe due to this property, Siribinha fishermen use the liver’s content (“fel”) of the puffer fish (Colomesus sp.; UV = 0.11) for the treatment of toothache, although it has a lethal venom.

In relation to crustaceans, chemists from the Federal University of Ceara, in Brazil, have developed three products that are extracted from discarded crusta- cean shells (lobsters, shrimps, and crabs). The biopolymers chitin, chitosan, and glucosamin are used to combat cholesterol and obesity, and regenerate cartilage and burnt tissues (Nogueira 1999).

Echinoderms are another group of marine organisms that yield a rich source of potential chemicals. Pharmacological studies have shown important active com- pounds in species, such as Actinopyga agassizi (anti-tumor), Acanthaster planci (antiviral), and Asterias forbesi (anti-inflammatory) (Finkl 1984). According to Alino et al. (1990), there ae studies pane: that the use of sea urchins as vermifuge might have a scienti Echinoids have been also used to make artificial blood veins (Russel 1978, cited in Mallmann 1996). In the folk medicine of Siribinha’s fishermen, teas made with the powder of the toasted sand dollar (Mellita sp.), starfish (Luidia senegalensis), and sea urchin (Echinometra lucunter) are commonly recommended for the treatment of asthma. These resources are very commonly used, although their UV was only 0.16.

n 1995, scientists from the Servier Research Institute in Paris discovered the way sea turtles reduce their cerebral activity while submerged in order to save oxygen (Anonymous 1995). This physiological process was referred to as the ‘turtle effect’. This study on the cerebral activity of sea turtles may lead to the develop- ment of drugs against apoplexy.

Threatened species as zootherapeuticals.— Of the 39 animals considered as sources of medicine to the Siribinha fishermen, only four are officially listed as threatened species by the Brazilian Institute of the Environment and Renewable Natural Re- sources (IBAMA, 1989). These are represented by the sea turtles, which are also cited by the Convention on International Trade of Endangered Species (http:/ / www.cites.org). Five of the world’s eight known species use the Brazilian coast- line as hatcheries or feeding ground (http://www.wwf.org.br/wwfeng/ wwfpr35.htm). In the Bahian territory, four species used to be widely caught and killed for both food and income by the coastal peoples, indigenous and artisanal fisherman communities, who have inhabited its coastline. Sea turtles were cap- tured specially during the reproductive season when females come up onto the beach to lay their eggs. According to the interviewees, several turtles were hung alive for a couple of days before being killed. The eggs themselves were also taken, Hees their shells were used to make glasses, combs, rings, necklaces, and brace- ets.

Indeed, coastal peoples all over the world fish sea turtles for subsistence and commerce. For example, the Miskito Indians from the eastern coast of Nicaragua are very dependent on green turtles (Chelonia mydas). They catch sea turtles for their meat, leather, shell, oil, and calipee, a gelatinous substance that is the base

Summer 2000 JOURNAL OF ETHNOBIOLOGY 101

for turtle soup (Nietschmann 1974). Among the coastal people of northern Mindanao and the Visayan islands of Bohol, Cebu, and Negritos in central Philip- pines, the flesh of green turtle and hawksbill turtle (Eretmochelys sp.) is cooked with vinegar, soy sauce, garlic, laurel leaves, pepper corn, and salt, and then eaten for the treatment of asthma (Alifo et al. 1990).

This extractive pressure has increased in such a way that all eight species of sea turtles are now on the endangered species list. Population growth and ad- vanced harvesting technologi de possible the enlarg t of the fishing power as well as overexploitation of natural resources. Nietschmann (1974) stated that the industrial fishing nets that were introduced in the Miskito’s culture helped to kill a greater number of turtles than when these Indians solely used harpoons. Siribinha fishermen have generated their own theory to explain the diminishment of sea turtles and sharks— the shrimp trawl] fishery. This fishery is primarily done by outside, motorboat fishermen, who go to their fishing grounds to trawl shrimp. Vincent and Hall (1996) have already reported the negative impacts of trawling on marine organisms. In addition to this direct human persecution, sea turtles are also the victims of environmental changes and pollution.

Due to the drastic decline of wild populations, conservation measures have been taken and legislative laws now prohibit sea turtle fishing. In Brazil, the TAMAR (Tartarugas Marinhas) project was set up in 1980 to protect sea turtle feed- ing grounds and hatcheries along the Brazilian coast (about 1,000 kilometers of beaches). This is a worthwhile project since only a few tens of hawksbill turtles breed in the northeast of Brazil, specifically on the Bahian coast. At present, TAMAR has 22 bases in eight Brazilian states, at strategic nesting and feeding points along the coast.

This relatively recent intervention, however, has raised an exogenous taboo, institutionalized by rules that have been imposed on local communities by both government and conservation agents. These rules have resulted in the disconnec- tion of a strong human/animal interaction that has a long history. Although fishermen fear being caught by IBAMA agents and taken to jail, they undertake transgressions and break the rules. Some turtles are indeed captured and eaten, or have their eggs harvested. We had the chance to document the cultural scene in which three young fishermen butchered an adult green turtle (C. mydas), and ex- tracted its flesh and oil as food and remedy respectively. According to their explanation, the turtle appeared floating in the estuary and apparently died a few hours previously due to the bite of a shark on its right flipper ( aba ). However, this incident occurred in a very concealed way because collecting wild animal species is not considered a bailable crime in Brazil.

We would also like to call attention to the importance of seahorses. Although they are not included on the Brazilian list of endangered species, they were listed by IUCN in 1996. The alleged cause of this endangerment has been the great de- mand of seahorse specimens for Chinese traditional medicine, aquariums, and curios (Vincent and Hall 1996). Due to their worldwide use, seahorses are becom- ing rare in some regions of the globe. For example, in Indonesia the population of some species has dropped to half since 1990, with pregnant males being the most

common prey.

102 COSTA-NETO & MARQUES Vol. 20, No. 1

The number of seahorses is indeed diminishing in Siribinha, as can be seen in the following informant’s assertion:

We took them to the market to be sold. People bought them for remedies. We fished a lot of them. My husband used to catch several of them, which were sun-dried and tied together in a string. He sold them at the market. But you do not see my son? Even they have disappeared that we do not find them anymore. (Mrs. Zulmira, 90 years old)

We would ca that a biostatistical atiy on the wild populations of seahorses and their fo arried out in order to develop reliable conservation measures that are both scientifically and culturally oriented.

Zootherapy and its sustainability — Instead of sending the practitioners of zootherapy to prisons, or creating policies which force them to abandon such practices, deci- sion-makers should view this human/nature interaction within its cultural dimensions. The value of animal-based medicines are very significant; they are usually the main available resources for the majority of the human population with limited access to official medicines and medical care. Since people have been using animals for a long time, suppression of their use will not save them from extinction. Considering the sea turtles, Carr (1996: 127) stated that “people have been eating turtles pretty steadily for as long as they have had the wits to get them out of their shells”. In addition, millions of hatchlings are caught annually for the pet market.

A growing body of literature shows that the cultural aspects of a given hu- man/nature interrelation should be taken into account in all debates related to sustainable development (Morin-Labatut and Akhtar 1992; Sachs 1993; Agrawal 1995; Zwahlen 1996). This cultural perspective includes the way people perceive, use, allocate, transfer, and manage their natural resources (Johannes 1993). As Alcorn (1995: 20) states, “Conservation is a social and political process.” In this way, discussing the relationship between food provided by the environment, their trophic use, the physiological consequences that result from their being eating, as well as the social-economical structures that support them within the multidimen- sionality of the sustainable development is one of the key elements to achieve sustainability (Bahuchet 1997).

Researchers should recognize that the sustainable use of natural resources due to their medicinal value is one of the ways by which biodiversity is used (Celso 1992). According to Kangas (1997), sustainable development is tied indirectly to biodiversity through the need to maintain overall esdbasamuaaeas highndte rl ever, the demand for natural products from m problem if collectors overexploit the typically sessile « organisms 8 1993). Yet, we have to realize that the negative impacts on biological diversity should not be restricted only to the traditional users, but should be extended to the use by phar- maceutical industries (Marques 1997)

Summer 2000 JOURNAL OF ETHNOBIOLOGY 103

ACKNOWLEDGEMENTS

We would like to thank Prof. T. Calado, W. Leahy, and P. Duarte for the specimens taxonomic identification, the anonymous referees for their helpful comments and suggestions on the manuscript, and to all informants for allowing reproducing their knowledge and their kind collaboration. Without them this work would not have been possible.

NOTES ‘This paper is part of a larger ethnoichthyological study carried out by Eraldo Medeiros Costa Neto as his master dissertation at Alagoas Federal University, and which was guided

by Professor Jos Geraldo Wanderley Marques.

LITERATURE CITED

ADEODATO, SERGIO. 1997. Os santos remédios do mar. Globo Ciéncia (Apr):

AGRAWAL, ARUN. 1995. Indigenous and scientific knowledge: some critical comments. Indigenous Knowledge and

servation Biology, D Ehrenfeld (editor). Blackwell Science,

Oxford.

ALINO, PORFIRIO M., GLORIA J.B. CAJIPE, EDNA T. GANZON-FORTES, WILFREDO Y. LICUANAN, NEMESIO E. MONTANO and LUIS M. TUPAS. 1990. The use of marine organisms in folk medicine and horticulture: a preliminary — stud Seaweed Information Center Newsletter.

ANGELETTI, LUCIANA R., UMBERTO

£3

MARIANI-COSTANTINI. Healing rituals and sacred serpents. The Lancet 340:223-225.

ANONYMOUS. 1995. “Efeito tartuga” pode combater derrames. Folha de Sao Paulo, 05 Apr. 1995.

ANTONIO, TANGO M.F. 1994. Insects as remedies for illnesses in Zaire. The Food Insects Newsletter 7(3):4-5.

BAHUCHET, SERGE. 1997. Ethnoécologie. Http://lucy.uck.ac.uk/Sonja/RF/ Divdocs/Ethnoecologie_S._Bahuchet. html.

BEGOSSI, ALPINA. 1992. Food taboos at Buzios Island (Brazil): their significance and relation to folk medicine. Journal of

Ethnobiology 12(1):117-139.

ver.

BISSET, NORMAN G. 1991. One man o another man’s medicine. ral of Ethnopharmacology 32:71-81.

BOTSARIS: ALEXANDROS S. 1995. Fitoterapia Chinesa e Plantas Brasileiras. Editora Icone, Sao Paulo.

BRANCH, LYN C. and MARLENE F. SILVA. 1983. Folk medicine in Alter do Chao, Para, Brazil. Acta Amazonica 13:737-797.

BUT, PAULP.-H., YAN-KIT TAM, and LAI- CHIN LUNG. 1991. Ethnopharmacology of rhinoceros horn II: Antipyretic effects of prescriptions containing rhinoceros horn and water srcina urnal f

Ethnopharmacology 33: 45-50.

CARR, ARCHIE. 1996. Handbook of Turtles. The turtles of the United States, Canada, and Baja California. Cornell

ork.

University Press, pape! SC 972. Dicionario do

Folclore Steen recone Rio de

Jane

104 COSTA-NETO & MARQUES

nee R. 1992. Criagdo de condigédes e neentivos para a conservagao loc local de biiderend ad. Pp. 92-93 in heap obal da Biodiversidad. Diretri Agaéo par Estudar, Salvar e Cae de Maneira Sustentavel e Justa a Rae rey Biética da Terra, J.C. Speth, M.W. Holdgate and M.K. Tolba (editors). WRI/UICN/PNUMA, Rio de Janeiro. CENTRO DE ESTATISTICA E INFORMAGOES (CEI) (BAHIA). 1994. pee basicas — ips eossoe te ianos: regido litoral no Estado da Bahia, Saar COLWELL, RITA R. 1997. Microbial biodiversity and biotechnology. Pp. 77- 98 in Biodiversity Il: Understanding and Protecting our Biological Resources, Marjorie L. Reaka-Kudla, Don E. Wilson and Edward O. Wilson (editors). Joseph enry Press, Washington D.C. pag boa bale ant E. and JOSE M. M. PINO. 1988. The utilization of insects in the eine medicine of ancient Mexicans. Journal of Ethnobiology

1): :

COSTA-NETO, ep M. 1996. Faunistic resou $ medicines by an Ato Diwsiinn sepeinarern from Chapada Diamantina National Park, State of Bahia, Brazil. Sitientibus 15:211- 219.

1999a. Recursos animais

utilizados na medicina tradicional dos

a Pankararé que habitam no

ordeste do estado da Bahia, Brasil. Actualidades Bioldgicas 21(70):69-79

. Healing with animals in

Feira de Santana city, Bahia, Brazil. Journal of Ethnopharmacology 65:225-

COTTON, CATHERINE M. 1996. Ethnobotany: Principle and Applications. John Wiley and Sons,

York.

ew CROFT, J.E. 1986. Relief from Arthritis: The Natural Way. A Safe and Effective Treatment from the ee Thorsons

li tain. FAULKNER, SAV) 19 1992. “Buxnailicn uses for natural marine products. 35

Oceanus 35(1):29-35, FERREIRA, SERGIO H. 1993. A descoberta

Vol. 20, No. 1

FINKL, CHARLES W. 1984. Os medicamentos do mar. Pp. 74-75 in Enciclopédia dos Mares, vol. 1, or. anes (editor). Salvat,

Rio de Jan

FUSETANI, NOBUHIRO. sa ae substances from marine spon oie . ‘Toxicology-Toxi mere 15(2):15

GU DGER, ‘EUGENE Ww. 1925. Stitching

beetles. Journal of the American Medical Association 84: 1862-1864 HAMADA, MINORU and TADASHI NAGAI. 1995. Inorganic components of bones of fish and their advanced utilization. Journal of Shimonoseki University of Fisheries 43(4):185-194. HUXTABLE, RYAN J. 1992. The pharmacology of extinction. Journal of Ethnopharmacology 37:1-11. IBAMA. 1989. Official List of Fauna reatened with Extinction in Brazil. Instituto Brasileiro do Meio Ambi e dos Recursos Naturais Renovaveis, Brasilia. JOHANNES, ROBERT E. 1993. Integrating traditional ecological knowledge and a with environmental act assessment. Pp.33-39 in Traditional re Knowledge: d Cases, J.T. Inglis (editor).

IDRC, Ottawa KANGAS, PATRICK.

ies Tropical sustainable and

Kudla, Don E. Wilson and Edward O. Wilson (editors). Joseph Henry Press, Washington D.C.

KUNIN, WILLIAM E. and JOHN H. LAWTON. 1996. Does biodiversity matter? Evaluating the case for conserving species. As 283-308 in Biodiversity: A Bi Numbers and Differences, K. J. Gaston (editor).

Blackwell Science, : LAGES-FILHO, JOSE. 1934. A Medicina

Popular em Alagoas. Separata dos

a do Instituto Nina Rodrigues,

Sa

LAUNET, EDOUARD. 1993. Dans les férets, a la des médicaments pg maha Science et Vie 904:86-91.

Summer 2000

LAZARUS, LAWRENCE H. and MARTTI ATTILA. 1993. The toad, ugly and venomous, wears yet a precious jewel re his skin. Progress in Neurobiology

:473-507.

MALLMANN, MARIA L.W. 1996. A Farmacopéia do Mar: — Marinhos de Interesse Médic Etnomedicina Alagoana. Monografia (Especializagdéo em Zoologia), “9h aaaaal Federal de Alagoas,

MARQUES, JOSE G. W. 1994. A fauna medicinal dos indios Kuna de San Blas

zooterapica. P. 618 in Anais. Sociedade Brasileira ey 0 Progresso da Ciéncia. 46" Reuniao Anual Vitéria

; 1997, —— medicinal: recurso do ambien ou ameacga 4a biodiversidade? Mutim 1(1):4. RIN-LABATUT, GISELE and AKHTAR, SHAHID. 1992. — Traditional

manage and share. se 4:24-30. NIETSCHMANN, BERNARD. 1974. When

NOGUEIRA, DANIELLE. 1999. As qualidades — dos crustaceos. Ciéncia Hoje

NORSE, ELLIOT A. “editor. 1993. Global

Making. Island Press, Washington D.C.

PHILLPS, O. and A.H. GENTRY. 1993. The useful plants of Tambopata, Peru: I. Statistical hypothesis tests with a new quantitative technique. Economic Botany 47(1):15-32.

JOURNAL OF ETHNOBIOLOGY 105

PISO, GUILHERME. 1957, Histéria Natural do Brasil lustrada. Companhia Editora Nacional, ~ de Janeiro.

ROSNER, FRED. 1992. Pigeons as a remed (segulah) for jaundice. New or State

Journal of Medicine 92(5):189-19.

RUSSEL, F.S. 1978, Marine eee and human ped Ady. Mar. Biol. 15:233-

48.

2

SACHS, IGNACY. 1993. Estratégias de eee 9 ey Sa o a lo XXI. Nobel / FUN

SALTE, rs K. NORBERG. and O, R. ODEGAARD. 1996. Evidence of a protein c-like anticoagulant system in bony fishes. Thrombosis Research

83(5):389-397, VAN HUIS, ARNOLD. 1996. The —-

Proceedings of eterna pr Applied Entomology, .E.V Amsterdam 7:3-20.

VINCENT, A.C. and H.J. HALL. 1996. The threatened status of marine fishes. Tree 11(9):360-361.

WEISS, HARRY B. 1947. Entomological medicaments of the past. Journal of t New York Entomological Society 55:155-

WERNER, DAVID. 1970. Healing in the Sierra Madre. The Natural History

79(9):61-66. ZWHALEN, ROBERT. 1996. Traditional — ee + é, 2 i ttt.

Indigenous Knowledge and Development Monitor 4(3):1-7.

106 COSTA-NETO & MARQUES Vol. 20, No. 1

APPENDIX.- Folk prescriptions of zootherapeutic resources used by artisanal fishermen from Siribinha Beach, state of Bahia, Brazil.

MOLLUSKS

Octopus, “Polvo” (Octopus cf. variegatus) Cephalopoda, Octopodidae et the “arms” (tentacles), toast them, and then grind to make a tea that is drunk for curing asthma; The powder of the toasted “lixa” (shell rudiment ?) is mse to make a tea, which is drunk to treat “doenca do vento” (stroke Get the hide (skin), burn it, and breathe the smoke to ne headache.

Squid, “Lula” (Loligo sp. ?) Cephalopoda, Lolliginidae Get the “stone” (internal shell), toast it, and make a tea for curing asthma.

ECHINODERMS Sand Dollar, “Estrela-da-costa” (Mellita sp.) Echinoidea, Mellitidae The whole toasted starfish is turned into a tea to treat asthma. Sea Urchin, “Pinatina” (Echinometra lucunter) Echinoidea, Echinometridae Idem Starfish, “Estrela-do-mar” (Luidia senegalensis) Asteroidea, Luidiidae Idem CRUSTACEANS

Giant Land Crab, “Gaiamum” (Cardisoma guanhumi) | Decapoda, Gecarcinidae Get the “fel” (fat?), make a plaster, and put it on wounds to help their cicatrization.

Ghost Crab, “Grauga” (Ocypode quadrata) Decapoda, Ocypodidae Get a ghost crab, toast it, and then grind to make a tea, which is drunk to treat asthma.

Hermit Crab, “Caranguejo-ermitao” (Species not determined) Anomura, Paguridae Get a hermit crab, take it away from the shell, toast it and grind to make a tea; then drink it to treat women’s hemorrhage.

Jellyfish Crab, “Caranguejo-da-dgua-viva” (Specimen not collected) ? The whole crab is turned into a tea, which is useful for asthma.

Mangrove Crab, “Uca” (Ucides cordatus) Decapoda, Ocypodidae Get the fat, filter it, mix it with white wine, and then drink it for treating women’s hemorrhage.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 107

FISHES

Atlantic Tarpon, “Cangurupim” (Tarpon atlanticus) Elopiformes, Elopidae Get a scale, burn it, and breathe the smoke for curing “doenca do vento” (stroke?), headache, and asthma.

Get a scale, toast it, then grind and make a tea to treat asthma.

Cascarudo, “Caboge” (Callichthys cf. callichthys) Siluriformes, Callichthyidae Someone who has faith split into its mouth three times and then free it alive in the river to be healed from asthma; Eat one in cases of umbilical hernia.

Coco Sea Catfish, “Bagre-fidalgo” (Bagre bagre) Siluriformes, Ariidae Rub the globe of the eye over the area that was injured by its spur to alleviate pain.

Cod, “Bacalhau” (Gadus cf. marhua) Gadiformes, Gadidae Put the hide (skin) on furuncles.

Croak, “Curvina” (Micropogonias furnieri) Perciformes, Scianidae Get the otolith and make a tea.

Curimata, “Xira” (Prochilodus sp.) Characiformes, Prochilodontidae Get the fat and make a plaster to treat boils.

Electric ray, “Peixe-elétrico” (Narcine brasiliensis) Torpediniformes, Narcinidae Put the fat on the tooth to treat toothache.

Grunt, “Bonome” (Haemulon sp.) Perciformes, Haemulidae Rub the fat of the liver over swollen areas.

Long-Snout Seahorse, “Cavalo-marinho” (Hippocampus reidi)

Gasterosteiformes, Syngnathidae

Get one, let it to be sun-dried, then toast it and grind to make a tea, which is drunk in cases of asthma.

Marine Catfish, “Bagre-urutu” (Sciadeichthys luniscutis) Siluriformes, Ariidae Rub the globe of the eye over the area that was injured by its spur to alleviate pain.

Pufferfish, “Baiacu-xaréu” (Colomesus sp.) | Tetraodontiformes, Tetraodontidae Put the content of the liver (“fel”) in the tooth to alleviate toothache;

Get the hide (skin) and cover the wounds with it.

Queen Triggerfish, “Capado” (Balistes vetula) Tetraodontiformes, Balistidae Get the scale, burn it, and breathe the smoke for curing stroke, Get the fat of the liver, put it in a piece of cotton, and then introduce it inside the ear; it is useful for curing earache; ae Get the spur, toast it, and then grind to make a tea which is drunk against its own venom; The toasted fat is drunk against stroke; ; Get the hide (skin), burn it, and breathe the smoke to treat asthma and stroke; Rub the “goga do olho” (globe of the eye) over the area that was stung by its spur.

108 COSTA-NETO & MARQUES Vol. 20, No. 1

Remora, “Pegador” (Echeneis naucrates) Perciformes, Echeneidae Get the sucking disk, let it to be sun-dried, then toast it and grind to make a tea which is drunk for curing bronchitis and asthma.

Scalloped Hammerhead (Sphyrna lewint) Carcharhiniformes, Carcharhinidae Toast the liver, get the fat and drink it to treat as ; Massage fat on rheumatic parts of the body; Rub the fat over wounds.

Sharpnose Shark, “Cacao-rabo-seco” (Rhizoprionodon sp.) | Carcharhiniformes, archarhinidae Toast the liver, get the fat and drink it to treat asthma; Massage fat on rheumatic parts of the body; Rub the fat over wounds.

Sheepshead Porgy, ” ‘Peixe-pena” (Calamus pena ?) Perciformes, Sparidae Toast the “pena” (fin ?) and grind it to make a tea for curing asthma

Smalltail Shark, “Cacao-gaia-preta” (Carcharhinus porosos) | Carcharhiniformes, Carcharhinidae Toast the liver, get the fat and drink it to treat asthma; Massage fat on rheumatic parts of the body; Rub the fat over wounds.

Snook, “Rubalao” (Centropomus undecimalis) Perciformes, Centropomidae Get the fat, toast it, and rub it over swollen legs.

Stingray, “Arraia” (Myliobates sp.) Sianeli Myliobatidae Toast the spur, grind it and make a tea for curing asthm Put the powder of the toasted spur in a broken tooth to amas the

Pats Toast the fat and drink it against asthma; Put the fat over wounds.

Swamp eel, “Mucum” (Synbranchus marmoratus) Synbranchiformes, Synbranchidae Split into its mouth and free it alive in order to treat bronchitis; Rub a live fish over an infant child’s legs to make him/her walk sooner.

Toadfish, “Niquim” (Thalassophryne nattereri) Batrachoidiformes, Batrachoididae Get = globe of the eye and rub it over the area that was injured by its

Rub the Resrnaeod (soft part of the head) over the injured area in order to ke the pain away;

Eat three esiiees roasted toadfish to abi feeling pains in the next

time when someone gets injured by its spur.

Trahira, “Traira” (Hoplias malabaricus) Characiformes, Erythrinidae Get the fat, toast it and use it for curing toothache, asthma, bleedings, boils, and wounds; The raw fat is recommended as an antidote against snakebites; Rub the fat over the eyes in order to treat conjunctivitis.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 109

Two-spot Astyanax, “Piaba-mirim” (Astyanax cf. bimaculatus) | Characiformes, Characidae Get three live fish and grind them; then put the resulting mass in a white rum bottle and bury it for a period of five days. After this, make someone drink it in order to stop drinking.

White Sea Catfish, “Bagre-do-mangue” (Netuma barba) Siluriformes, Ariidae Rub the globe of the eye over the area that was injured by its spur to alleviate pain.

REPTILES

Green Turtle, “Tartaruga-verde” (Chelonia mydas) Testudines, Cheloniidae The fat is toasted and rubbed over wounds and bangs; Put the fat in a piece of cotton and apply it on painful tooth; Get a small piece of the shell, burn it, and breathe the smoke for curing asthma; Get the rear foot, burn it, and breathe the smoke to treat stroke; Toast the shell of the egg, grind it, and put the powder to be boiled; then cover it for a while and drink it to treat asthma; A cooked egg is eaten for the treatment of diabetes; Drink the toasted fat in cases of headache, cough, bronchitis, hoarseness, and asthma; Mix the fat with honeybee and drink it to treat asthma and flu.

Hawksbill Turtle, “Tartaruga-de-pente” (Eretmochelys imbricata) Testudines, heloniidae

Idem

Loggerhead Turtle, “Tartaruga-cabecguda” (Caretta caretta)Testudines, Cheloniidae Idem

Olive Ridley, “Tartaruga-de-couro” (Lepidochelys olivacea)Testudines, Cheloniidae Idem

MAMMALS

Whale (Specimen not determined) Cetacea

Put the fat on a teaspoon, warm it, and drink it against asthma; Sit down on a vertebra in order to treat backaches.

110 BOOK REVIEWS Vol. 20, No. 1

Responding to Bioprospecting: from biodiversity in the South to medicine in the North. Hanne Svarstad and Shivcharn S. Dhillion ( Eds). Spartacus Forlag As, Oslo. 2000. Pp. 220. $ 17.95. ISBN 82-430-0163-8

Responding to Bioprospecting, provides an even handed look at the cultural, economic, an g ethical and unethical biodiversity prospecting. This volume is poignant in its anecdotes and discussions that range from issues of biopiracy, to what is needed for truly ethical bioprospecting in Af- rica. As one moves through the fourteen essays which immediately offer a crash course in “bioprospecting 101” and beyond, the reader will be immersed in not just the polarization that influences this issue, but many of the historical nuances that come to bear on bioprospecting. The volume begins with a powerful “picture this” example, when on vacation in Norway in 1969, a Swiss scientist collects a few soil samples. After some 30 years, including 11 years of research and develop- ment, that hand full of soil has yielded a pharmaceutical company US$ 1.2 billion in sales (page 9).

Part one of Responding to Bioprospecting comprises the first seven chapters which provide an in depth examination of the origins of bioprospecting. The first section is silent in its analysis of the differences between bioprospecting and biopiracy. In addition, the issues confronting the commercialization of medicinal plants and concerns regarding the phytomedicine industries use of “best prac- tices” are covered. Part two of the volume offers examples of engagement between the scientific community and traditional healers. Of particular interest are Chap- ter Eight’s insights on a working model of bioprospecting reciprocity in Mali. The final section covers Chapters 12 through 14, which discuss legal issues that ham- per denied development nations in their efforts to be equal partners in bioprospecting. While the essays in Responding to Bioprospecting are diverse and objective, missing is the voice of bioprospecting opponent Vandana Shiva. While Shiva is often cited in the articles, she does not have a contributing essay in the volume. Shiva would enhance Responding to Bioprospecting had it included a piece by her.

It remains to be seen whether or not there can be an ethical and fair en- gagement within the bioprospecting arena. Yet, in Responding to Bioprospecting the reader is presented with the often-used model of INBIO and Merck & Co in Costa Rica. There is a great deal of success and technological exchange in this particular arrangement that can be applied to other NGO/Industry agreements. But it should be understood that the Merck/Costa Rica deal is essentially a contract between a government and an industry leader and that bioprospecting takes place on gov- ernmental land reserves, hence “no work in agricultural lands or areas under

t of groups” takes place (page 54). Similar situations rarely exist elsewhere. Instead, most bioprospecting efforts take place in indigenous pauaere

hough many indigenous groups have failed to benefit from past Se cca efforts, the volume does provide some pias ed of current or re- cent agreements in which indigenous groups and ethnic minori with bioprospectors. An example of this cooperation appears in Fate 10, which examines the relationship between a rural community and Shaman Pharmaceuti-

Summer 2000 JOURNAL OF ETHNOBIOLOGY 111

cals. This chapter illustrates the trial and error process that many traditional com- munities currently find themselves in when dealing with bioprospecting interests The question that often arises is, can these communities withstand this trial and error process, and is it ethical to have unwitting communities enduring a process that can be avoided? History shows that if equitable arrangements are not agreed upon, communities find their own traditional knowledge becomes contested.

No single essay in Responding to Bioprospecting ties together all the vari- ables that keep the usually one-sided bioprospecting structures in place. But by the book’s end it is very apparent that this collection outlines the bi-polar posi- tions on the issue of bioprospecting, and it details many of the undercurrents and inner-workings of this industry. To this volume’s credit, Responding to Bioprospecting provides thorough coverage of biodiversity conservation issues. Unfortunately, it does not provide enough material regarding issues such as the loss of cultural diversity. Far too little attention is given to the cultural alterations that belie the aftermath of bioprospecting be it ethical or not. Also missing from the volume is examinations of the role advances in technology have played in making bioprospecting a lucrative industry in the last decade, and the pressure this these technological advances have on denied development countries. The failure on the part of the editors to include an essay on the role of bioprospecting and its links to health issues in both the nations of the North and South should not be seen as blind omission. Such oversights are often corrected in second editions. Given the broad array of topics covered in this volume, Responding to Bioprospecting is well worth the read for ethnobiologists interested in this topic.

Phoenix Savage

Department of Sociology and Anthropology University of Mississippi

University, MS 38677

112 BOOK REVIEWS Vol. 20, No. 1

Aboriginal Plant Use in Canada’s Northwest Boreal Forest. Robin J. Marles, Christina Clavelle, Leslie Monteleone, Natalie Tays, and Donna Burns. UBC Press, Vancouver. 2000. $29.95 (paperback), $75.00 (clothbound). Pp. 368. ISBN 0-7748-0738-5 (paperback), 0-7748-0737-7 (clothbound).

Aboriginal Plant Use in Canada’s Northwest Boreal Forest describes over 200 of the traditional plants and plant products (foods, medicines, and materials for handicrafts or technologies) of the Cree, Dene, and Métis peoples living in the northwest boreal forest regions of central to Northern Manitoba, Saskatchewan, and Alberta. The book is a compilation of original ethnobotanical fieldwork and supplemental information derived from the literature (on etl tritive and medicinal plant value, and ecological impact and economic potential of commercial plant development).

It attempts more than simply a list of useful plants by including an extensive introductory section detailing research methods, and ecological and cultural back- grounds. Sources of related literature on boreal plant uses are also included. In addition to mee plant uses, cag aiectine of the fieldwork included train- ing Aboriginal student h within their own communities, identifying ao with the potential for sustainable harvesting and economic development, and preparing Aboriginal peoples for a role in the develop- ment process. The authors indicate their concern for maintaining Aboriginal rights to intellectual properties, and briefly outline how they attempted to respect these and other related issues (e.g., including only information that elders wished to have shared, and maintaining confidentiality by using a code to identify contributors).

The majority of the book is a listing of boreal plants and their traditional uses, organized into sections as follows: Fungi, Lichens, Nonvascular Plants (i.e., Mosses), Vascular Spore-Producing Plants (i.e., Horsetails, Clubmosses, Ferns), Gymno- sperms, and Angiosperms. Plants are organized alphabetically within each section by scientific name, but English and local Aboriginal names are also included. For each plant, a 1-2 page description of the plant (including photograph), its habitat, and uses for food, medicine, technology and ritual is provided. Properties of each plant (based on the literature) that may be relevant to documented and/ or poten- tial uses are listed, and a brief assessment of the poter tial for is provided. The appendices contain a list of contributors (coded by number, gen- der and cultural affiliation) and a list of voucher specimens. A glossary, list of references cited, and index are also included.

Overall, the book is informative, well-written, highly readable, and serves a broad range of academic, government, commercial, and local interests. The plant use information will be a handy resource for anyone interested in culturally-im- portant plants of the Canadian northwest boreal forest, while the additional research methodologies, cultural information, ecological background and development potential provide a context that will enable the reader to appreciate some of the wider aspects related to documenting Aboriginal plant knowledge and use.

Kelly Bannister

Department of Botany University of British Columbia Vancouver, B. C. V6T 1Z4

Journal of Ethnobiology 20(1): 113-127 Summer 2000

FOLK CLASSIFICATION AND CONSERVATION OF BAMBOO IN XISHUANGBANNA, YUNNAN, SOUTHWEST CHINA

WANG KANGLIN, XU JIANCHU, PEI SHENGJI, AND CHEN SANYANG Department za Ethnobotany, Kunming Institute of Botany he Chinese Academy of Sciences pt 650204, P. R. China aie TEE Indigenous knowledge systems concerning the soneiarrilime vation of bamboo resources in ethnic fon pay an important none in sustainable sien sense Folk seinen en China.

Indigenous peoples classify ea based primarily on plants’ economic uses, orms, growth ha bitats, as well a sociocultural

values, which are strongly nent in biophysical and In addition, indigenous conservation practices that result from long-term mt, ee 1 rt = mA 1 P| 4 1 Me | hl

and efficient production systems for the sustainable utilization of bamboo resources Key words: Folk classification, conservation, bamboo, Xishuangbanna, China

RESUMEN.-Los sistemas de conocimiento indigenas referente a la clasificacién, detec Lael ag ae = An haste mera REA AS

podria ser un factor feeriertanie en cuanto al desarollo sostenible. La clasificaci6n

usada por los campesinos toma una posici6n significante en los sistemas de

identificacién de plantas de las minorias nacionales de la China. Los indigenas

clasifican plantas basdndose primordialmente en su uso econémico, sus

caracteristicas morfolégicas, sus formas de vida, sus habitats de crecimiento,

ademas de sus valores socioculturales que estan fuertemente radicados en 7 ee Pi $5 e_* . 14 1 En adicid 4 ] eee 27

1c10n y ihe eG one’ P ae bles y tes F la utilizacion

ig 1 sostenible de los recursos del bambu.

RESUME.-Les systémes de connaissance indigéne concernant la classification, Videntification, la nomenclature et la protection des ressources en bambou dans les régions ethniques ont un réle important 4 jouer dans le maintien du

aaah La classification is saree est une sassitadertin ie aml du systeme ee Les

“oe “at 1 }

commerciales, leurs caractéristiques es, leurs formes de vie, leurs habitats ainsi que sur des valeurs socioculturelles fortement enracinées eae les environnements et socioculturels. | En outre, les pratiq g e de protection assurent d’une fagon t effi l intien de l'utilisation des ressources en bambou qui proviennent ellesméme d’une interaction a long terme entre les peuples et la nature.

114 KANGLIN et al. Vol. 20, No. 1

INTRODUCTION

1 1 1

Bamboo, together with several groups of I d grass, is clas- sified by taxonomists as the subfamily Bambusoideae within the grass family Gramineae (Poaceae). As ornamental pean and sources of raw material for pa- permaking, textiles, basketry, matting, rope, house truction, furniture, bridges, and fishing equipment, bamboo provides a greater diversity of uses in Asia than any other group of closely related plants (McClure 1956).

Bamboo is an important non-timber forest product (NTFP) with a high com- mercial value. As an important resource, bamboo has been exploited and utilized by various institutions. Heightened attention has resulted from greater recogni- tion of the need for sustainable use of natural resources, and the need to maintain biodiversity while pursuing economic development (Williams et al. 1991). How- ever, increasing demand for the world’s bamboo resources is related to a series of threats to bamboo diversity, and has led to the extinction of a number of bamboo genetic resources. Over-exploitation and habitat destruction of bamboo genetic resources increases these threats. For example, in the Indian Himalayan region, twelve species of bamboo have been marked as rare and endangered due to biotic pressure coupled with biological phenomena such as periodic flowering, poor seed setting and indiscriminate exploitation (Biswas et al. 1997). Qiongzhuea tumidinoda in Yunnan is one bamboo species known for its beautiful culms (stems) and has been exported to south Asia since as early as the ancient Han Dynasty (1,200 years ago). The shoots of this bamboo species are exported to Japan and other countries every year. ae to the over-exploitation of this bamboo for various ornamental,

d handicraft purposes, it is now one of two species of Bambusoideae on the list of Beene Preserved | Plants.

In contrast, indi as the Hani, Dai and Jinuo in Yunnan, nes China have been using traditional methods and strategies of bamboo exploitation that lead to the sustainable utilization and development of bamboo resources. It is therefore important to study the indigenous knowledge systems that relate to the classification, identification, utilization, management and con- servation of bamboo resources.

OBJECTIVES

The objective of this paper is to survey, describe and evaluate bamboo species that are classified and conserved by indigenous communities in Xishuangbanna, unnan Province, China utilizing ethnobotanical methods, and to propose ap- proaches for the sustainable management and conservation of bamboo resources by g relevant indig owledge. Three objectives were therefore identified i in rcs study: (1) to survey and collect indigenous knowledge on classi- fication, nomenclature, and conservation of bamboo resources in indigenous communities; (2) to describe and discuss the folk classification system of bamboo; and (3) to determine and describe the conservation practices of bamboo resources by indigenous communities in Xishuangbanna, Yunnan.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 413

STUDY AREA

The study area is located in Xishuangbanna, Yunnan Province in southwest China. Yunnan Province is an inland and remote province in southwest China, located within 21°8' - 29°15' N and 97°32' - 106°12' E. Xishuangbanna is located in the southwest of Yunnan Province, bordering Myanmar and Laos (Figure 1).

G is located at the southeast end of the Hengduan mountains- the eastern appendages of the Himalayas. Xishuangbanna lies within 21°10' - 22°40' N, and 99°55' - 101°50' E with a total area of 1,922,300 hectares, of which 94% consists of mountainous and hilly terrain, with river valleys making up the remaining area. The elevation is low in the south and high in the north; from the south to the north the elevation rises from 420 m to 2,800 m above sea level. The annual rainfall ranges from 1,138 to 2,431 mm, the annual mean tem- perature varies from 15°C in winter to 22°the C in the summer, and the annual mean relative humidity is between 70 - 80%. The unique landforms and complex physical conditions make Xishuangbanna a diverse ecological environment with various ecosystems. Tropical forests account for 33.8% of the total area cover. The biological resources are so plentiful that Xishuangbanna is known as “the king- dom of wild flora and fauna.” Xishuangbanna is home to the vast majority of

am COUNTY wt JIANGCHENG <1 CUUNLY

LANCANG ~ es iia COUNTY “n

; LEGEND « Mi - Study site e- Town Mengla - Seat ef Mengla County 1 - Menggang Jinghang - Seat of.linghang County 2. Menghux LAOS Menghai-SeatofMenghai County 3 - Gadong

FIGURE 1.- Map of the study area and research sites in Xishuangbanna, Yunnan Province, China

116 KANGLIN et al. Vol. 20, No. 1

plants and animal species found in China. A total of 4669 higher plant species, subspecies or varieties belonging to 1697 genera of 282 families have been recorded in Xishuangbanna (Li et al. 1996). Nearly one sixth of the total species of China (30,000 species) can be found here, although it constitutes only one five hundredth of China’s total land area.

Xishuangbanna is a multi ethnic area. It is comprised of 13 different ethnic groups which include Dai, Hani (also called Aini), Mes Bulang, “A; hes Yao, Wa, Hui, Bai, Zhuang, Miao, Buyi, and other unid (e.g., Kemo, Kemie, and Kongge). Each ethnic group has its own language and folk knowl- edge, especially concerning the utilization, management, and conservation of natural resources. Bamboo is a useful plant and its cultivation is widespread, be- ing used by all of the ethnic communities.

METHODOLOGY

Literature search.— The first stage in this study included a background search for information, specimens, and documentation related to the taxonomy of bamboo and indigenous knowledge systems of folk classification, identification and con- servation of bamboo resources.

Site visits and interviews.— The second stage involved visiting sites and interview- ing local people. In this study, the sites were selected based on three criteria: species, cultural diversity, and economic diversity. The research team visited twenty three study sites (Figure 1) were visited. More than 100 local people were interviewed concerning the distribution, habit, ecological condition, and regeneration of bam- boo, as well as indigenous knowledge systems of folk classification, utilization management and conservation of bamboo. Interviews with local people were con- ducted in three phases. In the first phase the local leaders and experienced persons were asked to recall important and/or memorable bamboo species in the study area. This phase was usually done at night. The interviewers used a previously prepared questionnaire with simple and clear questions (Table 1, part I). The sec- ond set of interviews involved general and specific questions that were related to the species mentioned in the first interview, and were used to serve as a guide in the field survey (Table 1, Part II). The third and final phase of interview stage involved a group interview in which respected members of the local community were asked to assess the nature of use and confirm the local names of bamboo species. The study included community elders as they have a historical perspec- tive of the use of the bamboo resources.

Field observation and specimen collection. Information on culm habit (strictly erect, pendulous or climbing), rhizome system (sympodial or monopodial), culm char- acteristics (height, diameter, branching, node, internode, etc.), culm sheaths (the fifth culm sheath was characterized based on the general appearance, size, texture and shape of sheath and their blades), leaves, and inflorescence (by presentation) was recorded. This field investigation included some accurate and detailed infor- mation, such as local names and their meaning, diagnostic characteristics, distribution, special utilization and conservation practice, and so on. In addition, market data were recorded in detail.

Summer 2000 JOURNAL OF ETHNOBIOLOGY 117

TABLE 1.— Bamboo interview instrument

Part I: General Investigation

Village: Name of Farmer:

Collector(s): Date: Collection No.:

How many bamboo species are in your area? (Lists of local names): 1) Wild: 2) Cultivation:

Which bamboo do you like to cultivate? (Lists of local names):

Where is the bamboo cultivated?: 1. home garden, 2. swidden field, 3. bamboo garden, 4. bank of channel, 5. forest-land, 6. Other

How much bamboo is harvested in one year?: Culms__; Shoots:____ kg

How do you propagate bamboo?: 1. rooted cutting, 2. rhizome-offset, 3. culm cutting, 4. branch cutting, 5. seedling, 6. Other

Remarks:

Part II. Specific Bamboo Species Investigation

Vernacular Name ; Meaning Important Characteristics (for identification): Habitat: 1. cultivated, 2. disturbed, 3. partly disturbed, 4. other Rhizome Types: 1. sympodial (dense clump), 2. monopodial (scattered clump) Culm Habit: 1. erect, 2. arching over 3. decumbent, 4. scandent / climbing

Culm: 1. height m, 2. diameter cm, 3. thickness cm, 4. internode length cm, 5. surface: a) glaucous, b) glabrous, c) hairy, d) spinules present, e) striate, f) color, g) other

Number of Branches: 1. single, 2. two, 3. three, 4. multi-branching: a) with main branch, b) without main branch Culm Sheath: 1.characteristics: a) persistent, b) caducous, c) deciduous d) other 2. texture: a) soft, b) hard, c) leathery, d) others 3. surface: a) glaucous, b) glabrous, c) hairy, d) spinules present and e) other ____ Sheath blade: 1. erect 2. horizontal 3. reflexed, 4. hairy Leaf: 1. large, 2. medium, 3. small, 4. other (specify) Uses: 1. shoot: a) sweet, b) bitter, c) fresh use, d) dry use, e) pickled shoot, f) sour shoot 2. building material: a) pillar, b) scaffolding, c) framework, d) wall or ceilings 3. weaving materials; 4. furniture; 5. agriculture tools, 6. ornamental, 7. medicine 8. music instrument, 9. folk belief, 10. other (specify) Wood Quality Parameters: 1. strength: a) hard, b) soft, c) durable 2. elasticity: a) very good, b) good, c) poor, d) other (specify) 3. smoothness: a) smooth, b) rough, c) other Market Potential: 1. very good, 3. good, 3. poor, 4. other Remarks:

118 KANGLIN et al. Vol. 20, No. 1

Identification and examination of herbarium specimens.— Consultation and examina- tion of herbarium materials of related bamboo species was conducted in various institutions and universities. The collected specimens were identified and exam- ined based on folk classification and scientific taxonomic MaaWIeUS An ethnobotanical inventory was carried out based ont The inventory included scientific names, vernacular names, uses, distribution, and voucher specimens, although the inventory is not included due to the space limi- tations of the paper

FOLK CLASSIFICATION

It is common knowledge that when plants are utilized for any purpose, un- derstanding their value and characteristics are very important. Folk classification has a very important role in the identification system of Yunnan’s and China’s minorities. In China, especially in the indigenous communities of Xishuangbanna, local people classify plants mainly based on local language, production practices, social customs, legends, economic utilization, morphological characteristics, and growth habits, which have very important economic and functional values (Wang and Hsueh 1990). Different ethnic communities may have different folk classifica- tion systems. For example, Hani people in Mengsong of Xishuangbanna use and recognize bamboo through traditional knowledge of bamboo habit, utilization and other characteristics. They understand the differences between erect and climbing bamboo. Erect bamboo is called Ag or Al (in Hani) as the first name, and climbing bamboo is called Hag. Second, they give bamboo lower taxonomic rankings ac- cording to the morphological or utilization characteristics (Tables 2 and 3).

Dai folk classifications systems are more integrated and closer to Western sci- ence than the Hani system. For example, Dai people assign small-leaved se ete spp. to the group Maisang, glaucous-culmed ones to the group Maihe, and group culmed ane hes: sashes spp. to Maishua (Table 4). The Dai folk classification system of bam d exactly to Linnaeus’ taxo- nomic system, but it is similar in the concepts of group, species and subspecies.

In comparison with scientific taxonomy, indigenous folk classification offers some important benefits. (1) Folk classification is often faster to use and simpler than the modern scientific taxonomy. Folk classification names a kind of bamboo

TABLE 2.— Folk classification of bamboo as related to use in Hani Communities of Mengsong, Xishuangbanna

Hani Name Meanings Bamboo Species aqqyul qyul ‘sweet’, bamboo shoot is sweet = Dendrocalamus hamiltonii alhaq hag ‘bitter’, bamboo shoot is bitter Indosasa singulispicula almal mal ‘flute film’, bamboo film used to Phyllostachys mannii make the bamboo flute

aqmiov miov ‘not seen’; it is said “a pig that |= Cephalostachyum fuchsianum eats this bamboo seed will die”

Summer 2000 JOURNAL OF ETHNOBIOLOGY 119

TABLE 3.— Folk cassification of bamboo according to habit and character in Hani Communities of Mengsong, Xishuangbanna

Hani Name Meanings Botanical Name aqpeel peel ‘big’: big bamboo Dendrocalamus giganteus aqbaol baol ‘wild’: wild bamboo Dendrocalamus sp. alnml Inml ‘thorn’: internodes have thorns, _ Chimonocalamus fimbriatus and young shoots are good, abundant _ var. ligulatus and dense

alquq qugq ‘thorn’: but the younger shoots Chimonobambusa yunnanensis grow sparsely var. glabra

agjul jul ‘smooth’: culm is smooth Dendrocalamus strictus

aqljul Ijul: similar to jul, buteven smoother — Fargesia sp.

aqxao-aglan xao ‘stripe’, lan ‘grayish white’: culm —Gigantochloa nigrociliata

gray with stripe

aqxao-xeel _xeel ‘green’: so culm green with stripe Dendrocalamus membranaceus

var. striatus aqyeq yeq ‘black’: culm is black Dendrocalamus sp.

haqgeeq(-mal) hag ‘climbing’, geeq ‘wild’: climbing | Melocalamus compactiflorus amboo with wide culm-node haggeq-aqzaog aqzaog ‘narrow’: climbing bamboo Melocalamus sp. 1 with narrow culm-node haggeq-aqxeel agqxeel ‘thin’: climbing bamboo with Melocalamus sp. 2 thin culm-wal Jeiqnav bamboo shoot very dense and abundant _Indosasa sp.

based on direct observation and evaluative characteristics, whereas scientific clas- sification often requires herbarium study. (2) The local name or folk classification can facilitate communication between local people and researchers. However, skilled ethnobotanists must know and understand the local names of related spe- cies, then use the names and morphological characteristics to match the folk taxon with a scientific name or names. (3) Folk names are often related to use, and to characteristics and first-hand experience. Folk classification can offer important clues to the exploitation of the resource by local people, and thus is useful for commercial and government planning, and to scientific researchers. .

However, folk classification, in comparison with scientific classification, has a number of disadvantages. (1) Heterogeneity: similar bamboo species within a vil- lage or other villages may be referred to by different name (Table 5), or different bamboo species may be given a similar local name. (2) Limitation: bamboo folk classification, like all other forms of indigenous systems, has its limitations. Hani people divide bamboo into only two types: erect bamboo (Aq or Al) and climbing bamboo (Haqgeeg).

120 KANGLIN et al. Vol. 20, No. 1

TABLE 4.— Group and individual names for bamboo species in a Dai Community

Group name Individual Name Bamboo Species

Maisang l

Maisanghe D. membranaceus f. fimbriligulatus

Maisanglai D. membranaceus f. striatus Maisang:Dendrocalamus spp. Maisanghuan D. membranaceus f. crinitus with smaller leaves Maisanghai D. membranaceus f. bigemmatus

Maisangkou D. membranaceus f. pilosus

Maisangdaben D. membranaceus var. sulcatus

Maisanglan D. barbatus

Maisangbo D. albostriatus

Maithegai D. semiscandens

Maihelao D. brandisii

Maihelan D. brandisii f. hispiatus Maihe: Dendrocalamus spp. Maihezhang D. hookeri with glaucous or white haired Maihelong D. hamiltonii var. serratus culm. Maihegaihao D. longiligulatus

Maihemen D. longiligulatus f. lacanus

] D. longiligulatus f. striatus

Maishua Gigantochloa nigrociliata Maishua: Gigantochloa spp., Maiheshua G. felix with striated culm Maishuahei G. sp. 1

Maishuanai G. sp. 2

INDIGENOUS PROPAGATION PRACTICES

Cultivation of anumber of bamboo species around houses, villages and fields is a tradition of many ethnic communities and individuals in Xishuangbanna. For example, the Hani families in Mengsong cultivate a bea number of bamboos (Table 6, based on semi-structured interviews and sam A local community usually prefers to plant bamboo ‘and rattan together, because the bamboo supports the rattan. In recent years, ginger and tobacco have been intercropped with younger bamboo clumps for economic and ecological reasons.

Rooting of culm cuttings is traditionally used in propagating bamboo by eth- nic communities of Xishuangbanna. However, Hani people in Mengsong commonly plant branch-cuttings of Aqqyul (Dendrocalamus hamiltonii) in swidden fields. They believe that this method results in slowing the growth of bamboo.

Bamboo plays an important role in the local economy. Table 7 shows economic statistics from the Menglong town government. Although these figures are not exact, they show the role of bamboo in indigenous communities. Other case stud- ies have shown that bamboo and rattan together occupy the seventh position in the local economy based on semi-structured questionnaires (Wang 1998).

Summer 2000

JOURNAL OF ETHNOBIOLOGY

121

TABLE 5.— Names with translation, of Indosasa singulispicula in different ethnic

communities Ethnic communities Local Name Meanings Dai Commmunity Maihong Mai ‘bamboo’, hong ‘bitter’ Hani Community Alhag Al ‘erect bamboo’, hag ‘bitter’ Aini Community Raha Ra ‘erect bamboo’, hag ‘bitter’

Yao Community

Laoying-zhang

Kucong Community | Wakada

Lao ‘bamboo’, ying ‘bitter’, zhang ‘small’

Wa ‘bamboo’, ka ‘bitter’, da ‘shoot’

TABLE 6.— The status of cultivation and utilization of bamboo in Mengsong Villages

Item sample No. of groves No. of species No. of shoots/ No. of culms used per household groves/year per household 1 20 6 17.5 75 2 9 3 - “ 3 12 4 10 75 4 30 4 20 30 5 20 4 20 50 6 20 5 25 100 7 i? 5 5 25 8 25 - 12 60 9 25 + 10 75 10 10 6 10 50 11 10 4 15 50 12 150 6 20 100 ig 35 3 3 100 14 40 50 15 100 15 10 5 10 50 16 50 2 20 100 17 80 6 20 50 18 7 3 8 35 19 100 5 10 50 20 15 5 20 40 21 10 5 10 35 Ze 18 5 10 20 23 8 + 10 45 24 9 ) 8 50 25 6 2 10 20 26 50 5 15 50 27 20 5 18 100 28 30 5 30 100 29 100 - = 50 30 45 6 10 225 31 10 3 10 20 Total 991 140 426.5 1993 Average 32 4.5 14.22 64.33

122 KANGLIN et al. Vol. 20, No. 1

TABLE 7.- Statistics of bamboo production in Mengsong, Xishuangbanna

No. of cut or harvested culms Dry shoots No. of weaving Total value Year Large Small k products* US$) 1985 71100 - afae 648 53,192 1986 12920 - 3896 940 12,534 1987 13997 7550 3832 2035 18,130 1988 12920 7550 3896 970 15,182 * Note:

or

CONSERVATION

Practices.— All ethnic communities in Xishuangbanna have a long tradition of plant- ing, managing and conserving bamboo. Bamboo forests are classified into three types of property: national, community and family/individual bamboo forests. These are managed or protected by different agencies or individuals.

Based on the social, economic, cultural, and ecological surveys and studies, some efficient and reasonable management and conservation of bamboo genetic resources in pe oe are described below.

Pa Bd 1 ee | <

In-situ conservation.— and national policies of China warrant the conservation of genetic resources of bamboo in China. Especially in minority areas, in situ conservation, through cultivation of bamboo plants among or along the village boundaries through establishment of bamboo gardens, and the maintenance of National Nature Reserves, are efficient methods of protecting and conserving the bamboo germplasm of China.

Community based conservation. Many bamboo species, naturally distributed and cultivated, are protected in community bamboo forests. These community forests are especially maintained by local religious or belief groups. Sangpabawa (the community protected forest in Hani communities), Nong Man or Nong Meng (community Holy Hills), Nong Ban (village Holy Hills in Dai ethnic groups), Ba Hao and Lao Ben (Graveyard forests, in Dai and Hani communities, respectively), Gai Mei San Ha (Water-source forests in Hani communities) are protected and well managed. There are also many Buddhist communities and Buddhist temples in Xishuangbanna that conserve bamboo resources. There are four requirements for establishing a temple (Wa in Dai language) which are as follows: (a) a statue of Sakyamuni (Pagodama-Zhao, in Dai language), the founder of Buddhism, (b) a pagoda in which Sakyamuni’s ashes can be preserved, (c) at least five monks, and (d) the presence of some specified “temple-yard plants” (Pei 1991). Based on these requirements, many plants have been cultivated inside the temple, some of which include bamboo species such as Thyrsostachys siamensis, Bambusa sinospinosa, Dendrocalamus hamiltonii, and Phyllostachys mannii.

Family /individual based conservation. Home gardens, bamboo gardens (Aqpeya, in Hani community), agroforestry practices, and swidden farming form a series of traditional systems for sustainable cultivation, management and con- servation of bamboo resources. The bamboo clumps or forests belonging to every family are cut, managed, and conserved by the families. For example, the Dai people plant the multi-purpose bamboo species in the home garden or near the village,

Summer 2000 JOURNAL OF ETHNOBIOLOGY 123

and Hani people establish a bamboo garden (Aqpeya) or cultivate bamboo in the swidden system, and in Sangpabawa.

In general, more valuable bamboo species, such as Dendrocalamus giganteus, D. hamiltonii, D. barbatus, D. membranaceus, D. calostachyus, Indosasa singulispicula, Bambusa lapidea, Cephalostachyum pergracile, and Schizostachyum funghomii are planted inside home gardens or bamboo gardens.

Swidden cultivation is practiced among all the mountain ethnic groups in the eastern Himalayan region. In Mengsong of Xishuangbanna, the Hani people have developed an agroforestry system. They cultivate bamboo in the swidden lands, and plant rattan using the bamboo clumps as support for the climbing rattan. Tobacco is intercropped among the bamboo clumps. The bamboo branch or culm is sometimes burned and the ash is used to fertilize tobacco. In other cases, bam- boo is intercropped with maize, beans, vegetables and other crops in the agroforestry system of the Hani community of Mengsong. Short, medium and long-term income is obtained from the bamboo. The sloping swidden land is used for long-term agriculture products, thanks to the conservation of soil by the well developed roots-system of bamboo plants.

State conservation. National bamboo forests are grown and maintained in China’s National Nature Reserves. These bamboo species, along with other con- served plants in the reserves, are protected from commercial activities.

Ex situ conservation. Ethnobotanical information is considered useful to add to the information on collected plants in living-plant collections, seed-germplasm collections and herbarium collections. Data on vernacular (local) name of plants, indigenous uses of plants, ecological knowledge of plants, and local management practices of those plants are valuable. New information, which is generated from ethnobotanical expeditions, helps botanical gardens make plant lists for new col- lections (Pei 1994). In Yunnan and Xishuangbanna, the bamboo germplasm collection in the Xishuangbanna Tropical Botanic Garden is important for ex situ conservation of bamboo resources. According to this study, more than 100 species, varieties, and forms of bamboo belonging to 19 genera have been planted in this garden and are growing well.

Control of over-harvesting.— There are 91,800 ha of natural bamboo forests, or bam- boo/tree mixed forests in Xishuangbanna that belong to the state, communities, and individuals (Wang and Hsueh 1992). These areas are managed and harvested (bamboo shoots and bamboo culms) to protect the watersheds and conserve bam- boo gene pools. The traditional harvesting of bamboo shoots and culms from forest lands has been sustainable when these lands have been harvested for villager’s home consumption, rather than for commercial purposes.

CONSERVATION ISSUES Destruction of environment and natural habitat. Destruction of the environment and

natural habitat due to logging, shifting cultivation, and other land uses have led to the extinction of some bamboo species and depletion of bamboo forest resources.

124 KANGLIN et al. Vol. 20, No. 1

In some parts of Xishuangbanna, shifting cultivation has primarily destroyed bam- boo resources as a result of unscrupulous cutting of large areas of bamboo forest for agricultural land use (Wang and Hsueh 1992).

e conspicuous case of habitat destruction was the establishment of rubber plantations on more than 43% of Jinghong farms, which were previously natural bamboo forest or mixed bamboo/tree forests. Moreover, large bamboo areas have been converted to plantations of economic crops and plants, such as banana, cof- fee, tea, and fruit trees.

Over-exploitation and limited distribution.— In many cases, over-exploitation and lack of artificial propagation are threatening bamboo resources. For example, it is well known that the raw materials for papermaking, production of chopsticks and woven products in Xishuangbanna are supported only by the natural bamboo forests (Wang et al. 1993) he other hand, unreasonable and inefficient exploitation and utilization of bamboo resources have also led to the degeneration of natural bamboo stands. New riesgo workers who work for national farms or are employed. By local Hunan and other d shoots

mostly from the natural bamboo forests. Insufficient knowledge on how to use and manage bamboo resources can lead to inefficient utilization and overexploitation. Rhizome cutting which deviates from the traditional methods of propagation, and culm cutting and shoot gathering by choosing only the large culms and stocky shoots, lead to the deterioration of bamboo species resources.

CONCLUSION

Indigenous knowledge concerning the identification, classification, nomen- clature and conservation of bamboo plants have several important roles that are reviewed below.

Vernacular and scientific names.— The local plant nomenclature existed previously in oral tradition (Kelly and Dickinson 1985). Vernacular names of plant species are very important to plant resource inventories. They are based on indigenous people’s productive practice, social customs, legends, and economic utilization. Habit and morphology of plants also have economic and functional significance. It should be noted, however, that sometimes those vernacular or local names are not con- sidered very ns for id bamboo because they are often not reliable. Therefore, great caution is required in the interpretation of vernacular names (Dransfield and Widjaja 1995).

Validity of folk classification— There is a need to recognize the value and scientific validity of folk classification. Recognition and methods of use of plant resources by indigenous people are similar to those of the scientific world. For example, indigenous peoples classify plants, usually using morphological characteristics and physiological attributes, with consideration of their growing and reproduc- tive habits.

Deep knowledge and utilization of indigenous knowledge.— It is important to recognize

Summer 2000 JOURNAL OF ETHNOBIOLOGY 125

that there is more detailed knowledge contained in some folk classification sys- tems as compared to scientific classification. For example, local soil taxonomy of African farmers is based on soil characteristics as they relate to specific crops and,

traditionally, provides the insight and ecological knowledge required for making good use of available agricultural resources (Richards 1985, cited in Dialla 1993).

Dialla (1993) believes that the Mossi farmers classify soils in terms of cropping potential. The indigenous soil mecca May serve as a a complementary tool to scientifically based systems, an allows Mossi farmers to make appropriate use of their land. Conklin (1954) concluded that plant names are significant to the Hanunoo (Mindoro Island of the Philippines), not only as convenient labels for recognized plant segregates but also for the semantic asso- ciations of the names employed. The application of traditional knowledge could substantially increase previous scientific information related to flora, fauna, and an saa The Hanunoo could identify approximately 1,600 different varieties of plan ad recorded only 1,200 species (Conklin ene This difference reflects the fact that the Hanunoo taxonomy employs differ- ent principles from those followed in Linnaeus’ classification, grouping plants according to life form rather than in terms of genetic relationships (Rambo 1984). Similarly, Dai in Xishuangbanna divided a bamboo species (Dendrocalamus membranaceus Munro) into two types soft and hard types according to culm tex-

Conservation.— Indigenous knowledge or practices and indigenous communities have played a significant role for in situ conservation of bamboo resources. Home gardens, bamboo gardens and swidden cultivation systems demonstrate impor- tant conservation practices. In addition, recognition of the rights of local communities is a very important step in achieving the reasonable use, efficient conservation, and sustainable management of bamboo resources.

RECOMMENDATIONS

The following recommendations are made for future research and develop- ment.

Survey.— More interviews with indigenous communities and field surveys in other remote areas of China are necessary not only for species classification, identifica- tion and distribution of bamboo, but also to understand the cultural knowledge and uses of this important resource.

Conservation— Plans and programs on the conservation of natural resource and cultural diversity should be prepared on a national scale, but with specific recom- mendation for each tribal group. Development plans should emphasize efficient conservation and sustainable exploitation of bamboo resources. The involvement and cooperation of different levels of beneficiaries is also necessary.

Dissemination of technology. Some modern management methods and technolo-

gies can improve the efficiency of resource use, increase income to indigenous communities, and promote the improvement of well being in these communities.

126 KANGLIN et al. Vol. 20, No. 1

Py hy Pe

It is therefore imperative to new systems that will link indigenous knowl- edge and modern technologies for more sustainable resource development. For example, improved/modern methods of bamboo propagation (e.g. culm cutting, branch cutting, and attendant silvicultural practices) may be introduced to the communities.

Financial support for in situ and ex situ conseroation.— Conservation of bamboo ge- netic resources is very important for in the future, but may be difficult due to the complexity of implementation strategies that may be af-

fected by the diversity of resources, cultures, and financial incentives.

ACKNOWLEDGMENTS

Wang Kanglin would like to acknowledge the late Professor Hsueh Chiju, a botanist, ecologist, and forester of Southwest Forestry College (SWFC), who initially instilled in the mind of the author = science ce art of collection, classification, identification and

nd inspired him to write and publish the research results. He also expresses his gratitude to the guidance and considerations by Dr. Mercedes U. Garcia, Dr. Norma O. Aguilar, Dr. Juan M. Pulhin, Dr. Ernesto P. Militante and Dr. Myrna G. Carandang (University of the Philippines Los Banos). He would also like to thank the financial support by World Wildlife Fund (WWE - USA), and Therese Grinter for the revision of this paper.

We would like to thank all of people in indigenous communities of Xishuangbanna who have shared their extremely important knowledge with us.

LITERATURE CITED

BISWAS, SAS, S. CHANDRA and S. S. JAIN, 1997. En of

Rao d J. T. Williams (editors). ieienniiel Network for Bamboo and Rattan, India.

CONKLIN, HAROLD C. 1954. The Relation of Hanunoo Culture to the Plant World. PhD dissertation at Yale university. Microfilm-xerography in 1970 by University Microfilms, Inc., Ann Arbor, ig" U.S.A.

————-» 1957, Hanunoo Agriculture: =

Report onan

Cultivation in the Philippines. FAO

Forestry Development Paper No.

Food and Agriculture peace

Rome. (Reprinted by Elliot’s Books,

Northford, Connecticut, 1975).

DIALLA, BASGA E. 1993. The se in Burkin

Faso. Indigenous eis pe

Development Monitor a 17-18

DRANSFIELD and E WIDJAJA (editors). 1995. yee pleas of South-East Asia, No. 7: Bamboos. Backhuys Publishers, Leiden.

KELLY, D. L. and T. M. DICKINSON. 1985. Local names for vascular plants in the ni Crow Mountain, Jamaica.

ae Botany 39(3): 346-362.

LI YANHUI et al. (editors). 1996. List of orang in Xishuangbanna. Yunnan ait Rina a of National Minorities,

ane. McCLURE, F. A. 1956. Bamboos in the nomy of Oriental people. Economic Bony 10(4): 335-361.

PEI, SHENGJI. 1991. Conservation of biological diversity in temple-yard and holy hill by Dai ethnic minorities of China. Economic Botany 3: 27-35

Summer 2000

———— . 1994. Ethnobotany in the new development of botanical garden. Pp. 16-22 in Collected Research Sige on the Tropical Botany, Z. F. Xu et al. (editors), Yunnan nieces Press,

unming. RAMBO, A. TERRY. 1984. Human ecology research by social scientists on tropical agroecosystems. Pp. 39-59 in An n to Human Ecology Research on Agricultural System in Southeast Asia, A. T. Rambo and P. E. Sajise (editors). University of the Philippines Los Banos, Philippines. RICHARDS, 19 Indigenous Agricultural Revolution: Ecology and Food Production in West Africa. Westview Press, Boulder. WANG, JIANHUA. 1998. Traditional Culture and pega dard Management

of Mountain Ethni Group in Xishuangbanna: A case study of Mengsong ani community.

Unpublished Master thesis, Kunming Institute of Botany, the Chinese cademy of Sciences, Kunming. Chinese with English abstract. WANG, KANGLIN and CHIRU HSUEH. 1990. oe studies of bamboo resource engs Xishu angi ste, " Weewiart China Bamboo Research 10(2): 32-40

JOURNAL OF ETHNOBIOLOGY 127

atari 2. An approach to the emaiunetin of bamboo food of minority i

Natural Resources 1992(5): 67-72

, XUE JIRU, CHENG SANYANG, PELs SHENGII AND AI KANLA. 1993. Ethnobotanical puns of bamboo resources in Xishua nna, Yunnan, China. Pp. 47-65 in Collected Research Papers on the Tropical Botany, No. 2, Xu Zaifu et al. (editors). Yunnan University Press, Kunming (Chinese with English

abstract).

WILLIAMS, J. T., J. DRANSFIELD, P. M. GANAPATHY, W. LIESE, SALLEH M. NOR and C. B. SASTRY. 1991. Research needs for bamboo and rattan to the year 2000. Report Submitted to IDRC by the Tropical gra International Fund for Agricultural Research.

LA

__& W783

NOTICE TO AUTHORS - The Journal of Ethnobiology’s revised “ ahaa for Authors” appears in volume 19, ———— == 7% a (winter 1999). If you need a copy of these Guidelines you may request a copy from the Editor.

Careful scrutiny of recent issues of the ioe should provide appropriate stylistic models for any manuscript you may wish to submit.

Those submitting manuscripts for consideration for publication in the Journal should send three hard copies and one copy ona high ad a. oe in ee or mote format with original camera-ready figures. M ppror d format will be

returned.

Manuscripts should be sent to:

WILLIAM BALEE Editor, snails of Ethnobiology Department of Anthropology ae te rsity ew Orleans, ey aoe USA (wbalee@mailhost.tcs.tulane.edu)

BOOK REVIEWS We welcome suggestions on books to review or actual reviews from readers of the Journal. If you submit a book review for consideration please send two hard copies and one on diskette. Send suggestions, comments, or reviews to:

MICHAEL K. STEINBERG Book Review Editor, Journal of : nieces hy-Anthropology

(mstein@usm.maine.edu)

SUBSCRIPTIONS

Subscriptions to the Journal of Ethnobiology should be addressed to Virginia Popper, UCLA Institute of Archaeology, Box 951510, Fowler A-210, Los Angeles, CA 90095-1510 USA. Subscription rates are $60 for institutional subscriptions; $25 for students; $35 for regular individual subscriptions except for Latin America, for which subscriptions are $25. Joint members (with spouse, receiving a single copy of the Journal) pay $35. For postage outside of the USA, Canada, or Mexico, add $10. Make checks payable to the Journal of Ethnobiology. Defective or lost copies will be eeplaced if a written re roguest is Teceee heiomaaii one ties of issue. For information on back issues contact Cecil Brown,

gy, Northern Illinois University, DeKalb, IL 60115 USA, (815) 753-0246.

r

CONTENTS

ETHNOBIOTICA 2

MATSES INDIAN RAINFOREST HABITAT CLASSIFICATION AND MAMMALIAN DIVERSITY IN AMAZONIAN PERU Fleck and Harder 1

ETHNOECOLOGY OF WHITE GRUBS (COLEOPTERA: MELOLONTHIDAE) AMONG THE TZELTAL MAYA OF CHIAPAS Gomez et al. 43

IDENTITY AND CURRENT ETHNOBOTANIC KNOWLEDGE OF FRANCISCO HERNANDEZ’S “CICIMATIC” Ochoterena-Booth 61

MANAGEMENT OF TREES USED IN MURSIK (FERMENTED MILK) PRODUCTION IN TRANS-NZOIA DISTRIC, KENYA Mureithi et al. 75

FAUNISTIC RESOURCES USED AS MEDICINES BY ARTISANAL FISHERMEN FROM SIRIBINHA BEACH, STATE OF BAHIA, BRAZIL Costa-Neto and Marques 95

FOLK CLASSIFICATION AND CONSERVATION OF BAMBOO IN XISHUANGBANNA, YUNNAN, SOUTHWEST CHINA Kanglin et al. 113

BOOK REVIEWS a7, #2, 110, 112