JOURNAL (of) ETHNOBIOLOGY Alune Arachnophagy and Approaches to Spiders Among an Eastern Indonesian People - Healey and Florey “Head of Snake, Wings of Butterfly, and Body of Cicada”: Impressions of the Lantern-fly (Hemiptera: Fulgoridae) in the Village of Pedra Branca, Bahia State, Brazil - Costa-Neto and Pacheco Markets and the Use of Wild Animals for Traditional Medicine: A Case Study Among the Tsimane’ Amerindians of the Bolivian Rain Forest - Apaza, Godoy, Wilkie, Byron, Huanca, Leonard, Peréz, Reyes-Garcia, and Vadez The Grammar of Snchitsu'umshtsn (Coeur d'Alene) Plant Names - Palmer, Kinkade, and Turner Shellfish Assemblages from Two Limestone Quarries in the Palau Islands - Fitzpatrick The Use of Driftwood on the North Pacific Coast: An Example from Southeast Alaska - Lepofsky, Lyons, and Moss Wood Remains from Andean Argentina: The Use of Prosopis sp. L. in Hut Construction - Capparelli, Zagorodny, and Balesta Volume 23, Number 1 Spring/Summer 2003 JOURNAL STAFF EDITOR: eee EF. Miller, University of gal gies Museum-MASCA, 33rd and Spruce Sts., Philadelphia, PA 04 (nmillerO@sas.upenn.e BOOK ieueae EDITOR: Darro 5 Colina L tin American and Caribbean Secretariat, World Wildlife Fund, 1250 24th t., NW, Washington, DC 0037-1193 (darron.collins@wwfus.or SOCIETY OFFICERS PRESIDENT: Jan Timbrook, Santa Barbara Museum of Natural History, Santa Barbara, CA PRESIDENT-ELECT: Eugene Hunn, Department of Anthropology, University of Washington, Seattle, WA SECRETARY/TREASURER: C. Margaret Scarry, Department of Anthropology, University of North Carolina, Chapel ill, NC (scarry@email.unc.edu) BOARD OF TRUSTEES al apes Johnson, University of Alberta, Edmonton, Alberta, Canada John peth, University of Michigan, Ann Arbor, MI Dana pps Simon Fraser University, Burnaby, B.C., Canada SOCIETY COORDINATORS Sr aclpareisto COORDINATOR: Mollie S. Toll, Museum of New Mexico, Office of Archaeological Studies, Box 504 7, Santa Fe, NM 8 WEBSITE COORDINATOR: Douglas Trainor (trainor@ethnobiology.org) Ex Officio Past — Steven A. Weber, Amadeo M. Rea, Elizabeth S. Wing, Paul E. Digciag Cecil Brown, Catherine S. owler, Nancy J. Turner, Deborah M. Pearsall, and Karen R. Adam a Pa ppd Steven D. Emslie Editor, conference coordinator, and website co-ordinator EDITORIAL BOARD Eugene N. Anderson, University of California, Riverside, CA: ethnobotany, China, Maya Scott — ace Paris, France: Tiana it classification, ethnoecology, cognition, evolutionary psychology, history of - ence, Maya, Middle E William Balée, Tulane aare New mangas we oe ea ee orn Brent Berlin, University of Georgia, Athens, G botany, May Robert A. Bye, Jr., Jardin Botanico, aimee Nacional Aubin de Mision, México, D.F.: i wanes Mexico - George Estabrook, University of Michigan, Ann Arbor, MI: Mediterranean peasant agriculture, Portuguese speaking world, culturally informed technology, computational methods Nina Etkin, ——— of Hawaii, Honolulu, HI: medical ethnobotany, ethnoph l thnobiology of food, West r oY v °o o Pacific Gayle Fritz, Washington University, St. Louis, MO: paleoethnobot gricultural origi d developments, North Am Chris Healey, pages for Indigenous Natural and Cultural Resource Management, Northern Territory University, arwin, Australia: ethnobiological classification, tetas Melanesia, iesncnscen ae Brian suse: chang grad of Alabama at Birmingham, Birmingham s, Near East, Andes Eugene Hunn, University of hid, oo Seattle, WA: ethnobiological classification, ethnoecology, Sahaptin and Zapotec ages, subsistence use of pa Timothy joke Macdonald College of ai University, Montreal, Quebec: Africa and the Americas, ethnobotany, medicinal plants, nutrition Harriet V. Kuhnlein, McGill University, Quebec, Canada: Ethnoflasman — sani yas so Peoples David L. Lentz, Chicago cog en seman i Mes erica A Brien Meilleur, CNRS, Paris, e: ethnoecol ional knowled Elizabeth Reitz, University is Cons Athens: wansieiucs ation, pastoral sys- ov f vs entral Asia plant tion, Alpine Europe, Polynesia Editorial consultant: Elia San Miguel (Spanish), Real Jardin Botanico de Madrid, Spain The Journal of Ethnobiology is published semi-annually by the a of ahem Sew ethienbiolegy: org): Manuscripts for publication and book reviews should cover of this issue. © Society of Ethnobiology ISSN 0278-0771 COVER ILLUSTRATION: Stone money disk (Fitzpatrick, this issue Figure 2). MISSOURI BOTANICAL AUG 1 4 2093 GARDEN LIBRARY Journal of Ethnobiology VOLUME 23, NUMBER 1 SPRING/SUMMER 2003 ie een Sa PRE ii ced 0 ae die eit Be teal anal ee ie ne i CONTENTS ETHNOBIOTICA ALUNE ARACHNOPHAGY AND APPROACHES TO SPIDERS AMONG AN EASTERN INDONESIAN PEOPLE Christopher Healey and Margaret Florey “HEAD ei SNAKE, a OF BUTTERFLY, AND BODY OF CICADA’: S OF THE LANTERN-FLY (HEMIPTERA: cies IN THE VILLAGE OF PEDRA BRANCA, BAHIA STATE, BRAZIL Eraldo Medeiros Costa-Neto and Josué Marques Pacheco MARKETS AND THE USE OF WILD ANIMALS FOR TRADITIONAL MEDICINE: A CASE STUDY AMONG THE TSIMANE’ AMERINDIANS OF THE BOLIVIAN RAIN FOREST Lilian Apaza, Ricardo Godoy, David Wilkie, Elizabeth Byron, Tomds Huanca, William R. Leonard, Eddy Peréz, Victoria Reyes-Garcia, and Vincent Vadez THE GRAMMAR OF SNCHITSU’UMSHTSN (COEUR D’ALENE) PLANT NAMES Gary B. Palmer, M. Dale Kinkade, and Nancy J. Turner SHELLFISH ASSEMBLAGES FROM TWO LIMESTONE QUARRIES IN THE PALAU ISLANDS Scott M. Fitzpatrick THE USE OF DRIFTWOOD ON THE NORTH PACIFIC COAST: AN EXAMPLE FROM SOUTHEAST ALASKA Dana Lepofsky, Natasha Lyons, and Madonna L. Moss WOOD REMAINS FROM ANDEAN ARGENTINA: THE USE OF PROSOPIS SP. L. IN HUT CONSTRUCTION Aylen Capparelli, Nora Zagorodny, and Barbara Balesta BOOK REVIEWS 23 101 125 143 155 Vinge ETHNOBIOTICA The 2003 annual meeting provided a wonderful opportunity to see old friends and meet new colleagues at the beginning of spring in Seattle. For many of us, spring floods are a herald of change. Unusually high, they stand for destruction; as cyclical occurrences, they signify renewal. For example, 5000 years ago, the waters of the Tigris and Euphrates allowed Mesopotamian civilization to flourish. To this day, for people living in Iraq (“birthplace of civi- lization’), floods are more than just metaphors—the waters of the rivers and the sea have a direct impact on life. Gulf War II began a few days before our meeting; juxtaposed against the conference theme, ‘’Ethnobiology and Sustainability,’ I couldn’t help thinking about the different faces of change—negative, positive, and cyclical. In lower Mes- opotamia, the Marsh Arabs’ traditional management of the marshes at the head of the Persian Gulf persisted for millennia, the epitome of sustainability. The draining of the marshes in the 1990s was disastrous for the people and the en- vironment. Perhaps peace will now allow some part of this unique cultural habitat to be reestablished as a sustainable system.* ~~ “Three men making baskets, and one man ene a brave of ey | in front of native hut on Sunday.” Nippur, Mesopotamia (modern Iraq), Ottoman Empire, 1899. Photograph by John Henry Haynes. (Courtesy University of Pennsylvania Museum, UPMAA neg. G5-6911 * For information and background: UNEP. 2003. The Mesopotamian marshlands of the Tigris-Euphrates delta. United Nations pcg Programme. [Online: http://www.grid.unep.ch/activities /sustainable/ is/marshlands/, March 22, 2003] (verified May 8, 2003). tenes ager, Edward. 1998. Life on the edge of the marshes. Expedition 40(2):29-39. Thesiger, Wilfred. 1964. The Marsh Arabs. E.P. Dutton, New York. pee ADTYOR oy errail aati RS tisiew nog ‘ 938 egies 2 pet ry my ayn art Of denwoli & erst ast + ~-e1oriqatom tent nedt ssom sig 2! SS aye wal 6 nayod i stew Yiu 2 Dn \(goloidien wee + fags Bs Sn RD Hd PAS ar Siri? peach —tegetety te est treseolit > act tax ds — yitieng. fears | ATT oe Grek sigs nO? ay ah ane oh Nene tess spina: ail} to Pracy oes vat, veal we WT 20 Sle aqrni jortib fi stalk Roe Journal of Ethnobiology 23(1): 1-22 Spring/Summer 2003 ALUNE ARACHNOPHAGY AND APPROACHES TO SPIDERS AMONG AN EASTERN INDONESIAN PEOPLE CHRISTOPHER HEALEY? and MARGARET FLOREY® * Centre for Indigenous Natural and Cultural Resource Management, Northern Territory University, Darwin, NT 0909, Australia > School of Languages, Cultures and Linguistics, Monash University, Victoria 3800, Australia ABSTRACT.—Despite the close association between spiders and human com- munities, ethnobiologists and anthropologists have paid little attention to know]l- edge and use of spiders. This paper briefly reviews ethnographic information on human-spider interactions. We contend that spiders are intrinsically no less edible or otherwise useful than many other arthropods, and that the lack of reporting on spiders in the ethnobiological literature may reflect inattention on the part of observers rather than actual indigenous non-utilization of a potential resource. The paper then provides an example of ethnobiological knowledge and use of spiders among the Alune-speaking people of Seram Island, eastern Indonesia. Several Alune taxa of spiders are considered edible. While their contribution to the diet is undoubtedly minor, we suggest that spider eating is an element of a ‘small foods’ tradition of a mixed horticultural and foraging subsistence-oriented economy. Knowledge and use of spiders are components of Alune identity as forest-dwelling people, and of traditional patterns of livelihood that retain rele- vance in the contemporary context of widespread social unrest. The Alune ex- ample suggests the possible value of attention to spiders as dietary and medicinal items in the subsistence strategies of other indigenous cultures. Key words: Alune, Seram, arachnids, entomophagy, ethnozoological classification. RESUMEN.—A pesar de la estrecha asociaci6n entre las comunidades de arafias y las humanas, los etnobidlogos y antropdlogos han dedicado escasa atencion al conocimiento y uso de las arafias. Este articulo revisa brevemente la informacién etnografica sobre interacciones entre humanos y arafias. Sostenemos que las ara- fias no son intrinsecamente menos comestibles, o titiles en otros sentidos, que muchos otros artr6podos, y que la ausencia de informacién sobre las arafias en la bibliografia etnobiolégica puede reflejar mas una falta de atencién por parte de los observadores que un desuso real por los indigenas de este recurso poten- cial. El articulo proporciona seguidamente un ejemplo de saber y utilizacién et- nobiolégica de las arafias por los hablantes Alune de la Isla de Seram, en el este de Indonesia. Varios taxones Alune de arafias se consideran comestibles. Si bien su contribuci6n a la dieta es sin duda secundaria, sugerimos que la ingesti6n de arafias es un elemento mas de una tradicién de “‘pequefios alimentos’ dentro de una economia mixta de horticultura y recoleccién orientada a la subsistencia. El conocimiento y el uso de las arafias son componentes de la identidad de los Alune como habitantes del bosque, y de unos patrones tradicionales de vida que man- tienen su relevancia en el contexto actual de inestabilidad social generalizada. El ejemplo Alune sugiere la posible importancia del estudio de las arafias como 2 HEALEY and FLOREY Vol. 23, No. 1 elementos alimenticios y medicinales en las estrategias de subsistencia de otras culturas indigenas. RESUME.—En dépit du fait qu’araignées et étres humains cohabitent, ethnobio- logistes et anthropologues ont rarement étudié la conception qu’ont les indigénes de ces insectes et comment ils les utilisent. Cet article examine brievement nos informations ethnographiques relatives aux interactions entre étres humains et araignées. A notre avis, les araignées sont fondamentalement aussi comestibles et aussi utiles que beaucoup d’autres arthropodes. Le manque d’études ethnobiolo- giques consacrées a ce sujet est probablement di plus 4 |’indifférence des obser- vateurs qu’a l’absence véritable d’utilisation de ces insectes par les indigénes. Cet article présente ensuite un exemple d’exploitation et de connaissance ethnobio- logiques des araignées par les habitants de langue Alune de L’ile de Séram en Indonésie orientale. Plusieurs taxons d’araignées sur le territoire des Alune sont considérés comme comestibles. Leur contribution au régime alimentaire est cer- tainement mineure. Nous suggérons cependant que du point de vue alimentaire, les araignées fait partie des ‘‘nourritures accessoires” traditionnelles d’une éco- nomie de subsistance mixte cueillette/horticulture. La connaissance et l’exploi- tation des araignées font partie de l’identité des Alune en tant qu’habitants de la forét et des schémas traditionnels de moyens d’existence qui sont toujours im- portants dans le contexte contemporain d’une agitation sociale diffuse. L’exemple des Alune suggére que des études devraient étre consacrées a la fonction alimen- taire et médicinale des araignées dans les économies de subsistance d’autres cul- tures indigénes. INTRODUCTION Spiders often live in close association with humans, and are variously objects of interest, danger and aversion. Few ethnobiologists, however, have paid much serious attention to human knowledge of, and interactions with spiders. As a contribution towards an ethnographic understanding of human-spider relations, this paper documents the uses and ethnotaxonomy of spiders among the Alune people of Seram Island in Indonesia. It is now widely recognized in the anthropological literature that insects have formed an important part of the diet of human communities in many parts of the globe. Academic attention to the consumption and other uses of insects has been subject to the filtering lenses of Judaeo-Christian traditions and European gastronomy, which, with a few celebrated exceptions (honey, crayfish and escargots come to mind), eschew the consumption of non-marine invertebrates and their products. The Biblical vision of John the Baptist living a wandering existence in the desert, subsisting on locusts and wild honey, is a symbolic representation of his separation from the established social order. His diet combines the sublime honey, a pure insect product of nature, with a notionally inedible insect, which is, furthermore, inimical to subsistence in the Middle East through its plague attacks on crops. European observers of the subsistence habits of other peoples have tended to assume that the consumption of insects is an indication of destitute circumstances, such as The Baptist voluntarily endured, rather than a matter of choice. As Sahlins (1972:2ff) has pointed out, this ethnocentric view contributed Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 3 to the long delay in realizing that a hunter-gatherer existence was by no means as harsh as had been presumed. There is abundant evidence that in various times and places, insects have achieved an important place in patterns of human exploitation of the environment (Bergier 1941; Bodenheimer 1951; Menzel and D’Aluisio 1998; Meyer-Rochow 1978/79; Posey 1986). Among horticultural people in particular, there is an often detailed knowledge of the biology and ecology of insects which are pests of crops (Posey 1986). Insects and other terrestrial invertebrates figure widely in the everyday diet in many societies with subsistence-oriented economies (e.g., Bodenheimer 1951; Meyer-Rochow 1976; Posey 1986). In many instances, insects form an occasional or incidental part of the diet (e.g., Setz 1991), but are relished as side dishes (e.g., Ruddle 1973). Insects are often consumed in particular by children and women, as among the Kapauku of Irian Jaya (Pospisil 1963), Aché of Paraguay (Hill and Hawkes 1983) and Tukanoa of Columbia (Dufour 1987). This restriction by age and gender itself is suggestive of the relatively low status generally assigned to most insect foods. In other cases insects may be highly esteemed as feast food. In these instances, the particular species so regarded are usually only seasonally or periodically available, such as the bogong moths (Agrotis infusa Lepidoptera-Noctuidae) of highland southeast Australia (Flood 1980). Often, they are soft-bodied forms, es- pecially the immature stages of growth, as in grubs of certain beetles, such as the ‘sago beetles’ (Curculionidae), which are widely consumed in Melanesia and Southeast Asia, and bruchid beetle larvae (Bruchidae) of Amazonia (e.g., Dufour 1987; Ellen 1993b; Ernst 1978; Ruddle et al. 1978). Aside from the nutritional value of grubs, which are high in protein and fats (Dufour 1987; Ruddle et al. 1978:59), it is perhaps of significance that insects which feature as prestige foods are mostly pupating or burrowing forms, found in conveniently large numbers massed together either in the earth, encased in nests or cocoons, or in the moist pulpy interior of stems, fruit or rotting logs. The symbolic associations between such grubs and cultural representations of human sexuality have been explored by anthropologists (e.g., Ernst 1978; Gell 1975). And, of course, honey is also esteemed in many cultures for its qualities of taste, texture, and its symbolic properties (Levi-Strauss 1973; Posey 1983). Various insects have been recorded to have other uses, including therapeutic and hallucinogenic applications (e.g., Britton 1984; Groark 1996) and as sources of poisons, as among the Kalahari ‘Bushmen’ (San), who use exudate from the larvae of the Chrysomelid beetle Diamphidia simplex as arrow poison (Silberbauer 1981:76, 207). In New Guinea the brilliantly colored shards and heads of scarab beetles (Scarabaeidae) are threaded on sticks as headdress decorations, valuables and trade items (e.g., Healey 1990:70, 135, 366; Meyer-Rochow 1978/79). Posey (1986) briefly reviews the place of insects in myth, folktales and ritual practice in ‘tribal’ cultures. While there are numerous references in the literature to insects, there has been much less attention to knowledge, consumption and other uses of spiders. In some instances, this may be a product of the folk categories employed by 4 HEALEY and FLOREY Vol. 23, No. 1 ethnographers themselves. In scientific terms, ‘insects’ and ‘spiders’ are subcate- gories (along with crabs, centipedes and their allies) at different taxonomic levels within the phylum Arthropoda, the largest phylum in the animal kingdom— organisms characterized by a hard outer skin with flexible joints, bilateral sym- metry and a segmented body with paired limbs. Spiders constitute a single order Araneae within the class Arachnida (which includes related orders of scorpions, mites and ticks, harvestmen, etc.). Insects, on the other hand, are classified into many orders, which in turn are distributed in several classes grouped into the superclass Hexapoda. There is clearly no close equivalence between the English- language categories of ‘insect’ and ‘spider’ and the scientific classification of these arthropods. For some, ‘insects’ and ‘spiders’ are distinct and roughly equivalent categories, while for others, the folk English category ‘insect’ may include ‘spi- ders’. While the terms appear to be employed in a scientifically technical sense in many reports it may be that English-speaking investigators have actually failed to elucidate the boundaries of ethnotaxa glossed as ‘insects’. In at least some cul- tures, besides the Alune case discussed further below, spiders nonetheless con- stitute distinctive recognized zoological and linguistic categories (see, e.g., Ellen 1993b on Nuaulu; Hunn 1977 on Tzeltal; Posey 1981 on Kayapo; Taylor 1990 on Tobelo; also, Healey elicited similar responses from Maring in 1974). Spiders are commonly objects of mild unease to acute fear in Anglo-European culture. Arachnophobia is a recognized psychological disorder, and negative rep- resentations of spiders appear widely in European folktales, literature, film and children’s stories. Spiders and their webs have a prominent place in the European iconography of horror. Many cultures appear to share the ambivalent attitude towards spiders of Western European cultures. Spiders are widely regarded as objects of fear, aver- sion or as pests. The fact that many spiders have painful, poisonous, and even potentially lethal bites no doubt contributes to the common anxiety many people feel towards spiders. However, most spiders are no less intrinsically edible or otherwise useful than many other arthropods and insects which are widely eaten in many subsistence- oriented societies. Hunn (1990:167ff) is the only source we have found which specifically notes that spiders are not eaten, in this case (along with most other invertebrates), by the Sahaptin Indians of the Columbia Plateau. There are very few explicit references in the ethnographic literature to the edibility of spiders, or to sie uses to which they are put. The following section reviews the available record. In the absence of clear attention to the issue of the utilization of spiders by human cultures we are led to ask: Does the lack of information on the use of spiders for food or other purposes reflect widespread non-utilization of this po- tential resource, or is it a consequence of inattention on the part of ethnobiologists to actual indigenous practices in relation to spiders? ETHNOGRAPHIC RECORD OF USE OF SPIDERS The negative attitudes to spiders outlined above is by no means universal. There are a few scattered reports in the literature of spiders being used for food, Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 5 medicinal practices, as providing raw materials for items of technology, and as positive markers of social identity. In his world survey of the use of insects as food, Bodenheimer (1951:65, 67— 68) also briefly reviews the use of spiders as food and for medicinal or magical purposes, citing a number of sources from the eighteenth century onwards. Bo- denheimer’s data are fragmentary. The ethnicity or location of spider-users is not always clearly identified. Meyer-Rochow’s more recent survey (1978/79) suffers from the same defects. Bodenheimer reserves greatest attention to cases of arach- nophagy in Europe, perhaps out of a sense of the grotesque. It is clear from his account that instances of spider-eating attributed to particular individuals in eigh- teenth- and nineteenth-century France and Germany are “perversion[s] of taste” (Bodenheimer 1951:65), rather than common cultural practice. Several of these arachnophagists likened the taste of spiders to hazelnuts. Table 1 summarizes results of our review of the available material that men- tions the use of spiders as food and for other purposes. In a number of Melanesian cultures, spiders (mostly unidentified) are roasted in embers and eaten as snack foods (see Kapauku, Yopno, Pawaia, Maring, Kilimeri, New Caledonia in Table 1). The apparent prominence of spider use among peoples of Oceania relative to those of other regions may merely be an artifact of small sample size. The Khmer of Cambodia are reported to eat large tarantulas (Theraphosidae) deep-fried in oil and served on skewers. They are reputed to enhance virility (Menzel and D’Aluisio 1998). Similarly, the Yanomamo of Venezuela extract Theraphosa leblondi (Theraphosidae) tarantulas from their burrows to eat, roasting them on the fire (Menzel and D’Aluisio 1998). This species is the world’s largest spider and con- tributes a substantial amount of meat to a meal. The reports in Table 1 attributed to Bodenheimer should be treated with some caution, as none of the sources he cites are by recent credible ethnographers. A more exhaustive search might yield further examples from other cultures, and in particular we would expect more cases from foraging peoples, particularly of tropical and subtropical America, Africa and Asia. Spiders and their webs are also used for purposes other than food. Again, there is very little of substance in the literature beyond passing references. The Nuaulu of central Seram, who are culturally related to the Alune, use the com- pacted web masses of Nephila species (Araneidae) as bait in line-fishing for nee- dlefish, a practice Ellen (1993b:203) considers they must have learnt from other people, as, like the Alune, the Nuaulu are traditionally an interior people. Speiser (1996 [1923]:241) reports the use of compacted web mass of unidentified spiders to construct purses and ritual masks in Malekula, Vanuatu. The Ngarinman of northwest Australia used spiders’ web to fashion small purses (Healey, unpub. fieldnotes). Spiders are reported to be used for medicinal purposes in a number of areas. Hunn (1977:310-312) notes that among the Tzeltal of Chiapas, Mexico, tarantulas (Mygalomorphae) are used in a cure for tumors, with the spiders induced to bite the affected area. Bodenheimer also reports the medicinal or magical use of spi- ders. De Walckenaer notes that “in Brazil certain [unidentified] spiders are be- lieved to be strong aphrodisiacs ..., and the same quality is ascribed to them in folk medicine throughout the world’ (Bodenheimer 1951:68). In the Kamchatka TABLE 1.—Sources on use of spiders. Ethnolinguistic group/location Source Uses Australasia/Pacific Anindilyakwa (N Australia) Amiyenggal, Larrakia, Malak Malak, Wadji- giyn, Werat, Wulna (N Australia) Ngarinman (NW Australia) Kapauku (western New Guinea) Yopno (Morobe Province, Papua New Guin- ea Pawaia (Eastern Highlands, Papua New Guinea oo. oT Highlands, Papua New Kilimeri ‘(West Sepik, Papua New Guinea) Malekula (Vanuatu) ew Caledonia Asia Thailand aie: (Cambodia) Nuaulu (Seram, Indonesia) Kamchatka (E Russia) Africa San (South Africa) Maniana (Gambia) Merina (Madagascar) Cameroon (W Africa) Azande (Central Africa) Waddy (1988) a sa (1907), Foelsche (1881-2), Healey (unpubl. field notes), Mackil- lop (1892-93), Spencer (1914) Healey (unpubl. field notes) Poe ey (1963) er Schmid (pers. comm.) Kocher Schmid (pers. comm.) Healey (unpub. field notes) Kocher Schmid (pers. comm.) Speiser (1996) Bergier (1941) Bodenheimer (1951) Menzel & D’Aluisio (1998) Ellen (1993) Bodenheimer (1951) Bodenheimer (1951) Bodenheimer (1951) Bergier (1941) Zeitlyn (1993) Evans-Pritchard (1937) lan totems Web of ?Nephila used as ligature to sever fin- ger joint of young females, in at least some eg in accordance with mythical prescrip- Web of unidentified species used to make No details Unidentified species roasted and eaten by chil- Unidentified species roasted and eaten by chil- ren Various unidentified species, including ?Nephi- la sp. roasted and eaten, mostly by children Funnel-web spiders (?) roasted and eaten by women Web used to construct ritual masks Araneus edulis eaten grilled on coals Egg bags of spiders eaten Tarantulas (Therap oe ea beg fried Web of Nephila used as fis Unident. species eaten by women ats enhance fertility, ease labor Eaten. No details Eaten. No details Nephila madagascariensis eaten fried in oil or fat Unident. species provide divination in legal cases Unident. species manipulated in curing rites AFUOTA Pue APTVAH L “ON ‘€Z “TOA TABLE 1.—Continued. Ethnolinguistic group/location Source Uses Americas Tzeltal (Chiapas, Mexico) Guaharibos, Piaroa (Paraguay) Brazil Yanomamo (Venezuela) Europe France, Germany England S Italy Hunn (1977) Bodenheimer (1951) Bodenheimer (1951) Menzel & D/Aluisio (1998) Bodenheimer (1951) Black (1883) is (1991) Tarantulas (Mygalomorphae) used in cure for tumors Tarantulas eaten Unident. species eaten as aphr Theraphosa leblondi ihenpioaas) peel Unident. species eaten (personal idiosyncrasy?) Unident. species used to relieve fever Lycosa sp. associated with Tarantula cult €00z Journ /Sur1dg XOOTOISONHLA JO TVNANO[ 8 HEALEY and FLOREY Vol. 23, No. 1 Peninsula of eastern Russia eating unidentified spiders is said to confer fertility on sterile women, and ease labor (Bodenheimer 1951:68). The seventeenth-century English antiquarian, astrologer, and solicitor, Elias Ashmole, provides us with evidence of another medicinal use of spiders in his diary for 11 April 1681: “I tooke early in the Morning, good dose of Elixir, & hung 3 Spiders about my Neck & they drove my Ague away, Deo gratias’’ (Ash- mole 1966:1680).! This was no Ashmolean idiosyncrasy; live spiders encased in a nutshell and worn about the neck were believed to relieve fever (Black 1883:59- 60). One of Bodenheimer’s possibly idiosyncratic German spider eaters spread them on bread in place of butter “as a purge’ (Bodenheimer 1951:68). Spiders may also serve positive symbolic uses. Among the Anindilyakwa of Groote Eylandt, northern Australia, spiders of any kind are among clan totems (Waddy 1988). Also in Australia, a number of related Aboriginal cultures of north- western Northern Territory (see Table 1) traditionally used the web of a large orb spider (Nephila species, Araneidae) to tightly bind the top joint of the first finger of young girls. This cut the circulation and in time the top joint dropped off (Basedow 1907:8; Foelsche 1881-2:7; Mackillop 1892-93; Spencer 1914:10). This practice follows a tradition associated with a large earthquake in the creation period (Lindsay et al. 2001). It was believed that by driving the stump of the severed finger into the ground women would be able to quell earth tremors (Hea- ley, unpub. fieldnotes 2001). Spiders were employed in divination and curing in various African cultures. Zeitlyn (1993) discusses the use of spiders as infallible agents of divination in courts administered by traditional rulers in several Cameroonian societies. Among the Azande of central Africa, a spider was sometimes produced by sleight of hand in curing rites by ‘witch doctors’ who claimed the spider contains the sick person’s blood (Evans-Pritchard 1937:236). Lewis (1991) reviews the Tarantismo spider cult of southern Italy, where wom- en are typically believed to be recruited to the cult by seeking a ritual cure after supposedly being bitten by a tarantula spider. Tarantulas of the genus Lycosa are regarded as the actual counterpart of a dangerous spirit which inflicts harm on humans. These fragmentary accounts of the mystical and medicinal aspects of spider use suggest an association of spiders with dangerous states. These include recti- fying reproductive dysfuntion in individuals by enhancing libido, conferring vi- rility, and overcoming infertility; divining causes of social misfortune; and guard- ing against a repeat of mythical cataclysm. Even some of the reports of arach- nophagy indicate that they are perceived primarily as medicines rather than as foods. It is tempting to speculate that this association of spiders with mystical dangers and bodily dysfunction may stem from a widespread cultural ambiva- lence towards spiders, perhaps stemming from the combination of the perceived delicacy of their webs and bodies and the remarkable strength of webs and the poisonous bite of at least some species. at our review does indicate is that not only is the number of reported cases of use of spiders low, but the level of available detail is also generally limited. Our own material on the Alune is also limited in detail. It is offered here as a Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 9 A Wes t 1 i) 10 aaa” Name kilometres ‘AUS IA Taniwel Piru }® e Lchiasapalewa WEST SERAM FIGURE 1.—Indonesia and West Seram. Alune villages are indicated by circles (not all shown), towns by squares. preliminary contribution to a neglected area of human interaction with the bio- logical environment. THE ALUNE The Alune are an Austronesian-speaking people living in twenty-five villages in north west Seram. Seram is one of the larger islands of central Maluku (the Moluccas or ‘Spice Islands’), roughly midway between New Guinea and Sulawesi (Celebes) in eastern Indonesia (see Figure 1). The ancestral Alune domain is in the mountains of the interior of west Seram. The natural vegetation of this region is lowland and mid-montane rainforest. Large tracts of the rugged interior be- came depopulated when many Alune villages relocated to the narrow coastal belt 10 HEALEY and FLOREY Vol. 23, No. 1 in northwest and southwest Seram, as detailed below. As a consequence, much of the natural vegetation remains in the rugged interior, with earlier settlement and gardening sites reverting to heavy forest. The Alune were traditionally one of a number of indigenous tribal popula- tions of Seram living in autonomous village communities that maintained sepa- rate domains. The local community is composed of several localized intermar- rying descent groups. Principal sources on traditional Alune social organization and religion include Ekris (1867), Jensen (1948), Jensen and Niggemeyer (1939), Sachse (1907), Stresemann (1923), and Tauern (1918). Alune social organization, ecology and subsistence are similar to that of the Nuaulu as described by Ellen (e.g., 1978; 1988). Alune subsistence was based on a combination of shifting cultivation of hill rice (Oryza sativa) and other food crops, processing of sago starch (Metroxylon sp.) from scattered family- and lineage-owned stands within the rainforest, and ex- tensive gathering and hunting (Wolff and Florey 1998). Consistent with the sea- sonal and geographic distribution of these activities, the typical settlement pattern was one of a loose cluster of hamlets and widely dispersed family garden houses and more temporary encampments in the forest at sago groves and hunting and gathering grounds. Often groves and foraging grounds coincided, allowing for a diversity of activities based at a camp site; in other parts of the domain they were more widely scattered. Traditional Alune religion focused on the placation of ancestral and local nature spirits, whose goodwill was regarded as necessary for ensuring the health and vitality of the living, and the productivity of the environment. Virtually all Alune in inland villages are now adherents of the Protestant Church of Maluku. The church is a significant source of pressure for cultural change, with its official expectation that traditional beliefs and practices will be abandoned by its adher- ents. It also broadly subscribes to the ideology of ‘development’ espoused by the Indonesian government, which regards ‘traditional culture’, often glossed by the Indonesian term adat, as essentially ‘backward’ and an impediment to progress. In practice, however, traditional beliefs nonetheless remain strong in many Alune villages, albeit seldom openly articulated in the presence of church or government officials and other outsiders. There is still a strong belief in the presence of an- cestral spirits which will punish failure of the living to abide by the customary norms of moral behavior, and belief in the efficacy of certain magical formulae and practices (Florey and Wolff 1998). All members of the local community are expected to share the same cosmological beliefs and ritual practices, and failure to do so is considered to invite ancestral anger and retribution. Paradoxically, the strength of commitment to the Protestant Church of Maluku is itself a product of a continuing principle of adat, which has seen those who have chosen to follow dissenting Christian churches excluded from the community (Florey and Healey 2002). In that sense, wholesale conversion of villages to the same variant of reli- gious observance masks a continuation of adat belief and practice, which is sup- pressed in its outward expression, while at the same time re-affirmed as funda- mental to the maintenance of social order. The influence of the Dutch colonial government was for the most part con- fined to the suppression of hostilities and fostering conversion to Christianity. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 11 From the early twentieth century a policy of encouraging resettlement of villages along the narrow coastal plain of north and south Seram served these interests well. However, a number of Alune villages resisted such moves, retaining occu- pation of their ancestral domains in the rugged interior. One such village was that of Lohiasapalewa, the site of fieldwork on which this paper is based. The pace of change increased following Indonesian independence in 1945 and more effective state control over Alune domains. In response to pressures from the state, Alune settlement patterns have been modified with the establishment of nucleated villages and more relocations to the coastal belt. Of particular relevance for the focus of this paper, post-colonial authority has influenced changes in the patterns of Alune subsistence practices and other in- teractions with the environment. But it has also periodically resulted in temporary reassertion of more traditional pursuits. Official encouragement of consolidation of residence in villages, combined with the demands of the church for regular participation in religious practice centered on the village church, have concen- trated subsistence activities within easy reach of the village. Recent decades have seen an increase in small-scale production of tree cash-crops such as cloves (Syzygium aromaticum), cacao (Theobroma cacao), and coffee (Coffea canepora) planted in small scattered groves in old garden sites mostly close to the village. As a consequence of political, religious and economic changes, extended res- idence in distant garden shelters or forest camps is less common than in the past, and gardens are more likely to be sited closer to the village. Although hill rice is still a preferred crop, cassava (Manihot esculenta) has increasingly supplanted it in the everyday diet. Furthermore, hunting and gathering in the forest, especially the more distant parts of village domains, is less common now than in the past. Nonetheless, at least twice over the last fifty years, the Alune of Lohiasapa- lewa have fled the village to take refuge in the forest. In the 1950s and 1960s, many villagers sought refuge in the forest to evade the conflicts between the Indonesian military and guerrillas of the ill-fated separatist Republic of the South Moluccas, or RMS (Republik Maluku Selatan). Then again in 1970, Lohiasapalewa villagers fled to the forest to evade intense pressure from the local administrative officer to relocate the village to the coast. There are unconfirmed reports that villagers have again taken refuge in the forest in early 1999, since the outbreak of sectarian violence between Muslims and Christians in Ambon city, which rapidly spread to many other parts of Maluku. Sporadic violence continues to the present (August 2002), and has included attacks on Alune villages in the coastal plains and raised more general fears of escalation of violence. The significance for this paper of these retreats into the forest is that recent historical circumstances forced Lohiasapalewa villagers to revert to a subsistence economy centered on hunting and foraging in the forest. The present unrest fur- ther underscores the potential importance of the forest as a safe haven and source of subsistence foods. Thus, a close dependence on wild resources, including in- sects and spiders, is part of the shared experience of virtually all Alune today, and goes at least part of the way to explaining the retention of a considerable body of detailed ethnobiological knowledge. Changes in subsistence patterns are 12 HEALEY and FLOREY Vol. 23, No. 1 dealt with in more detail by Florey, Healey and Wolff with Manakane (in prep.) and Wolff and Florey (1998). METHODOLOGY The data on which this paper is based were collected over the course of sev- eral field trips to the inland Alune village of Lohiasapalewa in 1993-95. This research was part of a larger project on Alune ethnobiology and language change, involving collaboration between Florey, Healey, botanical consultant Dr. Xenia Wolff, and Alune subsistence farmer and expert on local natural history Mr Wem- pi Manakane. The research team drew upon the substantial linguistic and eth- nographic knowledge of Alune gained by Florey over several fieldwork seasons in various Alune villages totaling over 24 months since 1988. Alune knowledge of spiders was recorded in a number of different contexts. These included the following: focused discussion sessions on local fauna with individuals and small parties, mostly men, but sometimes also children and wom- en; observations of spiders during excursions into different habitats in and around the village attended by Manakane and other Alune advisors and informants; ex- amination and discussion of specimens of spiders and other invertebrates collect- ed by Manakane and brought to Healey in evening work sessions, in which other adults of both sexes usually participated; chance observations of Alune behavior towards or talk concerning spiders. No collections of specimens were made. However, we are building a photo- graphic record of specimens. Most of the scientific identifications noted below were provided by Dr. Graham Brown of the Museums and Art Galleries of the Northern Territory, Darwin, Australia, based on photographs, as acknowledged in the text. Other identifications are our own tentative suggestions and should be treated with caution. Ethnobiological knowledge is not evenly distributed within Alune commu- nities, but varies with age, gender and personal interest. Despite his comparative youth, Mr Wempi Manakane, who was aged 25 years in 1995, is widely acknow]- edged by residents of Lohiasapalewa as a local expert in a broad range of tradi- tional knowledge, including ethnobiological domains. The richness of the ethno- biological knowledge that he was able to convey to us, compared to that of many other villagers, certainly supports this reputation. He is also a highly articulate exegete of local knowledge, fluent and literate in Indonesian and the Ambonese Malay lingua franca. He has also adopted Florey’s orthography for Alune and became fully literate in his own hitherto unwritten language. On a number of occasions, other Alune deferred to Manakane’s opinion on various aspects of eth- nobiological lore, or referred us to him as an appropriate person to consult. These referrals were not merely acknowledgement of the close working relations that developed between Manakane and us, as villagers also freely referred us to other recognized experts on diverse aspects of Alune culture and ecology. Alune freely admit when they do not know the answer to questions, as in- dicated by their willingness to refer to other experts. Manakane himself is clearly conscious of the limits of his own ethnobiological knowledge, and referred us to others, or himself made enquiries of older people who are recognized as more Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 13 knowledgeable in various areas of indigenous knowledge. He made these referrals both in our presence when other experts were also present, and independently by seeking out advisors when he had been unable to provide us with information we sought. We made no systematic tests for consistency of identification of specimens or other details of ethnobiological information between informants. Indeed, we argue against the methodological value of rigorous testing of consistency of information or responses to elicitation stimuli as being contrary to naturalistic discourse. Con- sistent with the general methodological orientation outlined above, our focus has been on eliciting Alune identifications and knowledge of organisms in the mean- ingful contexts within which they are encountered in the course of everyday life. One context of encounter with organisms in the environment is not the same as another. Wherever possible, we allowed our Alune hosts, advisors, and informants to determine the contexts in which we encountered and recorded their ethnobio- logical knowledge, rather than seeking to impose an alien framework of our own. It was, for example, Manakane’s idea to collect specimens of invertebrates and bring them to Healey in the evening as an aid to discussions. Because we sought information contingently in contexts meaningful to Alune we have undoubtedly not encountered all possible contexts, and our compendium of information is therefore probably incomplete. However, we argue that the re- sulting record more closely represents the Alune world view than more formal elicitation techniques. ALUNE ETHNOBIOLOGY OF SPIDERS The Alune regard spiders as just another element of the diverse fauna of their region. Alune knowledge and exploitation of insects are extensive (Florey et al., in prep.). Spiders are classified separately from insects, but they are treated in very similar ways, and attract much the same character of attention. Alune regard most spiders as eminently edible. Like other small invertebrates they are usually lightly roasted beside the fire. With an increasing orientation of the local economy to subsistence horticulture and cash cropping, and a residential focus on the village, rather than a more forest-oriented economy of foraging, hunting and cultivation, spiders no longer appear to be commonly eaten or oth- erwise exploited, even by children. Certainly, this was the view of several of our informants, but we cannot be sure that the various opportunistic snacks we ob- served being consumed around hearths did not contain the odd spider. We have no data on the nutritional contribution of spiders and other inver- tebrates to the Alune diet. The only available study of Alune diet (Novotny et al. 1996) focused on adult women, and is compromised by apparently recording only household meals, and neglecting the possible nutritional significance of between- meal snacking. Our observations indicate that Alune consume sometimes consid- erable amounts of food as snacks away from the main residence. Snacks range from wild and cultivated fruits, nuts and vegetables to invertebrates and small vertebrates cooked in garden houses or at impromptu small fireplaces, especially by small bands of children. The views of informants indicate that spiders were more commonly exploited 14 HEALEY and FLOREY Vol. 23, No. 1 in the past, but we have no evidence that this was other than on an opportunistic basis. It is therefore likely that spiders have only ever been a very minor com- ponent of the diet, and the nature of other human-spider interactions limited. Nonetheless, in part because of the recent history of refuge in the forest, and the continuing interest in forest resources, many Alune preserve considerable knowledge of older patterns of exploitation of spiders and other invertebrates. Spiders form part of a wider pattern of close and intense interactions the Alune maintain with the natural environment. Exploitation of arthropods continues a long standing form of interaction with the environment in respect of that part of the diet that is based on a foraging component of the economy. Although recent social change has led to an increasing prominence of cassava and sago as the core component of main daily meals, there is a persistence of longer-established pat- terns of the foraging sector of the economy that yields considerable variety to the diet through frequent snacking, as an element of what we call a ‘small food tra- dition’. This pattern of impromptu snacking, and the foraging behavior by which it is sustained, maintains a strong meaningful and subjective association with the forest as a core element of Alune perceptions of their ethnic identity as ‘forest people’. This identity has been reinforced by their recurrent retreat into and de- pendence on the forest in periods of social upheaval. We do not mean to suggest that exploitation of spiders is itself definitive of Alune identity, but certainly this activity and the knowledge on which it is based are an integral part of a specif- ically Alune style of interaction with the environment at a subjective level. The Arachnid fauna of Seram is poorly known. The abundance and species diversity of arthropods are high in central Seram (Stork and Brendell 1993). There are no good faunal surveys of the arachnid fauna of Seram that might be a guide to the number and identity of species that could be expected from the Lohiasa- palewa area. Ellen (1993b) records ten terminal categories applied to spiders by the Nuaulu, another Austronesian-speaking people of south-central Seram. These are assigned to seven biological species, although the actual number is likely to be considerably greater. This compares well with our own data for the Alune; we have recorded a total of eight terminal categories applied to spiders, each tentatively ascribed to separate biological species. This suggests that there may be few, if any, Alune categories that we have not yet discovered. Alune class all spiders in a single category termed ‘wala’wa.? The term is applied to a set of categories which are all regarded as morphologically similar. Possession of eight legs is a key characteristic. ‘wala‘wa thus designates a well- defined natural category of organisms not further included in a higher order taxon. Despite the small number of terminal taxa making up the category, we consider “wala’wa can reasonably be designated as a ‘life form’, equivalent to other Alune life forms, such as manue, ‘birds’. The label ‘wala‘wa is cognate with the Nuaulu term wala-wala recorded by Ellen (1993b:201). Interestingly, however, the Nuaulu term is applied to spiders’ webs and to certain web-spinning spiders, rather than to spiders in general. In Alune spiders’ webs are designated by the noun classifier for fibrous matter, bueti. Spiders’ web is termed ‘wala‘wa esi bueti (‘spider its fibrous mass’). We have no record of any uses of spiders’ webs. The Nuaulu equivalent of the Alune ‘wala’wa Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 15 is kahunekete, which contains ten named terminal categories (Ellen 1993b:200- 204). Nuaulu consider scorpions (Scorpionida) and harvestmen (Opiliones) to be closely related to spiders, and some of Ellen’s informants suggested that harvest- men may be a kind of spider (Ellen 1993b:203). Alune opinion is much the same; at least some of our informants regard harvestmen somewhat ambiguously as possibly spiders, and certainly similar to spiders. While they are generally not included in the category ‘wala‘wa, they are discussed here because of their per- ceived close similarity. Scorpions are more clearly classified in a separate category, as they are among the Nuaulu (Ellen 1993b:203). Alune liken them to freshwater crabs, as indicated by the term ‘li’we ta’unui (‘li’we is the general term for freshwater crabs, ta’unui means ‘likeness’ or ‘resemblance’, but also ‘spirit of the dead/ghost’). Crabs are relished as food, but scorpions are not eaten.* The ambiguous meaning of the term for scorpions (‘resembling a crab’ and ‘ghost crab’) is perhaps appropriate for these venomous creatures. Alune regard the named categories outlined below to be morphologically dis- tinct. Taxa are thus distinguishable by perceptual criteria, although there are also behavioral and ecological characteristics which informants highlight. The eight terminal categories applied to spiders and one applied to harvestmen appeared to be used with some discrimination, having been offered as identifications of specimens which were also morphologically distinctive to us. However, as already noted, we are unable to assess the consistency with which these category names are applied in instances of identification of specimens. The number of biological species present is undoubtedly well in excess of the forms we have observed and the Alune terms may have fairly wide application. Ellen (1993b:203) has noted that Nuaulu categories appear to be ‘wide-ranging and overlapping.” While our own data on identification and classification are insufficient to corroborate this conclusion in respect of spiders, our data on other biological domains indicate that Alune identifications of naturally-occurring specimens and classifying be- havior is similar to that reported by Ellen (1993a, 1993b) for the Nuaulu. Several spiders, such as Nephila sp(p)., are said by Alune informants to have a venomous or painful bite. However, while Alune handle dangerous spiders care- fully they do so with confidence, and without apparent anxiety or squeamishness. In what follows we briefly outline Alune knowledge of spiders and harvest- men by named terminal category within the higher order category of ‘wala’wa. We also indicate uses and other cultural significance of spiders and harvestmen. ‘wala’wa. The term applied to spiders collectively is also used as a label for a constituent terminal category. According to informants, the terminal category ‘wala’wa is applied to spiders described as of medium size, and which shelter in curled leaves suspended within their webs which resemble a wheel. These serve as the focal category of spiders in general, and ‘wala’wa as a terminal category probably is applied to a number of species of similar appearance and web-build- ing. Focal ‘wala’wa make a nest around a curled leaf in which the eggs are laid. These spiders are common in vegetation about homesteads and gardens, and are edible. We obtained no photographic record of specimens and the species remain 16 HEALEY and FLOREY Vol. 23, No. 1 unidentified. We tentatively identify them as members of the leaf-curling spider group, family Araneidae. ‘wala luma. This term means ‘house spider’, and is applied to huntsman spiders, Heteropodidae, found in houses. Several different species of Heteropodidae are represented in our photographic record, but unfortunately they have not been identified beyond family. Only one kind of these huntsmans (G. Brown, pers. com.) in the photographic record, a large grey spider encountered in houses, was identified as ‘wala luma (Figure 2). However, it is possible that identifications of ‘wala luma are based primarily on habitat, rather than morphological features, and this term may therefore apply to several similar species encountered in hous- es. Because ‘wala Iuma frequent houses and eat cockroaches and other objects of disgust they are considered inedible. ‘wala metene. This term, meaning ‘black spider’, was applied to huntsman spiders found in fallow regrowth, a photograph of which has been ascribed to a second species of Heteropodidae. The large egg cases of this spider, lightly roasted by the fire, are regarded as delicious. These yield hundreds of tiny young when hatched. ‘wala porole. This term was applied to another huntsman spider found in fallow regrowth. This has been identified as a third species of Heteropodidae (G. Brown, pers. com.). It is called labalaba kuning in Ambonese Malay. Both names mean ‘yellow spider’ and refer to the light brown color of this huntsman spider. This spider is edible. ‘wala si’oli. This term is applied to the very large and common orb-weaving spiders, identified as Nephila, Araneidae (G. Brown, pers. com.) (Figure 3). The large females are highly distinctive and visible spiders, because of their size, glossy black and yellow bodies, and massive strong and sticky webs built across open spaces. Males are usually very small and seldom encountered with females except in the mating season (G. Brown, pers. com.), and it is not clear if Alune are aware of this extreme sexual dimorphism. Alune provided no etymology of si‘oli, but it can be translated loosely as ‘it falls to the ground’, from (e)si ‘it’ and ‘oli ‘to slip and fall onto the ground (from a height)’. This is perhaps a reference to the spider’s habit of constructing its webs high above the ground. One can tell if other people have already passed by on a track as the webs will have been cleared away. This is the closest we have come to recording a functional use for spider web. Note that it is webs of the same genus that are used by the related Nuaulu as fishing lures. ile this spider is recognized to have a harmful bite, it is readily handled with care. It is considered edible, and spiders are roasted whole in the fire or steamed in bamboo tubes. Nephila species are also eaten in Madagascar and prob- ably New Guinea (see Table 1). ‘wala susu. This term was applied to distinctive hunting spiders encountered in undergrowth. We have no photographic record or identification of specimens, but they are possibly wolf spiders, Lycosidae. The translucent white color of these Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 17 J i E 2 on FIGURE 2.—‘wala‘wa luma, Heteropodidae. Huntsman spider. (Head and body length approx. 3 cm.; based on color transparency 94/4.23 20 June 1994 by C. Healey) spiders is likened to breast milk, hence the specific name, susu meaning ‘breast/ milk’. ‘wala susu are edible. ‘wala tasie. This term was applied to medium sized robust black spiders, prob- ably Araneus species, Araneidae, found in leaf nests set in webs in scrubland. The spider is characterized by a hairy body and a painful bite. It is considered edible and has a rather pleasantly salty taste, hence its name: tasie meaning ‘salty’, Mem- bers of the same genus are also eaten in New Caledonia (see Table 1). ‘wala munine. This term was applied to medium sized dark wheel-web spinning spiders, possibly of the Family Araneidae (G. Brown, pers. com.). The spider is considered to be distinctive in having four clearly discernible eyes and in catching and eating other spiders and insects considerably larger than itself. It is likened to human sorcerers, munine, hence its name, because it kills others of its kind. Also, like human sorcerers, it can see behind itself. It is considered inedible, pre- sumably because of its association with sorcery. Idori. This term is applied to harvestmen (G. Brown, pers. com.), which constitute the Order Opiliones (also referred to by the older name Phalangida). These are small, round-bodied arachnids with eight immensely long, very thin legs. These resemble spiders in their general form, except that the two body parts of head- thorax and abdomen separated by a distinct ‘waist’ in spiders are fused into a single body part. Harvestmen also lack fangs, and unlike most spiders, they do 18 HEALEY and FLOREY Vol. 23, No. 1 [2 jt j2 cm FIGURE 3.—‘wala’wa si ‘oli, Nephila sp. Orb-spinning spider (Head and body length approx. 3.5 cm.; based on color transparency 94/1.6 18 June 1994 by C. Healey) not make webs. In discussing several live specimens brought by Manakane to Healey one evening, informants expressed uncertainty as to whether harvestmen should be classified as ‘wala’wa or in a category of their own. Manakane sought to resolve this issue by counting the number of legs, reasoning that if they possess eight legs they should be classed as ‘wala’wa. The fragile specimens had all lost varying numbers of legs, and it was not possible to determine the normal com- plement. Manakane and others present therefore opted to cautiously ascribe idori to an unaffiliated category. This indicates a strong objectivist basis to classificatory behavior, at least in this instance. Idori are found under leaves in undergrowth, always two or three or more together. They run quickly when disturbed. We were offered no etymology of idori, but it can be translated as ‘it falls over on the ground’, from i ‘he/she’, and dori ‘to fall over on the ground’. Perhaps this is a reference to the somewhat clumsy, gangling movements of harvestmen, and their tendency to be found close Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 19 to the ground. A secondary meaning of dori is ‘to live alone and unhappily’, although this does not seem readily applicable to the sociable harvestmen. In the past harvestmen were used as a cure for bed-wetting at night in children who should have outgrown the behavior. A specimen was rubbed on the child’s belly on being put to bed at night.‘ No objective, symbolic or metaphysical explanation was offered as a basis for this use. Idori are not considered edible. CONCLUSION All names for terminal categories of spiders and harvestmen are polymor- phemic, with the exception of the focal category of leaf-curl spiders (‘wala‘wa). Three names refer to color or appearance (“wala metene, ‘wala parole, ‘wala susu), one to taste (“wala tasie), one to habitat (“wala luma). These names thus refer to objective criteria that distinguish categories. The last three names refer to aspects of behavior descriptively (‘wala si‘oli, idori) or metaphorically (‘wala munine). Six of the eight terminal taxa applied to spiders are regarded as edible. Ellen makes no mention of views of the edibility of spiders among the Nuaulu, although he records many other invertebrates as edible (1993a, 1993b). The other two cat- egories are deemed to be inedible, because of their association with dirt or mys- tical danger. While the recorded number of Alune categories of spiders is very small, and probably applied widely to diverse biological species of spider, it is notable that the edibility of all spiders is unambiguously marked. Alune categories are designated as positively edible (even highly esteemed) or definitely inedible for clearly articulated reasons. Harvestmen are more neutrally inedible, but have a medicinal application. Such marked definition of edibility is not a feature of all life forms in Alune ethnobiology. For example, while many insects are defined as edible, and others clearly proscribed, many more have neutral edibility, being neither considered worthy of eating nor specifically regarded as inedible. Similarly, as far as we are aware, all birds are considered edible but only a few are particularly esteemed as food. Even the smallest bird, which can be fairly easily killed with a lucky shot with bow and arrow or taken from the nest, yields more meat than a spider, but attracts less notice for its edibility. Spiders are evidently an insignificant source of food, given their small size and the solitary habits of most species, which precludes collecting large quantities readily. Alune ascribe no remarkable edible qualities to spiders other than pleas- ant taste. Functionally, spiders on their own can never have assumed any signif- icance in the Alune diet or adaptation to their rainforest habitat. Yet, taken in conjunction with a highly varied intake of other animal foods, from ant and spider eggs to deer and pigs, they form an integral part of the ‘small food tradition’ in a diversified subsistence economy of forest foraging, swidden cultivation, sago processing and hunting. The Alune example suggests the possible value of atten- tion to spiders as dietary and medicinal items in the subsistence strategies of other indigenous cultures of similar economy. 20 HEALEY and FLOREY Vol. 23, No. 1 NOTES 1 We are grateful to Margaret Healey for this information on Ashmole’s use of spiders. 2 We adopt the orthographic convention of using an apostrophe to represent the glottal stop, which occurs in the Central Alune dialect spoken in Lohiasapalewa (Florey 2001). 3 Lest it be assumed that scorpions are intrinsically inedible, we note that Menzel and D’Aluisio (1998) report that in Guangzhou province, China, scorpions are raised in homes and specially constructed basement ‘farms’ as wholesale business. They are added to soups and used in medicines. 4 See Florey and Wolff (1998) for an overview of Alune ethnomedical concepts and practice. ACKNOWLEDGMENTS A version of this paper was presented in a poster session at the 22nd Annual Con- ference of the Society of Ethnobiology in Oaxaca, Mexico, 10-13 March 1999. We thank participants for their positive — to our paper and helpful comments. We also thank i e Department of Anthropology, University of en at Canterbury, for information on use of spiders in Papua New Guinea which appears in Table 1. We also thank three anonymous referees for the Journal of Ehrotiobea, editor Naomi F. Miller and former editor, Bill Balée, for helpful comments on earlier versions of this paper. We acknowledge the Wenner-Gren Foundation for Anthropological Research for Grant 5683 in support of Florey’s research titled “The Alune Language of Central Maluku’ and an Australian Research Council Small Grant administered by Northern Territory University in support of a joint project by Healey and Florey in 1994-95. We thank Dr. Graham Brown, Chief Entomologist of the Museums and Art Galleries of the Northern Territory, Darwin, Australia, for providing scientific identifications from photographs of specimens. Our greatest debt is to our Alune hosts, collaborators, advisors and informants. In particular, we wish to thank Wempi Manakane for his tireless assis- tance, advice, and inf ; Yusup and Leonorci Lumaupuy and Anderius and Blandina Kwalomine for their help and hospitality in Lohiasapalewa; and Herman and Fransina Marwa and the late Albert Nikijuluw and Emmy Nikijuluw of Ambon. REFERENCES CITED Ashmole, Elias. 1966. Elias Ashmole (1617- insectes comestibles: étude sur les moeurs 1692): pe Puasa and Historical de l'homme et de l’insecte. Imprimerie Ru- Notes, H: respondence, and Other Con- lliére Fréres, Avignon. posed Sue Relating to His Lifeand _ Black, WG. 1883. Folk-medicine: A Chapter in Work: Volume IV Texts 1673-1701, ed. the History of Culture. Folk-lore Society, C.H. Josten. The Clarendon Press, Ox- ondon. Bodenheimer, ES. 1951. Insects as Human ord. Basedow, Herbert. 1907. Anthropological Food: A Chapter “i the Ecology of Man. W. notes on the western coastal shies re Junk, The Ha the Northern Territory of South Aus Britton, E.B. 1984. A pointer to a new ha- lia. Proceedings of the Royal ee of llocinogen of insect origin. Journal of South Australia 31:1-62. : : Bergier, Emile. 1941. Peuples entomophages et Dufour, Darna L. 1987. Insects as food: a Spring/Summer 2003 case study from northwest Amazon. American Anthropologist 89:383-397. Ekris, A. van. 1867. lets over het Ceramsche ) Seg bahia Tijdschrift voor Indische Land- en Volkenkunde 16:290-315. Ecology: An Approach to the Environmen- tal Relations of an Eastern Indonesian Com- munity. Nijhoff, The Ha ———. 1988. Foraging, sta ner extraction and the sedentary lifestyle in the low- land rainforest of central Seram. In His- tory, Evolution and Social Change in Hunt- ing and Gathering Societies, eds. J. Wood- burn, T. Ingold, and D. Riches, pp.117-— 134. Berg, London. . 1993a. The Cultural Relations of Clas- sification: An Analysis of Nuaulu Animal Categories from Central Sera bridge University Press, Cambridge. 1993b. Nuaulu Ethnozoology: A ne tematic Inventory. Centre for Social An- thropology and Computing Monograph 6, reed of Kent at Canterbury, Canterbury Ernst, Thomas M. 1978. Aspects of mean- ing of exchanges and exchange item among the Onabasulu of the Great oie uan Plateau. Mankind 11:187-197. Evans-Pritchard, EE. 1937. be greseee woe 2 i 2) —) | o9 : Alps. Australian pean of Aboriginal Studies, Canberr. pied Margaret. 2001. Verb and valence in Alune. In Explorations in Valency in Aus- tronesian Langu Barry Blake, and Margaret Florey, FP. 73-120. La Trobe Papers in Linguisti Volume 11. Melbourne. Florey, Margaret and Christopher Healey. 2002. ‘Work well and guard your hon- our’: temporary labour migration and the role of adolescent women in eastern te Asian Studies Review 26:355— Florey, Margaret, Christopher J. Healey, nice oa a Wolff, with Wempi Mana- In preparation. From the Edge of the Forest Alune Ethnobiology in a Chang- Ing Florey, aca J. and Xenia Wolff. 1998. Incantations and herbal medicines: Alu- ne ethnomedical knowledge in a con- JOURNAL OF ETHNOBIOLOGY ai text of change. Journal of Ethnobiology 18: 39-67. Foelsche, Paul. 1881-2. Notes on the origines of north Australia. oil of the Royal Society of South Australia 5: -18. cal, hing 1975. Metamorphosis of the Cas- Umeda Society, Language and Ritual “Athlone Press, London Groark, Kevin P. 1996. Ritual and therapeu- nativ California. Journal of Ethnobiology 16:1 ‘a a: ype Christopher J. 1990. Maring So hghg Traders: Production and Exchan the Papua New Guinea Highlands. Uitiver- sity of California Press, Berkeley. Hill, Kim and Kresten Hawkes. 1983. Neo- Hames and William T. Victors, pp. 139- 188. New York: Academic Press. Hunn, Eugene S. 1977. Tzeltal Folk Zoology: The Classification of Discontinuities in Na- ture. Academic Press, New York. with James Selam and family). 1990. Neh’i-Wana “The Big River”: Mid- Columbia Indians and Their Land. Univer- sity of Washington Press, Seattle. Jensen, Adolf E. 1948. Die Drei Stréme: Ziige dem geistigen und religidsen Leben der Wemale, einem primitiven Volk in dem Mo- lukken. J.W. Goethe-Universitat, Frobe- nius-Institut, Leipzig. Jensen, Adolf E. and H. Niggemeyer. 1939. Hainuwele: Volkserzaihlungen von der Mo- lukken-Insel Ceram. J.W. Goethe-Univer- sitat, Forschungsinstitut ftir Kultur- morphologie, Frankfurt. Levi-Strauss, Claude. 1973. From Honey to Ashes. Cape, London Lewis, I.M. 1991. The spider and the pan- golin. Man 26:513-525 Lindsay, Biddy Yingguny, Kitty Waliwara- rra, Frances Miljat, Helen Kuwarda, lakmalak and Matngala Plants and Ani- mals: Aboriginal Flora and Fauna Knowl- edge from the Daly River Area, Northern Australia. Northern Territory Botanical Bulletin No. 26, Parks and Wildlife 22 HEALEY and FLOREY —— of the Northern Territory, ees Donald. 1892-93. Anthropolog- ical notes on the Aboriginal tribes of the Daly River, north Australia. R Society of South Australia Tacnsactiihe and Proceedings 17:254-264. Menzel, Peter and Faith D’Aluisio. 1998. Man Eating Bugs: The Art and Science of Eating Insects. Ten Speed Press, Berke- ley. Meyer-Rochow, V. Benno. 1976. The use of insects as human food. Food and Nutri- tion 33:151-152. . 1978/79. The diverse uses of in- sects in traditional societies. Ethnomedi- zin 5:287-300. Novotny, Rachel, Fooi Ling Ng, Fred Ru- malatu, and Pick Yee Tang. 1996. Nu- ased foraging mment in eastern Indonesia. Cakalele, Maluku Research Journal 7:9-16. Posey, Darrell A. 1981. shi a basa and fearless men: ethnoentomology Kayap6 Indians of eat Brazil. pale of Ethnobiology 1:165-174. ———. 1983. Keeping of stingless bees by the Kayapo’ Indians of Brazil. Journal of Ethnobiology 3:63-73. ———. 1986. Topics and issues in ethnoen- tomology with some suggesti for of Pospisil, Leopold. 1963. Kapauku Papua Economy. Yale University Publications in in Anthropology, no. 67, New Hav Ruddle, K. 1973. The human use of insects: examples from the Yukpa. Biotropica 5: 94-101. Ruddle, K., D. Johnson, P.K. Townsend, and J.D. Rees. 1978. Palm Sago: A Tropical Starch from Marginal Lands, yesh Australian National University Pres Sachse, FJ.P. Peas Het eiland Seran en nie bewoners. Brill, Leiden Sahlins, Marshall. 1972. Stone Age Econom- ics. Athlone, London. Setz, Eleonore Zulnara Freire. 1991. Ani- mals in the Nambiquara diet: methods of collection and processing. Journal of Ethnobiology 11:1-22. Vol. 23, No. 1 —— George B. 1981. Hunter and Habitat in the Central Kalahari Desert. Suinbaddes University Press, Cam- brid ridge. Speiser, Felix. 1996. Ethnology of Vanuatu: arly Twentieth Century Study. inally published 1923 as Ethnographische materialien aus den Neuen Hebriden und den Banks-Inseln. C.W. Kreidel’s Verlag, Berlin. Spencer, Baldwin. 1914. Native Tribes of the Northern Territory of Australia. McMillan, London Stork, Nigel E. and Martin J.D. Brendell. ay of Seram, Maluku, Indonesia, eds. 1.D. Ed- wards, A.A. Macdonald and J. Proctor, pp. 115-130. Raleigh International, An- dover. Stresemann, E. 1923. Religidse Gebrauche auf Seran. Tijdschrift voor Indische Taal-, Land- en Volkenkunde 62:305-424. — Odo D. 1918. Patasiwa und — a Seran und seinen idea Voigtlander, Leipzig. Taylor, Paul Michael. 1990. The Folk Biology of the Tobelo People: A Study in Folk Clas- sification. aes eage Institution Press, Washingt ! Waddy, J.A. 1988. Classification of Plants and tional University North —— Re- search Unit Monograph, Wolff, Xenia and Margaret J. sn 1998. voreges AARNE A and culinary practices among the Alune of west Se- ram, with j implications for the changing significance of cultivated plants as pp. 267-321. Centre for Resource and Environmental Studies, ANU, Canber- ra. Zeitlyn, David 1993. Spiders in and out of 0 styles of spider divinatio: ciological contexts. Africa 63:219-240. Journal of Ethnobiology 23(1): 23-46 Spring/Summer 2003 “HEAD OF SNAKE, WINGS OF BUTTERFLY, AND BODY OF CICADA’: IMPRESSIONS OF THE LANTERN-FLY (HEMIPTERA: FULGORIDAE) IN THE VILLAGE OF PEDRA BRANCA, BAHIA STATE, BRAZIL ERALDO MEDEIROS COSTA-NETO® and JOSUE MARQUES PACHECO? * Departamento de Ciéncias Biolégicas, Universidade Estadual de Feira de Santana, Km 3, BR 116, Campus Universitdrio, CEP 44031-460, Feira de Santana, Bahia, Brazil E-mail: eraldont@mail.uefs.br > Departamento de Biologia Evolutiva e Ecologia, Universidade Federal de Sao Carlos, Rod. Washington Luis, Km 235, Caixa Postal 676, CEP 13565-905, Sao Carlos, Sao Paulo, Brazil E-mail: djmp@power.ufscar.br To the memory of Darrell Addison Posey (1947-2001) ABSTRACT.—Four aspects of the ethnoentomology of the lantern-fly (Fulgora la- ternaria L., 1767) were studied in Pedra Branca, Brazil. A total of 45 men and 41 women were consulted through open-ended interviews and their actions were observed in order to document the wisdom, beliefs, feelings, and behavi lated to the lantern-fly. People’s perceptions of the external shape of the insect influence its ethnotaxonomy, and they may categorize it into five different ethnosemantic domains. Villagers are familiar with the habitat and food habits of the lantern- fly; they say it lives on the trunk of Simarouba sp. (Simaroubaceae) by feeding on sap with the aid of its ‘sting’. The culturally constructed attitudes toward this insect are that it is a fearsome organism that should be exterminated whenever it is found because it makes ‘deadly attacks’ on plants and human beings. Local ideas about the origin of the lantern-fly, the metamorphosis process, as well as its transformation into another organism were also recorded. The insect inspires feel- ings of fear and aversion which create obstacles to developing an efficient strategy for the conservation of Fulgora species. Environmental education can play a sig- nificant role in changing these negative attitudes. Key words: ethnoentomology, folk knowledge, Hemiptera, Fulgoridae, Fulgora Ia- ternaria. RESUMO.—O artigo refere-se 4 etnoentomologia da jequitiranabéia (Fulgora later- naria L., 1767), baseando-se nas quatro dimensG6es conectivas que Os seres hu- manos podem manter com o inseto. O trabalho de campo foi realizado no po- voado de Pedra Branca entre os meses de fevereiro a maio de 2001. Foram con- sultados 45 homens e 41 mulheres através de entrevistas abertas e observac6es comportamentais, com o objetivo de registrar os conhecimentos, as crengas, 0s sentimentos e os comportamentos relacionados com o inseto. Os resultados de- monstram que a percep¢4o que os moradores tém da aparéncia externa do animal influencia em sua etnotaxonomia, uma vez que foi categorizado em cinco domi- nios etnossemanticos distintos. Os conhecimentos nativos referentes ao habitat e a ecologia trofica da jequitiranabdia revelam que ela vive nos troncos de Simarouba 24 COSTA-NETO and PACHECO Vol. 23, No. 1 sp. (Simaroubaceae), alimentando-se da seiva por meio do ‘ferrao. As atitudes culturalmente construidas com relacéo ao inseto colocam-no como um ser que deve ser exterminado ou temido sempre que encontrado devido a crenga do ‘ataque mortifero’ a plantas e seres humanos. Impress6es locais sobre a origem da jequitiranabdia, o processo de metamorfose, bem como de sua transformacao em um outro organismo também foram registradas. Os sentimentos de medo e aversao ao inseto representariam obstaculos para a realizagdo de uma estratégia eficaz de conservacao das espécies de Fulgora. Dai, o papel significativo da edu- cacgdo ambiental para modificar essa visao. RESUME.—Ce rapport étudie quatre aspects de |’éthnoentomologie du fulgore porte-lanterne (Fulgora laternaria L., 1767) 4 Pedra Branca au Brésil entre les mois de février 4 mai de 2001. Les auteurs ont consulté 45 hommes et 41 femmes au total en utilisant un systeme de questions ouvertes. Ils on observé leur réactions pour documenter leurs connaissances, croyances, points de vues, et comporte- ments relatifs au fulgore porte-lanterne. Les résultats montrent que les indigénes classifient l’insecte en cing différents domaines ethnosémantiques, et fondent l’ethnotaxonomie sur leur interprétation de l’apparence extérieure de l’insecte. Ils connaissent |habitat et les habitudes alimentaires du fulgore porte-lanterne: ils disent que l’insecte vit sur le tronc du Simarouba sp. (Simaroubaceae), et se nourrit de la séve avec son ‘dard’. Pour la culture locale, c’est un insecte redoutable a exterminer oti qu'il soit, 4 cause de ses attaques mortelles contre les plantes et les étres humains. Cet article document également les croyances locales relatives a l’origine du fulgore porte-lanterne, 4 son processus de métamorphose, et a sa transformation en un différent organisme. L’insecte inspire des sentiments de peur et de dégofit qui génent la mise au point de vue d’une stratégie efficace pour la protection de l’espéce. Une éducation en matiére d’ environnement pour- rait grandement modifier ces attitudes hostiles. INTRODUCTION Jequitiranaboia damned snake. The reason for thy pains is in the name. Death is what you shall expect, insect! Bug without a delineated shape First a cicada, then a snake, then a moth. orthless even as medicine. Does nothing but wilt trees And disturb the countryfolk. Costa-Neto, 2001 Insects of the genus Fulgora L., 1767 are commonly known as lantern-flies and alligator-headed or peanut-headed insects. Folk beliefs about them abound, es- pecially due to their unusual shape. Since the first European colonization of the New World, chroniclers, travelers, and natural historians have recorded native impressions of these strange insects (Hogue 1993). The species Fulgora laternaria L., 1767 (= Laternaria phosphorea L., 1764), for example, is supposed to bear a devastating poison that dries up those trees on which it feeds or rests, and also kills both men and animals (Carrera 1991; Costa Lima 1942; Fonseca 1926, 1932; Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 25 Janzen and Hogue 1983; Neiva and Penna 1916; Poulton 1928; Wied 1940). This belief is widespread from the Atlantic to the Andes, and is shared not only by the simple and superstitious but by persons of higher education (Poulton 1932). In the northeastern Brazilian State of Cearé the insect’s folk name is synonymously used to describe a terrible person (Lenko and Papavero 1996) and is applied to any individual who has lost his good reputation. In Peru, the chicharra machacui, as it is locally known, is as dreaded as a serpent because people believe its sting is equally mortal (Dourojeanni 1965). In Costa Rica, peasants believe that the insect’s huge, peanut-shaped head is full of poison. If someone is stung by the insect, he or she must have sexual intercourse within twenty-four hours. Other- wise, he or she will die (Ross 1994). According to Ross, urgency of treatment varies: one woman told him that a ‘cure’ would be necessary within 15 minutes, and that, for a man, a virgin would provide the best antidote. This legend is partly blamed for Costa Rica’s soaring birthrate. It is not surprising that, in Co- lombia, the colloquial expression picado por la machaca (stung by the lantern-fly) is applied to a person who has a great sexual appetite (Anzola 2001). However, this seems to be more of a ruse invented by local men and used to their personal advantage than a valid folktale (Hogue 1993). Due to their significance in legends, lantern-flies are represented in the graph- ic and plastic arts, as well as in the music of different South American countries. In 1987, on the occasion of the fiftieth anniversary of the Brazilian Society for Entomology (BSE), the Brazilian Post Office issued a set of two commemorative stamps. One of them depicted the species E servillei Spinola, 1839 (actually Fulgora laternaria), which is the symbol of the BSE. Similarly, the Colombian Society for Entomology has the anecdotal periodical La Machaca as one of its newsletters. In the folkloric music of Ecuador and Colombia, the fast cumbia rhythm is said to reflect the emotions that follow from the insect’s bite (Ross 1994). The insect is still regarded as a tourist attraction and has value as a souvenir (Hogue 1984). In 1964, a specimen was sold for nine dollars in Tingo Maria, Selva Central, Peru (Organizacgao dos Estados Americanos 1987). Fulgora spp. belong to the order Hemiptera, suborder Fulgoromorpha, super- family Fulgoroidea, and family Fulgoridae (Bourgoin and Campbell 2002). Ful- gorids may be distinguished by a combination of the second hind tarsomere with a row of apical spines and both apical and anal area of hind wings with cross veins (O’Brien and Wilson 1985). According to these authors, Fulgoridae is com- prised of 108 genera and 543 species, which are distributed in the following geo- graphical zones: nearctic (16 species), neotropical (242 species), Ethiopian (104 species), Oriental (180 species), and Australian (18 species). These numbers, how- ever, need to be updated. The genus Fulgora ranges from southern Mexico to northern Argentina, and is represented by eight species (O’Brien 1989). The ge- neric name probably owes its origin to the ancient Roman goddess Fulgora, who protected houses against lightning and terrible storms. Fulgor is the Latin word for lightning, brightness (Ross 1994). Although fulgorids are notable for their size (some species are 7 mm in length, but some are 95 mm [O’Brien and Wilson 1985]), bizarre forms, brilliant colors, and wax secretions, there is very little scientific information about the biology and life cycle of the large neotropical members (Hogue et al. 1989). The exceptions are 26 COSTA-NETO and PACHECO Vol. 23, No. 1 for those species reported to be of economic importance, such as Phrictus diadema (L.) Spinola, 1839 on cocoa trees (Theobroma cacao L.) in Brazil and Pyrops candelaria (L.) on longan (Euphoria longana Lam.) and mango (Mangifera indica L.) in Asia (O’Brien 1989). There are, however, some initiatives expanding our knowledge of their biology, such as the project “Biodiversity and Evolution of Fulgoromorpha: a Global Research Initiative,” by Bourgoin and Hoch (1999). The present work investigates the knowledge, beliefs, feelings, and behaviors that are related to the lantern-fly in the village of Pedra Branca, Bahia State. It is hoped that ethnoentomological knowledge will contribute to better scientific un- derstanding of this group. METHODOLOGY Data presented here are part of a broader research project that aims to record the ethnoentomology of Pedra Branca’s villagers. A former settlement of the Kiriri Indians that was established by the Portuguese pioneer Gabriel Soares de Souza in the sixteenth century, the village is located at the Middle Paraguacgu, west central region of Bahia State, northeastern Brazil (Paraiso 1985). It is inside the municipality of Santa Terezinha (which is also the capital), but it is about 13 km away from it. It is situated at the base of the Serra da Jibéia, a mountain range of about 225 km? of area whose peak elevation is 805 m above sea level. It lies between 12°46’ south latitude and 39°32’ west longitude (Junca et al. 1999). In 1991, the resident population in the county of Santa Terezinha was 8,851 individuals (Centro de Estatistica e Informacg6es 1994). The actual population in the village of Pedra Branca is nearly 400 persons (about 80 families according to the local Health Assistant), who depend on cassava cultivation (Manihot esculenta Crantz.) as their main economic activity. Livestock is also important, mainly cattle and goats. This region, which is totally included in the Drought Polygon, has a semi- arid climate with a mean annual temperature of 24.3°C and a mean annual rainfall of 582 mm. The rainy period lasts from November to January. The vegetation of the Serra da Jib6ia includes campo rupestre savannas on the peaks; dense, ombro- philous Atlantic coastal forest in the valleys and on the slopes; semi-deciduous forest at the base; and arboreal Caatinga in the north. The soil is good for agri- cultural activities and not bad for livestock-raising (Centro de Estatistica e Infor- mac6es 1994). Fieldwork was carried out over 64 days from February to May 2001 by one of the authors (EMCN), who also did the translations into English. Both open- ended interviews followed ethnoscientific principles (Posey 1986b; Sturtevant 1964). Informal observations of behavior related to lantern-flies were also record- ed. Forty-five men and forty-one women, whose ages ranged from 13 to 108 years old, constituted the sample universe. This sample accounts for just those inter- viewees who provided information about the lantern-fly. Interviews were con- ducted in Portuguese since the villagers are Portuguese-speakers. Both individual and collective interviews were done to elicit native impressions on the insect, and people talked freely about other insects as well. Most of the interviews were re- Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY rag corded in microtapes; semi-literal transcriptions are deposited at the Laboratory of Ethnobiology of the Universidade Estadual de Feira de Santana (UEFS). Data were analyzed by using the union model (Marques 1991), which involves considering all available information on the surveyed subject. Controls were per- formed both through consistency checking tests and reply validity tests, which make use of repeated inquiries in synchronic and diachronic conditions, respec- tively. One tests consistency by asking different people the same question within a very short time period. Reply validity is tested by asking the same question to the same person at different times. Two undergraduate volunteers, who have been in the village three times, helped the authors interview the subjects. During the fieldwork period just one specimen of lantern-fly was collected by a villager, when it suddenly appeared in the village one night. This allowed us to conduct projective tests. These consisted in displaying both the photograph and the specimen itself to the informants in order to prompt them to talk about the insect. Their reactions and those of the rest of the members of the community (many of whom had never seen the insect before) were recorded during the in- terviews. The specimen, which was identified as Fulgora laternaria L. 1767,! was handled in accordance with the usual patterns for scientific collections and was deposited in the entomological collection at UEFS. RESULTS AND DISCUSSION The relationship between the Pedra Branca villagers and the insect has four dimensions: cognitive, ideological, affective, and ethological. With regard to the cognitive dimension, native knowledge about the lantern-fly’s ethnotaxonomy, habitat, feeding ecology, and its transformation into another being were recorded. The way people behave toward it (ethological dimension) results from the way they perceive it (ideological dimension) and how they feel about it (affective di- mension). All the interactive processes that occur between villagers and the lan- tern-fly (and the rest of the biotic elements from the surroundings as well) pass through these four dimensions. Despite being cryptic, nocturnal, solitary, silent, and rare, Fulgora laternaria stands out as one of the insects that has a cultural importance to these villagers. Its importance is not utilitarian, since this insect is neither a food nor a medicinal resource. Rather, it is ‘good to think’ in the Lévi- Straussian sense (Lévi-Strauss 1989). Some of the gender-based differences related to the ethnodiagnostic criteria (morphological, biological, and noxious criteria) which were attributed to the lantern-fly are shown in Table 1. In the village of Pedra Branca, Fulgora laternaria is known by at least six dif- ferent names. Twenty-five interviewees called it a jitiranabéia; eleven referred to it as a jitirana; nine treated it as a cobra-de-asa; eight referred to it as a tiranaboia; three termed it a cobra-cega. A single informant called it a serra-velha. Several synonyms are found throughout Brazil. These are: gitirana, jitirana, jaquiranabéia, jaquitiranabéia, jequitiranabéia, jiquitiranabéia, jaquitirana, jequitirana, ji- tiranabéia, tiranabéia, tirambéia, cobra-voadora, cobra-do-eucalipto, cobra-de-asa, cobra-do- ar, cobra-cigarra, serpente-voadora, gafanhoto-cobra, cigarra-doida, cigarra-cobra, and ja- caré-namboya (Becker 1976; Buzzi 1994; Cascudo 1972; Lenko and Papavero 1996). Thering (1963), however, says that the term jaquiranabéia is the original term. Et- 28 COSTA-NETO and PACHECO Vol. 23, No. 1 TABLE 1.—Gender-based differences of the diagnostic criteria used to describe the lantern- fly (Fulgora laternaria) (Hemiptera, Fulgoridae) during open-ended interviews performe with 86 residents of the village of Pedra Branca. Gender Male Female Diagnostic criteria (n = 41) Percentage (n = 45) Percentage Morphological criteria Head’s conspicuousness 7 8.1 10 11.6 Absence of eyes 7 8.1 8 93 Absence of mouth 2 2.3 ) 0 Presence of stin 13 15.1 - 8.1 Presence of eye spots 2 20 1 ie f 1 1.1 0 0 Biological criteria Reproduction 1 1.1 0 0 Habitat (Serra da Jibdia) 8 9.3 9 10.4 ding habit 1 44 3 3.4 Host tree 8 9.3 2 ia Change to another being v4 po 0 0 Noxious criteria It kills/dries trees 13 jie i 22 Zoo It kills/dries people iZ 13.9 15 17.4 It causes blindness 0 0 1 1 It is venomous* 12 13.9 19 22.0 * This noxious characteristic includes others like ‘angry’, ‘bad’, ‘dangerous’, ‘harmful’, and ‘fierce’. ymologically, the word jaquirana comes from the Tupi-Guarani language: fiakyra means cicada (Sampaio 1995). In the 1926 issue of Revista do Museu Paulista the term jakiranaboia appears. According to Cruz (1935), it is a corruption of andira- naboia, which means a bat-like animal (andira) with a snake body (mboia). Tastevin (1923) and Carrera (1991) corroborate the Tupi-Guarani origin for the word, which can be glossed as snake-like cicada (yaki ‘cicada’, rana ‘similar’, mboya ‘snake’). By using this folk name, indigenous peoples have recognized the resem- blance between Fulgora species and cicadas. Both are jumping, free-feeding he- mipterans. In folk biological classification systems, names that cross the bound- aries of communities and extend to a larger region have gained great cultural significance (Berlin 1992). In the nomenclature system of the Jibaro-Aguaruna Indians, the lantern-fly is known as manchi dapi (Guallart 1968). Among the Bororo Indians, the term arée eporéu is the generic designation given to these insects; it means an insect similar in its external shape to a corpse wrapped up in mats (Albiseti and Ven- turelli 1962). The Xerente people call it anquecedarti, which means flying-snake (Posey 1986a). The Canela Indians who inhabit the south of Barra do Corda, Maranhao refer to it both as ka-no-iard and heganunui. Unfortunately, the ety- mology of these words has not been provided (Vanzolini 1956-58). The Significance of the Insects External Features for Naming and Folk Perception.—The abundance of terms currently used to designate the insects of the genus Fulgora Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 29 FIGURE 1.—Lateral view of Fulgora cf. laternaria L.; specimen is 50 mm long. (Drawing based on photograph of a specimen collected in the study area). presumably results from their awesome appearance (Figure 1). When the infor- mants talked about the insect, they mentioned the most prominent character— the head—whose shape reminded them that of a snake’s or a cayman’s head. A similarity to a chestnut was noted too, as can be seen in the following interview- ees’ assertions: The head is strange. It looks like a chestnut. (Mr. E., 62 years old) Its head looks like a cayman’s head. (Mrs. E., 34 years old) The head is very ugly. It is like a snake. (Mrs. V., 58 years old) I am explaining that its head reminds [me] of an alligator’s head. Have you ever seen it? It has a closed mouth and its head is spongy, very spongy, isn’t it? It has nothing inside. And it is horrible. (Mrs. N., 38 years d) It looks like a moth, but its head is like a snake's. (Mr. P., 54 years old) Much as people native to the region remark on the similarity of lantern-fly to snakes and crocodiles, scientifically-trained observers do, too. Gilmore (1986) comments on the insect’s “swollen face, which is fantastically like a cayman’s head, [and] even reproduces its protuberant eyes and sharpened teeth.”’ Spix and Mar- tius (1938) had already noticed this resemblance as they wrote the folk name jacaré-mamboya, the cayman-like snake. When Poulton (1932) described two spec- imens coming from the Brazilian Amazon, he reported that the entire visible surface of the insect in an attitude of rest (except the wings) reminded him of a cayman. As O’Brien and Wilson (1985) stated, members of Fulgora have a head that resembles a peanut (dorsal view) or the head of an alligator or cayman (lateral view). The scientific name given to E crocodilia Brailovsky and Beutelspachen, 1978 from Mexico reveals the resemblance of this species to a cayman (Brailovsky and Beutelspachen 1978). A certain likeness with the head of snakes can be admitted, especially if the following features are taken into account: the lateral square mac- ulae to the labial scales and pits of boids, and in some Fulgora species, a black spot between the false eye and nostril to the loreal fossa of arboreal pit vipers of the genus Bothrops (Hogue 1984). The insect has also been compared to a winged dragon (Cascudo 1972), and Hogue (1993) introduced the common name dragon- headed insect for F laternaria based on the shape and mimetic pattern of the large 30 COSTA-NETO and PACHECO Vol. 23, No. 1 head protuberance that he believed actually simulates the upturned head of a medium-sized, arboreal lizard. According to Fonseca (1926), the structure of the cephalic appendix in this and other fulgorid genera of the Fulgorini tribe (Phrictus Spinola, 1839 and Ca- thedra Kirkaldy, 1903) is owed to the “extraordinary development of certain re- gions of its surface (vertex, front, faces, and so on), which extended forward like a bladder and constitute a gibbosity.’” Someone said the head is usually hollow, but there is a sac which is connected to the digestive system (Grassé 1952)— presumably sap can be stored there for later digestion.” It is believed that the frontal region’s protuberance, which in some fulgorids is extended like a volu- minous process similar to those of membracids’ thoracic structures (Grassé 1952), is a defense against natural enemies (birds, lizards, and small mammals). How- ever, there are no reports that confirm the protective advantage of this formation (Hagmann 1928). Birds, for example fly-catchers, are predators of other fulgorids, as shown by the analysis of stomach contents and photographs.° Two folk species of lantern-fly seems to inhabit the region of the Serra da Jibdia. According to a single interviewee who provided that information, the true species of jequitiranabéia possesses a round head, whereas the false one is slender Mr. T., 34 years old). Actually, one might hypothesize the existence of more than one species of Fulgora living sympatrically in this area, since three other species are found within the state of Bahia: FE lampetis Burm., 1845, EF graciliceps Blanchard, 1849, and EF lucifera Germar, 1821 (O’Brien 1989). Thus, further ethnotaxonomic studies are urgently needed. Perhaps accurate recording of trees on which eggs are laid would help.* The ‘sting’ that everybody fears is nothing but the piercing-sucking stylet located in the middle line of the body and folded between the legs (Santos 1987). It is only when the insect is going to suck the phloem from plants that this long ‘murderer dart’ (Cruz 1935) is extended. Dukinfield Jones, who has spent many years in Brazil, corroborated the statement about the native superstitions by not- ing the insect had a poisoned spine or point at the end of its head that is capable of flying at a man’s chest and inflicting a wound (Poulton 1928). When people talk about the head, they always refer to the ‘sting’ as well. They think it is the vehicle the insect uses to ‘inject the mortal poison’: It has a sting in its belly. If it strikes a tree it dries up. It can be a jackfruit tree [Artocarpus integrifolia L.], it can be a coconut tree [Cocus nucifera L.], whatever. Even if it strikes a person he/she will die. (Mr. M., 57 years old) Its sting is beneath [the body]. In the moment it is going to sit on a person, then it stretches the sting out. (Mr. L., 41 years old) It has a mischievous sting. When it drives the sting against the tree it kills the plant. (Mrs. S., 82 years old) It is said that the danger is when it is furious. When it flies it extends the little beak (‘sting’) forward. Wherever that beak touches ... Cause it is venomous ... It is not fierce when it is calm. (G., 22 years old) It doesn’t have a mouth but a sting. (Mr. E., 80 years old) Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 31 Another morphological characteristic that was emphasized by the informants refers to a presumed lack of eyes. Of 86 interviewees, 15 have called attention to the insect’s blindness. One of the main reasons for the widespread panic when a lantern-fly is seen near the community is this supposed blindness. As stressed by Fonseca (1926), when the insect “flies in the middle of the living beings it slaughters lives without distinction of class.” The following testimony describes the dread people feel about its zigzag flight: ‘That tiranabdia is like this. For example, people must have a lot of defense because .. . If it comes flying, where it ... Because it is blind. It strikes. The tree dies. If a person, it is also said that [if] it strikes, [he/she] dies” (Mrs. E., 52 years old). However, the apparent blind- ness of the lantern-fly has been questioned by one of the informants: ‘People say it is blind. But what! Once I killed one and I saw two eyes like those of a cicada’ (Mr. E., 62 years old). Besides the head, the wings deserve some attention because of the eye spots: “Tt has marks like these on the wing. It looks like two eyes that we see when it flies’ (Mr. E., 62 years old). On another occasion, this same informant said: ‘I know it is beautiful when the insect is flying because there are two eyes beneath the wings.’” One female informant compared the eye spots with those markings on the peacock’s feathers. The literature records that Fulgora laternaria resembles an owl butterfly (genus Caligo, the ‘witch’ in the local perception) because the hind wings, shorter and wider than the fore wings, present large marks that look like the owl’s eyes (Ihering 1968; Penny and Arias 1982). Such eye spots would seem to serve a startle or warning function as well (O’Brien 1989). As Ross (1994) states, false eyes are much more frightening when revealed unexpectedly, causing hesitation or delay, in a nervous predator’s decision to attack. Interviewees stressed the presence of ‘ash’ released by the insect. Fulgorids are known by their wax secretion, whose white filaments solidify in contact with the air and take the aspect of a substance resembling flakes of asbestos (Ihering 1968). Some species (e.g., Cerogenes auricoma Burmeister, 1835) produce elaborate trailing plumes of white wax from the abdomen. Fulgora laternaria do not develop this trailing plume, but the thin, white, powdery wax is often so abundant that it covers part of the insect’s body. This helps the insect to look like the lichens or scars on the bark (Carrera 1956). In fact, fulgorids’ primary defense is their ability to be homochromous with the substrate on which they live (Robinson 1982). The white powder that covers its body has been regarded as a strong emetic. The simple inhalation of it was enough to provoke vomiting (Burmeister 1952). This secreted wax is considered highly aphrodisiac in Colombia (Anzola 2001). How the Lantern-fly Was Categorized—The way the villagers of Pedra Branca per- ceive the jequitiranabéia’s external morphology plays a preponderant role in their ethnoentomological classification system. The shape of the head, the presence of eye spots on the hind wings, the presence of a ‘sting’, the wax production, and the presumed absence of mouth and eyes are all salient features that contribute to the imaginary construction of an animal potentially deadly to men, animals, and plants. Depending on the way informants perceived the lantern-fly it could be assigned to five distinct ethnosemantic domains. About 47% of the 86 inter- viewees classified it as a snake, 10% of them regarded it as a moth, 8% classified 32 COSTA-NETO and PACHECO Vol. 23, No. 1 it as a cicada, 3% considered it as a beetle, and 1% thought about it as a grass- hopper. This ethnocategorization appears in the local expression ‘‘It is a kind of.” The other 32% of the respondents gave no information related to the insect’s folk classification. Some examples of the informants’ statements concerning the insect ethnotaxonomy are cited below: It is a beetle, but it has the shape of a snake. (E., 24 years old) It imitates a moth when its wings are folded. (Mr. A., 56 years old) It is a large insect resembling a moth. It has a caterpillar’s face. (Mrs. M., 55 years old) A brave beetle. It is not a snake, but a beetle. (Mr. Q., 33 years old) It is a snake, isn’t it? A winged-snake. (Mr. Z. P., 108 years old) It is a kind of grasshopper ... (Mrs. L., 78 years old) People say it is a venomous snake and it is a kind of cicada. (Mr. D., 78 years old In Pedra Branca, the continual inclusion of Fulgora laternaria in the ‘snake’ domain and the strong aversion to it was observed through the projective tests. When a villager captured one specimen, he did not touch it and he was followed by a small group of curious people who wanted to see the weird creature more closely. On that occasion, they warned that a winged-snake should not be han- dled! One informant, who wondered about the presence of ‘feet’ (legs) as she was looking at a picture, questioned the insect’s classification as a venomous snake: ‘Is this the winged-snake? Even on photograph I had no knowledge about it. A footed-winged snake? It resembles much more an insect, a thing. With leg and everything! Snake creeps” (Mrs. T., 68 years old). As it was noted, ‘snake’ was the ethnosemantic domain used by the majority of the informants to classify the jequitiranabéia. Yet snake (the animal itself) can be also considered as a kind of insect, since the lexeme ‘insect’ includes other taxa beyond Insecta in the ethnobiological classification systems. For example, the Pan- kararé Indians from Brejo do Burgo village, northeastern Brazil, view snakes as ‘insects’ because they cause damage to people and domestic animals (Costa-Neto 1997). However, the boa (Boa constrictor L., 1768) is not considered an ‘insect’ because it is useful (they eat it as food). Costa-Neto (2000a) has explained the way human societies construct the ethnocategory ‘insect’ through the Entomo- projective Ambivalence Hypothesis: human beings tend to project attitudes and feelings of harmfulness, danger, irritability, repugnance, and disdain toward non- insect animals (e.g., toads, rats, scorpions, vultures, snakes, bats, lizards, earth- worms, spiders, among others), by associating them with the culturally defined category ‘insect. The idea of ambivalence comes from sociology and relates to the attitudes that oscillate among diverse, and sometimes, antagonistic values. Pro- jection results from the psychological processes by which a person attributes to another being the reasons for his/her own conflict and/or behavior. Accordingly, ‘insects’ can be seen as a representational category since they become metaphor- ical realizations of other beings or their qualities (Greene 1995). Nolan and Rob- Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 33 bins (2001) state that the organization of ethnozoological semantic domains (‘mammal’, ‘snake’, ‘bird’, ‘fish’, ‘insect’, etc.) is influenced both by the emotive meaning and the culturally constructed attitudes toward these domains. Indeed, the way people perceive, identify, categorize, and classify the natural world chang- es the way they think, act, and feel in relation to the animals. As Posey (1986b) points out, folk biological classification systems do not al- ways fit in classificatory schemes that biology artificially attempts to organize. Thus, cognitive categories cannot be considered as universal and must be inferred through a methodological approach that allows the researcher ‘to discover’ the conceptual paradigms instead of impose them on the society under study (Posey 1987). For example, in their folk entomological classification system the Kayap6 Indians from the Brazilian State of Para categorize animals with shells and no flesh as equivalent to insects (Posey 1983). To the Ndumba, an ethnic group that lives in the highlands of New Papua Guinea, tovendi is an ethnocategory that refers to all insects and arachnids (Hays 1983). In some contexts, however, tovendi can be assigned to inedible animals (e.g., some types of toads), while in other contexts it can mean any creature considered disgusting (e.g., snakes). Considering Berlin's principles of categorization (Berlin 1992), the term ‘insect’ and its similar (emic) equivalents usually represent the level of classification as- sociated with a life-form category. This level of ethnobiological classification is, according to Berlin, the broadest classification of organisms in groups that are apparently easily recognized on the basis of innumerable morphologic characters. Studies of Brazilian ethnoentomology have shown, however, that in folk zoolog- ical classification systems the life-form ‘insect’ is identified and described based not only on morphologic and biological characters, but especially on the psycho- emotional criteria, which are very important when someone is naming the or- ganisms. In other words, folk taxonomies are based not only on the knowledge of biological characteristics (cognitive dimension), but also on feelings (affective dimension), beliefs (ideological dimension), and behaviors (ethological dimen- sion). Traditional Knowledge Concerning the Lantern-flys Ecological Aspects.—Informants’ folk entomological knowledge based on habitat has revealed that Fulgora laternaria lives on the trunk of a tree species that appears to be more common in the Serra da Jibéia than in the other hills. According to two key informants, the jequitir- anabéia ‘stays on the pau-paraiba’’ (Simarouba sp., Simaroubaceae). As one of them has said, “[One] can go anytime and find it. Sometimes, two or three are on the same wood” (Mr. E., 62 years old). The other has added: “Now, through the bushes, there is a wood that is said it is where it stays more. It is on that pau- paratba (.. .). Who knows, sees and says: ‘That is the jitiranabéia over there’ ’’ (Mr. Q., 64 years old). Eight other informants have confirmed the association between the insect and this tree. Because the insect is always seen on the trunk of this tree, people generally associate its emergence with spontaneous generation; that is, they think the insect is born naturally from the wood: “It comes from the pau- paraiba’’ (Mr. Q., 64 years old). At the end of the nineteenth century along the Bahian south coast, lantern- fly was already known as bicho do pau parahy’ba because it frequented this tree 34 COSTA-NETO and PACHECO Vol. 23, No. 1 (Branner 1885). Bondar (1931) stressed that F laternaria lives on Simaba versicolor St. Hilaire in Bahia State. Lantern-flies and several other species of Fulgoridae were observed and collected on trunks of Simarouba amara Aublet in Santarém, Para, and in the region of Marmoré River, in Bolivia (Poulton 1932). The preference for a given tree species has been proved by Johnson and Foster (in Hogue et al. 1989). In a study carried out in a period of five years at Santa Rosa National Park in Costa Rica these authors observed that 98 of 100 adult F laternaria were seen on trunks of Hymenaea sp. (Fabaceae). This preference has a scientific explanation. Fulgoromorpha are generally closely associated with their host-plants that give them food, shelter, and protection against predators (Penny and Arias 1982). Different plants are known hosts to different species of Fulgora. It has been proposed that these plants serve as hosts to Fulgora spp. because they either produce and concentrate resins, oils, or bitter substances in their sap, pos- sibly generating allelopathic chemicals: Simaroubaceae—Simarouba amara Aublet, Simaba versicolor St. Hilaire; Fabaceae—Hymenaea oblongifolia Lee and Langenheim, H. coubaril L., Myroxylon balsamum (L.) Harms; Rutaceae—Zanthoxylum sp.; Lecy- thidaceae—Lecythis sp.; Vochysiaceae—Vochysia tucanorum Martius; Bignoni- aceae—Jacaranda acutifolia Humboldt and Bonpland; Apocynaceae—Aspidosperma tambopatense Gentry; Euphorbiaceae—Hura crepitans L.; Myrtaceae—Eugenia oer- stedeana Berg., Eucalyptus sp. (Cruz 1935; Hogue 1984; Johnson and Foster 1986; Lenko and Papavero 1996; Poulton 1932). Informants have also mentioned that the jequitiranabdia stays fixed to the tree when it dies: “In the place it sits, it stays. There it fixes the sting and does not get out. Then, it dies in that place’ (Mrs. L., 66 years old). Another said: ‘The ancient ones told that the insect had a manner of sitting on green wood. It sat for a long time. Then, it weakened and died” (Mr. M., 68 years old). This fixation to the tree trunk due to its death has been reported also by Francisco Peres de Lima, in 1938 (Lenko and Papavero 1996). Another fulgoroid species was collected on trunks of Simarouba sp. while we looked for Fulgora specimens at the upper slopes of the Serra da Jib6éia. In those occasions, a key informant considered the insects as the lantern-fly’s ‘daughters’. (These are currently under taxonomic analysis.) Hogue (1984) recorded the pres- ence of over 20 specimens of Lystra lanata (L., 1758) as he was examining Simarouba amara in the vicinity of Iquitos, Peru. Informants also told us about the insect’s origin. Individuals from older gen- erations believe that the jequitiranabéia has come from the sertéo (Brazilian arid midland): “There wasn’t that kind of snake here. We only knew it through stories. Because they [people] say it is from the arid midland” (Mrs. E, 52 years old). According to another informant, the lantern-fly has come in the Serra da Jibdéia because “it accompanied the herds of cattle that came from the sertdo, from distant places” (Mrs. M., 62 years old). A third informant has stated that this insect has come from the South (of Bahia State?). The notion that this fulgorid comes from this arid environment was once used by the lexicographer Candido de Figueiredo in his incongruous entry: ‘‘Venomous butterfly from the sertio’”’ (Santos 1987). Traditional and Scientific Knowledge of Jequitiranabéia’s Feeding Habits.—Villagers of Pedra Branca referred to the insect’s feeding habits: “It feeds on the humidity of Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY a5 the wood” (Mr. E., 88 years old) because it ‘‘sucks from the plant’ (Mrs. M., 30 years old). This ‘“humidity’’ to which the informant referred can be interpreted as the sap, since fulgorids feed exclusively on trees and woody shrubs. They introduce their mouth apparatus (‘sting’ or ‘beak’) through the plant's bark and feed on phloem by turgor pressure (O’Brien and Wilson 1985). Apparently, the informants did not recognize the host tree (pau-paraiba) as being the source of food for Fulgora laternaria. Johnson and Foster (1986) reported that the phloem of Simarouba amara lies just below the smooth, thin outer bark. These authors stressed that this species possesses a phagostimulant on its trunk called simarolide, which is a quassinoid that probably is responsible for the insect’s great attraction to this tree. When hemipterans feed on phloem with an imbalance (for insects) of amino acids, they are able to obtain the food they need through symbiotic associations with microorganisms that live inside specialized cells known as mycetocytes (Chapman 1994). All Fulgoromorpha appear to have more than one type of my- cetocyte microorganism and, in Fulgoridae, both yeasts and bacteria are present. Some species have as many as six different symbionts. The Terrible Effects of its Sting. —The alleged ‘deadly attack’ on plants and human beings was the most cited and the best known of the jequitiranabéia’s behaviors. Since the insect is often perceived and categorized as a snake, people analogically confer on it the same fear they feel for the ophidians. Thus, the following testi- monies were recorded: I have heard people talking about the jitiranaboia. That it is too venomous. I have heard my mother saying that the plant died whenever it was sat on. And there wasn’t that snake here. (E., 24 years old) People fear it because it stings. It is like a snake. The poison that a snake carries it carries too. (Mrs. M., 55 years old) It is a dangerous snake. If it strikes a person it kills her. If it stings even wood the plant dies. (Mrs. E., 82 years old) Although many informants mentioned the danger posed by the jequitiranabéia, there were individuals who questioned the risks attributed to having any dealings with it: “I don’t know. If it was like that many [trees] have already died in the forest’ (Mrs. G., 41 years old); “People say it is venomous, but F (19 years old) took a look at the dictionary and found it is not” (Mrs. E., 52 years old). Carrera (1991) points out that the damages caused to the plants by its sting are insignif- icant and never result in death. Furthermore, these insects are too scarce to be harmful to trees (Ross 1994). Some fulgoroids, however, produce honeydew. This is a sweet, watery excrement that serves as the substrate for the growth of sooty mold. This blackens the leaf, decreases photosynthesis activity, decreases vigor, and often causes disfigurement of the host (Kessing and Mau 2001). Planthoppers also damage plants by ovipositing in plant tissues and by feeding in the phloem, sometimes spreading a variety of plant pathogens. At least three species of the Fulgoromorpha family Cixiidae are suspected vectors of the lethal yellowing of Canary Island date palms in Texas (Meyerdirk and Hart 1982). Considering our 36 COSTA-NETO and PACHECO Vol. 23, No. 1 informants’ testimonies, many of which have stated that trees passed away due to the injuries brought about by the lantern-fly, it could be deduced that this insect probably bears some kind of deleterious virus or bacteria. Maybe one can make a case for carrying out a phytosanitary investigation on those trees struck ‘dead’ by it. : On the other hand, Janzen has seen Fulgora ovipositing and feeding for years on the same trees in Costa Rica, students at La Selva in Costa Rica can point out trees that have had Fulgora on them for several years, and that is true also in Belize in Rio Bravo Conservation Area.° When asked about the occurrence of possible cases of injury and/or death caused by the ‘attack’ of a jequitiranabdia to any member from the village of Pedra Branca, or elsewhere, the informants replied that no real incident has ever been registered. Even so, the belief persists: It is spoken that if someone is stung he will die. But nobody ever saw anyone die. (Mrs. E., 34 years old) Here, when a tree dies, then they say soon: ‘’That tree over there has died because the jitirana has rested on.”” But people have not died here. (Mrs. V., 58 years old) I’ve never heard [about any case of death], but we feel as soon as we see some trees in the forest, all dried up, with wrinkled leaves, and com- pletely lifeless without any reason. It’s just been caused by the insect itself. (Mrs. P., 80+ years old) In the mid-nineteenth century, the lantern-fly was thought to kill animals and trees. Branner (1885) recorded that along the Amazon, when a monkey suddenly came tumbling down dead from the forest canopy without any apparent cause, it was said that he had been struck by the fatal jequitiranabéia. Branner cited a Spanish-American newspaper published two years before, which reported that this insect was said to be destroying the cattle of Brazil in the grazing country of the southern provinces. The idea that Fulgora is very poisonous is so deep-rooted in common attitudes that even an entomologist from the Rio de Janeiro National Museum blamed contact with the animal when he felt bad (Lenko and Papavero 1996). Stories of dramatic and tragic encounters abound in the literature. Bates, an eminent British entomologist who collected insects for eleven years along the Amazon River in the nineteenth century, was once told that one of these ‘dan- gerous’ creatures suddenly emerged from the forest and attacked and killed eight of a nine-member boat crew (Bates 1943). Apparently, the evil attributed to the insect is not a simple belief at all. Ac- cording to Hagmann (1928), Fulgora laternaria may sting when carelessly handled. And incidental circumstances may render it toxic, as when it feeds on sandbox tree (Hura crepitans L., Euphorbiaceae) or other plants that produce toxic or nox- ious allelopathic compounds. Then, it extracts those chemicals and makes itself a bearer of fatal toxins (Orico 1975). It is known that certain insects sequester toxic secondary plant compounds and store them in their bodies, and in this way gain protection from predators and pathogens (Engel 2002). Fortunately, no case of Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY a7 death resulting from the attack of a jequitiranabéia has been found in the scientific literature. If the insect is inoffensive, then what is the basis for such a terrifying tradi- tion? Surely, the physical resemblance with ophidians is a reasonable explanation. But the origin for this fear may be also found in indigenous myths and legends. However, little if any information concerning its presence in native mythology is available. In an Amazonian legend about the Matintaperera, the lantern-fly is used as an instrument of torture (Lenko and Papavero 1996). Because of its anomalous morphology medicine men of many Amazonian tribes regard the insect as mag- ically powerful and carry it (dead) in their amulet bags around their neck (O’Brien and Wilson 1985). me indigenous groups seem to consider the insects (or at least some of them) to be the tangible manifestation of ominous principles; these principles sometimes are attributable to the activity of some malevolent medicine men (Ces- ard et al. in press). To the Palawan people, aggressive and poisonous animals such as dlupjan (centipede), bancanawa (scorpion), katimamang kdtimamang (bird-spider), and sédli (snake) are said to be owned by malevolent non-human agents such as Langgam to whom they are totally obedient and friendly (Novelino 2002). The Munducuru Indians regard lice as the true materialization of the will- ingness of some animals to cause illnesses. The Yora/Yaminahua Indians of the Peruvian Amazon attribute a great number of illnesses to the malevolent spirits of noxious invertebrates such as wasps, which are blamed for gastrointestinal conditions, and a caustic millipede known as xaco, which is associated with re- spiratory conditions. These Indians also blame urinary tract infections on termite spirits: “If one urinated on a termite mound, the termite would take vengeance and cause painful urination’ (Shepard 1999). In fact, the belief in vengeful spirits of stinging insects is part of the Amerindian societies’ folklore, which associates wasps and bees with a variety of mythical forces (Shepard 1999). Different reasons for the consistent human aversion towards insects and other invertebrates, especially among many Westerners, have been proposed in the sci- entific literature (e.g., Kellert 1993). One suggests that people have an innate fear of potentially dangerous insects, which was generalized to include other animals. Another explanation is the association of invertebrates with illnesses and human habitation. A third is suggested by the notion of human alienation to creatures so different and distinct from our own species. To Laurent (1995), the general shape, the morpho-ethological aspects, and the negative sensations people attri- bute to the animals (e.g., disgust, revulsion) are reasons that explain man’s aver- sion to the invertebrates, particularly to the insects. In general, more positive attitudes towards invertebrates can be found when these animals possess aes- thetic, utilitarian, ecological or recreational values (Kellert 1993). In contrast, East Asian peoples have a more balanced perspective regarding insects, where most of them are considered to be aesthetically pleasing, good to eat, interesting pets, subjects of sport, enjoyable to listen to and useful in medicine (Pemberton 1999). Although a genetically inherited process cannot be ruled out, there are a number of theories which allude to cultural and social transmission of some common animal fears (Davey 1994; Matchett and Davey 1991). Due to the socially constructed behaviors toward the jequitiranaboia, people of 38 COSTA-NETO and PACHECO Vol. 23, No. 1 Pedra Branca regard it as an organism that must be exterminated or dreaded whenever it is found. Such an affective representation, which occurs in the brain’s limbic and neocortical organs (Soulé 1997), is done through images, stereotypes, and interpersonal myths. Since individuals are acting for ‘rational’ reasons (al- though scientifically incongruous), it can be said that the set of knowledge (= corpus) about the lantern-fly may be characterized as a kind of cognition that Anderson (1996) calls ‘“hot cognition.’” According to him, the ‘‘hotter’’ the cog- nition of a given object is the better individuals will tend to think, know, speak, and act upon it. It is precisely because the lantern-fly represents a “potential danger” to human beings that it deserves some attention. That is why people generally know something about it, even though they have never seen the insect either in situ or in vivo. As Anderson (1996) emphasizes, emotional factors drive cognition. Traditional Knowledge of the Lantern-flys Reproduction.—With regard to the tradi- tional knowledge related to the lantern-fly’s reproductive behavior, we have just recorded information on the moulting process of the juveniles into adults. A key informant mentioned: ‘The daughters are black. Then, they transformed into large [insects] and change the color. Now, it changes its shape while it is growing. It is this same kind’ (Mr. E., 62 years old). Another said: “The wood raises a beetle that originates it’ (Mr. E., 88 years old). Although these informants know some- thing about the metamorphosis process involving these insects, the ‘daughters’ actually were the adults of another fulgoroid species. At first sight we might think that such a classificatory relationship between two different species is a perceptive anomaly, but this parent-offspring relationship is based on a belief about ontogeny or origin which indicates a close similarity between them (Ellen 1985). It is known that hemipterans develop through paurometaboly, which means their metamorphosis is gradual and inconspicuous (Kessing and Mau 2001). Ap- parently, Hagmann (1928) was the first scientist who described the nymph of Fulgora laternaria. He referred to it as a larva, very weird due to the shape of its long, cylindrical head (Figure 2). It resembles the adult in the possession of the inflated head structure but is wingless and much smaller. There is little scientific knowledge about the jequitiranabéia’s reproduction and life history. Fonseca (1926) stated that ‘both sexes show the same color, design, and size, so that unless by the genitalia characters no superficial difference exists.” The female has a reduced ovipositor, externally smaller that the male external genitalia (O’Brien and Wilson 1985). Literature records data concerning mating and oviposition, which occur on the host plants. Eggs are laid in masses on the surface of bark and glued together with a collateral fluid and covered with wax secreted from the abdomen (O’Brien and Wilson 1985). According to R. W. Hings- ton (in Hogue 1993:240), this structure is similar to a mantid egg case. Local impressions of its transformation into another animal have been also recorded: ‘’People say it turns into a cobra-de-cip6 [maybe Philodryas sp., Colubri- dae}’’ (Mr. C., 32 years old). But this was a misconception, since the jequitiranabéia has been mistaken for the praying mantis (Mantodea). In the local classification system, mantids and phasmids are thought to arise from the branches and twigs of verbena (Lantana camara L., Verbenaceae) and change into snakes. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 39 (GURE 2.—Lateral and top views of the nymph of Fulgora laternaria L.; specimen is 68 mm long. (Redrawn from Hagmann 1928). Other Behavioral Patterns.—Lantern-flies can drum their heads against the trunk of a tree if molested.° The informants have not commented on such behavior. The phenomenon of bioluminescence, which was first recorded by Nehemiah Grew in late 1681 and corroborated by Maria S. Merian’s book Metamorphosis Insectorom Surinamensium published in 1705 (Ross 1994), was not mentioned by them, either. It is interesting to note that Grew erroneously attributed light produced by beetles of the genus Pyrophorus to Fulgora. Many discussions have followed since then. Ridout has studied that aspect with fast frozen specimens, and could not get a response using the chemical components of all known biological luminescence systems.” However, a luminescence in Fulgora may be observed and it is owed to the occasional, and generally deadly, appearance of pathogenic bacteria that de- velop on the abdomen and into the anterior intestinal caecum that is accommo- dated in the cephalic prolongation (Grassé 1952; Ihering 1968). The Amazonian peasants still believe that F laternaria produces a type of prolonged sound in the evening similar to the whistle of a train. However, it is the cicada Quesada gigas Olivier, 1790 that produces this stridulation (Lenko and Papavero 1996). When persistently and sufficiently molested, Fulgora species may emit a vol- atile, fetid defensive chemical released as a ‘’skunk-like spray”’ (Janzen and Hogue 1983). However, no glands specifically for the production of noxious odors seem to exist in the insect’s body. Hogue (1984) suggests that such volatiles could reside in the body’s covering of wax. Additional information on its behavior is found in Fonseca (1926), who writes: They stay motionless, phlegmatic, for hours in one spot, by placing them- 40 COSTA-NETO and PACHECO Vol. 23, No. 1 selves in a manner that their heads are always turned toward the top of the tree. I have never seen them at another position. They are neither brave nor noisy like cicadas. They let anyone get close to their immediate prox- imity and extend the hand to catch them. Then, they move slowly and cautiously to the other side of the trunk. When very bothered they lei- surely rise over their legs, and by impelling the body with the former legs, they fly to another nearby tree, producing a muffled sound with the vibration of the wings during the flight. Of Lantern-flies, Storms, and Electricity—Fulgora laternaria’s rare appearances in the village of Pedra Branca are linked to the storms and rains (what an amazing relationship with the Roman goddess Fulgora!), and invariably it is seen resting on electric poles. As informants say: In times of thunderstorms, of strong thunders, [in] the other day you can look [and] you find it on electric poles. (Mr. J., 78 years old) As soon as the electric light has come more than 50 [insects] have ap- peared on electric poles. (Mrs. M., 36 years old) It gets down from the ‘Serra da Jibéia’ when it is raining. (Mr. E, 40 years As soon as the light has come, people [the parents] didn’t allow anybody to get out. They said: “The snake is crazy! The snake is crazy!’ (Mr. V., 36 years old) The insects of the family Fulgoridae are luciphilous;§ artificial light spots often attract them (Poulton 1932). Fonseca (1926) has noted that “sometimes these in- sects look for light, at night, landing on electric poles or entering through the windows wherever there is some clarity.” After the introduction of electric energy in the village of Pedra Branca, people came into more regular contact with in- vertebrates. Dozens of different insects (e.g., moths, beetles, katydids) and their natural enemies, attracted by luminosity inside the houses, came in. In fact, the establishment of electricity caused great cultural changes. An informant men- tioned that electricity was the reason they felt apprehensive about the insects. The electric poles were placed along the village’s main street, so most of the trees that bordered it have been cut down in order to avoid harboring during the day the insects that were attracted to the lights at night. Of the jequitiranaboia’s nocturnal activity, one key informant has said it ‘walks’ only at night (Mr. E., 62 years old). Hogue (1984) says that specimens of Fulgora laternaria typically rest during the day on the trunks of trees. They position them- selves vertically with their anterior protuberance uppermost and elevated at an angle away from the substratum. As Johnson and Foster (1986) pointed out, the vertical position may be a conservative characteristic of the family Fulgoridae. Hogue (1984) sees in this posture a mimetic correspondence similar to that as- sumed by certain arboreal Iguanidae lizards. According to him, these insectivo- rous lizards probably are the lantern-fly’s closest predators; thus, the insect tries to resemble them. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 41 Conservation Status of Fulgora spp—Considering the actual environmental situation in the region of the Serra da Jibdia, it could be thought that the local subpopu- lation of Fulgora laternaria might be particularly at risk of extinction. The two main anthropogenic causes of forest fragmentation and associated loss of entomofauna are the expansion of cattle-raising and the extraction of wood, which still occurs clandestinely. Conspicuous species, due to their associated ecological specializa- tions, often live in closed or sedentary populations that are considered to be es- pecially threatened by habitat fragmentation (van Hook 1997). Considering the conservation status of Brazilian primary forests, it is reasonable to expect that some species of Fulgora may be present in some red list of threatened animals in the near future. In Venezuela, F laternaria is already listed as one.’ In 1932, Poulton noted that this species was rarer than it was 20 years before. Gabriel Mejdalani,’° a researcher from the Rio de Janeiro National Museum and specialist in leafhop- pers (Hemiptera, Cicadellidae), believes that lantern-flies may be vulnerable since they inhabit ‘the interior of primary forests on the thickest trunks of the oldest trees.” Actually, they are relatively rare because they exist in low population den- sities. To O’Brien, the conservation of Fulgora spp. is conditioned by the mainte- nance of forest preserves.!! It is widely known that public support for conservation continues to rest on emotional rather than intellectual motives, and has been garnered primarily by cute and cuddly vertebrates (van Hook 1997). As van Hook points out, humans most readily learn about, care about, and make sacrifices for animals that are visible, familiar, aesthetically appealing, and that demonstrate positive benefits to nnate fear of insects may also create obstacles to their conservation, especially when species are inconspicuous, unattractive, and economically un- important (Kellert 1993). Thus, as a main contribution of the present research for the conservation of EF laternaria and their kin, we would suggest an environmental education program especially built on emotive basis in order to change, or at least diminish, people's feelings of fear and aversion towards fulgorids. It is hoped that the data now available will be incorporated into a curriculum by those researchers interested in biology conservation and ethnobiology as well. CONCLUSION The set of knowledge, beliefs, feelings, and behaviors that individuals from the village of Pedra Branca possess related to Fulgora laternaria shows that it has some cultural importance. Although people fear it, they think about it and put it in their oral literature. In general, local knowledge of its ethnotaxonomy, ecology, feeding habits, and behavior is in agreement with the scientific entomological knowledge. According to the ethnotaxonomic classification system, more than one species of Fulgora may live sympatrically in the area of the Serra da Jibéia. A further systematic taxonomic survey would clarify this point. The way local people behave toward the jequitiranabéia results from their per- ceptions of and feelings about it. Because F laternaria is categorized into different ethnosemantic domains, especially ‘snake’, the entomoprojective ambivalence hy- pothesis is reinforced. Although it is perceived as deadly poisonous, no actual 42 COSTA-NETO and PACHECO Vol. 23, No. 1 case of injury or death has been recorded. Even so, the culturally constructed attitudes toward it make people kill it whenever they find one. The subpopulation of Fulgora laternaria living in the area of the Serra da Jibdia might be at risk of extinction due to anthropogenic factors. Local people should be involved if we are to achieve an efficient strategy for the conservation of Fulgora and other species. Thus, folk entomological knowledge would not only assist re- searchers in their understanding on the ecological role played by insects, but also would help them to comprehend native cultures (Blake and Wagner 1987). Ad- ditionally, decision-makers would be able to apply proper conservation programs and management practices only if they recognized that the cultural perspective is to be taken into account in every debate focused on biological conservation policy (Costa-Neto 2000b). NOTES 1 Tts taxonomic identification deserves more attention, since other three species inhabit the Atlantic rain forest in Bahia State. 2 Dr. Lois O’Brien, Florida A & M University, Tallahassee, personal communication, 2001. 3 Dr. Lois O’Brien, personal communication, 2001. * Dr. Lois O’Brien, personal communication, 2001. ° Dr. Lois O’Brien, personal communication, 2001. °Fulgomorpha Lists on the Web [on-line: http://fl jussieu.fr/introduction/fulgores- en.html] (verified December 17, 2002) 7 Dr. Lois O’Brien, personal communication, 2001. * Entomologists’ most common way of collecting Fulgoridae is by hanging a night light or an ultraviolet light in front of a white sheet hung on a line between trees. Lois O’Brien, personal communication, 2002. * Fundacion Polar, Caracas [on-line: http://www.fpolar.org.ve /librorojo/insectos.htm] (verified January 13, 2003) '© Dr. Gabriel Mejdalani, Departamento de Entomologia, Museu Nacional, Rio de Janeiro, letter dated July 5, 2001. "Dr. Lois O’Brien, personal communication, 2001. ACKNOWLEDGMENTS The authors are very grateful to all informants who provided cultural data and who allowed us to disseminate their traditional wisdom. Special thanks to Dr. Lois O’Brien for sharing interesting comments. We would like to thank the two undergraduate volunteers, Daniela Ramalho Serra and Silvia Leticia Cerqueira, for helping collect data; the biologist, Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 43 Ivan Castro, for drawing the figures; and the editor, Dr. Miller, for her comments and suggestions. This paper is part of a larger ethnoentomological study carried out by Eraldo Medeiros Costa-Neto for his doctoral dissertation (Etnoentomologia no Povoado de Pedra Branca, Municipio de Santa Terezinha, Bahia. Um Estudo de Caso das Interacgées Seres Humanos/ Insetos) at Sado Carlos Federal University, and which was guided by Professor Josué Marques Pacheco. REFERENCES CITED Albiseti, César and Angelo Venturelli. 1962. Enciclopédia Bororo, volume 1, Vo- cabulério e etnografia. Museu Regional Dom Bosco, Campo Grande. Anderson, Eugene N. 1996. ee of the Heart: Emotion, Belief, and Environ- ment. Paros University st Oxford. Anzola, denies R. 2001. El zooldgico col- ombia pi Available at http:// ocean. se edu/~nanzola/colzoo. html vetied ory 16, 2001). Bates, Henry W. 1943. The Naturalist on the River snot . we Dent & Sons Ltd., London. Becker, Claudio J. 1976. Lendas e curiosi- dades sobre insetos: I—jequitiranabéia. Natureza em Revista 1:16-19. Berlin, Brent. 1992. Ethnobiological Classifi- cation: Principles of Categorization of Plants and Animals in Traditional Socie- ties. Princeton University Press, Prince- on. ee Elizabeth A. and Michael R. Wagner. 1987. Collection and consumption of Pandora moth, Coloradia pandora lindseyi o La Bulletin of the Entomological Society of America 33(1):23-27. Bondar, G. 1931. Madeiras oo Cor- reio Agricola 9(7):149-15 Bourgoin, and sah C. Campbell. 2002. yariiiegte a ih ie for Hemip- tera: falling in trap. Denisia 4, peta Kataloge des OO “nay ti neue Folge 176: 67-82. Linz, Austri Bourgoin, ‘sme gen and Hannelore Hoch. 1999. The BEFRI project: biodiversity dol bach.snv.jussieu.fr/befri (verified Janu- ary, 13 2003). Brailovsky, Harry and Carlos R. Beutel- - ng . 1978. Una nueva especie de Fulgora ‘Linneo (Homoptera: Fulgori- dae) de México. Annales del Instituto de Biologia de la UNAM, série zoologia 49: 175-182. Branner, John C. 1885. The reputation of the fly. The American Naturalist 19: Burmeister, Hermann. 1952. Viagem ao Bras- il através das provincias do Rio de Janeiro e Mi ' ahiciaioe Livraria Martins Editora, Sao Pau. Buzzi, er 1994. Coletdnea de nomes po- pulares de insetos do Brasil. J. Z. Buzzi, Curitiba. Carrera, Messias. 1956. Entomologia para vocé, Chacaras e Quintais, S40 Paulo. ———. 1991. Insetos, lendas e historia. The- saurus, Brasilia. Cascudo, Luis da C. 1972. Diciondrio do folc- lore brasileiro. Ediouro, Rio de . Centro ecole e Informacdo. 1994. Inform — sobre os municipios avo Sul. Governo do tado da Thomas. Peeters-Selaf, Paris. Chapman, Reginald F 1994. The Insects: Structure and Function. 4th ed. Cam- bridge University Press, Cambridge. Alfredo da. 1942. Insetos do Elisabeth Motte-Florac and Jacqueline M. C. en — Escola Nacional de Agron- a, Sao Paulo. Pree “et M. 1997. Etnotaxon- omia zoolégica do grupo indigena Pan- kararé do Raso da Catarina, Bahia. In Resumos do 11 encontro de zoologia do nordeste, p. 84. Universidade Federal do Ceara, Fortaleza. 00a. The significance of the cat- egory ‘insect’ for folk biological classi- fication pana ‘agile of Ecological An- thropology 4: 2000b. Seepacae development and traditional knowledge: a case study in a Brazilian artisanal fisherman's wee ger ye Sustainable Development 8(2):89— Cruz, nis 1935. Na terra das igacabas: contos, mythos e folk-lore da Amazonia. Of- ficinas Graphicas do Instituto “Dom Macedo Costa”, Belém Davey, Graham C. L. 1994, The “disgust- ing” spider: the role of disease and iones in the perpetuation of 3 spiders. So- ciety and Animals 2(1):17- Dourojeanni, Marc J. 1965. Denomina- ciones vernaculares de insectos y algu- nos otros invertebrados en la selva de Pert. Revista Peruana de Entomologia 8(1): 131-137. Ellen, Roy F. 1985. Species transformation and the expression of resemblance in Nuaulu ethnobiology. Ethos 1(2):5-14. Engel, Cindy. 2002. Wild Health: How Ani- mals Keep Themselves Well and What We Can Learn From Them. Houghton Mifflin Company, New York. Fonseca, José P. da. 1926. As espécies bras- ileiras do género “Laternaria’’ (Fulgori- deos). Revista do Museu Paulista 14:471- 501 . 1932. Novos subsidios para 0 con- hecimento do género Laternaria L. (Homopt., Fulgoridae). Revista de Ento- mologia 2(1):1-5. Gilmore, Raymond M. 1986. Fauna e etno- zoologia da América do Sul tropical. In Suma etnolégica brasileira, volume 1, Et- Grassé, Pierre-Paul. 1952. Traité de zoologie. Anatomie, systématique, biologie, volume 10, fase. 2: insectes supérieures et hémip- 3(1). [Online]. Available at sel // www.psyeta. sa reve 1/greene.html (verified ci ay 4,1 Guallart, José M. 1968. ena ji- baro-Aguaruna de la fauna del Alto COSTA-NETO and PACHECO Vol. 23, No. 1 Marafion (invertebrados). Biota 7(56): 195-209. Hagmann, Godofredo. 1928. A larva da La- ternaria i iy L. Boletim do Museu Nacional 14(3):1-6 Hays, Terence E. 1983. Ndumba folk biol- and American ett ons Sai 85:592-611. Hogue, Charles L. 1984. singe on the plant hosts and possible mimic models of “lantern bugs’ (Fulgora PF) (Homoptera: Fulgorid es Revista sian ners 32(1):45- siete merican on and En- alan Univecsity of California Press, Berkeley. Hogue, Charles L., T.W. Taylor, A.M. Young, and M.E. Platt. 1989. Egg mas- ses and first instar nymphs of some gi- ant neotropical planthoppers (Homop- tera: Fulgoridae). Revista de Biologia Tropical 37(1):221-226. Ihering, Rodolpho von. 1963. nossos animais: fauna do mund, South Pasadena. . 1968. Diciondrio dos ce do Bras- il. Ed. da UnB, Sao Pau Janzen, Daniel H. and Charles L. Hogue. 1983. Fulgora — (machaca, pea- nut-head bug, lantern fly). In Costa Ri- can Natural History, ck Daniel H. Jan- zen, pp. 726-727. University of Chicago Press, Chicago. Johnson, Leslie K. and Robin B. Foster. 1986. Associations of large Homoptera (Fulgoridae and Cicadidae) and trees in a tropical forest. Journal of the Kansas En- pipe sng Society 59:415-422 Junca, a A., Marco Freitas, Liliane Al- Da vida dos Brasil. Roter- do nord: este, p. 412. Universidade Estadual de Feira de Santana, Feira de Santana. Kellert, Stephen R. 1993. Values and per- ceptio invertebrates. Conserution Biology 7:845-853. gine Jayma L.M. and Ronald EL. Mau 001. Lamenia caliginea (Stal.): a derbid etadthiopes [Online]. Available at http://www. extento. hawaii. eae kbase/ 2 /Ty htm (veri- fied June 2001). Laurent, “web 1995. Definition and cultur- Spring/Summer 2003 al representation of the category mushi in Japanese culture. Society and Animals . [Online] i ( 97). tc — and Nelson Papavero. 1996. se s no folclore. Pléiade/FAPESP, Sao wras bisbe Straaués: Claude. 1989. O pensamento seluagem. Papirus, Campinas Marques, José Cz, W, 3991. Aspectos ecold- gicos na etnoictiologia dos sig do complexo estuarino-lagunar Mun- daui-Manguaba. a Ba Diestetion at ill Universidade de Campinas, mpin Matchett Georg and Graham C.L. Davey. : test of a disease-avoidance nee ia animal phobias. Behaviour Re- anh and Therapy 29(1):91-94. Meyerdirk, D.E. and WG. Hart. 1982. Sur- vey of Auchenorrhyncha (Insecta: Ho- moptera) associated with the Canary Is- land date palm in southern Texas. Flor- ida Entomologist 65:327-334. Neiva, Arthur and Belisario Penna. 1916. Viajem cientifica pelo norte da Bahia, sudoeste de Pernambuco, sul do Piauhi e de norte a sul de Goiaz. Memérias do Instituto Oswaldo Cruz 8(3):74-224. Nolan, Justin M. and Michael C. Robbins. nitive organization of ethnozoological domains. Journal of Linguistic Anthropol- ogy 11(2):240-249. age isi Dario. 2002. The relevance of yths and worldviews i in Palawan clas- Biocultural Diversity, eds. Richard Stepp and E Wyndham, pp. 189-206. Univer- sity of Georgia Press, Georgia. O’Brien, Lois B. 1989. New World Fulgori- dae, part I: genera with elongate head processes. Great Basin Naturalist Memoirs 12:135-170. O’Brien, Lois B. and S.W. Wilson. 1985. Planthopper systematics and —— Plehthopperss eds. L. R. Nault an aj. G. aay tea pp: pete’ John Wiley and Sons, New York. Greamamnes dos Estados Americanos. 1987. Minimum conflict: guidelines for planning the use of American humid tropic environments. [Online]. Available at JOURNAL OF ETHNOBIOLOGY 45 http:/ /www.oas.org /usde/ publications/ crap oea37e/begin.htm (verified May 24, 2001 Orico, Oswaldo. 1975. Mitos amerindios e crendices ee Civilizacaéo Brasi- ii Rio de Jane Paraiso, Maria H.B. 1985, Os Kiriri Sapuyd Pedra Branca. Centro de Estudos Baia- nos da UFBA, Salvador. Pemberton, Robert W. 1999. Insects and other arthropods used as drugs in Ko- rean traditional ‘sei Journal of Eth- nopharmacology 65:207-216. — ogi pe D. and a R. Arias. 1982. s of an Amazon Forest. Columbia iversity Press, New York. Posey, gic A. 1983. Ethnomethodology as an emic guide to cultural systems: the case ge the insects and the Kayapo Indians of Amazonia. Revista Brasileira de Zoologia 1(3):135-144. . 1986a. Etnoentomologia we tribos indigenas da Amazonia. In Sum er égica brasileira, ae Etnobiolog ia, Darcy Ribeiro, pp. 251-271. Ping Fi- nep, Petrépolis ——. 1986b. Etnobiologia: teoria e prati- ca. In Suma etnol6gica brasileira, volume 1, Etnobiologia, ed. Darcy Ribeiro, pp. 15- 25. Vozes/Finep, Petrépolis. . 1987. Tépicos e inquiri¢gGes em et- noentomologia: algumas sugestées quanto a geracao e teste de hipéteses. Boletim do Museu Paraense Emilio Géeldi, série antropologia 3(2):99-134. Poulton, Edward B. 1928. Notes on Later- naria (Fulgoridae) and other “‘lanthorn- flies.” Proceedings of the Royal Entomolog- ical Society of London 2: 86-8 1932. The alligator-like head and thorax of the tropical American Later- naria laternaria, L. (Fulgoridae, Homop- on Proceedings of the Royal Entomolog- ical Society of London 7:68-70. Robinson, Michael H. 1982. Defensa contra depredadores que cazan por medios visuales. In Evolucién en los Trépicos, eds. Georgia A. de Alba and Roberta W. Rubinoff, pp. 57-76. cgpanaaoagy Trop- ical Research Institute, Panam Ross, Edward S. 1994. cabana Fulgora. Pacific Discovery 47(3):19- Sampaio, Mario A. 1995. Palavras indigenas no linguajar brasileiro. Sagra, DC Luzzat- to, Porto Alegre. 46 COSTA-NETO and PACHECO Santos, Eurico. 1987. Histéria, lendas e folc- ~ sb nossos bichos. Itatiaia, Belo Hori- shales Glenn H., Jr. 1999. Pharmacog- nosy and the Senses in Two Amazonian Societies. Ph.D. Dissertation (Medical Anthropology), University of Califor- nia, Berkele Soulé, Michael E. 1997. Mente na biosfera, mente da biosfera. In Biodiversidade, ed. Edward O. Wilson, pp. 593-598. Nova Fronteira, Rio de Janeiro. Spix, Johann B. v. and Carl E P. v. Martius. 1938. Viagem pelo Brasil, Volume 2. Im- rensa Nacional, Rio de Janeiro. Sturtevant, William C. 1964. Studies in eth- Vol. 23, No. 1 noscience. American Anthropologist 66: 99-131. Tastevin, pam 1923. Nomes de plantas e animais em lingua tupy. Re- vista do Museu Paulista 13:687-763. van Hook, Tonya. 1997. Insect coloration and implications for conservation. Flor- ida Entomologist 80(2): +5 fle [Online]. Available at http: ww.fcla.edu/ FlaEnt/fe80p193.htm (verified Novem- ber 9, 2000). Mann. Paulo E. 1956-58. Notas sobre a oologia dos indios Canela. Revista do jt Paulista 10:155-171. Wied, Maximilian von. 1940. Viagem ao Brasil (nos anos de 1815 a 1817). Com- panhia Editora Nacional, Sao Paulo. Journal of Ethnobiology 23(1): 47-64 Spring/Summer 2003 MARKETS AND THE USE OF WILD ANIMALS FOR TRADITIONAL MEDICINE: A CASE STUDY AMONG THE TSIMANE’ AMERINDIANS OF THE BOLIVIAN RAIN FOREST LILIAN APAZA,? RICARDO GODOY,’ DAVID WILKIE, ELIZABETH BYRON,“ TOMAS HUANCA,? WILLIAM R. LEONARD; EDDY PEREZ,? VICTORIA REYES-GARCIAs and VINCENT VADEZ®* * Departamento de Biologia, Casilla 10077—Correo Central Campus Universitario, c/ 27 Cota Cota, Universidad Mayor de San Andrés, La Paz, Bolivia » Sustainable International Development Program, MS 078, Heller School for Social Policy and Management, Brandeis University, Waltham, MA 02454-9110 ‘Wildlife Conservation Society, 18 Clark Lane, Waltham, MA 02451-1823 4 Department of Anthropology, University of Florida, Gainesville, FL 32611 © Independent researcher, San Borja, Bolivia ‘Department of Anthropology, Northwestern University, Evanston, IL 60208 8 Tropical Conservation and Development Center, University of Florida, Gainesville, FL 32611 ABSTRACT.—The use of animal parts for traditional medicine is growing in Af- rica and developed nations of Asia, and persists among African and Asian im- migrants in developed nations. The practice undermines wildlife conservation. We contribute to studies of the use of wild animals for traditional medicine by: 1) focusing on an Amerindian society (Tsimane’) in the rain forest of Bolivia, 2) using a large sample (n = 508) of households selected at random, and 3) using multi- variate regressions to test hypotheses about how markets affect the use of wild animals for traditional medicine. We find that the average adult uses only a few wild animals and obtains only a few parts to treat only a few human ailments. Markets exert unclear effects on people's use of wild animals for traditional med- icine. Some proxies of markets (e.g., wages) correlate with a lower probability of using animals for medicine, but other proxies (e.g., cash income) correlate with a higher probability. Animal abundance in the village, income, and prices for mod- ern medicines correlate with a higher probability of using animals. In the Bolivian rain forest, the use of animals for traditional medicine does not threaten conser- vation as it does elsewhere in the world. Key words: medicines, Amerindians, animals, Tsimane’, Bolivia. RESUMEN.—El uso de animales para medicina tradicional esta creciendo en Afri- ca y en los paises desarrollados de Asia, y persiste entre los inmigrantes asiaticos y africanos que viven en paises desarrollados. Esta practica dafia la conservacion de la vida silvestre. Este articulo contribuye a la investigacién sobre el uso de animales silvestres para medicina tradicional porque 1) estudia una sociedad ame- rindia (los Tsimane’) en la selva tropical de Bolivia, 2) utiliza una gran muestra de hogares seleccionados al azar (n = 508), y 3) usa regresiones multivariables para comprobar hipotesis sobre el efecto del mercado en el uso de animales como 48 APAZA et al. Vol. 23, No. 1 medicina tradicional. Los resultados muestran que tipicamente un adulto usa po- cas partes de pocos animales silvestres para tratar pocas enfermedades humanas. El mercado ejerce un efecto ambiguo en el uso de animales para medicina tra- dicional. Algunos estimadores del mercado (e.g., el salario) se correlacionan con una menor probabilidad de usar animales para medicina, mientras que otros es- timadores (e.g., los ingresos monetarios) se correlacionan con una mayor proba- bilidad. La abundancia de animales en la comunidad, los precios de las medicinas modernas, asi como salarios correlacionan con una mayor probabilidad de usar animales. En la selva tropical de Bolivia, el uso de animales para medicina tra- dicional no representa una amenaza para la conservaci6n, como ocurre en otras partes del mundo. RESUME.—U utilisation de parties animales dans la fabrication de remédes tra- ditionnels augmente en Afrique et dans les pays en développement d’Asie, et persiste parmi les immigrants asiatiques et africains habitant dans les pays dé- veloppés. Cette pratique porte atteinte aux efforts de protection de la faune sau- vage. Pour cette étude nous avons suivi la démarche suivante: 1) travail centré sur une société amérindienne (les Tsimane’) dans la forét tropicale bolivienne, 2) utilisation d’un large échantillon de foyers sélectionnés au hasard (n = 508), et 3) utilisation de régressions a plusieurs variables pour vérifier les hypothéses con- cernant l’effet des marchés sur l'utilisation d’animaux sauvages dans la médecine traditionnelle. Les résultats montrent qu’un adulte moyen n’utilise que peu d’animaux sauvages, et seulement quelques parties de l’animal, pour quelques maladies seulement. Le marché exerce un effet incertain sur l'utilisation d’ani- maux dans la fabrication de remédes traditionnels. Certains estimateurs du mar- ché (ex: les salaires) correspondent a une probabilité inférieure, d’autres (ex: re- venus en espéces) a une probabilité plus élevée. Dans les communautés ot les sont abondants, les salaires bas, et le prix des médicaments modernes hes la probabilité est plus grande. Dans la forét tropicale de Bolivie, contrai- rement a d’autres régions du monde, l'utilisation de parties animales dans la fabrication de remédes traditionnels ne représente pas une menace pour la con- servation de la faune. INTRODUCTION Amerindian societies in the tropical rain forests use a wide range of native plants to treat human ailments (Reyes-Garcia 2001), but there is limited evidence to suggest that they also use wild animals to treat human ailments. Costa-Neto and his colleagues have documented the widespread use of fish, insects, and terrestrial animals to treat human ailments among fishermen and town dwellers in the semi-arid regions of northeast Brazil (Costa-Neto 1998; Costa-Neto and Marques 2000; Costa-Neto and Oliveira 2000). As part of a broader study on social integration among the Kiriri Indians of the semi-arid zone of the state of Bahia, Brazil, Bandeira (1972) found that the Kiriri used only 13 animals for medicine. Alexiades (1999) found that the Esse Eja Indians in the tropical rain forest of southern Peru used about 50 animal species for medicine. In a recent comparative study among four lowland Amerindian groups, Wil- kie and Godoy (2001) show that as incomes increased, consumption of fish de- clined, but consumption of terrestrial game rose and then fell, overshadowed by Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 49 the increasing consumption of meat from domesticated animals. At higher levels of income, Amerindians switched to domesticated animals for most of their ani- mal proteins. Here we build on this prior line of research and ask: “How does integration into a cash market economy with a consequent rise in cash income among households of tropical lowland Amerindians affect their traditional use of wild animals to treat human ailments?” The question of how integration into a market economy affects the use of wild animals for traditional medicine has implications for conservation policy. If integration into the cash market reduces the human use of wild animals for tra- ditional medicine, game conservation will improve, but if greater participation in a cash market economy increases the use of wild animals for medicines, conser- vation will worsen. Most studies of the use of wild animals for traditional medicine have been done in Asia or in Africa. Studies by Bolze et al. (1998), Servheen (1993, 1996), Callister and Bythewood (1997), Marshall (1998), Kritsky (1987), Mainka and Mills (1995), and Banks (1998) suggest that the expansion of markets induces people to make greater use of wild animals for traditional medicine and that the practice has spread in developed nations of Asia and the Pacific (e.g., Taiwan, Australia). But other research suggests that the increasing use of animals for traditional med- icine can also take place without economic prosperity. For example, Kritsky (1987) says that the use of insects for traditional medicine in China increased during the Cultural Revolution. Based on an ethnozoological survey of the use of medicinal birds, Joseph (1990) concludes that the use of birds to treat human ailments in- creased in Madhya Pradesh, Central India, because people could not afford mod- ern treatments. To help explain the conflicting findings we draw on a supply-demand frame- work from microeconomics. To answer the query of how markets affect the use of animal parts, we use information from a relatively large sample of households (n = 508) and villages (n = 59) of Tsimane’ Amerindians, a society of horticul- turalists and foragers in the tropical rain forest of the Bolivian lowlands. In work- ing with a large sample, we avoid inaccuracies from a small sample. We also avoid the bias of relying on traditional medical experts who may not know how the rest of the population uses wild animals. aioe Our goal is to estimate how various direct indicators of integration into a cash market economy—such as cash income, distance to the nearest town, and modern forms of human capital such as literacy and schooling—affect the probability that people will use wild animals for traditional medicine. We omit domesticated an- imals from the analysis because the policy debate in conservation centers on how the human use of wild animals for traditional medicine affects conservation (Brau- tigam et al. 1994; Huxtable 1992; Khoshoo 1997; Malik et al. 1996; Pui-Hay But 1995). Animals and animal parts are heterogeneous commodities. Integration into the cash market economy can increase the demand for some animals but not for others, and it can increase the demand for some animal parts but not for other parts. We estimate the correlation of a household’s cash income, wealth, and var- ious other indicators of integration to the market, and the age, gender, and size 50 APAZA et al. Vol. 23, No. 1 of the household head with the household’s use of wild animals for traditional medicine. We recognize three different effects on the use of wild animal parts for tra- ditional medicine resulting from a change in cash income. We call the animal part a “superior” good if a one-percent rise in cash income increases consumption of the animal part by more than one percent. We call the animal part ‘‘normal” if a one-percent rise in cash income increases the consumption of animal parts for traditional medicine by less than one percent. We call the animal part “inferior” if a one-percent rise in cash income reduces consumption of animal parts because people seek other treatments. An increase in income could increase demand for parts of an animal, but it could also lower demand for parts of another animal, producing an unclear net effect on conservation. An animal part might be a normal good when people have low income, but it might become an inferior good when people have higher in- come. Other factors besides income may change the use of wild animals for tradi- tional medicine. A greater abundance of wild animals in a community might reduce the costs of extracting animals for traditional medicines; if their use is limited by cost, then it will rise with increased abundance. If people view modern and traditional medicines as substitutes for each other, then a rise in the price of modern medicines should cause an increase in the use of traditional medicine. If people use modern and traditional medicines together, then a rise in the price of modern medicines should cause a decrease in the use of both medicines. The last case is a theoretical possibility; we do not know of any empirical case supporting this point. ETHNOGRAPHIC SETTING The Tsimane’ are a foraging and horticultural Amerindian society of approx- imately 7,000 people living in about 100 villages in the rain forests at the eastern foothills of the Bolivian Andes in the Ballivian and Yacuma provinces of the department of Beni (Ellis 1996; Godoy 2001; Government of Bolivia 1995; Huanca 2000; Reyes-Garcia 2001). The Tsimane’ are culturally and historically related to the Mosetene Amerindians, who form part of the Awaruna linguistic group (Reyes-Garcia 2001). Tsimane’ villages are spread over several parks and political jurisdictions, including the Pilén-Lajas Biosphere Reserve, the Parque Nacional Isiboro-Sécure, Territorio Multiétnico, and Territorio Uno. Tsimane’ villages are typically inhabited only by Tsimane’. The villages sur- veyed were in Territorio Uno (along the river Maniqui and along the dirt road linking the towns of San Borja and Yucumo), Territorio Multiétnico, and in the Pilén Lajas Biosphere Reserve. Transport in the region is along the river and by footpaths in the forests. Several logging roads cut across the Tsimane’ territory. During the rainy season travel by roads is slower; at that time, logging roads and even public roads linking different towns in the department of Beni become im- passable. The Tsimane’ territory spreads from the foothills of the Andes to the north- east, reaching the edges of the Moxos savanna (14°35’-15°30’ south latitude; Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 51 66°23'-67°10' west longitude). Habitats in Tsimane’ territory range from wet to moist subtropical and gallery forest; some of their territory abuts savannas (Mi- randa 1995). Some Tsimane’ inhabit areas of tropical rain forest at the foothills of the Andes, about 500 meters above sea level. Other Tsimane’ communities are in moist forest in the savanna region at 150-250 meters above sea level. This type of forest is similar to Amazonian wet forest but is less diverse and lacks some of its typical species such as rubber (Hevea brasiliensis (Willd.ex Adr.Juss.) Muell.Arg.) or brazilnut (Bertholletia excelsa Humb. and Bonpl.). Drainage is deficient in the area owing to low variation in relief. Part of the Tsimane’ habitat is made up of savanna, which is found in areas with alluvial soils. The floristic composition of the savanna changes with relief and flooding. In the plains, the main vegetative formation is flooded forest with relatively low tree diversity, probably because of low pH soils with limited avail- ability of nutrients (Killeen et al. 1993). Poaceae and Cyperaceae (grasses and sedges) predominate on the savannas of lower altitude and are used as natural forage for cattle. Islands of forest occur with more frequency in the savannas of higher elevation (Killeen et al. 1993). The climate of the Tsimane’ area is moist, with a 4-5 month dry season (May to September). During one 15-month period of research, the average temperature was 26°C, ranging from a minimum of 10°C during May-June, when cold winds, locally called “Sur,” arrive from the south, up to 42°C during August. The mean maximum temperature was 32°C, the mean min- imum temperature was 20°C, and the average annual precipitation was 1,924 mm. Tsimane’ show large variation in social, educational, and economic attributes. Some Tsimane’ in the upper Maniqui river are nomadic, live in small communities without schools, are monolingual in Tsimane’, and rely on shifting cultivation, hunting, fishing, plant foraging, and barter of forest products. Down the river Maniqui, close to the town of San Borja (pop. approx. 16,000), Tsimane’ are bilin- gual in Tsimane’ and Spanish, live in large settlements reachable by road, and are more likely to live in villages with schools (Reyes-Garcia 2001). Besides sub- sistence agriculture and foraging, Tsimane’ in modern villages also work for wag- es and grow rice as a cash crop. METHODS The quantitative information for the analysis comes from a survey done be- tween June and November 2000 among 508 households in 59 Tsimane’ villages in the department of Beni. During May and June 2000 we tested the survey in Tsi- mane’ communities close to the town of San Borja. The design and the adminis- tration of the survey drew on and were informed by a year of fieldwork by five researchers in two communities, one close (San Antonio) and one far (Yaranda) from the town of San Borja, during 1.5 years. For the survey, we selected villages in the main Tsimane’ regions, including the Pil6n-Lajas Reserve, Territorio Multiétnico, and Territorio Uno. In each region, we selected villages close and far from towns and villages in between. In each village, we first did a population census of all households, assigned a number to each household in the village, wrote each number on a piece of paper, and we then selected numbers at random from the folded pieces of paper to decide which 52 APAZA et al. Vol. 23, No. 1 households to interview. If the household chosen was absent, we picked another number. In each village, we surveyed between 12 and 15 households. Once we had selected a household, we flipped a coin to decide whether to interview the female or the male household head. According to the census of lowland Bolivian Indians, most indigenous households are nuclear (76%) or extended (22%) (Censo Indigena 1994-1995), so having one of the two household heads answer the ques- tions of the survey captured one of the most important decision makers of the household. If the subject knew Spanish, we did the interviews in Spanish, but if the subject was a monolingual speaker of Tsimane’ we used a translator. From the household heads, we got information on the uses of wild animals for traditional medicine and on demographic and socioeconomic attributes of that person and that person's household. From village leaders we collected information on village-level attributes, such as prices, wages, and animal abundance. Table 1 contains a complete list of all the variables we used in the statistical analysis, including definitions, number of households answering each question, and the mean, standard deviation, maximum, and minimum values for each variable. The actual data used in the analysis can be found in an Excel format on the web page of the Center for International Development at Brandeis University, which also contains the data dictionary and details of the definition and construction of var- iables." We express household size in adult-equivalents, and use different consump- tion weights from the 1.5-year study of the two Tsimane’ villages. As children, women, and men consume different amounts of food owing to differences in nutritional requirements, expressing the total size of the household in adult equiv- alents rather than in the number of persons captures with more accuracy the total effect of household needs. To capture competence in arithmetic and in literacy we relied on people’s own assessment of skills in reading and in arithmetic. In the arithmetic test we asked people whether they knew how to add, subtract, multiply, and divide, and scored each positive answer as one point. To determine literacy, we asked subjects whether they knew how to read. We opted to ask rather than to test people about their competence in arithmetic and in literacy because objective tests create tensions in a cross-sectional study; those tests are more ap- propriate when done with a person who has been followed over time and with whom researchers enjoy rapport. We measured income by asking about the cash value the entire household had earned from the sale or from the exchange of all goods and services during the two weeks before the interview. We measured the consumption of bush meat and wild fish consumed during the two days before the day of the interview by asking household heads about the consumption of wildlife. In each village, we asked village leader to assess the abundance of six com- mon wild animals to obtain a subjective measure of animal availability; from this information we constructed an index of animal abundance. The six animals in- cluded the spider monkey (Ateles chamek Humbolt), the giant anteater (Myrmeco- phaga tridactyla L.), South American tapir (Tapirus terrestris L.), white-tailed deer (Mazama americana Erxleben), collared peccary (Pecari tajacu L.), and neotropical river otter, also known as lobito de rio (Lutra longicaudis Olfers). When asked about the abundance of animals, subjects could respond “none,” “few,” “average,” Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 53 wt sh “many,” “animals have never been seen,” or “animals used to exist in the past, but no longer available.” The response for each animal was recoded as one if village leaders said there were “many” or “average,” and as zero otherwise. The recoded responses for each animal were added to arrive at an index of animal abundance that ranged from zero to six. We also asked village leaders about other village-level information, such as the price of modern medicines, the price of game, the number of traditional and modern health workers in the village, and the prevailing wages in the village. Researchers used a Global Positioning System receiver to measure the town-to- village distance in a straight line and decided for themselves whether or not the village had access to a road. one-time cross-sectional survey such as the one used here will produce a lower estimate of true use if respondents feel outsiders devalue traditional med- icine. Open-ended, initial questions about what animal parts people used to treat human ailments might also miss the many insects that lowland Amerindians populations have been known to use for medicines (e.g., Costa-Neto 1998; Costa- Neto and Nogueira de Melo 1998; Costa-Neto and Oliveira 2000; Marques and Costa-Neto 1997). To analyze the information we used univariate, bivariate, and multivariate analyses. We use univariate analysis to describe the types of animals used for medicine, the animal parts used, and the ailments treated (Table 2). We use bi- variate analysis to compare the socioeconomic and demographic attributes of households that reported using and not using animals for medicine (Table 3). We use multivariate regressions to estimate the correlation of selected variables on the probability of using an animal for medicine. As we shall see, many of the results from the bivariate analysis gain strength and others become weaker once we control for the role of third variables. A multiple regression is a statistical technique in which one predicts a de- pendent variable as a function of several explanatory variables. We chose to use one called a probit regression, a type of multiple regression appropriate when the dependent variable is categorical. In this study, the dependent variable took the value of one if the household head said the household had used an animal part to treat an ailment and it took the value of zero otherwise. Probit regression allows us to estimate the probability that the dependent variable will take the value of one when the explanatory variable increases by: [A] one unit above its mean value or [B] when the explanatory variable increases by a half standard deviation above its mean value. In both [A] and [B], all the other explanatory variables remain constant at their mean value. We decided to report the results of the probit regression in two forms, [A] and [B], for two reasons. First, it is customary in some social sciences to report coefficients from probit regressions when the Pike Me variable increases by one unit above its mean value. Second, since the units in which we measured explanatory variables are arbitrary and differ—e.g., income is measured in boli- vianos, age is measured in years, distance is measured in kilometers—we decided to standardize their effect on the dichotomous variable by estimating probabilities when the explanatory variable increased by a half standard deviation. We used a probit regression to estimate the effect of explanatory variables on 54 APAZA et al. Vol. 23, No. 1 TABLE 1.—Definition and summary statistics of variables used in regression analysis. Name Definition Obs Mean Stddvt Min Max Dependent variable Household uses Household uses Lie elo se UU. Uae (CU 1 medicinal an- animal: 0 = r; 1 = at ima least once in villi of household head Explanatory variables: community level Price-medicine Bolivianos (Bs)/box of gener- 508 4.11 LES 10 al-purpose ointment* Price-game ee price (Bs/kg) for the 508 8.73 6B. anor EY f 18.33 white-tailed deer. For scien- tific name of animals, see Table 2. Distance Distance to closest town in 508 33.86 23.24 0 101 tioning System receiver. Caregivers Number of health workers in 508 0.18 LR i 5 1 village | Road 1 = village has ee 508. 0.492... 0.500....0 1 access by road; 0 = Animal abun- Bstimated abundance of lobito 508 1.94 1.95 0 6 dance i pir, spider sig col- lared peccary, and series deer (arbitrary cale 0-6) Wage Value of daily wage, Bs/day, 508 20.90 5.864 10 50 wi or unskilled worker in village Explanatory variables: household level Wildlife Weight in kg of game and =508 =: 11.905 Ss 3.500 _—0 54.369 fish brought into house- hold/adult equivalent in last two days. Adult equiv- alent refers assigns a high- er weight to adults than to Income Bs/adult lat eamed WS. 722 7085. 0 546 b Wealth Value in Bs of 16 assets/adult 507 600.68 445.93 70.588 3429 equivalent. Assets included such things as chickens, ri- fles, fishnets Household size Adult equivalents in the 508 4.0 1.9 0.62 IZ.13 household; see under vari- able wildlife Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 50 TABLE 1.—Continued. Name Definition Obs Mean Stddvt Min Max Explanatory variables: personal level Education Maximum education attained 508 0.89 1.64 0 13 (in years Arithmetic Self-assessed ability in arith- 507 0.317 0.589 0 2 metic: 0 = none, 1 = a lit- tle 2: Reading Self-assessed reading ability: 508 0.338 0.611 0 2 0 = illiterate, 1 = a little, 2 = reads well Age i ars 504 36.5 14.5 13 125 Gender Gender of subject: 1 = fe- 508 = 0.48 es slate © 1 male; 0 = male * 6.03 bolivianos = 1 US dollar at time of field work. the probability of using an animal for medicine. The advantage of the probit regression over a linear probability model is that the probit regression keeps val- ues of the dichotomous dependent variable between zero and one without arbi- trarily setting them there, as is the case with a linear probability model. The probit model draws on maximum likelihood to estimate the effect of regression param- eters on Z scores for specific units of observations (in this case households). It then draws on the cumulative density function from the standard normal distri- bution to get an estimate of Pr(Y=1) conditional on changes in the underlying Z score. Two alternatives to probit models are logit and logistical models; since all three types of regression models produce similar results when estimating changes in the probability from changes around the mean value of explanatory variables, we decided to use the probit model. Aldrich and Nelson (1984) and Long (1997) provide details of how probit and logit regressions work. Before using the probit regression we tested and rejected the assumption of constant variance in the error term at the 10% confidence level. We therefore use Huber-White robust standard errors. Robust standard errors are necessary when running a regression with heteroskedastic or non-constant error terms (Gujarati 1995). We did not have instrumental variables to control for possible biases from endogeneity or reverse causality. An instrumental variable is a variable that is highly correlated with the endogenous explanatory variable, but is uncorrelated with both the error term and with the dependent variable. Because we cannot establish the direction of causality in unambiguous ways, we speak of correlation when discussing the regression results. In the context of probit regressions, dis- cussed later in this paper, a correlation refers to the probability of using a wild animal for medicine when the explanatory variable increases by one unit or by a half standard deviation above its mean value while holding constant all other explanatory variables at their mean value. Since we did not collect repeated measures of people over time, we cannot control for unobserved fixed heterogeneity in endowments and preferences of people or biological attributes of villages (other than animal abundance) that could influence both the use of animals for medicine and the covariates used in 56 APAZA et al. Vol. 23, No. 1 TABLE 2.—Animal part last used and ailments treated among Tsimane’ Amerindians, Bolivia, 2000. Animal Parts Human ailments treated 1. Brazilian tapir (Tapirus fat (1) cough (1) terrestris) (3) nail (1) “bad wind” (1) liver (1) general body pain (1) 2. brown agouti (Dasy- bile/gall bladder (1) _— childbirth (1) Senet baile (1) . brow mon-_ bile/gall bladder (1) _ eye infection (1) ey (Cebus ae (1) Tayas- fat (1) cold (1) ow i Oo [2 [or] 2 as ie) 2.8. ao) oO io) ie) © common opossum (Di- fat (6) cut ithe fever (2), cold (2), leg pain t anteater (Mirme- nail (1) lt body pain (1) cophaga tridactyla) (4) hair (2) snake bite @), urinary problem (2) liver (1) - pain 7. jaguar (Panthera onca) _ fat (1) ough (1) 8. kinkajou (Potos flavus) _ penis bone (1) earache (1) 1 9. paca (53) (Agouti paca) — fat (1) general body pain (1) bile/gall bladder (42) | stomach pain (1), leishmaniasis (5), snake bite (26), rheumatism (1), heart pain (1), pain in bones (2), liver pain (1), mayo (7), fever and Spoinpeia teeth (8) child birth (7), snake i. (1) 10. South American coati _ fat (6) cold i cough (2), leg pain (1) jes nasua) (17) thorns (1) cou hair (2) wounded mt (1) bile/gall bladder (1) ane bite penis bone (5) iste oe ed pain (1) 11. spider monkey (Ateles _ fat (2) ver (1), cough (1) chamek) (8) bile/gall bladder (5) vo (1), cold (1), shoulder pain (2), sleeping problems (1 bone (1) leishmaniasis (1) 12. stingray (Potamotrygon — fat (10) stomach pain (2), cold eh , cough sp.) (13) (2), ayes be te bones (1), leg in (1). liver (3) pain in bones 3), genet body (1) 13. tortoise (Geochelone sp.) fat (3) oe a infection (1), rheumatism (2), peer es At (1) 14. white-tailed deer (Ma- carcass (1) Id (1) zama americana) (2) In parentheses are the number of households io part or ailment. Sometimes number of Rel are under animals parts used is less than the number of households for the animal because people could not recall the part of the animal they nee used. The same applies to the number of ailments mentioned and the animal part; sometimes people mentioned an animal part, but could not recall for bebo ailment they had used it. Sometimes the number of ailments treated with a part exceeds the nv of households under “parts”; this happens when households use one part to treat several Peron Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY oF TABLE 3.—Comparison of mean values of explanatory variables between households that used and did not use animals for medicines, Tsimane’ Amerindians, Bolivia, 2000. Medicinal use of animals by households that Did not use animals —_ Used animals Test of equality of mean House House- Variable holds Mean holds Mean Statistic p value Price medicine 363 4.01 145 4.37 t = —2.36 0.02 Price game 363 8.7 145 8.82 t = —0.42 0.68 Distance 363 RF be) 145 36.64 t = —1.08 0.09 Caregivers 363 0.16 145 0.23 t = 1.80 0.07 oa 363 0.45 145 0.6 x? = 10.69 0 Animal abundance 363 1.77 145 2.38 t = -3.21 0 age 363 Pa RS 145 19.82 t = 2.64 0 Wildlife 363 1.6 145 2.67 t = 3.13 0 Income 363 41.46 145 61.65 = —292 Wealth 363 589 - 144 628 t = 0.88 0.38 Household size 363 4 145 4.09 t = 0.53 0.6 Education 363 0.94 145 0.76 t= 1.14 0.25 Arithmetic 362 0.32 145 0.3 t = 0.34 0.73 Reading 363 0.34 145 0. t = 0.02 0.99 361 35.78 143 38.43 t = 1.85 0.06 Female 363 0.49 145 0. xX? = 0.10 0.32 the analysis. For example, some villages may contain greater biological diversity and may, therefore, have greater abundance of wild animals for traditional med- icine, so people in those villages may be more likely to use wild animals for traditional medicine. If we had observations over time, we could control for fixed, unseen, attributes; since we collected information only at one time, unseen fixed attributes of communities, households, and people will bias our estimates in un- known directions. We focus our discussion of the regression analysis on results that are statis- tically significant at the 90% confidence level or higher. We discuss explanatory variables at the community, household, and personal levels. Table 1 lists the explanatory variables by level. Table 4 contains the regression results; we estimate the probability that a household will use a wild animal for medicine when an explanatory variable increases while all other explanatory var- iables remain constant at their mean value. In column A the explanatory variables increase by one unit and the probabilities reported are the resulting marginal probabilities. For example, we know from Table 1 that the average household size in the sample was 4.025 adult equivalents; if the household size increases by one adult, then the probability that a household will use an animal for traditional medicine increases by 3.6%; results are significant at the 95% confidence level. Note that 3.6% is the marginal increase in the probability. In column B the explan- atory variable increases by a half standard deviation and we report the overall probability of using an animal for traditional medicine (30.1%). Since the mean predicted probability when all explanatory variables are held constant at their mean value is 26.7% (see bottom of Table 4), a half increase in the standard 58 APAZA et al. Vol. 23, No. 1 TABLE 4.—Results of multivariate regressions. Probabilities of using wild animals as medicine among Tsimane’ Amerindians, Bolivia, 2000. Probability of using animal’ when explanatory variable increases by following above its mean value while holding constant all other variables: One unit 1/2 standard Explanatory variable* [AP deviation [B] Price medicine Ts" 28.5 Price game* 0.7 26.7 Distance* here 22.5 Caregivers 28.0 iw 29.9 Animal abundance OY ded 31.4 ge* ~~ 2772 23.9 Wildlife 2.6 28.1 Income* 8.0** 29.8 Wealth* 4.7 28.1 Household size ied 30.1 Education be Ot eae a4 A Arithmetic 6.8 28.6 Reading 10.6 29.9 g 0.1 pt Go Female 2.4 26.0 Pseudo R? 0.1 Number of households 449 Predicted probability at mean values of a 26.7 Probability from raw #s 145 users/508 total sample = 28.5 phi probability when all Xs in- ease by 1/2 standard deviation 41.0 aaa teenie is wate! dichotomous categorical variable (1 = household used wild animal for medicine; ? Column [A]: rot ein with robust standard errors. For units of each explanatory variable see second column of Table 1. Asterisks: *, **, and *** significant at the 90%, 95% and 99% confi- dence level. 3 Explanatory variables with an asterisk (*) are in logarithms; 100 bolivianos added to income to avoid producing missing values when taking logarithms. deviation of household size produces a change in the marginal probability in the use of medicinal plants of +3.37% (3.37% = 30.12%-26.75%), roughly comparable to the 3.6% of column A. The slight difference arises because the unit increase in column A (one adult) differs from the half standard deviation increase in column B (0.9 adult). The last five rows of Table 4 contain information about the overall model. The pseudo R’ is measure of goodness of fit of the model. The number of observations is 449, lower than the number of observations in Table 1 because of missing values that arose from taking the logarithm of the variable on wildlife consumption/ adult equivalent. We re-estimated the probit models adding a constant (0.01) to households without wildlife consumption and iD that the results did not change from those reported below in a significant way. The bottom three rows deserve a brief comment. The row called ‘ ase probabilities at mean values Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 59 of all Xs” represents the overall probability (26.7%) of using an animal when all Xs are held constant at their mean value; the probability is estimated using the 449 observations used in the regression. The penultimate row represents the share of the households that used an animal for medicine and is derived by dividing the number of households that used animals (145) by the total sample (508). The last row contains the overall probability of using an animal when all explanatory variables increase simultaneously by a half standard deviation. RESULTS Few Tsimane’ said they had used wild animals for traditional medicine. In response to the question ‘“‘when was the last time you have used an animal to treat an ailment?” about three quarters of the household heads (71.4%) said they had never used an animal part to treat human ailments. In a companion study we show that the Tsimane’ make widespread use of plant and plant products for many purposes, including the treatment of human illnesses (Reyes-Garcia 2001). The same does not appear to be true with the use of animal parts. Of the Tsimane’ who said they had used a wild animal for traditional med- icine, most said they had relied on only a few animals. The survey yielded a total of 14 animals used to treat human ailments (Table 2). All animals were mammals except for a stingray (Potamotrygon sp.) and a tortoise (Geochelone sp.). The mam- mals used belonged to five orders: Carnivora, Primates, Artiodactyla, Perissodac- tyla, and Rodentia. According to one source from Bolivia (Tarifa 1996), among the animals used for medicines by the Tsimane’ five species are threatened or vulnerable to extinction: Brazilian tapir, spider monkey, giant anteater, collared peccary, and jaguar (Pantera onca L.). Of these, only the spider monkey appears in the IUCN list of vulnerable species. Although Tsimane’ mentioned a total of 14 animals, four of them—paca (Agouti paca L.), South American coati (Nasua nasua L.), stingray, and spider monkey—accounted for 79.2% of all animals used. None of the Tsimane’ surveyed mentioned more than one animal. Again, this contrasts with the use of medicinal plants. Reyes-Garcia (2001) found that Tsimane’ used wild plants mostly for traditional medicine. She found that most medicinal plants had at least two unrelated medicinal uses. Out of the 169 medicinal plants registered during the longitudinal study in the two Tsimane’ villages, Reyes-Garcia found that only 27 medicinal plants had only one use, 58 had 2-3 unrelated uses, and 18 medicinal plants had at least four uses. The total number of wild animals (14) used for traditional medicine by the Tsimane’ is similar to the total number of wild animals (13) used by the Kiriri Indians of Brazil (Bandeira 1972), but considerably lower than the total number of animals (50) used by the Esse Eja Indians of southern Peru (Alexiades 1999). Tsimane’ used few animal parts to treat human ailment. Fat, bile, and gall bladders ac- counted for most of the animal parts used. Other parts included such things as teeth, livers, and bones. Tsimane’ obtained fat for traditional medicine mostly from the South American coati, common opossum (Dipelphis sp.), tortoise, and stingray, and they obtained bile mostly from the paca and from the spider mon- key. 4 The last column of Table 2 suggests that the Tsimane’ use animal parts to 60 APAZA et al. Vol. 23, No. 1 treat many different human ailments. For example, the Tsimane’ use the bile and gall blader from the paca to treat nine ailments. Although the Tsimane’ mentioned many different types of ailments, they were most likely to have used an animal part to treat snake bites and, to a lesser extent, to treat coughs, colds, and urinary tract problems. In Table 3 we compare the socioeconomic and demographic attributes of households that used animals for medicines and households that did not. A com- parison of mean values between the two samples suggests that households that used animals for medicine lived in communities with higher price for modern medicines and were farther away from market towns. Those villages had more health care workers, animals in the commons, and lower wages, and were more likely to have access to a road. Household heads that used medicinal animals were slightly older and lived in households with more income and wealth than households that did not use animals for medicine. The results of the multivariate regressions shown in Table 4 suggest that at - the community level, animal abundance and greater access to roads correlated positively with a greater probability of using an animal for traditional medicine, but wages and distance to the nearest market town correlated negatively. A one percent increase over the sample mean of 33.85 km in the distance from the nearest market town correlated with a 7.6% (p<0.05) lower probability of using an animal for traditional medicine. Access to a road correlated with a 13.1% (p<0.001) higher probability of using an animal for traditional medicine. Last, a one percent increase in the village wage correlated with a 20.7% (p<0.05) lower probability of using a wild animal for traditional medicine. We found support for the idea that animal abundance correlated positively with the probability of using a wild animal. A one-unit increase in the index of animal abundance around the village was associated with a 4.7% (p<0.001) high- er probability of using wild animals for traditional medicine. If we focus on the three following indicators of integration to the market— distance to the nearest town, access to roads, and wages—we find ambiguous results about how markets affect the use of wild animals for traditional medicine. If we equate integration to the market with proximity to market towns or with access to roads, we find that villages closer to towns and villages with access to roads were more likely to use wild animals for traditional medicine. The results suggest that markets correlate positively with increases in the demand for wild animals for traditional medicine. But if we equate integration to the market with wages, we find that Tsimane’ living in villages with higher wages were less likely to use wild animals for traditional medicine. Last, the sign of the coefficient for the price of modern medicine was positive, suggesting that modern medicines and wild animals for traditional medicines are substitutes. A doubling in the price of modern medicines increased the probability of relying on wild animals for traditional medicine by about 12.5% (p<0.10). Two of the four household-level variables were significantly associated with the probability of using wild animals for traditional medicine. Household size correlated positively with a greater probability of using wild animals for tradi- tional medicine. Each additional adult in the household above the 4.0 mean for the sample correlated with a 3.6% (p<0.05) greater probability of using a wild Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 61 animal for traditional medicine. A one boliviano increase in income above the sample mean of 47.22 bolivianos correlated with a 8.0% (p<0.05) greater proba- bility of using a wild animal for traditional medicine (6.03 bolivianos = 1 US dollar at the time of fieldwork). The result suggests that among the Tsimane’ parts of wild animals used for traditional medicine may be normal goods—demand in- creases as incomes rise. Last, the coefficients of personal-level variables suggest that only education correlated with a change in the probability of using wild animals for traditional medicine. An additional year of education above the sample mean of 0.886 years of education correlated with a 7.7% (p<0.001) lower probability of using wild animals for traditional medicine. DISCUSSION The central question motivating this article was: How does integration into a cash market economy with a consequent rise in cash income among households of tropical lowland Amerindians affect their traditional use of wild animals to treat human ailments? We made the point at the outset that relatively little was known about human uses of wild animals for medicines among indigenous peo- ple of the New World. The picture of the Tsimane’ that emerges from the survey is that of a tropical lowland Amerindian society in which the average household head uses only a few wild animals, to obtain only a few animal parts, to treat only a few human ailments. The finding is consistent with the low use of wild animals to treat human ailments reported by Bandeira (1972) among the Kiriri Indians of Brazil. The low use contrasts with the high use in northeast Brazil reported by Costa- Neto and his colleagues (e.g., Costa-Neto 1998; Costa-Neto and Marques 2000) and in southern Peru reported by Alexiades (1999). It also contrasts with the widespread use of wild plants to treat human ailments among the Tsimane’ and other lowland Amerindian groups and with the well-documented importance of wild animals as a source of animal proteins among lowland Amerindians. In answer to our central query of how integration into a cash market economy affects the use of wild animals to treat human ailment, a complex picture emerges. For example, we found that two proxies of a market economy—wages and school- ing—correlated with a lower probability of using wild animals for traditional medicine. On the other hand, two indicators of participation in the market econ- omy—proximity to market towns and access to roads—correlated with a higher probability of using wild animals for traditional medicine. The findings must be read with caution owing to shortcomings in the infor- mation collected. Because we relied on a one-time cross-sectional survey, we could not control well for unseen, fixed attributes of households and people. We relied on the perception of village leaders about animal abundance rather than counting wild animals in the field. Last, we did not collect information on the actual quan- tity of wild animals used for traditional medicine, so we could not estimate the income elasticity of consumption—the percentage change in the consumption of wild animals for traditional medicine from a percentage change in income. The low use of wild animals as medicines suggest that policies to enhance 62 APAZA et al. Vol. 23, No. 1 wildlife conservation among the Tsimane’ should probably be targeted at reducing the use of wild animals as a source of food (whether by indigenous people or by outsiders) rather than at reducing the use of wild animals as a source of traditional medicine. Unlike some of the case studies of Asia and Africa reviewed earlier, among the Tsimane’ the use of animals for medicine is not a significant threat to conservation. If policy-makers decide they wish to improve the conservation of fauna by reducing the use of wild animals for medicine, then the results of this analysis hint at two possible policy options: wages and the price of modern medicines. We have presented evidence to suggest that higher wages from greater employ- ment opportunities correlates with a lower probability that rural populations will use wild animals for traditional medicine because the opportunity cost of foraging rises. Policies geared at creating greater employment opportunities might have beneficial, direct effects on the conservation of wild animals for medicine. Simi- larly, we found that a doubling in the price of modern medicines correlated with an increase in the probability of using wild animals, suggesting that public health policies that deliver medicines and medical services to rural areas at lower prices will likely enhance the conservation of fauna in a direct way. NOTE ' www.heller.brandeis.edu/sid/research/bolivia.asp ACKNOWLEDGMENTS The programs of Cultural Anthropology and Human Dimensions of Global Change of the National Science Foundation (USA), the Conservation, Food and Health Foundation, and the MacArthur Foundation financed the research reported in this article. Eraldo Me- deiros Costa-Neto, George Estabrook, and anonymous reviewers provided valuable com- ments on earlier drafts. We would like to thank Mary Brooks for editorial assistance. REFERENCES CITED Aldrich, J-H. and ED. Nelson. 1984. Linear and D. Krakower. 1998. The availability Probability, Logit, and Probit Models. Sage of tiger-based traditional Chinese med- Publications, Newbury Park, California. icine products and public awareness Alexiades, M. 1999. Ethnobotany of the Ese about the threats to the tiger in New Eja: Plants, Health and Change in an York City’s Chinese communities: A pi- Amazonian Society. Ph.D. Dissertation lot study. Wildlife Conservation Society enti opolee City University of Working Paper No. 12, New York. w York, New York. Brautigam, A., J. Howes, T. Humphreys, aise MLL. 1972. Os Kiriri de Mirandela: and J. Hutton. 1994. Recent information um grupo indigena integrado. Universi- on cathe status and utilization of African dade Federal de Bahia, — pangolins — Bulletin 15(1):15-22. Banks, RB 1998. Tigers and traditional med- Callister, DJ. and T. Bythewood. 1997. Of s—a study. In Proceedings of the EU Tiger Treatments & Rhino Remedies: Trade Wildlife Law Enforcement Workshop, pp. in Endangered Species Medicines in Aus- 93-95. Department of the Environment, oe it New ei World Wildlife Transport and the Regions, London. Fund, Canberra, Australia Bolze, D., C. Chetkiewicz, Q. Mingjiang, | Censo inetd 1994-1995, Censo Indigena Spring/Summer 2003 del Oriente, Chaco y Amazonia, 1994— 1995. Secretaria de Asuntos Etnicos, de Genero, y Generacionales, Ministerio de Desarrollo Humano, Gobierno de Boliv- ia, La Paz. Costa-Neto, E.M. 1998. Folk taxonomy and cultural significance of ‘abei’ (Insecta, Hymenoptera) to the Pankararé, north- eastern Bahia State, Brazil. Journal of Ethnobiology 18:1-13. Costa-Neto, E.M. and J.G.W. Marques. 2000. Faunistic resources used as med- icines by artisanal fishermen from Siri- binha beach, State of Bahia, Brazil. Jour- nal of Ethnobiology 20:93~109. Costa-Neto, E.M. and M. Nogueira de . 1998. Entomotherapy in the County of Matinha dos Pretos, State of Bahia, northeastern Brazil. The Food In- sects Newsletter 11(2):1-3. Costa-Neto, E.M. and M.V.M. Oliveira. 2000. Cockroach is good for asthma: zootherapeutic practices in northeast yeas Research in Human Ecology 7(2): Deaton, te de 1997. The Analysis of House- hold Data. The Johns Hopkins University Press, Baltimore, Maryland. Ellis, R. 1996. A Taste for Movement: An Dissertation pote net St. An- drews University, Scotland Godoy, Ricardo. 2001. Indians, “Markets, and Rain Forests: Theory, Methods, Analysis. Columbia University Press, New York. Government of Bolivia. 1995. Censo Indige- na del Oriente, Chaco y Amazonia, Bolivia: Secretaria de Asuntos Etnicos, de Ge- nero y Generacionales, Ministry of Hu- man Development, La Paz. Gujarati, Damodar N. 1995. Basic Economet- rics. McGraw-Hill, Inc., New York. Huanca, Tomas. 2000. Tsimane’ Indigenous Knowledge, Swidden Fallow Manage- ment, and Conservation. Ph.D. Disser- tation (Anthropology), University of Florida, Gainesville. Huxtable, R.J. 1992. The pharmacology of extinction. Journal of Ethnopharmacology 37:1-11. Joseph, A.N.T. 1990. The relevance of tra- ditional bird drugs in relation to mod- ern primary health care in Madhya Pra- JOURNAL OF ETHNOBIOLOGY 63 desh, India. Records if the Zoological Sur- vey of India 86:193-21 Khoshoo, T.N. ie Coniation of India’s endangered m : ae ive 8 ; Gui de arboles de Bolivia. La Paz, Bolivia, Herbario Nacional de Bolivia y Missou- ri Botanical Gardens. Kritsky, G. 1987. Insects in traditional Chi- nese medicines. Proceedings i - Indi- ana Academy of Science 96:289- Long, J.S. 1997. Regression Models om Cate- gorical and Limited Dependent Variables. Sage Publications, Thousand Oaks, Cal- ifornia. Mainka, S.A. and J.A. Mills. 1995. Wildlife and traditional Chinese medicine—sup- ply and demand for wildlife species. ietlipe 2 Zoo and Wildlife Medicine 26(2): Malik, ca ce Wilson, R.J. Smith, D.M. La- vigne, and B.N. White. 1996. Pinniped penises in trade: a molecular-genetic in- vestigation. Conservation Biology 11(6): 1365-1374 Marques, J.G.W. and E.M. Costa-Neto. 1997. Insects as folk medicines in the state of Alagoas, Brazil. The Food Insects Newsletter 10(1):7-10 Marshall, N., ed. 1998. Searching for a Cure: TRAFFIC International, Cambridge, UK Miranda, Carmen. 1995. The ae aoe Reserve, Bolivia. UNESCO, P. Pui-Hay But, P. 1995. Tigers, eis, and ese medicine. In American Zoo and Aquarium Association, Annual Conference Proceedings, 1995. Woodland Park Zoo- logical Gardens, pp. 68-73, Seattle, Washington. Reyes-Garcia, Victoria. 2001. Indigenous Bolivia. Ph.D. Dissertation (Anthropol- By) University of Florida, Gaines- ville. Servheen, C. 1993. The trade in bears and bear parts. In Bears: Status Survey and onservation Action Plan, eds. C. Serv- heen, S. Herrero, and B. Peyton, pp. 33- 38. IUCN/SSC, Gland, Switzerland. 64 APAZA et al. Vol. 23, No. 1 . 1996. Aspects of the Asian trade in bears and bear parts. Journal of Wildlife Research 1(3):301-303. Tarifa, T. 1996. Mamiferos. In Libro rojo de los vertebrados de Bolivia, eds. P. Ergueta S. and C. de Morales, pp. 23-48. Centro de coon Para la Conservaci6én, La Paz, Boliv Wilkie, David and Ricardo Godoy. 2001. In- come and price elasticities of bushmeat demand in lowland Amerindian socie- ties. Conservation Biology 15(3):1-9. Journal of Ethnobiology 23(1): 65-100 Spring/Summer 2003 THE GRAMMAR OF SNCHITSU’UMSHTSN (COEUR D'ALENE) PLANT NAMES GARY B. PALMER,? M. DALE KINKADE? and NANCY J. TURNER* * Anthropology & Ethnic Studies, University of Nevada at Las Vegas, Las Vegas, NV 89154-5003 » Department of Linguistics, University of British Columbia, Vancouver, B.C. V6G 2L2 © School of Environmental Studies, University of Victoria, Victoria, B.C. V8W 2Y2 ABSTRACT.—This paper analyzes 106 known plant names of Snchitsu’umshtsn (Coeur d’Alene), a Salishan language of northwestern North America whose eth- nobotany has not been previously described. Grammatical analysis of plant names reveals semantic motivations, the structure of classification, and the position of Snchitsu’umshtsn among Salishan languages. A five-level botanical taxonomy cor- relates only partially with the levels defined by Berlin (1992) and Brown (1984). The morphological structure of plant names shows that classification is only part of the motivation for their construction. Many describe appearances and other sensory qualities that facilitate identification. Utilitarian concerns play a role, but not the dominant one. Snchitsu’umshtsn names are compared to those of other Interior and Coast Salish languages. A cline of decreasing cognate frequencies appears as one moves from Snchitsu’umshtsn in the east to the Coast Salish lan- guages in the west. The 16 terms with cognates in at least six of the seven lan- guages include names for eight trees (including six conifers), three berry bushes, one edible bulb and two edible taproots. Reasons for this distribution are dis- cussed. We include a listing of plant terms with Salishan cognates, tables describ- ing the morphological analysis of terms, and a table of cognate incidence in Sa- lishan languages. ‘ . : , Key words: ethnobotany, categories, lexicon, plant names, Salish, Coeur d’Alene. RESUMEN.—Este trabajo analiza 106 nombres de plantas en Snchitsu’umshtsn (Coeur d’Alene), una lengua Salish del noroeste de Norteamerica cuya etnoboté- nica no ha sido descrita. El andlisis gramatical de los nombres de plantas revela los temas seménticos, la estructura de la clasificaci6n, y la posici6n de Snchitsu- ‘umshtsn entre las lenguas Salish. La taxonomia botanica, en cinco niveles, se correlaciona sélo parcialmente con los niveles definidos por Berlin (1992) y Brown (1984). La estructura morfolégica de los nombres de las plantas muestra que la clasificacion es sélo parte del motivo para su construccién. Muchos nombres des- criben la apariencia y otras cualidades sensoriales que pueden facilitar la identi- ficacién. Los conceptos referentes a la utilidad desempefian un papel, pero no es el dominante. Los nombres Snchitsu’umshtsn se comparan a los de otros idiomas de los grupos Salish del Interior y Salish de la Costa. El numero de nombres semejantes decrece a medida que crece la distancia desde el Snchitsu’umshtsn, en el este, hacia las lenguas de la Costa en el oeste. Los 16 términos que tienen palabras semejantes en al menos seis de las siete de las lenguas Salish comprenden los nombres de ocho Arboles (entre ellos seis confferas), tres arbustos con bayas, un bulbo comestible y dos raices comestibles. El trabajo discute las razones de 66 PALMER et al. Vol. 23, No. 1 esta distribuci6n. Incluimos una lista de términos sobre plantas relacionados con términos semejantes en otras lenguas Salish, las tablas que describen el andlisis morfolégico de los términos, y una tabla de la incidencia de semejanza entre términos en los dialectos Salish. RESUME.—Cet article analyse 106 noms de plantes Snchitsu’umshtsn (Coeur d’Alene), une langue Salishan du nord-ouest de l‘Amérique du Nord dont I’eth- nobotanie na pas encore été étudiée. L’analyse grammaticale des noms de plantes révéle les nuances sémantiques, la structure de la classification, et la position du Snchitsu’umshtsn au sein des langues Salish. Une taxonomie botanique 4 cing niveaux ne correspond que partiellement aux niveaux définis par Berlin (1992) et Brown (1984). La structure morphologique des noms de plantes indique que cette classification n’explique qu’en partie leur construction. Beaucoup de noms décri- vent |’aspect et autres qualités sensorielles qui facilitent l’identification. Les con- sidérations utilitaires jouent un réle, mais elles ne sont pas déterminantes. Cet article compare les noms Snchitsu’umshtsn a ceux d’autres langues Salishan de l’intérieur et de la c6te. Un cline oti la fréquence des cognates diminue apparait quand on passe des Snchitsu’umshtsn 4 l’est aux langues Salishan de la céte ouest. Les 16 termes avec cognates qu'on trouve dans au moins six des sept langues comprennent les noms des plantes suivantes: huit arbres (y compris six coniféres), trois arbustes a petits fruits, un bulbe comestible, et deux racines pivotantes co- mestibles. Cet article examine les raisons de cette répartition. Sont également in- clus dans cet article: une liste des noms de plantes avec les cognates Salishan, des tableaux de l’analyse morphologique des mots, et un tableau de l’incidence des cognates dans les langues Salishan. INTRODUCTION Previous researchers studying the ethnobotany of the Salishan languages of northwestern North America have used plant names to understand botanical clas- sification, grammatical conventions of naming, and relationships among cognate languages. No previous studies have focused on the plant names or the ethno- botany of Snchitsu’umshtsn' (Coeur d’Alene), a language of the Interior Salish division of Salishan (Figure 1). This paper analyzes the grammar of plant names in Snchitsu’umshtsn. The grammatical analysis reveals new information on bo- tanical classification and the relationship of this language to other Salishan lan- guages. The analysis includes 106 names for plants at the genus and species level (Appendix 1). A few of these terms have not yet been correlated to taxa identified in English. A few terms for higher-level categories are also included, revealing a botanical taxonomy with five levels which correlate only partially with the ranks defined by Berlin (1992) and Brown (1984). Though we describe the Snchitsu’umshtsn taxonomy, the emphasis in our analysis is not on discovering taxonomic principles, but rather on describing and analyzing the linguistic structure of plant names. We find that when a plant name has internal morphological structure, this often reflects perceptions of the plant that are specific to the language and culture. Our findings suggest that classifi- cation is only part of the motivation for the construction of plant names and that another important motivation is the description of appearances and other sensory qualities that are salient or that enable plants to be readily identified. Some plants Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 67 Secwepemc Stl’atl’imx Nicola (Athabaskan) Nlaka‘pamux Kalispel Flathead Snchitsu’umshisn FIGURE 1.—Interior Salish speech community territories ca. 1850, principally after Elmen- dorf (1965). are named in more than one way according to which structural part (leaves, trunk, etc.) is most salient at the time of speaking. Utilitarian concerns play a role in plant naming, but not the dominant one. We find that many terms have lost linguistic structure and original meanings have become partially or totally ob- scured. This is the case with nearly half (47) of the terms. We also record six names borrowed from English and French. We discuss our findings in relation to comparable data from other Interior and Coast Salishan languages and we de- scribe the cline of decreasing cognate frequences in plant names as one moves from Snchitsu’umshtsn in the east to the Coast Salishan languages in the west. Theoretical Approaches.—There are several perspectives from which one can analyze a language's botanical terminology, or more specifically, its names for plants. One approach is to look for universal hierarchies of categorization, called taxonomies. For example, Brown (1984:1) wrote, “For speakers of American English, white oaks, pin oaks, and post oaks are kinds of oak; oaks, walnuts, and maples are kinds of trees; and trees, vines, and bushes are kinds of plants. Such a system of inclusive relationships forms a folk biological taxonomy”’ [italics in original]. Berlin (1992) proposed a taxonomic framework of ranks, starting with the most inclusive category “kingdom” and descending through “‘life-form,” “intermediate,” “ge- neric,” “specific,” and “varietal,” but he noted that strictly taxonomic presenta- tions of ethnobiological material have been questioned, first by Bright and Bright (1969) and later by others (Ellen 1986; Hunn 1976; Randall 1976). The shift of emphasis away from taxonomy received further support from Turner (1987, 1989). A second approach is to investigate what the naming of plants reveals about ethnically distinctive classifications of the botanical world. For example, Bright and Bright (1969) found that many plant names of two northwestern California 68 PALMER et al. Vol. 23, No. 1 tribes—Yurok and Smith River—do not necessarily fall into any hierarchy. In- stead, consultants often identified a plant as being “like such-and-such.’”’ They also reported that “where generic terms exist, they may also refer to a specific member of the class. ... Thus Yurok tepo’ refers to ‘fir tree’ or ‘tree’ in general” (1969:70). They concluded that “the aboriginal taxonomies of northwestern Cali- fornia can be represented more faithfully by a kind of ‘sphere of influence’ model,” a conclusion echoed by Hunn (1985). Thus, one of the problems consid- ered in this paper is the extent to which the structure of plant names reveals taxonomic categorization as opposed to other types, such as the identification of family resemblances, or functional (metonymic) relationships among plants. A second problem is to determine whether the structure of plant names is in fact intended to categorize by relating one kind to another, or simply to describe salient perceptual characteristics of plants. Alternatively, plant names may be studied from a historical perspective in which cognates in neighboring languages are taken as evidence for common or- igins, borrowing, or language change. For example, Fowler (1972:109) found that plant names provided “ecological clues to early homeland situations.” Examining plant name cognates among northern Uto-Aztecan languages, including those of the Numic, Tiibatulabalic, Takic, and Hopic groups, she was able to conclude that their ancestors must have lived in a territory that was diverse in elevation and probably in or near desert zones. Based on the distribution of pinyon, prickly pear, ephedra, chia, lycium and cholla, as well as various animals, she could place the homeland area in the Sierra Nevada mountain range south of 36°30’ north latitude. Hinton (1994:87-90) followed a similar approach in her investigation into the origins of the Wintun people in Northern California. The areal ethnohistory approach using cognate distributions requires the examination of terms in all the members of a group of related languages together with data on the distributions of the named plants and animals. In this study we simply compare frequencies of cognate forms in other Salishan languages to determine closeness of relation- ship to Snchitsu’umshtsn. Yet another approach to the study of plant nomenclatures examines their ap- pearance in other domains of culture. For example, plant names may be used in the names of mythical persons, as in the Snchitsu’umshtsn story about Ylmikhwm Asp’ukhwenichelt “Chief Child of the Root,’ who taught each of the animals how to live. The name of the mythical actor is composed of ylmixem ‘chief’ and a-sV p’éxvené-ilt ‘child of desert parsley (Lomatium macrocarpum)’ ( Xx . x : x 08 Te ? WAN 1Vd I ‘ON ‘€Z “TOA TABLE 4.—Continued. # Snchitsu’umtsn name English or Latin name PS PI LI TH SH ev CM KA 35 p ex*pux® any camas : x ?us us (NP) : x : 37 sp ‘éxenc ae parsley x x x 38 péqai* barestem ar (NP) 39 pt biscuitroo : x 40 st ue wild caraway x x x 41 dm ni? yarro . . : 42 igacut “ea?al’gs ha sg’arpm a Rl etc. . ; , 43 ‘u? orthern wormwood x Xx : x 44 q’al ‘q’al ‘mnitp big sagebrush : . a . x 45 smuk*a?cn* balsamroot x x Xx x x 46 tek “tuk “t wild thistles x x Xx . : 47 riupa* wavy-leaved thistle ‘ . . : 48 nel ‘cil ’x°qi pineapple w x 49 xal sg’arpm dandelion : 50 q’éyu? Oregon-grape x x 51 spittena:* paper birch . . . x 52 q ip ’x"e? hazelnu x x x x x x x x 53 sampaqn* black twinberry : : : : : . . 54 c ‘6k’ ak blue elderberry x x x x x e 55 tamtamni?etp snowberry . . x x x 56 stichtskhw red willow x x x 57 tfafaxc’e? cantaloupe . . . 58 ni?Sarusi?utm quash 59 smax*nélp* silverberry . : : é : sx®tism soapberry * x x x x x x 61 Pile kinnikinnick . x x x x : : : 62 st ‘eq ‘tn dwarf blueberry x : : . 63 st ‘asastq black huckleberry x x 64 paqpagaxn huckleberry x ; 65 : garden pea €00Z souIUNs /3utids XDOTOISONHLA JO IVNMNOL TABLE 4.—Continued. # Snchitsu’umtsn name English or Latin name PS Fi LI SH CV CM KA 66 sn¢ tmasmas frasera ' : ' 67 sc ‘érus golden currant > x x yarcen* squaw currant ; : x x x 69 hnt ‘it ‘me ?lps* wild gooseberry x Xx : x x 70 waxi?tp mock orange x x x x 71 naq ‘nag ‘tetp Canada mint : x x Xx 72 lag ’mx°* pring beau , ' ‘ 73 sq’étm* Claytonia sp : : ’ : 74 sp it’em bitterroot Xx x x x Xx 75 stné?rmn sagebrush buttercup . x x x 76 sté‘i:hdyus buttercup . : slaq serviceberry x x 78 keela* red hawthorn ‘ : 79 sx°u ?nec* black hawthorn : x x x 80 scaq°m wild strawberry ; 81 macmaci?etp oceanspray ? x x x x x 82 sPaplsalq” apple , : ; . . 83 plamsalq® plum . pacté bitter cherry : x x “ x : 85 tax*lax* chokecherry ? , : x x 86 picus ach : : : 87 t ‘asilépa* Prunus sp ' . . 88a = sx“d:yapa?qn™ wild rose : . x x x 89 gal ‘xelp wild r x x : : 90 eniee: ce? wild en . . ; . . x 91 ack lackcap x x x x x x x 92 oe olqn thimbleberry : : : Xx x Xx 93 sok xve e tit-tel-l’mx® boysenberry . . . 94 tittell ’mx® us sp 95a : : : ‘ ° . . 96 cottonwood. x x x x 4 muls v7 daretdiitdutp trembling aspen Te 2 YAW T1Vd L ‘ON ‘€Z ‘TOA TABLE 4.—Continued. # Snchitsu’umtsn name English or Latin name PS PI LI TH SH CV CM 98 dutdutp trembling aspen dél’etp willow : 100 q ‘6lsalq® willow . x . Xx x : 101 smi?lx® wild tobacco Xx x Xx x x * x 102 pa:taq : : 103 mdsmas edible valerian 1042 ‘ . 105 piceltisa* a root 106 sk “axk “axelk’a? a water plant : 107 taxtaxitp black birch x x 108 t eptptetp a berry “Numbers are keyed to item scenes < Appendix 1 Shuswap (Secwepemc), KA = K . CM = Columbian (Nxa?amxcin), CV = Okanagan-Colville (Nsilxtsin), LI = Kalispel, and Flathead dialects), TH = Thompson (Nlaka’pamux), PI = Proto-Interior Salish, PS = Pro Salish. Terms (95) and (104) aeined as io are Spokan Lillooet (Stl’atl’imx), — _ €00Z seuruINg /ButIds XDOTOISONHLA JO TVNANOL 84 PALMER et al. Vol. 23, No. 1 derosa pine) and (63) st’Sastq, / s-V t’as-astq/ (black huckleberry). As in these ex- amples, there may also be a nominalizing prefix and/or one or more spatial prefixes and/or a stem-forming suffix, such as -t ‘inherent’. Reduplications of the linguistic root are common. The plant names display a more limited set of spatial prefixes than are found in the domains of place names and anatomical terms. A variety of substantive suffixes occur. The largest categories, involving 29 of the 103 simple lexemes, establish a division into terms with the suffixes -atp ‘plant’ (19 terms) and those with -alq” ‘tree~bush’ (10 terms). However, the structure of one term—éléatpalq” ‘kinnikinnick’—that combines the two suffixes suggests that -atp ‘plant’ may be the more general classifier, having the sense of ‘’green or leafy plants.” It appears in the names of herbs (cow parsnip, Canada mint), small or low shrubs (snowberry, silverberry, big sagebrush, northern wormwood) as well as several larger shrubs or bushes (wild rose, mock orange, ocean spray, willow) and trees (Rocky Mountain juniper, ponderosa pine, Douglas-fir, cottonwood, black birch). The suffix -alq” specifies plants that take the form of a tree or a bush, more often the former, or perhaps it is simply applied to those for which the notion of pole is most salient. The corresponding suffix -dlekw translates as ‘wood’ in Nsilxtsin (Okanagan-Colville) (Turner et al. 1998). This is one respect in which classification in Snchitsu’umshtsn may differ from other Salishan languages. In the Salishan languages having the ‘plant’ suffix, this is most often applied to various species of trees and shrubs, especially those with berries or other impor- tant cultural resources. Fourteen of the simple lexemes in Snchitsu’umshtsn have anatomical suffixes, including seven instances of -qgn (~gi) ‘head’, which, like English, has a meta- phorical extension to ‘top’. Non-hierarchical taxonomic relations are rare among the Snchitsu’umshtsn plant terms, but one instance of a plant as the descendant of blackberry vine occurs in a complex term. Only two terms have locative suffixes other than the anatomical suffixes, which can often be regarded as locative. No- tably absent from the classificatory suffixes of Snchitsu’umshtsn is -usa? ‘berry, face, eye, round thing’, which can be found in neighboring Salishan languages (Palmer 1998b). Snchitsu’umshtsn does possess the related suffix -us ‘face, eye’, but it does not occur in the recorded plant names except as a pun in (86). It appears that there is a term that stands for conifers in general, and that is term (13) yatqvetp. This term also has the more specific referent ponderosa pine. The general term for any tree is sydlalq’, a term which suggests generalization of an earlier term limited to conifers (s-yél-alq’ NOM-pitch-tree~log~pole). The gen- eral term for berries is stsastq, which is also the term for (63) black huckleberry. The general term for a bush or shrub is eedel’ There seems to be no free lexeme that covers all trees, shrubs, and herbs, only the suffix -alp (~ -atp, tp). This is a common pattern among Interior Salish lan- guages (Turner 1987, 1988). The suffix is found in all the languages and used in many names (see, for example the tree names in Turner et al. 1998). In Stl’atl’imx (Lillooet) it coexists with a more common form -az’, which apparently has the same meaning (Ex. Stl’atl’imx céx-az’ /c’q’’-atp ‘Engelmann spruce’). The ‘log, pole’ suffix (-alq’) occurs much less frequently in the Interior Salish tree names, but it is found in clearly recognizable form in all but Stl’atl’imx and Nlaka’pamux (Thompson). The latter has the possible cognate forms -alx, -dygq”, and -yaq”. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 85 a w leafy plants st’éde? (trees and shrubs) w/o woody stems (grasses) sy6lalq” eedel’ (trees) (shrubs) yatq”etp non- st’Sastq other (conifers) conifers (berries) shrubs yatq” other st’Sastq, st’$A other (ponderosa pine) conifers (black huckleberry) berries FIGURE 2.—Taxonomy of plant forms in Snchitsu’umshtsn. (Fungi and lichens not includ- ed. Dotted line indicates hypothetical inclusion.) In Columbian the only tree name that has the -alq” suffix is ‘jack pine’. Oth- erwise it is found on terms for ‘fruit trees’, ‘bump into a tree’, ‘go under a log’, ‘peck at a tree’, ‘grave marker’, ‘train’, ‘twisted tree’, ‘a tree hit by lightning’, ‘logs’, ‘round like a pole’, ‘roll up one’s sleeves’, ‘cradle of a scythe’, ‘tall’, and others. It even occurs on words for ‘short’. The suffix -afp occurs on many tree and plant names, and comes closest in Columbian to being a suffix for ‘tree’. However, it is also the suffix to indicate the plant on which particular berries grow, as opposed to the berries themselves. In a few cases it may not be divisible, that is, the root to which it is attached is not found elsewhere in the language, as in terms for ‘juniper’ and ‘spruce’. It also occurs with ‘tall sunflower’, ‘skunk cabbage’, ‘pine grass’, ‘wild lupine’, ‘sand bur plant’, and three unidentified plants, and it is used for counting plants. This distribution suggests that the taxonomic diagram in Figure 2 must be evaluated with some caution, as the function of the -alp and -alq” suffixes seems as much classificatory (differentiating by form or part within a genus or species) as taxonomic (differentiating by genus and species). We have not explored the extensions of these terms with native speakers of Snchitsu’umshtsn in a systematic fashion, but given these facts and our under- standing of the suffixes -afp and -alq’, we can still posit a taxonomy something like that in Figure 2. This taxonomy agrees generally with the classification of plants implied by Okanagan mythology (Turner et al. 1980). There, the category of ‘‘bushes, flowers, and trees” subdivides into categories of ‘trees with leaves’ and “trees without leaves.” The chief of the latter is white pine and of the former, Rocky Mountain maple. In Snchitsu’umshtsn one can propose a taxonomic hier- archy of five levels, but the classifying suffixes (-alp and -alq”) that partially struc- ture the hierarchy appear in only a minority (29) of the terms, as described above. Describing Fraser River Lillooet, Turner (1974) found eight life forms, includ- ing ‘trees’ (divided into ‘with leaves’ and ‘evergreens’), ‘berries’, Bowers , grasses (and grass-like plants)’, ‘mosses’, ‘mushrooms and fungi’, ‘weeds’, and ‘roots (and underground parts, including poisonous types). Only trees, evergreens, berries, flowers, grasses, mosses, and weeds are given general terms. Trees with leaves are unnamed, as are roots and underground parts. The Snchitsu’umshtsn classi- 86 PALMER et al. Vol. 23, No. 1 fication depicted in Figure 2 appears to support Turner's observations in a general way. Trees are divided into conifers (evergreens) and non-conifers. There are gen- eral terms for trees, conifers, grasses, shrubs, and berries, but not, as in Fraser River Lillooet, for trees with leaves or for roots. In addition there are suffixes for plants in general, and for trees and shrubs as a single category. It is possible that Snchitsu’umshtsn also has or had general terms for mosses, mushrooms and fun- gi, and weeds, but we do not have the data to confirm it. Most terms whose derivations are clear are descriptive constructions involv- ing linguistic roots specifying some attribute of color or light (eight terms), taste, smell, shape, danger, motion, texture, or use. Two pertain to death and ghosts. Utilitarian concerns are present, but not primary. The infrequent usage of spatial prefixes (‘in’, ‘on, attached’, ‘on, distributed’, ‘amidst’, etc.) in descriptive terms suggests that conceptualizations of plant structure play little role in their naming. These findings are similar to those reported by Palmer (1998b) for Secwepemc, in which 45 of 144 recorded terms were descriptive. Of the 45 descriptive terms, 33 were based on the perceptual characteristics of appearance (30) and smell (3), with the remaining few terms classified as danger or irritation (6), usefulness (5), and behavioral (1). In general, the terms bear out Randall’s (1976) observation that, rather than storing elaborate taxonomic hierarchies directly in memory, peo- ple typically store only the perceptual characteristics of classes. However, utili- tarian concerns may be primary in the entrenchment and widespread distribution of a few names, that is, those with the greatest number of cognates in neighboring languages. The Snchitsu’umshtsn terms provide some support for Berlin's (1992: 21) generalization that ‘‘names for plants and animals commonly allude metaphorically to some typical morphological, behavioral, ecological, or qualitative characteristic feature of their referents’’ [italics in original], but this generalization is so inclusive as to be almost vacuous. At least six of the terms may involve borrowings from European languages, three of these apparently from French, reflecting the influence of Father Point and other missionaries, and three from English, probably reflecting experiences sub- sequent to 1876. These include terms for barley, pea, and potato from French and apple, peach, and plum from English.’* Terms for the crops wheat and alfalfa do not appear to be borrowings. Borrowings from English can provide an occasion for puns, as in the rendering of peaches in Snchitsu’umshtsn as pi¢-us, which can be construed as ‘peach face’. Interior Salish plant names that are cognate with Snchitsu’umshtsn forms are distributed along a cline of decreasing frequency from Kalispel in the east to tl’atl’imx in the west, providing support for the conclusions of Turner et al. (1998). The 16 terms with cognates in at least six of the seven languages include names for eight trees (six of which are conifers), three berry bushes, one edible bulb and two edible taproots. At first glance, size, value in manufacturing, and subsistence value appear to be the major factors in their wide distribution, but other factors, such as trade and continuity of distribution on the landscape may be involved as well. Such utilitarian factors may motivate the creation and use of plant names, but they do not appear to govern the grammatical structure of plant names as categorizing symbols. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 87 NOTES 1Snchitsu’umshtsn <_ s-n-Cicu?ums-cn, NOM-in-Coeur.d‘Alene.people-mouth.or.language, ie., ‘Coeur d'Alene language’. The stem term Schitsu’umsh (Coeur d’Alene) has been trans- lated by Lawrence Nicodemus as ‘discovered people’. In the Coeur d’Alene community orthography, the stress is usually indicated with an underline, e.g., Schitsu’umsh. Other names of ethnic groups are presented in their own community orthographies. We judged it to be an impossible task to resolve all the Salish orthographies into one. Names of plants discussed in the text are presented in the Americanist orthography, which is discussed in the Appendix. The names may appear in analyzed form, as in the Appendix (second entries), or the unanalyzed form as they appear in Table 4, column 2. The use of the Americanist orthography was necessary to enable discussion of morphemes that are often only partially represented in the community orthography. * Nlaka’pamux is a community spelling of /nla?kdpmax/. * An anonymous reviewer made this suggestion. * The list also contains two Spokane terms, bringing the total of plant names in Table 1 to 108. ° It is quite possible that plant names that are known to a few living speakers of Snchitsu- ‘umshtsn have not yet been recorded. ° Terms 1, 2, 9, 11, 17, 23, 25, 31, 33, 40, 41, 43, 45, 47, 53, 59, 62, 72-74, 76, 78, 87, 94, 100, 103, 105, 106. See Appendix 1. ’ The spatial prefix ni?- ‘amidst’ occurred in (58), but it is omitted here because (58) is a complex term by our definition. * The syllable lacking stress or having a reduced vowel is taken as the copy. * Apparently, at least at the time the field work was conducted, the term ilch or the longer form indicating a bush alchatpalqw could be used for either the wild cranberry or kinni- kinnick. Nicodemus (1975a:111) has “ilch, n. red berries, knick-knick berries, wild cran- berry.” In (1975b:355) he has “wild cranberry, n. ilch’” and “cranberry (wild), n. al- chatpalqw” (145). ‘© Ray Brinkman and Felix Aripa, personal communication, Language Preservation Pro- gram, Office of Education, Coeur d’Alene Tribe of Idaho, Plummer, Idaho, 2000. 4 Possible Nez Perce cognates or borrowings supplied by a reviewer. 2 Naomi F. Miller suggested that nors ‘barley’ might derive from French orge. ACKNOWLEDGMENTS We thank the Tribal Council of the Coeur d’Alene Tribe of Idaho and Schitsu’umsh elders both living and deceased for their cooperation with our project over the past two decades. The list of those who contributed terms and ethnobotanical information ened Catherine Paschal (Spokane), Felix Aripa (Schitsu’umsh), Lawrence Aripa (Schitsu’umsh), 88 PALMER et al. Vol. 23, No. 1 Lavinia Felsman (Schitsu’umsh), Lawrence Nicodemus (Schitsu’umsh), Lavinia Alexander (Spokane), Maggie Mullen (Spokane), Mary Michaels (Schitsu’umsh), Margaret Stensgar (Schitsu’umsh), Mary Tutu (Schitsu’umsh), Sarah Tonasket (Schitsu’umsh), Tilly Nomee (Schitsu’umsh), William (Billy) Howard (non-Indian), and William Mishell (Kalispel). Lawrence Nicodemus, Raymond Brinkman, and Felix Aripa assisted us in rechecking forms. Thanks are due to George Urioste, for assistance with the Spanish abstract, and to Margaret Harp and Marie-Odile Junker for assistance with the French abstract. Thanks also to Roy Ogawa for his close reading of an early draft. The research on Snchitsu’umshtsn was funded preponderantly by two grants from the UNLV University Research Council in 1978 and 1981, a National Endowment for the Hu- manities Summer Stipend in 1981, an Association for the Humanities in Idaho Fellowship in 1980, and a UNLV University Research Council Summer Fellowship in 1991. Some of the work on Stl’atl’imx, Nlaka’pamux and Secwepemc plant terms was supported by grants from the Social Sciences and Humanities Research Council of Canada (e.g., SSHRC General research grant # 410-2000-1166; NT P.I.), and with contributions from Marianne B. Ignace, Laurence C. and M. Terry Thompson, Jan Van Eijk and Randy Bouchard. REFERENCES CITED Berlin, Brent. 1992. Ethnobiological Classifi- cation: Principles of Categorization of Plants and Animals in Traditional Socie- ties. Princeton University Press, New Berlin, Brent, Dennis E. Breedlove, and Pe- folk biology. American Aiton? is 42. Boas, Franz. 1890. [Tillamook, pera and Siletz texts.] Manuscript, M. Edel Collection, Manuscript ete is The University of Meggett Libraries, Seattle, Washingt » LZ, Comparative Salishan vocab- ularies. D. 1628 str ips of paper. Boas augers American Philosophical So- cie Bright, ee O. and William Bright. 1969. Semantic structures in northwestern California and the Sapir-Whorf hypoth- esis. In Cognitive Anthropology, ed. Ste- phen A. Tyler, pp. 66-78. Holt, Rinehart and Winston, Inc., New York. Brown, Cecil H. 1984. Language and Living Things: Uniformities in Folk Classification ‘ Conklin, Harold C. 1969. Lexicographical treatment of folk taxonomies. In Cogni- tive Anthropology, ed. Stephen A. Tyler, pp. 41-59. Holt, Rinehart and Winston, c. York. Doak, Ivy. 1997. Coeur d'Alene Grammat- ical Relations. Ph.D. Dissertation (Lin- guistics), The University of Texas at ustin. Ellen, Roy FE 1986. Ethnobiology, hangar the structure of pre m: some crt cee notes. aipone of Eth- nobiolo 93. Elmendorf, Wiliam. 1965. Linguistic and geographic relations in the Northern Plateau area. Southwestern Journal of An- thropology 21(1):63-78. Fowler, Catherine S. 1972. Some ee, clues to proto-Numic homelan Great Basin Cultural Ecology: A Santi: Frey, Rather 2001. Landscape Traveled by oyote and Crane. The World of the Schi- tsu’umsh (Coeur d‘Alene Indians). Univer- oe of Washington Press, Seattle /Lon- Coser, Charles. 1846. Notes on vegetation & the Missouri botanical journ souri pepe the south-pass of the poy ountains, to the Pacific, during the years 1843, and 1844. Transcript copied from the London Journal of Botany 4:479- 524. Washington State University, man. Spring/Summer 2003 Gibbs, George. 1877. Dictionary of the iskwalli. Linguistics (Appendix to Tribe es of Western Washington and Northwestern Oregon). In Contributions to North American Ethnology, volume I, part IL, pp. 285-361. waver Print- ing ice, Washin: Giorda, J. 1879. A Dibtionary op iy ow 4 or Flat-head Indian Language. St. Ignati Print, St. Ignatius, Montana. Hinton, Leanne. 1994. Flutes of Fire: Essays ards a perceptual model of folk biological classification. American Ethnologist 3:508-524. . The utilitarian factor in folk biological classification. In Directions in Illingis Press, Urbana an ; . 1998. Mixtepec Zapotec ethnobio- logical classification: a preliminary sketch and theoretical commentary. Anthropologica 40:35—-48. Johnson, Robert E. 1975. The Role of Pho- netic Detail in Coeur d’Alene Phonolo- gy. Ph.D. Dissertation (Anthropology), Washington State University, Pullman. Kuipers, Aert H. 1975. A Classified English Shuswap Word-List. The Peter de Ridder Press, Lisse. 1983. ‘Shuswap-English Dictionary. _Peeters, Leuven. ttina, Anthony. 1987. Colville-Okanagan Diath University of Montana Oc- casional Papers in Linguistics 5. Uni- versity of Montana, Missoula. Nater, H.F. 1977. Stem List Ay the Bella Coola Language. PdR Press Publications on Sa- lish Languages 4. The Peter de Ridder Press, Lisse. ——. 1990. A Concise Nuxalk-English Dic- tionary. Canadian Ethnology Service, Mercury Series Paper 115. Canadian Museum of Civilization, Hull, Quebec. Nicodemus, Lawrence G. 1975a. Snchitsu- umshtsn: The Coeur dAlene Language, Volume I. University Press, Spokane, Washington. . 1975b. Snchitsu’umshtsn: The Coeur dAlene Language, Mabie 2. University s, Spokane, gton. Palme, Gley B. 1975. Shuawae Indian eth- tany. Syesis 8:2) 993. “‘Where deters are muskrats”: JOURNAL OF ETHNOBIOLOGY 89 the semantic structure of Coeur d’Alene . Linguistic Anthropology 294. oeur d‘Alene. In Handbook 13 Jr, EP. cy b. Foraging for patterns in In- A Salish semantic domain: tud- ies in Salish Linguistics: Current Perspec- tives, sae Ewa Czaykowska-Higgins . Dale Kinkade, pp. 349-386. Mouton de Gruyter, Berlin/New York. . 2001. Indian pioneers: the settle- ment of Ni’lukhwalqw (Upper Hangman Creek, Idaho) by the Schitsu’umsh (Coeur d’Alene “prueeg Oregon Histor- ical Quarterly 102:22-4 Palmer, Gary B. and pieces Nicodemus. 1985. Coeur d’Alene exceptions to pro- posed universals in anatomical nomen- clature. American Ethnologist 12:341-359. Dale Kinkade, and the Schitsu’umsh (Coeur d’Alene). Un- published ee in preparation. Randall, Robert A. 1976. How tall is a tax- onomic tree? Some evidence for dwarf- m. American Ethnologist 3:543-553. Reliab, gaeey 1938. Coeur dAlene. Handbook of American Indian Languages. J.J. Augustin, New York. Be svaiiae ie Stem list ie the Coeur d’‘Alene language. International Journal of Ameri- can Linguistics 10:92-108. Suttles, Wayne and William Elmendorf. 1963. Linguistic evidence for Salish pre- history. In Symposium Wallace L. Chafe, pp. 42-52. University of Washington Press, Seattle. me hase A. 1930. The Coeur dAlene. Bu- of American Ethnology, Anthro- alaeieal Records 55. Washington, D.C. Thompson, Laurence C. and M. Te Thompson. 1996. Thompson River Salish Dictionary: nle?kepmxcin. University of Montana Occasional Papers in Linguis- tics 12. Missoula, Montana. Thwaites, Rueben Gold, ed. 1906. Early Western Travels he nce The Arthur H. Clark Company, Cleveland. Turner, Nancy J. 1974. Plant asa sys- 90 PALMER et al. tems and ethnobotany of three contem- orary Indian groups of the Pacific Northwest (Haida, Bella Coola, and Li- llooet). Syesis 7:1-104. ———. 1987. General plant categories in lish pas we points © 7200-02. —. obotany of conifer ‘lbers in eis son and Lillooet fabrwe Salish of British Columbia. Economic Botany 42:177-194. Vol. 23, No. 1 kowska-Higgins and M. Dale Kinkade, pp. 387-417. Mouton de Gruyter, Ber- lin/New York Turner, Nancy J. and Dawn C. Loewen. 1998. The original “ te) Turner, Nancy J. and Ro A review of the Northwest Coast tobac- co mystery. Syesis 5:249-257. Turner, Nancy J., John Thomas, eet! z . 1989. All berries have relations. vincial Museum nr “ate a ries No. 24. Victo: Jy gout C. Thompson, Anni : Thompso Journal of Ethnobiology 9:69-110. aes Nancy J., Randy Bouchard and I.D. Kennedy. 1980. Ethnobotany of the esis -Colville Indians of British Co- lumbia and Washington. British Columbia Provincial Museum Occasional Paper No. 21. Victoria. Turner, Nancy J., Marianne Boelscher Ig- nace, and eed D. Compton. 1998. Sec- wer Hans. 1940. The Kalispel Language: An wepemc (Shuswap) tree names: key to Outline the past? In Studies in Salish ia ae lations, and Dictio Current Perspectives, eds. Ewa Cza denskaps Neat: i Oslo, Oslo. and Usage of Plants by the Thomp dians of British Columbia. Royal British ale Museum Memoir No. 3. Vic- ria. APPENDIX 1.—SNCHITSU’UMSHTSN PLANT TERMS AND COGNATES Plant terms in this appendix are divided into botanical groups (lichens, horse- tails, conifers, and flowering plants, the last including monocotyledons and di- cotyledons). Within these groups, they are alphabetized by botanical family, ge- nus, and species names. Identification of botanical genus and species may be positive (no marker), probable (marked with following *), or possible (**). Each unique Snchitsu’umshtsn name (or Spokane name in two cases) is given a number. The data format for each numbered name is as follows: scientific name (common English name) (#) Snchitsu’umshtsn name in community orthography, (morphological analysis), (morpheme glosses) Morphological analyses and morpheme glosses are not always possible. In the morphological analysis field, morphemes are separated by hyphens. Morphemes may be formed by reduplications (RDP), which generally operate on the roots by complete reduplication, reduplication of consonants with vowel reduction, or par- tial morpheme reduplications of either initial or final segments. Linguistic roots are prefixed with the V symbol. In a reduplication, if the first instance were stressed, it would be labeled as the root and the rpp marker would follow. For an explanation of Snchitsu’umshtsn reduplication, see Doak (1997:27-29). In the morpheme gloss field, the gloss for each morpheme is separated from Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 91 its predecessor or follower by a hyphen. Alternative glosses of a single morpheme are separated by a tilde (~). The words of phrases are linked by periods. For example, the expression on-RDP-wind ~wrapstring.evenly-skin has three morpheme glosses, if one does not count the reduplication. The gloss for the second mor- pheme has two alternatives: wind and the phrase wrapstring.evenly. The redupli- cation applies to the second morpheme, as will be evident from inspecting the phonetic form and the morphological analysis, which flags the linguistic root. Linguistic terms in glosses are abbreviated as follows: ART=article; AUG=augmentative; CONN=connective; CONT=continuative; DEM=demonstrative; DIM=diminutive; GLOT=glottalized; INC=inchoative; INH=inherent; INT=intensive; MDL=middle; NOM=nominative; POSS=possessive; PROX=proximate deictic; RDP=reduplication; REFL=reflexive; REM=remote deictic; TR=transitive; VB=verbal; VOL=volition. The Snchitsu’umshtsn terms recorded in this study appear in three orthog- raphies: a community orthography, a contemporary Americanist linguistic or- thography (a modified version of the International Phonetic Alphabet), and the linguistic orthography used by James Teit (1930). Stress is marked where the in- formation is available, but terms taken from documentary sources do not always indicate stress and stress could not always be determined in the field. Plant names are provided in the community orthography in the appendix for non-linguists. The contemporary linguistic orthography is used for precise phonetic description and for morphological analysis. Teit’s orthography is used for terms that he re- corded, but these are also presented in the other orthographies. The Snchitsu’umshtsn community orthography used by Nicodemus (1975a, 1975b) is generally consistent, but it omits reduced vowels [9] or [I]. Consequently, there is some ambiguity in the proper placement of glottals, which are written as apostrophes, but this can usually be resolved by resort to morphological analysis with concomitant reference to the English glosses. In the appendix, phonetic forms and morphological analyses reconstructed from the community orthogra- phy are flagged with a star (*) after the word; the star before a word indicates a proto-form reconstructed by means of the comparative method of linguistics. The Nicodemus orthography underlines vowels to indicate stress. The “(” character is pharyngeal [f]. When writing glottalized consonants and semivowels, apostro- phes are placed before sonorants—'l, ‘m, ‘n, “w, ‘'y, ( and ‘(w—but apostrophes follow the voiceless stop consonants k’, p’, q’, t. The phonemes are written a, D, ch, ch’, d, e, gw h, i, j, k, k’, kw k’w khw |, ft, m, ‘m,n, ‘N, 0, BP 4, qu qu gh, qhw, r, 7, s, sh, t, t’, ts, ts’, uw wy, 'y, ( (w (, (w This is also the sorting order, except that the parenthesis is ignored. In the Americanist orthography the phonemes are written as follows: (voice- less stops and affricates) p, t, c, ¢ k*, q, q®, 2; (glottalized stops and affricates) ae t’, c’, &, k”, q’, q’®; (voiced stops and affricate) b, d, g”, ji (voiceless continuants) s, t, 8, x”, x, x”, h; (resonants) m, n, |, , w y, £, f*; (glottalized resonants) m,n, lV’, f, w, y, £’, £*’: (vowels) i, e, a, u, 0, a. In order to facilitate comparisons to other languages and simplify the transcriptions, the Snchitsu’umshtsn mid-front vowel that is often written with epsilon € is here written with e; the open 9 is here written as 0. Teit’s (1930) phonetic transcriptions may be unreliable. He seems to have often 92 PALMER et al. Vol. 23, No. 1 failed to distinguish glottalizations, labialization of consonants, rounding of vow- els, and postvelar from velar consonants. Forms reconstructed from Teit’s orthog- raphy, like those reconstructed from the community orthography, are flagged with a star (*) after the word. Teit used a straight apostrophe after the vowel to mark stress. His d is [a], which is usually written e in contemporary Salish or- thography. His £ is schwa [2]. The alveolar and palatal affricates which Teit wrote as ts and tc are written as fs and ch in practical orthographies and c and ¢ in the Americanist linguistic orthography. The palatal fricative which he wrote as c is $ in the Americanist orthography. Due to the fact that plant names and other information were collected in the course of ethnohistorical and ethnolinguistic studies not specifically focused on ethnobotany, identifications are based on the authors’ prior knowledge of local species and no voucher specimens have been deposited for curation. LICHENS Bryoria fremontti (Tuck.) Brodo & D. Hawksw. (black tree lichen) (1) sech’echt, sée’-eét, ?-hand~branch ?Peltigera sp. (lichen) (2) sitsseetsiye, Vsic-sic-iyi? ?-AUG.RDP-playingly HORSETAILS Equisetaceae (horsetail family) Equisetum spp. (horsetails, or scouring rushes) (3) he st’ede’ te t’ukhwen, he st’ede? te t’ixen, POSS grass REM horsetail CONIFERS Cupressaceae (cypress family) Juniperus scopulorum Sarg.* (Rocky Mountain juniper) (4) puntp, Vpun-t, ?-plant Thuja plicata Donn. (western red-cedar) (5) k’wa'ysalqu, Vk dys-alq*, ?-tree Thuja plicata Donn.** (western red-cedar) (6) sk’ust, s-Vk’ us-t, NOM-ghost-INH Pinaceae (pine family) Abies lasiocarpa (Hook.) Nutt., A. grandis (Dougl.) Lindl. (subalpine fir and/or grand fir) (7a) maramtpalqw, V marim-alp-alq®, medicine-plant-tree (7b) stmarimtpecht, s-t-V marim-tp-ect, NOM-attached-medicine-plant-whole. hand ~branch Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 93 Larix occidentalis Nutt. (western larch) (8) tseqwlsh, Vcéq*-I8, pink-motion.in.horseshoe.curve Picea engelmannii Parry ex Engelm. (Engelmann spruce) (9) shaghshaghtatp, Sax-\/ sax-t-atp, AUG.RDP-?-INH-plant Pinus albicaulis Engelm. (white-bark pine) (10) suwistch*, suwistc* Pinus contorta Doug]. ex Loud.* (lodgepole pine) (11) qogo'li't, q°o-V g*ol’-i ?t, INT.RDP-?-source! Pinus monticola Doug}. ex D. Don (white pine) (12) t’ada’alqu, Vt'éde?-alq”, canoe-tree Pinus ponderosa rere ex. Loud.* (ponderosa pine, yellow pine) (13) ‘yatqwetp, s-V étq’-elp, NOM-?-plant Pseudotsuga menziesii (Mirb.) Franco var. glauca (Beissn.) Franco (Douglas-fir) (14) ts’aq’atp, V éag-atp, bunched~clumped-tree? Taxus brevifolia Nutt. (yew) (15) (atsech’nalqw*, V fac-iti-dlq”, tied-back-plant (bow-plant) FLOWERING PLANTS—MONOCOTS Alismataceae (water-plantain family) Sagittaria latifolia Willd. (wapato, arrowleaf) (16) sqigwts, s-Vqig*c, NOM-wapato Araceae (arum family) Lysichiton americanum Hultén & St. John (skunk cabbage) ; (17) gekhwaekhwlshiye’, gex®-V gex®-als-tye?, AUG.RDP-?stink-arc.motion-play- ingly (18) timu’, timu? Bromeliaceae (bromeliad family) Ananas comosa (L.) Merr. (pineapple) nee (19) hnt’aplts’e’entsotn, n-\t’ap-tc'e?-n-ciit-n, in-shoot-inside-TR-REFL-NOM; ‘what shoots self through inside’ Reichard (1938:222) Liliaceae (lily family) Allium sp. (onion) (20) quwliw’Ish, qg*aliwal’s, raw Allium sp., A. douglasii Hook.** (onion) (21) sisch, sis¢ Camassia quamash (Pursh) Greene (edible blue camas) (22a) etghwe', ?étx*e?, Pdtx”e? (22b) sqha’wlutqhwe*, s-V xiw-al-?itx”a?, NOM-raw-CONN-cooked.camas 94 PALMER et al. Vol. 23, No. 1 Lilium columbianum Hanson in Baker* (tiger lily or Columbia lily) (23) ch’awgh*, Cawax* Melanthiaceae (melanthium family) Veratrum viride Ait. (Indian hellebore), very toxic (24) slaq’mn, s-V/laq’-mn, NOM-?-used for? (25) sitsSechiye, Vsacsec-iya, ?-playingly Poaceae or Gramineae (grass family) A common Interior Salish form for grass resembles the Okanagan form swu- ptla’xw ‘ground hair’. Nicodemus (1975a:81) lists the cognate form gupu’Imkhw, but he defines it as a verb: ‘vt. It (ground) is covered with much grass’. Very likely it could have been nominalized with the s- prefix to sgupu’Imkhw. In Moses Co- lumbian, st’iya? is any tall grass, but short grasses are suwpul’ex”, literally ‘hair on the ground’. grass (various kinds of forage, including grasses and the legumes alfalfa and clover (26) st’ede’, s-V t’éde?, NOM-grass Digitaria sp.** (crab grass) (27) sq’i'ts‘u’‘lmkhw, s-V q'i?c’-ul’mx~, NOM-grow-on.the.ground. Hordeum vulgare L. (barley) (28) ors, nors, possibly from Fr. orge Triticum aestivum L. (wheat) (29) st’ada’qn, s-V t'ada ?-qn, NOM-grass-head Typhaceae (cat-tail family) cat-tail, or bulrush (Typha latifolia L.)* (30) q’wosq’ws*, Vq’*és-q’“as, AUG.RDP-gather FLOWERING PLANTS—DICOTS Aceraceae (maple family) Acer circinatum Pursh*™* (vine maple) or Acer glabrum Torr. (Rocky Mountain maple (31) sqwaxt Anacardiaceae (sumac family) Toxicodendron radicans (L.) Kuntze; syn. Rhus radicans L. (poison ivy) (32) p’utp’uttumsh, p’ul-V p’ul-t-tims, AUG.RDP-poison.ivy-INH-people Apiaceae or Umbelliferae (celery family) Heracleum lanatum Michx. (cow parsnip or Indian rhubarb) (33) ghoghtp, Vx°6x"a-tp, ?-plant Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 95 Ligusticum canbyi Coult. & Rose* ee lovage) (34) ghasghs, V xis-xas, good-AUG.RD Lomatium canbyi Coult. & Rose** (white camas) (35) p’ekhwp’ukhw* V p’éx*-p'ex, glow-AUG.RDP Lomatium cous (Wats.) Coult. & Rose* (cous, or biscuitroot) (36) ka’us, ké?us Lomatium macrocarpum (Nutt.) Coult. & Rose (desert parsley) (37) sp’ekhwench, s-Vp’éx*-ené, NOM-light~glow-belly ~bank Lomatium nudicaule (Pursh) Coult. & Rose** (barestem lomatium or Indian celery) (38) peqai*, péqai*, prob. equivalent to N.P. pégiy (L. triternatum (Pursh) Coult. & Rose var. triternatum) Lomatium sp. (biscuitroot) (39) piwye, piwye~pitiwea Perideridia idles (H. & A.) Mathias** as caraway or Indian carrot) (40) st’ugom*, s-V't’tig”-m®, NOM-?-MD Asteraceae or Compositae (aster or composite family) Achillea millefolium L.** (yarrow) (41) dmdmu’geyni’, dem- V dem- -u?-gin-i?, AUG.RDP-?old-?-head-NOM Antennaria spp., Erigeron spp., Aster spp.** snag kein fleabane, aster) (42) ghaln'n’nak’wa'a'lgs ha sgwarpm, xaln (n- -V nek’ -alqs)DIM.GLOT ha sg“arpm, lie.in.order (DIM.RDP-one-spur~ridge)DIM.GLOT POss bloom Artemisia frigida Willd.* (northern wormwood) (43) p’up’u'nelp, (p’u-V pun -itp)DIM.GLOT, INT.RDP-?-plant Artemisia tridentata Nutt. (big sagebrush) (44) quw’Iqw’Imnitp, q’al’-Vq°al’-mn-itp, AUG.RDP-dark-used.for-plant Balsamorhiza sagittata sion Senet (balsamroot or spring sunflower) (45) smukwa'tsn*, smiik*a?c Cirsium brevistylum Crong. and other spp. (wild thistles) and other spiny plants, e.g., Opuntia fragilis (Nutt.) Haw. and O. polyacantha Haw. (prickly-pear cactus) (46) tek’wtuk’wt, Viek’®-tek’®-t, AUG.RDP-barb-INH Cirsium undulatum (Nutt.) Spreng.* (wavy-leaved thistle) (47) mariupa*, mariupa* Matricaria matricarioides (Less.) Porter* (pineapple weed) (48) hnts'Itsi’Ikhwgi, n-cel’-cel’x-qin, in-AUG.RDP-?-head Taraxacum officinale Weber** (common dandelion) or Agoseris sp.** (mountain dandelion) 96 PALMER et al. Vol. 23, No. 1 (49) ghal sgwarpm, xal sg’arpm, lie.in.order bloom Berberidaceae (barberry family) Mahonia aquifolium (Pursh) Nutt.; syn. Berberis aquifolium Pursh (Oregon- grape) (50) sqweyu’, s-Vqéy-u?, NOM-blue.or.green-? Betulaceae (birch family) Betula papyrifera Marsh.* (paper birch) (51) spichtena, s-V pitten-dlq**, NOM-?leaf-tree Corylus cornuta Marsh.* (hazelnut) (52) q’ip’khwe’, q’ip’x”e? Cactaceae (cactus family). See (46). Caprifoliaceae (honeysuckle family) Lonicera involucrata (Rich.) Banks ex Spreng.* (black twinberry, or twinflower honeysuckle (53) sampqn*, sdmpgn* Sambucus cerulea Raf.* (blue elderberry) and/or S. racemosa L.* (red elderberry) (54) ts’ekukw, ts’ek’ukw, c’ék’ak” Symphoricarpos albus (L.) Blake (snowberry or waxberry) (55) tmtmni’elp, V tmtmni?-elp, corpse-plant Cornaceae (dogwood family) Cornus stolonifera Michx. (red willow or red-osier dogwood) (56a) stichtskhw, s-V tiécx”, NOM-? (56b) stichtskhwelp®, s-V titcx”-élp, NOM-?-?plant Cucurbitaceae (cucumber family) Cucumis melo L. (cantaloupe (57) t(a(aghts’e’, t-fa-V/ fax-c’e?, on-INT.RDP-wind~wrap.string.evenly-skin Cucurbita pepo L. (squash) (58) niSharusi’utm, ni?-V Sar-us-i?-ut-m, amidst- hang-fire-?-be.in.position- MDL Elaeagnaceae (oleaster family) Elaeagnus commutata Bernh.** (silverberry) (59) smqhwnelp*, s-V max*-n-elp*, NOM-?snowbound-?Nom-plant Shepherdia canadensis (L.) Nutt. (soapberry or soopolallie) (60) sqhusm, s-V x*tis-m, NOM-foam-MDL Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY Ericaceae (heather family) Arctostaphylos uva-ursi (L.) Spreng. (kinnikinnick) (61a) ilch, ?ilé (61b) alchatpalqw, V ?i:l¢-itp-alq’, wild cranberry-plant-tree~bush Vaccinium caespitosum Michx.* (dwarf blueberry) (62) st’eq’In, s-V't’eq'I-n*, NOM-?-NOM Vaccinium membranaceum Dougl. ex Hook. (black huckleberry) (63) stShastq, stSha, s-V t’a8-astq, NOM-sweet-crop Vaccinium sp. (huckleberry) (64) pagpaqaghn, paq-\ pag-axn, AUG.RDP-white-arm’ Fabaceae (pea family) Pisum sativum L. (garden pea) (65) lipowee, lipowe:, from the Fr. le pois Gentianaceae (gentian family) Frasera sp.** (frasera) (66) snch’tmasms*, s-nic’-t-masmas*, NOM-cut-CONN-mdsmas. See also (103). Grossulariaceae (gooseberry family) Ribes aureum Pursh** (golden currant) (67) sts’erus, s-V c’ér-us*, NOm-hurt-face~eye Ribes cereum Dougl.* (squaw currant) (68) yarch’n*, V yar- Angaur FIGURE 1.—Map of the Palau Islands. lifted reef systems formed during the Pleistocene that have been chemically and physically weathered over the millennia, forming a rough karstic topography with caves, overhangs, and solution fissures that harbor crystalline limestone deposits (Sem and Underhill 1994). In these many flowstone and dripstone formations stone carvers from Yap quarried their famous stone money (Fitzpatrick 2001). According to Yapese oral traditions, a navigator named Anagumang first found the stone in Palau and ordered his men to cut it into various shapes, in- cluding a fish and then a full moon (de Beauclair 1963; Gillilland 1975). They settled on a circular shape and perforated the large disks so they could be carried with wooden poles or beams (Figure 2). Transported back to Yap, 400 km north- east of Palau, these disks were highly valued exchange items. Their worth de- pended on the size, shape, quality of stone, and effort expended in quarrying and transporting the disk. Europeans who became involved in this exchange system introduced metal tools and transported disks back on larger vessels, often making stone money produced with traditional technologies (e.g., shell or stone tools; 104 FITZPATRICK Vol. 23, No. 1 FIGURE 2.—A stone money disk found on Orrak Island, Palau (photograph by author). canoes and rafts) even more valuable (see de Beauclair 1963, 1971; Einzig 1966; Fitzpatrick 2001; Gillilland 1975; Nero n.d.). Several stone money quarry sites have been identified, although the paucity of archaeological research has limited what we currently know of how Yapese used the sites and the social processes that surrounded this exchange system through time. Omis Cave and Metuker ra Bisech, two of the most well known quarries found thus far in Palau, were chosen to begin an intensive survey of stone money production. Analysis of shellfish recovered during excavation pro- vides a foundation for interpreting site formation processes (e.g., engineering ef- forts that include mass movement of limestone debitage and large stone money disks), and the importance these resources had in the overall subsistence econo- mies of site inhabitants. They also provide important information on the preser- vation of these and other archaeological remains from the two sites suggesting that a variety of taphonomic processes probably affected the assemblages. Omis Cave (B:OR-1:35).—Omis Cave, located on the east side of Oreor island (part of Koror State and overseen by Ngermid Village) near the smaller Rock Islands of Itelblong and Ullemetamel, is approximately 3.3 km south of the Airai side of the K-B ‘Friendship’ bridge which connects the islands of Koror and Babeldaob. The cave encompasses an area of 780 m? and is oval-shaped in plan-view. The entrance is at sea level and faces north into a small lagoon where several smaller Rock Islands are visible (Figure 3). Metuker ra Bisech (B:IR-2:24).—Metuker ra Bisech is a large inland site with nu- merous caves, overhangs, and stone architectural features. The interior of the site Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 105 Omis Cave B:OR-1:35 am * Fe-retaininge site datum wall stone money preform (7?) ey disk FIGURE 3.—Plan of Omis Cave. is accessed by a newly constructed dock and summerhouse along the southwest portion of Ngerusar, Airai State. A pathway built as part of an ecotourism project y the government in 1997 takes visitors from the dock up 20 m above sea level into the site where excavations took place (Figure 4). The dock is approximately 2.5 km east of Omis Cave, and 0.5-1.0 km northeast of the Ngedert and Omelochel Rock islands. The Toachel Mid Channel separates Metuker ra Bisech from Omis Cave, roughly 4 km apart. RADIOCARBON CHRONOLOGY Nineteen conventional and AMS radiocarbon dates have been obtained thus far from Omis Cave (12) and Metuker ra Bisech (7) (Table 1; Fitzpatrick 2001, 2002). The chronology of Omis Cave suggests that people used the site for over 2,000 years. Four dates range from roughly 360 B.C to A.D. 560 and seven others date to the late historic period. Several older dates are intermixed with ones from the historic period, a reflection of the sloping and highly mixed stratigraphy found throughout the site (Figure 5). Overall, it is apparent that a high level of activity 106 FITZPATRICK Vol. 23, No. 1 Metuker ra Bisech B:IR-2:24 4 QO: 5: 10m + N Ww E x Countour Interval: 2 meters s FIGURE 4.—Plan of Metuker ra Bisech. was taking place within the last few hundred years around the same time that people began intensifying stone money quarrying. These activities drastically af- fected soil deposits at the site. Radiocarbon dates from Metuker ra Bisech all post-dated roughly A.D. cal. 1700. In general, the artifact assemblages, quality of stone money disks (those obviously carved using metal tools), and radiocarbon dates indicate that Omis Cave and Metuker ra Bisech were primarily used during the historic period after European involvement when stone money quarrying intensified (Fitzpatrick et al. n.d.). METHODS Three test units were excavated at Omis Cave (two—2.0 < 1.0 m; one—0.6 X 0.5 m; total = 2.4 m? soil volume) and four at Metuker ra Bisech (one—2.0 X 1.0 m; three—1.0 X 0.5 m; total = 1.9 m® soil volume). Despite limited excavation at the sites, a substantial amount of shellfish remains was collected. This was in part due to the excellent preservation of shell in the coralline limestone environment. All excavated materials were water screened over %-inch mesh. Following the methods described by Claassen (1998) and outlined by Grayson (1984) and Gilbert and Steinfeld (1977), total number of fragments (TNF), minimum number of in- dividuals (MNI), and weight (g) were recorded for both assemblages. Shellfish taxa from Omis were identified using comparative collections at the University of Oregon’s archaeological laboratory following Hinton (1972), Dance (1974), Wells 107 JOURNAL OF ETHNOBIOLOGY Spring/Summer 2003 quioq-jsod quuog-jsod OZ1 (OZ) Oz DA L GV (OF) OOL Od JIIOFSTY OZ8L (OLZT) O89 AV OF8T (OLS) OOZL AV IMO}STY OS6L (OZ8T) OZZL GV S1IO}STY IIO}STY JLIO}STY (eursts [) dV/54 ‘TD quiog-3sod Ze + + 919 96 = LET a8e Dy painseayy ‘AWDBY SIN euOZIIy—yy ‘DARD SIUIQ—DO “ypasig er JayNjayy—gqy) pesiq PI Jayna] puke aaed spwiCQ wor sayep 61 eZ _ — LLS ‘ds vusvpisy 8960FVV 30 9'ZZ7— 9°0 02-09 z € [eooreyp L960bVV 30 9'97— 90 OF-0€ Z € Jeooreyp 9960PVV D0 L£9t- rgé 0€-0Z I € [eooreyp G960FVV 30 Zz O'Z 09-0¢ F Z ds wivpouy b960TVV DO o'sz— rl 0S-OF 7 Z [eooreyp OPPEPLVA 30 797— 90 OF-0€ € Z yeooreyp €960FVV aie) or da Gig OF-0€ € Z sndoddiy “yy SPPEPLVE 30 ee S71 O€-0Z z 4 aPPIQUIONS Z960FVV 30 LZ €Sl O€-02 € I aepripsies L960FVV 30 9°S7— 80 O€-0Z € I Teooreyp 0960bVV 30 zt ST 0Z-0 Zz I ds shumy> SS60FVV 90 7SZ— S'0 0€-0Z I I yeooreyp 6S60FVV 90 07 16 OF-0€ € 7 ds snua, GL60PVV aN TZ Ell O€-0Z € p ‘ds vivpouy PL60PVV aw rea ZI 07-01 I P ‘ds vanudh> €L60hVV AW 90 6 O1-0 I p ds vivpouy 7LOOPVV aN 6'T €TL 0€-0Z Z : ds vavidi5 IL606VV AW ¥'67— Ul 09-05 9 I yeooreyp OL60FVV a 9LZ— a O€-0Z Z I Jeoozeyo 69606 VV aw Da/Der one pPAa] wAe]T = tun Jelayeypy ‘ON qe’] aS ‘(uy ‘on ATeuy eJag—y¢q uogresoIpey—' |, FIGVL 108 FITZPATRICK Vol. 23, No. 1 Test Unit 1 North East South [_] limestone rock E=:] burned limestone rock FIGURE 5.—Stratigraphic profiles from Test Units 1 and 3 at Omis Cave. and Bryce (1988), and Wye (2000). Remains from Metuker ra Bisech were iden- tified in the Palau field laboratory with help from University of Oregon field school students and Palau Bureau of Arts and Culture (formerly the Division of Cultural Affairs) staff using the same reference works. ARTIFACTS The most common artifact type found at the two sites was limestone debitage, including over 7,100 flakes and chunks from Omis Cave and nearly 6,000 from Metuker ra Bisech. Pottery was also recovered from Omis Cave and petrographic analysis of 28 of the 51 sherds suggests that they were all made with Palauan clays and tempers (Dickinson 2000; Fitzpatrick et al. 2003). Two fragments from a Trochus sp. shell ring were found at Omis Cave in Test Unit 1 and are similar to ones recovered from the Chelechol ra Orrak stone money quarry in 2000 (Fitz- patrick 2001). The only clearly identifiable tools found at the sites were two iron hatchet or blade implements recovered from Metuker ra Bisech in Test Unit 2 (Fitzpatrick et al. n.d.). SHELLFISH REMAINS Omis Cave.—From the three test units at Omis Cave, over 2,600 shell fragments (7.2 kg) representing 256 individuals were recovered. At least 50 shellfish taxa were identified from 34 different families (Table 2). Nearly all remains were re- covered from Test Unit 1 and Test Unit 2. This is probably due to the location of Test Unit 3 adjacent to an unfinished stone money disk that had been placed upright for final carving and abrasion (Fitzpatrick 2001), an unlikely area for shell refuse deposits. Tridacnids (giant clams) are by far the most common family, representing 66% of the shell weight, 56% of the TNE, and 12% of the MNI. In other Rock Island assemblages analyzed by Carucci (1992), tridacnids were not Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 109 as well represented, comprising 54.3% of the overall weight at Tmasch, 30.6% at East Beach, and 8.6% at Uchularois; overall, tridacnids make up only 15.6% of those assemblages. Although we might expect tridacnids to be overrepresented in the overall weight of the assemblage given their large size, they are also quite robust and durable, one of the main reasons for their use in making tools in Micronesia and elsewhere (Moir 1986-87; Osborne 1979). The high fragmentation rate of tridac- nids is perplexing, especially given what is already known about tridacnid re- mains recovered from other shellfish assemblages in Palau (Carucci 1992; Masse 1989; Osborne 1979) where larger and fewer fragments are typically found. Given the restricted area within which Yapese were quarrying stone money at the site and the evidence of large-scale limestone carving and movement of debitage to make room for further quarrying activities (shown most dramatically by slightly vertical, mounded stratigraphic profiles in Test Units 1 and 3), it is likely that tridacnids and other mollusks were effectively crushed during the quarrying pro- cess at Omis Cave. Although other uses of tridacnids may have included slaked lime for betel nut chewing (Carucci and Mitchell 1990) or for tools, the sheer quantity of limestone debitage and stratigraphic representations suggest that the fragmentation is due primarily to engineering tasks that included the movement of debitage (e.g., raking, piling, or pushing) and stone money (e.g., pulling, push- ing, and lifting disks in armatures or rollers). I discuss alternate hypotheses re- garding shellfish remains in the discussion section below. Many shells found at Omis Cave are quite small, including whole and frag- mentary Carditidae, Limidae, Noetiidae, Pandoridae, Architectonicidae, Cerithii- dae, Littorinidae, Mitridae, Neritidae, Strombus mutabilis, Terebridae, and Trochidae Specimens. Significant wave action could be a factor behind the deposition of these specimens. Following the arguments of Hughes and Sullivan (1974) and Gagliano et al. (1982), Carucci (1992:149) notes that materials ‘‘redeposited by storm waves generally include tiny and juvenile shells, species not considered food, water worn and fragmented shell, rounded pebbles, and flotsam such as pumice and char- coal.” Consistent with Carucci’s (1992) findings in the southern Rock Islands, the smaller specimens from Omis Cave include those that appear to be water worn and fragmented. High concentrations of charcoal and a few pebbles observed in Site deposits could also be partly attributed to storm action. It is also possible that smaller taxa were boiled in soups or stews, or even entered the site attached to larger shells and other materials such as sea grasses. Metuker ra Bisech.—A total of 1,019 shellfish fragments (24 kg) from 432 individ- uals was recovered from the surface and in stratified deposits at Metuker ra Bisech (Table 3). We identified 24 shellfish taxa from 16 families. Surface survey of the main site suggested that shellfish refuse would be found almost exclusively around or adjacent to the larger caves. Excavation of various locations within the Site confirmed this pattern with only three marine shells recovered from areas away from these caves (Test Unit 1 and Test Unit 2W). : : To examine the spatial distribution and preservation of shellfish remains, a 31.5-m? grid (Grid 1) was placed near the opening and to the west ofa large cave along the southern portion of where the investigation was focused. Shellfish re- TABLE 2.—Shellfish from Omis Cave. TU1 Total TU2 Total Total (includes TU3)* Taxon TNF MNI- Wt.(g) Wt.(%) TNF MNI- Wt. (g) Wt.(%) TNF MNI- Wt.(g) Wt. (%) Bivalves Anomiidae 6 2 3.55 0.0 — a _ — 6 2 3.00 0.0 Arcidae 1 1 0.10 0.0 a a — = 1 1 0.10 0.0 Anadara spp. — — — — 10 Z 14.00 0.2 10 2 14.00 0.2 ardiidae 34 10 54.06 0.8 _- - — -- 34 10 54.06 0.8 Fragum spp. mee —- —_ —~ 1 1 5.10 0.1 1 1 5.10 0.1 Carditidae 2 es 1.77 0.0 _— — _ — 2 2 177 0.0 Chamidae — — — —_~ 1 1 0.70 0.0 1 1 0.70 0.0 Crassatellidae 2 2 1.30 0.0 — — _ a 2 2 1.30 0.0 Donacida = a _ — 1 1 4.10 0.1 1 1 4.10 0.1 Limidae 1 1 0.59 0.0 _ — — — 1 1 0.59 0.0 Mactridae 4 3 2.70 0.0 — — — — 4 3 2.70 0.0 Mactra spp — _ — ~— 3 3 4.10 0.1 a 3 4.10 0.1 Mytilidae 6 5 0.74 0.0 5 1 0.90 0.0 11 6 1.64 0.0 Noetiidae I 1 0.23 0.0 _ —_ a — 1 1 0.23 0.0 Ostreidae 6 2 2.46 0.0 5 4 4.20 0.1 11 6 6.66 0.1 Pandoridae 1 1 0.61 0.0 —_— — _— — 1 1 0.61 0.0 Pectinidae 11 6 27.00 0.4 : 1 1.60 0.0 14 7 28.6 0.4 Chlamys spp. _ — — — 4 3 3.20 0.0 4 3 3.20 0.0 Spondylidae 1 1 31.30 0.4 — — _ — 1 1 31.30 0.4 Tridacnidae Tridacna spp. 1067 10 3881.40 54.0 364 6 879.50 22 1431 16 4760.90 66.2 Hippopus hippopus 86 S 535.80 74 78 ‘4 468.40 6.5 164 12 1004.20 14.0 (non-cultural) 24 3 190.20 2.6 2 1 222.70 3.1 26 4 412.90 ow Tellinacea _— — — 108 34 97.80 1.4 108 34 97.80 14 Tellinidae 170 12 99.50 14 35 12 37.40 0.5 205 24 136.90 19 : 15 4 14.10 0.2 — — a — 15 4 14.10 0.2 Tellina virgata 1 1 0.10 0.0 — — —_— — 1 1 0.10 0.0 Veneridae 4 4 fo 0.0 — — 4 4 3.505 0.0 Unident. bivalve 142 — 76.00 ii 132 — 64.00 0.9 274 — 140.00 1.9 OLL MOMILV dZ.L14 T ‘ON “EZ “TOA TABLE 2.—Continued. TU1 Total TU2 Total Total (includes TU3)* Taxon TNF MNI- Wt. (g) Wt. (%) TNF MNI- Wt. (g) Wt. (%) TNE MNI Wt. (g) Wt. (%) Gastropods Architectonicidae 13 11 0.96 0.0 2 2 0.86 0.0 15 13 1.82 0.0 Buccinidae — — — — 1 1 0.40 0.0 1 1 0.40 0.0 Cerithiidae 3 2 0.07 0.0 —_ — — — s 3 0.07 0.0 Conidae 4 3 8.08 0.1 oe a= a _ 7 6 8.33 0.1 Conus spp. su 1 2.15 0.0 — — — _ 3 1 2.15 0.0 Conus pennaceus 1 1 1.87 0.0 — — — 1 1 1.87 0.0 Conus tessulatus 1 1 1.24 0.0 —_ — — — 1 1 1.24 0.0 Conus litteratus 2 2 117.69 1.6 a — _ — 2 2 117.69 1.6 Cypraeidae q Z 25.81 0.4 3 3 8.71 0.1 7 5 34.52 0.5 iotida 15 2 2.80 0.0 oe _ — — 15 2 2.80 0.0 Littorinidae _ a= — _ 4 4 0.65 0.0 4 4 0.65 0.0 Mitridae — — — _ 1 1 0.49 0.0 10 10 1.05 0.0 itida 5 3 2.98 0.0 1 1 0.21 0.0 2 2 0.35 0.0 Patellidae 1 1 0.33 0.0 _ — — — 1 1 0.33 0.0 Strombidae 1 1 110.92 1s 2 2 35.52 0.5 3 3 146.44 2.0 Lambis lambis 1 1 0.76 0.0 1 1 19.57 0.3 2 2 20.33 0.3 Strombus 6 6 16.86 0.2 -— — — 6 6 16.86 0.2 Strombus mutabilis —_ — —_— — — — — — 2 Zz 0.37 0.0 Terebridae 1 1 0.03 0.0 — — 1 1 0.03 0.0 erebra spp 1 1 .04 0.0 — — — 1 1 0.04 0.0 Turbinidae 2 2 0.06 0.0 — — —_ — 2 2 0.06 0.0 operculum a oa oo —_ 3 3 2.83 0 Z a 2.83 0.0 Turridae — _ —_ — 1 1 0.40 0.0 1 1 0.40 0.0 Trochidae 12 9 15.95 0.2 _ —_ - 12 9 15.95 0.2 Umbonium vestiarium 9 9 39 0.0 _ — — — 9 9 39 0.0 Undent. operuclum 5 5 0.25 0.0 7 7 3.98 0.1 12 12 4.23 0.1 Unident. gastropod 100 — 34.23 0.5 39 a 14.75 0.2 139 — 48.98 0.7 Total 1773 141 5270.51 73.3 817 103 1896.07 26.4 2602 256 7164.92 99.7 * TU3 is represented by only four taxa. Conidae: 3 TNE, 3 MNI, wt. 0.3 g, wt. % 0.0; Mitridae: 9 TNE, 9 MNI, wt. 0.6 g, wt. % 0.0; Neritidae: 1 TNE 1 MNI, wt. 0.1 g, wt. % 0.0; Strombus mutabilis: 2 TNE. 2 MNI, wt. 0.4 &, wt % 0.4. Total: 15 TNE 15 MNI, wt. 1.3 g, wt. % 0.0. €007 Joumuns /Sutidg XOOTOISONHLA JO TVNMNO[ TABLE 3.—Shellfish data from Metuker ra Bisech. CLL TU4 Total Grid 1 Total Total (includes TU1 and TU2W)* Taxon TNF MNI- Wt. (g) Wt.(%) TNF MNI Wt. (g) Wt.(%) TNF MNI- Wt.(g) Wt. (%) Bivalves Anomiidae a — 4 2 109.3 0.5 4 2 109.3 0.5 Arcidae 17 102.4 0.4 2 1 70.6 0.3 19 4 173 0.7 ra spp 147 78 5856.1 24.4 88 57 2969.1 12.4 236 136 8871.2 37.0 Cardiidae 2 67.8 a -- “= — 2 2 7 0.3 Fragum spp. 167 87 3617.1 15.1 — os — -— 167 87 3617.1 15.1 Acrosterigma sp. 2 2 61.7 0.3 — 2 61.7 0 Chamidae 4 4 350.8 iS | 3 352.4 15 8 7 703.2 2.9 Mytilidae 41 2 114.3 0.5 — 41 2 114.3 0.5 Septifer spp 4 2 4.4 0.0 oa a 4 2 4.4 0.0 Ostreidae 70 17 837.6 3.5 1 1 4.2 0.0 71 18 841.8 35 Pectinidae 2 2 91.9 0.4 3 2 102.1 0.4 5 4 194 0.8 Spondylidae pondylus spp. ? 6 452.3 1.9 1 1 107.8 0.4 8 z 560.1 23 Tridacnidae Tridacna crocea 95 50 2919.4 12.2 13 13 604.5 25 108 63 3523.9 14.7 Veneridae Venus spp. 11 a 723 0.3 1 1 41.7 0.2 12 4 114.0 0.5 Unident. bivalve 71 — 55.2 0.2 — —_ _ -- 71 _ 55.2 0.2 Gastropods Conidae —_ — -— — 2 2 88.0 0.4 2 2 88.0 0.4 Conus spp. 61 5 1047.2 4.4 19 14 474.2 2.0 80 19 1521.4 6.3 Conus litteratus 7 7 92.8 0.4 20 13 453.4 1.9 27 20 546.2 ya Cypraeidae 24 9 309.1 i 1 1 liz 0.0 26 11 378.4 1.6 Cypraea tigris 18 6 240.9 1.0 3 3 99.3 0.4 21 9 340.2 1.4 Neritidae — — — — _ —_ _ —— 1 1 Bao 0.0 Strombidae Lambis lambis 5 5 83.3 0.3 — — o = 5 5 83.3 0.3 Strombus spp. 2 2 0.05 0.0 — — — —- 2 2 0.05 0.0 Terebridae 65 22 168.1 7.0 — — _ — 65 22 1681.1 7.0 Trochidae —_ a — _ i 1 225.0 0.9 1 1 225.0 0.9 Unident. gastropod 28 — 82.1 0.3 3 — 133 0.1 31 — 95.4 0.4 _ Total 850 314 18139.9 757 166 15 5726.1 23.9 1019 432 23972.4 100.0 WORLLVdZ.L14 L ON ‘€Z “TOA * Tui *7TU1 had only one taxon, Neritidae: ve TNE, 1 MN wt. 2.3 g, wt. % 0.0; TU2W has only two taxa—Anadara spp.: 1 TNE 1 MNI, wt. 46.0 g, wt. % 0.2; Cypracidac: 1 TNE, 1 MNIL, wt. 58.1 2, wt. % 0.2 Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 113 covered from the grid area consisted predominantly of Anadara sp. intermixed with mostly Chamidae, Tridacna crocea, Conus litteratus, and other Conus spp., sim- ilar to what was observed in other areas of the site with concentrated shellfish remains. Nearly all of the specimens were whole with only a small percentage of fragments. To determine the depth of deposits, subsurface shellfish taxa, and taphonomic processes on these specimens, a small 1.0 X 0.5-m test unit was placed within the grid and excavated to a depth of 50 cm. A total of 850 individual fragments weighing 18.1 kg and representing 314 individuals was recovered. The assem- blage consisted mostly of Anadara sp., Conus sp., and Tridacna crocea, with a small amount of Chamidae and Conus litteratus. This is similar to what was found in the surface collection. A large number of specimens that were not well represented in the surface grid were also recovered: Cardiidae (especially Fragum sp.), Cypraeidae, Mytilidae, Ostreidae, Spondylidae, and Terebridae. Overall, excavation revealed extensive shellfish remains primarily concentrated in deposits less than 30 cm deep, numerous taxa that were not visible on the surface, and no artifacts. Other Shellfish and Faunal Remains—Although mollusks dominate the Omis Cave and Metuker ra Bisech faunal assemblages, remains of fish, crustacea, echino- derms, and hawksbill turtle (Eretmochelys imbricata) carapace were at Omis Cave, and only a few crustacea fragments found at Metuker ra Bisech. Fish remains identified thus far include parrotfish (Scaridae), wrasse (Labridae), grouper (Se- rranidae), and sea bream (Lerithinidae). Analysis of faunal assemblages is cur- rently underway at the University of Oregon and Kansai Gadai University, Japan. DISCUSSION The most common shellfish taxa at both sites would have been collected in intertidal zones or shallow reefs, including those from the families Arcidae (e.g., Anadara), Cardiidae (e.g., Fragum), Conidae (e.g., Conus SP., Conus litteratus), Te- rebridae (Terebra sp.), and Tridacnidae (e.g., Tridacna sp., Hippopus hippopus) (Table 4). Many of the species within these families are found in shell midden deposits elsewhere in the archipelago (Carucci 1992; Osborne 1979) and are food items still popular with present-day Palauans (Melson Miko, pers. comm.). Other gen- era, including Tridacna and Terebra, were used for making shell adzes (Osborne 1979), and Conus was modified for making beads, pendants, scrapers, peelers, or other ground objects (Carucci 1992:94; Fitzpatrick n.d.; Osborne 1979). Unfortunately, there is little published ethnographic or ethnohistoric infor- mation regarding shellfish collecting in Palau and it remains an understudied aspect of Micronesian subsistence activities. Semper (1873) and Kramer (1926) provide brief accounts of traditional Palauan shellfishing practices, but these are only minimally descriptive and lack quantification. A comparative study by Lebar (1963) in Truk (Chuuk) gives some insight into the importance shellfishing has in the Micronesian diet, but is contemporary in perspective and does not neces- sarily reflect past food collecting behaviors. Thus, archaeology must be used to determine shellfish abundances and preferences and how this reflects past cul- tural behavior through time. 114 FITZPATRICK Vol. 23, No. 1 TABLE 4.—Habitats of mollusks found at Rock Island limestone quarry sites, Omis Cave (O.C.) and Metuker ra Bisech (M.B.). Present Taxon Common name O.C. M.B. Habitat Bivalves Anomiidae Jingle shells XX usually attached to rocks or coral shallow water Arcidae Ark shells XX — muddy or coral sand or attached to rocks in intertidal zones Cardiidae Cockle shells oa’ muy or coral sand in intertidal Carditidae Cardita clams X Hiscned to the undersurfaces of rocks or coral in intertidal zones Chamidae Jewel boxes XX cements to rocks or reefs in inter- tidal zones or shallow water Crassatellidae Crassatellas x muddy or st bottoms Donacidae onax or Wedge X exposed sandy beaches in intertid- clams al zones te shallowly bur- ied) Limidae File clams nie rocks or coral in intertidal zones Mactridae Mactra clams entre or coral sand in intertidal Mytilidae Mussel shells A>, & ay attached to rocks or wood shallow water; some bore into rocks or burrow in sand or avel Noetiidae Noetias X muddy bottoms in shallow water Ostreidae True oysters XX __ cemented to rocks or shells in in- tertidal zones Pandoridae Pandoras x generally live in sandy or pebbly bott Pectinidae Scallop shells X X__ rock crevices or sandy areas in shallow water; avid swimmers Spondylidae Thorny oysters XX __ attached to rocks or coral i i Giant clams X X __ shallow waters of coral reefs; rest unattached to reef or lagoon bottoms; bore into coral pockets Tellinidae Tellins burrow in sandy or muddy areas in intertidal region: Veneridae Venus clams X muddy sand or clean coral sand Gastropods Architectonicidae Sundial shells X sandy areas in shallow water Buccinidae Whelks X rocky shores of intertidal zones Cerithiidae Ceriths shells X clean coral or grass hal- low waters of intertidal zones; also mud and grassy areas Conidae Cone shells X X_ under coral or in crevices; coral sand or hard reef Cypraeidae Cowrie shells X X_ coral reefs of shallow water Littorinidae Periwinkles X clings to rocks and grasses near de line; also inhabits mangrove Mitridae Miter shells X clings to rocks and grasses or in sandy areas near tide line Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 115 TABLE 4.—Continued. Present Taxon Common name O.C. M.B. Habitat Neritidae Nerites XX — under rocks or in crevices at high tide levels of intertidal regions; clings to wave-washed rocks in great numbers Strombidae Conch shells X X muddy, grassy, or clean coral sand in shallow waters of intertidal zones Terebridae Auger shells XX __ sandy floors in shallow water Turbinidae Turban shells xX intertidal reef zones; under coral rocks Turridae Turrid shells X attached to rocks in intertidal zones Trochidae Top shells X X_ grassy areas; sometimes attached to rocks in intertidal zones Despite the paucity of archaeological research dedicated to shellfish assem- blages in Micronesia, it is clear that shellfish were an important component of the diet to people living in the limestone Rock Islands of Palau and probably the volcanic islands too, although preservation bias precludes a better assessment. One of the difficulties in evaluating the significance of shellfish in the Rock Islands is that some specimens could have been carried by natural phenomena (e.g., wind, tide, and storm activities). Likewise, it is often difficult, if not impossible, to dis- cern which specimens, especially of the smaller taxa, were actually eaten by past inhabitants. And, all of these problems are exacerbated by the fact that different site inhabitants (e.g., Palauans, Yapese) during separate or even overlapping pe- riods of time may have had similar (or dissimilar) cultural behaviors that influ- enced shellfish collection strategies and responses to environmental stimuli. These are issues that can be partially resolved, however, by careful examination of the faunal assemblages and other site constituents, extensive radiocarbon dating, and complementary lines of evidence such as ethnography and ethnohistoric accounts. So, what are the main factors influencing the quantity and quality of shellfish assemblages at Omis Cave and Metuker ra Bisech? Despite both sites having both been used as stone money quarries, shellfish assemblages reveal some striking contrasts. In terms of taxonomic richness, Me- tuker ra Bisech has less than half the number of discrete shellfish taxa (48%) and families (47%) as those found at Omis Cave, despite the much larger sample from the former. As mentioned previously, one reason for this discrepancy is that Omis Cave is situated adjacent to the water and Metuker ra Bisech is far inland and at a much higher elevation. Wave and storm action could deposit smaller species in greater numbers and variety. Other predators such as birds may also contribute to this greater number of taxa. The gathering of larger shellfish (such as tridac- nids) or other resources, including sea grasses, may also introduce smaller species into archaeological deposits as “piggy backers”; however, this is difficult, if oie impossible, to quantify accurately. Several studies have addressed the sorting o 116 FITZPATRICK Vol. 23, No. 1 @ Metuker ra Bisech O Omis Total number of fragments (TNF)/weight (g) Mollusk Families FIGURE 6.—Total number of fragments (TNF)/fragment weight (g). Higher numbers in- dicate larger size fragments; lower numbers indicate smaller size fragments. faunal and artifactual material by weather-related phenomena or “‘piggy-backing”’ on other shells or objects (Bird 1992; Carucci 1992; Hughes and Sullivan 1974; Lever et al. 1964). These hypotheses, however, have not yet been tested at quarry sites in Palau. To better compare assemblages from both sites, quantitative measures of the total number of fragments (TNF) for each taxa, minimum number of individuals (or MNI, based on the unique identifiable elements for each taxa that is represen- tative of a single individual), and weight (g) were standardized (see Claassen 1998:117; Grayson 1984). Because the types and quantity of shellfish remains were different between sites, I used the seven most common families that included both bivalves (Tridacnidae, Arcidae, Cardiidae, Ostreidae) and gastropods (Conidae, Cypraeidae, and Strombidae) to standardize the data and give a measure of rel- ative abundance. The TNF were divided by total specimen weight for each site to illustrate differences in fragment size (Figure 6). The MNI were divided by soil volume to establish a comparison of shell quantity between sites (Figure 7). Total shellfish weight (g) for each site was divided by the weights for each family to give a comparative indication of their overall prevalence (Figure 8). The average weights for individual fragments indicate that shellfish are more highly fragmented at Omis Cave. Taphonomic studies of various mollusks dem- onstrate that under many conditions, bivalves are more likely than gastropods to become fragmented. The morphology and crystalline composition of bivalves makes them susceptible to breaking along structurally weaker points (Claassen 1998:56). This trend holds true for Omis Cave, where there is an increase in av- erage weight per fragment from bivalves to gastropods. At Metuker ra Bisech the Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 117 B® Metuker ra Bisech 2 $s 2 3 oom 2 & ee ce § °o 2.5 gq. 2 — 2 Eg is (4 & C4) ae Per SAG - iS Mollusk Families ” é FIGURE 7.—Shellfish density measure for each site using minimum number of individuals (MNI)/soil volume (m?). 350 300 250 - @ Metuker ra Bisech oo | to the overall site assemblage weight (g) Percent weight (g) of shellfish families relative i j i ili i he overall site FIGURE 8.—Percentage by weight of major shellfish families relative to t i assemblage vieiahit (. Higher bes indicate the relative abundance of each family within their respective site. 118 FITZPATRICK Vol. 23, No. 1 2 15 134 Z 13 ase e 11+ 3 8.55 ze Le @ Metuker ra Bisech 2 7 4 0 Omis Cave 5 i Phie a4) 25 2.02 2.12 9 1.09 a ae 2 Gastropods Bivalves Total FIGURE 9.—’Mean completeness’ of shellfish (TNF/MNI). Values approaching 1.0 are more complete, whereas numbers above 1.0 show increasing rates of fragmentation. reverse is true, except for Ostreidae. This is not surprising given how oyster shells, having a foliated structure, tend to fracture more easily than other types of bi- valves (Claassen 1998:56-57). The ratio of MNI divided by soil volume (Figure 7) follows a similar trend to that of total shellfish weights with a higher density of taxa from Metuker ra Bisech than Omis Cave. More whole bivalve halves were also recovered from Metuker ra Bisech, mirroring what both quantitative stan- dardizations show. Figure 8 simply illustrates the relative abundance of families between each site in terms of weight, but also gives an indication of how weight measures affect the samples. The seven shellfish families from Omis Cave com- prise less of the overall site assemblage weight compared to Metuker ra Bisech, with the exception of Strombidae and Tridacnidae. The data presented in these figures suggest that 1) shellfish are more highly fragmented at Omis Cave; 2) Omis Cave is taxonomically richer; 3) the overall assemblage weight is dominated by tridacnids at Omis Cave (not necessarily reflecting a higher percentage of use as seen by MNI counts); and 4) peoples living at Metuker ra Bisech had a pref- erence for bivalves, especially Tridacnidae, Arcidae, and Cardiidae with gastro- pods being far less important—the similarity in environment between the two sites makes it unlikely that these differences can be attributed to habitat. The ‘mean completeness’ of shellfish assemblages was also determined by dividing TNF (minus the indeterminate fragments) by MNI to give another es- timation of fragmentation (with values approaching one being more complete; see Grayson 1984 for a review of sample size effects; Figure 9). The completeness measure for Omis Cave gastropods was 1.09, bivalves 13.4, and total assemblage 8.55. Metuker ra Bisech gastropods were 2.5, bivalves 2.02, and total assemblage 2.12. Using this measure of completeness, the Omis Cave assemblage shows an overall fragmentation rate that is over four times that of Metuker ra Bisech. Al- though gastropod fragmentation is higher at Metuker ra Bisech than Omis Cave Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 119 and would appear to conflict with the overall trend, this can be attributed to the size of the gastropods found at each site. By dividing gastropod TNF with total weight, individual fragments at Omis Cave weigh 1.7 g, while Metuker ra Bisech weighs 19.0 g. These numbers reflect some of the smaller species or specimens (< 2 cm) in various gastropod families such as Architectonicidae, Cerithiidae, Neritidae, Patellidae, Terebridae, Turbinidae, and Trochidae; although they have a high MNI, they easily could have been washed or brought into the site by non- cultural processes. Two of the other major shellfish assemblages reported from Palau are from Peleliu (Osborne 1979) and sites in the southern Rock Islands such as Ngemelis (Carucci 1992). In general, the most common bivalves found in these studies were Atactodea and Fragum, occurring in about the same relative abundance (MNI) in both collections. Strombus gibberulus was the most common taxon (95% in Os- borne's and 89% in Carucci’s collection). Except for Fragum, which comprises 15% of the total weight and 20% of the MNI at Metuker ra Bisech, these data are quite different from what is found at the two Yapese stone money quarries. My analysis reveals a disparity in the shellfish taxa identified as well as pres- ervation between the two quarry sites. I suggest that the high shellfish fragmen- tation at Omis Cave is primarily due to the intensive engineering tasks associated with limestone carving in a smaller, more restricted area, whereas remains from the site of Metuker ra Bisech (most of which were almost completely intact), reflect the separation of habitation and quarrying areas at a much larger inland site. Widely separated radiocarbon dates and evidence of stone money quarrying at Omis Cave suggest substantial disturbance of the soil. I would not discount the possibility that the Yapese collected only certain taxa or restricted their collecting to certain locations, but this hypothesis is difficult to test given differences in location and preservation. Despite these questions, shellfish remains provide im- portant evidence about habitation, subsistence, quarrying, and other daily activ- ities of the Yapese quarry workers who lived at the two sites while they carved stone money disks. The diversity and abundance of shellfish demonstrates the importance these resources had for people living at the sites. We still do not know, however, the role shellfish actually played. Were they eaten, used for making tools and orna- ments, or did they serve some other purpose? In their analysis of Strombus gigas in the West Indies, Jones O’Day and Keegan (2001) suggest that “most gastropod and bivalve shells are so durable that they are unlikely to break into fragments without human intervention,” and that “there is no reason for large and heavy shells (e.g., S. gigas) to be brought to a settlement unless some further use was intended. Thus, we should assume that large shells and their fragments found in archaeological sites were the product of tool manufacture.” This may be true of West Indian sites, but not necessarily for others along the North American Pacific coast (see Erlandson et al. 1999 and Sharp 2000 for some California Ce, the Mississippi Valley (Peacock 2000), or stone money quarries. opti (197 also suggested a similar arrangement, noting that when tridacnid shells were no needed for tools, their meat was collected and the massive shells were left on the reef. However, some shellfish will remain fresh for days if left in shells and then transported over long distances. 120 FITZPATRICK Vol. 23, No. 1 There are several possible explanations for the relatively high presence of tridacnid shells (TNE, MNI, and weight) at Omis Cave and Metuker ra Bisech that may also address Jones O’Day and Keegan's (2001) and Osborne's (1979) hypoth- eses. The first is that radiocarbon dates and metal tools place quarrying activity at Metuker ra Bisech just prior to and after European contact. Tridacna crocea, the only species of tridacnid found at the site, was rarely, if ever, used for producing tools in prehistory and was probably only a food item. Because of its smaller size (< 15 cm in length), there would have been little effort expended to bring both the meat and shell up to the main site. The shell then could have served as a container or other unmodified utilitarian instrument. In addition, Yapese quarry workers were generally low caste, often providing corvée labor to Palauan clans or villages in exchange for quarrying rights (de Beauclair 1971; Einzig 1966; Gillilland 1975). It is unlikely that they would have had unrestricted access to any and all food items. Oral traditions describe Palauans bringing food to quarry workers (e.g., Holyoak and Miko 2000), but whether this was done on a regular basis is unknown. The Rock Islands have shallow, poor soils for cultivating crops like taro. Thus, shellfish could have been a nearby, easily captured, and important food source. The high fragmentation rate of tridacnids and other shells at Omis Cave is probably not explained by tool production either. No shell tools have been found at the site and only two small fragments of a Trochus shell ring were recovered during excavation (Fitzpatrick 2001). Shell midden deposits are quite common in other caves and rockshelters in Palau (Carucci 1992; Masse 1989), and it is likely that people living there brought larger shells into the site. The cave’s proximity to reef and lagoon habitats would have made this quite easy. Radiocarbon dates at Omis Cave also span more than 2,000 years from 300 cal. B.C. to the modern era. The slightly sloping stratigraphic profiles, radiocarbon dates, and lack of shell artifacts all point to shellfish assemblages having been affected by engineering activities. It should be noted too, that a fundamental difference between Strombus gigas and tridacnids lies in their size and morphology. The most common tridacnids used for tools in the Pacific, Tridacna maxima and Tridacna gigas (Moir 1986-87), are much larger than S. gigas, attaining lengths of 35 cm and 100 cm respectively (Rosewater 1965), whereas S. gigas rarely exceeds 30 cm in length (Dance 1974: 83). Tridacnids also have two halves and are often used as containers for water. S. gigas is a gastropod and its use as a vessel is not well documented, nor would it be as convenient to use as half of a larger bivalve shell. With such a complex mixing of cave deposits, it is difficult to assign specific faunal remains to a par- ticular cultural period. Nonetheless, it is reasonable to conclude, given the avail- able evidence, that engineering tasks associated with stone money quarrying in- troduced a variable that broke shell into fragments as a result of human inter- vention on a massive scale. CONCLUSIONS The analysis of shellfish remains from Omis Cave and Metuker ra Bisech suggests the following: Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 121 1) Radiocarbon dates and artifact assemblages indicate that Palauans used Omis Cave over a long period of time, perhaps intermittently as a temporary campsite, and by the Yapese for stone money manufacture. Metal tools and a lack of pottery at Metuker ra Bisech suggest that only the Yapese used this site within the last few hundred years for quarrying stone money. 2) The greater varieties of shellfish found at Omis Cave are likely a result of the site’s close proximity to the sea. Storms, tidal action, predators, and indirect transport of smaller taxa all likely influenced the kinds and size of shellfish found at the sites. 3) Nearly all shellfish taxa, including those shared by the two sites, are found in intertidal zones or shallow reef areas. This is not unexpected given that Rock Island ecological zones are within a barrier reef with numerous complex coral reef systems. 4) Shellfish remains are highly fragmented at Omis Cave and much less so at Metuker ra Bisech. In the case of stone money quarries, this is probably a result of multiple factors including site dimensions, quarrying activities in a restricted area, and engineering tasks associated with the carving or breaking of limestone, movement of debitage, and transport of stone money disks away from flowstone deposits. We should expect to find similar fragmentation rates in future investi- gations at quarry sites. 5) People living and working at stone money quarry sites showed a prefer- ence for bivalves, particularly tridacnids, although a wide variety of shellfish were exploited. Previously I thought giant clam shells found at quarry sites may have been used for producing tools or as utilitarian objects (e.g., water or food con- tainers) (Fitzpatrick 2001). Very few shell tools have been found at stone money quarry sites and none in direct association with quarry refuse (i.e., limestone debitage). This suggests that tridacnids were predominantly gathered for food, although their use as containers cannot be ruled out. The small number of shell artifacts in general suggests that shellfish at these sites are primarily food remains or incidental site constituents. 6) Shellfish varieties recorded at stone money quarries are dissimilar to those found at other sites in Palau (Carucci 1992; Masse 1989; Osborne 1979), suggesting that even though people took advantage of the rich diversity of faunal material in these environments, they were probably doing so with different dietary or technological preferences. Further research will help determine what these pref- erences were and if they represent cultural differences (Yapese vs. Palauan). In any case, this research reinforces the idea that shellfish were an important re- source for those living in the Rock Islands over a long period of time, as Masse (1989), Osborne (1979), and Carucci (1992) have suggested. : is study provides the first description of faunal remains recovered from Yapese stone money quarries. The data allow for a better understanding of how Yapese Islanders made use of these sites and provide a framework for developing hypotheses about settlement patterns, engineering tasks, and modes of activity at quarry sites. The evidence, especially in the case of the Omis Cave assemblage, suggests that proper sampling techniques using ’4-inch screen can help in recov- ering smaller taxa present, some of which may result from non-cultural processes. 122 FITZPATRICK Vol. 23, No. 1 ACKNOWLEDGMENTS I am indebted to all of the staff from the Palau Bureau of Arts and Culture for their on- going support of my research during the past several years. Rita Olsudong, Taurengel Eme- siochel, and Tamael Klouchelad have been especially generous with their time and I owe them a great deal of thanks. Vicky Kanai, Dave Orak, Walt Metes, Florencio Gibbons, and Lynda Tellames were all instrumental in setting up the logistical and administrative sides to the projects. Bethany Watson and Cassidy DeBaker served as co-trainers for the 1999 and 2000 field projects in Palau. Stephen Delaney, Brian Diveley, Ryan Hagerty, Bill Mcgivern, Kate Swanger, and Erika Zwarg also participated in the 2000 field project. Jenna Boyle and Sally Jones helped to identify shellfish remains in the University of Oregon archaeological laboratory. Kate Swanger aided in drafting the site maps. William R. Dickinson conducted thin-section petrography on the ceramics. The Omis and Metuker ra Bisech projects were supported by Sigma Xi, the National Science Foundation (SBR-0001531), and a grant from the Sasakawa Foundation for archaeological training projects in Micronesia awarded to Wil- liam S. Ayres. James Carucci, Jon Erlandson, Madonna Moss, Torben Rick, William F. Keegan, and two anonymous reviewers provided useful comments on previous drafts of this paper. Thanks go to Christophe Descantes and Carla Guerr6n-Montero who translated the abstract into French and Spanish, respectively. I especially thank Kate Moore (University of Penn- sylvania) who reviewed the final draft for content and consistency, and the editor of the journal, Naomi FE Miller, for working with me to improve various aspects of this manuscript. Of course, all conclusions and errors derived from this research are mine alone. REFERENCES CITED Bird, M.K. 1992. The impact of tropical cy- clones on the archaeological record: an Australian example. Archaeology in Oce- ania 27:7. Carucci, J. 1992. Cultural and natural pat- terning in prehistoric marine foodshell from Palau, Micronesia. Unpublished Ph.D. dissertation ee ~ ern Illinois University, Carbon dal Cavaae J. and S. Mitchell. 1990. Lime-en- crusted shell artifacts and the prehis- toric use of slaked lime. Micronesica Supplement 2:47-64. Claassen, C. 1998. Shells. Cambridge Uni- venaity Press, Cambridge. Dance, S.P. 1974. The Collectors Encyclopedia Shells. McGraw Hill, New York. de Beauclair, I. 1963. The stone mon Yap Island. Bulletin of the Institute 4 Ln nology, ia Sinica 16:147- . 1971. Studies on Botel Tobeg ago and Yap. In Asian Folklore and Social Life Monographs, ed. Lou Tsu-k’uang, pp. 183-203 Orient Cultural Service, Taipei. ickinson, W. 2000. Report on the petrog- raphy of Omis Cave pottery sherds. uscript in possession of the author. Einzig, P. 1966. Primitive Money in Its Eth- nological, Historical, and Economic As- pects, 2nd edition. Pergamon Press, Ox- ford. aires +, J.T, C. Rick, R.L. Vellanoweth, DJ. Kennett. 1999. Maritime sub- rear at a 9300-year-old shell midden on Santa Rosa Island, California. Journal of Field Archaeology 26:255-265. Fitzpatrick, S.M. 2001. Archaeological in- vestigation of Omis Cave: a Yapese stone money quarry in Palau. Archaeol- ogy in pe 36:53-62. 2. A radiocarbon chronology of Yapese stone erg quarries in Palau. Micronesica 34:227-2 ——. nd. Early aah burials at Che- lechol ra Orrak. Evidence for a 3000- year-old occupation in western Micro- nesia. Antiquity. (in press) Fitzpatrick, S.M., W.R. Dickinson, and G. Clark. 2003. Ceramic petrography and cultural interaction in Palau, Microne- sia. Journal of Archaeological Science. (in press) Fitzpatrick, S.M., A. Caruso, and J. Boyle. n.d. Metal tools and the transformation Spring/Summer 2003 of an Oceanic exchange system. Manu- he on file, Palau Bureau of Arts and Cultu (options: S. M., C.E. Pearson, R.A. Weins- tein, D.E. Wiseman, and C.M. Mc- Clendon. 1982. Sedimentary studies of prehistoric archaeological sites of the northern Gulf of Mexico continental G , nc. Gilbert, A.S. and P. Steinfield. 1977. Faunal remains from Dinkha Tepe, northwest- ve Iran. Journal of Field Archaeology 4: 9-51. | Seactey on C. 1975. The Stone Money of : 1 , Senisnecan S ngton D.C. Grayson, D. 1984. Quantitative pene oe logy. Academic Press, New Yor Hinton, A. 1972. Shells of New Guinea and the Central Indo-Pacific. Robert Brown and Associates and Jacaranda Press, Port Moresby. Holyoak, L. te M. — pis Oral his- tory of Omis Yap quarry, Ngermid Homie Rarer ‘State, Republic of Palau. Report on file at Division of Cultural Affairs, Republic of Palau. Hughes, PJ. and M.E. Sullivan. 1974. The re-deposition of midden material by storm waves. Journal and Proceedings of oe Bik Society of | New South Wales 107: banc! ~ Day, S. and WE Keegan. 2001. Ex- pedient shell tools from the northern West Indies. Latin American Antiquity 0 Kramer, A. 1926. Ergebnisse der Sudsee-Ex- pedition 1908-1910 II. In Ethnographie, B. Mikronesian, Band 3, ed. G. Thile- nius, Augustine Kramer: Palau, Theil- band 3. Stoffliches st Geistige Kultur. hear ger Ham Lebar, EM. 1963. The sae culture of experiment with artificial valves of Do- max ere Netherlands Journal of Sea Ge- ology 2: ‘ Masse, WB. 1989. The Archaeology and Ecology of Fishing in the Belau Islands, JOURNAL OF ETHNOBIOLOGY tee Micronesia. Ph.D. Dissertation (Anthro- pology), Southern Illinois University, Carbondale. oe T.L. 1991. Molluscan exploita- in the late prehistoric and early his- An thropology). Southern Illinois Univer- sity, Carbondale. Moir, B.G. 1986-87. A review of tridacnid ecology and some possible implications for archaeological research. Asian Per- spectives 27:95~121. Nagaoka, L. 1994. ae recovery of Pacific Island fish remains: evidence from the Moturakan oasis Aitu- taki, Cook Islands. Asian Perspectives 33: 1-1 Nero, K hav = ne oan linkages: to- wards regional histories. Manuscript on file, library of the Palau Bureau of Arts and Culture. Osborne, D. 1979. Archaeological test ex- cavations in the Palau Islands, 1968-69. ment 1. Bes teas lee 965. fam: mily Tridacni- he Indo-Pacific. Indo-Pacific Mol- ee 1: 34 7-396. Sem G. and Y. Underhill. 1994. Implications of climate change and sea level rise for sity of Papua New Guinea, Port Mores- sien K. 1873. The Palauan — in the Pacific Ocean. Translat om German by Mark L. Berg (1982). Sieeeesian ees Research Center, University of sat, ie 2000. Shellfish peng from a nta Cruz red abal idden: re- evaluate the marine nan Siettiesis, In The 5th California Island Symposium, eds. D. ape os Mitchell, and H. Cha- ney, pp. 56 . US. Department of the Interior ten oe Ser- vice. Pacific OCS Regio Wells, EE. and C.W. Bryce. 1988. Seashells of secesad ee Western Australia yee “Ki R. 2000. The a of Shells. Chartwell Books, Ediso so e Boe. 8h pone: Journal of Ethnobiology 23(1): 125-141 Spring/Summer 2003 THE USE OF DRIFTWOOD ON THE NORTH PACIFIC COAST: AN EXAMPLE FROM SOUTHEAST ALASKA DANA LEPOFSKY, NATASHA LYONS? and MADONNA L. MOSS» * Department of Archaeology, Simon Fraser University, Burnaby, B.C. V5A 156 » Department of Anthropology, University of Oregon, Eugene, OR 97403-1218 ABSTRACT.—North Pacific Coast societies were dependent on several tree species to supply them with fuel and raw materials used in construction and in making implements with utilitarian, social, and/or ceremonial purposes. Although much of the North Pacific Coast is blanketed with forested ecosystems, usable and ac- cessible wood was not always readily available. An archaeobotanical study of Cape Addington Rockshelter in southeast Alaska (49~-CRG-188) revealed that the site occupants relied on driftwood to supplement the live trees and dead wood in the forest. Both accessible and renewable, driftwood also supplied preferred fuel wood taxa that otherwise would have been available only through trade with other groups. A review of the North Pacific Coast ethnographic literature reveals that driftwood was an important source of wood for fuel and technology for several First Nation groups. Key words: Northwest Coast First Nations, paleoethnobotany, driftwood, Cape Addington, Alaska. RESUMEN.—Las sociedades de la Costa del Pacifico Norte dependian de diversas especies arbéreas para la obtencién de combustible y materias primas usadas en la construccién y en la elaboracién de articulos con valor utilitario, social y/o ceremonial. Aunque gran parte de la costa del Pacifico Norte esta cubierta por ecosistemas forestales, no siempre era facil obtener madera aprovechable y acce- sible. Un estudio arqueobotdnico en el sureste de Alaska revelé que los habitantes del sitio usaban madera de deriva como suplemento a los arboles vivos y madera muerta obtenidos del bosque. La madera de deriva, accesible y renovable, también proveia de ciertos taxones preferidos como lefia, que de otra manera solo habrian estado disponibles a través del comercio con otros grupos. Una revision de la literatura etnografica de la Costa del Pacifico Norte revela que la madera de deriva era para varios grupos indigenas una fuente importante de combustible y material para manufacturas. RESUME.—Les sociétés de la céte du Pacifique Nord dépendaient de plusieurs espéces d’arbres pour subvenir a leurs besoins en combustible et en matiéres premiéres qu‘elles utilisaient dans la construction et pour la fabrication d usten- siles A des fins utilitaires, socials, et/ou cérémonielles. Bien qu'une grande partie de la céte du Pacifique Nord soit recouverte d’écosystémes forestiers, les arbres utilisables et d’accés facile étaient parfois difficiles 4 trouver. Une étude archéo- botanique de Cap Addington Rockshelter au Sud Est de l’Alaska (49-CRG-188) montre que les occupants du site comptaient sur le bois de gréve pour complé- menter les arbres et le bois mort de Ia forét. A la fois accessible et renouvelable, le bois de gréve procurait également des taxons de bois de feu fort prisés qui rauraient été disponibles autrement que par des échanges commerciaux avec 126 LEPOFSKY et al. Vol. 23, No. 1 d’autres groupes. Un examen de la littérature ethnographique de la céte du Pa- cifique Nord révéle que le bois de gréve était une importante ressource pour les besoins en combustible et en technologie de plusieurs groupes de Premiéres Na- tions. INTRODUCTION North Pacific Coast peoples relied on wood.! Ethnographically, large quanti- ties of wood were used to construct and heat houses, to process foods, and as raw materials for manufacturing canoes, implements, and art and ceremonial ob- jects (Turner 1998). In the archaeological record, evidence of this high demand for wood can be seen in the size and extensive remodelling of plank houses (Ames et al. 1992; Lepofsky et al. 2000), the abundance of charcoal in most sites (Sten- holm 1992), the dominance of wooden artifacts and debris at wet sites (Bernick 1991), and the profusion of standing and dead culturally modified trees with evidence of bark and wood harvesting (Mobley and Eldridge 1992; Pegg 2000; Stryd 1997; Stryd and Eldridge 1993). In addition to sheer quantity, both the ethnographic and archaeological records indicate that North Pacific Coast peoples recognized different qualities of woods for fuel and technology, and sought par- ticular species for specific tasks (Friedman 1975; Lepofsky in press; Turner 1998; Turner and Peacock in press). Despite a well-developed system of ownership and management of trees (Stewart 1984:36-37; Turner and Peacock in press), the ethnographic record sug- gests that local forests could not always supply either the amount of wood or the particular species of wood needed by North Pacific Coast groups for fuel and technology. Heavy demands on firewood meant that some groups exerted con- siderable effort to collect fuel at some distance from their villages (Boas 1935 [1969]:7; Drucker 1951:107; Jewitt 1974:96). The effort involved in harvesting fuel and its overall value is further indicated by the fact that among the Coast Salish, the lower class were often required to supply firewood along with other essential supplies to dominant elite villages (Jenness 1955:86; Suttles 1987:5). In the more sparsely forested ecosystems of the northern portion of the coast, groups such as the Dena’ina had to re-locate their villages when local supplies of firewood were depleted (Kari 1987:15). Further, when specific woods for technology could not be obtained from forests within a group’s territory, people traded for wood or the finished products (de Laguna 1972:35, 413; Drucker 1955:61; Singh 1966:27; Turner 1998:43-44; Wennerens 1985:59). The archaeological record of culturally modified trees indicates that North Pacific Coast peoples also travelled consider- able distances from their settlements to harvest wood for technological purposes (e.g., Lepofsky and Pegg 1996). In addition to the wood supply in the surrounding forests and that obtained through trade, many North Pacific Coast peoples could obtain wood as driftwood. This paper explores the importance of driftwood as an accessible and renewable source of wood for fuel and technology. Our paleoethnobotanical investigation of the Cape Addington Rockshelter in southeast Alaska revealed that a significant proportion of fuel wood used at the site did not grow locally, and likely came from driftwood. Further, the results suggest that the inhabitants of the rockshelter Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 127 ry pees) a Yakutat Dena’ina Tlingit Chugach Alutiiq BRITISH Northern limit _-—* COLUMBIA of redcedar -—" Cape Addington site 2 Ma Haida —— - a" a ae | oS ree 4 kms Northern limit __.---"~ of Douglas-fir ee Me Kwakwaka’wakw " ‘ Vancouver ~ Nuu-chah-nulth crete. a oe Nitinaht \ H i Makah ee 4 : WASHINGTON } Ny Quieleute { Twana —— a Portland j Puyallup- ¢ eer Nisqually a H Northern limit of ~~ OREGON | i several California T A a ‘ tree species i i i i FIGURE 1.—The North Pacific Coast, showing location of Cape Addington Rockshelter, northernmost range limits of some major tree species, and cultural groups for whom there is an ethnographic wor of driftwood use (see Table 3). Tree range limits come from Pojar and MacKinnon (1994). selected preferred fuel wood taxa from among the drift. A review of the ethno- graphic literature for the region indicates that the Cape Addington Rockshelter inhabitants were not unique on the North Pacific Coast, and that driftwood was an important source of wood for fuel and technology for many groups. CAPE ADDINGTON ROCKSHELTER Cape Addington Rockshelter is located on Noyes Island, one of many small islands west of Prince of Wales Island in southeast Alaska (Figure 1). The site was excavated by Moss in 1997. The cultural deposits extend over a 20 10-m area on a slope that ranges between 5.5 to 9.0 m above mean high tide within an uplifted wave-cut rockshelter. The 279 cm deep shell midden at the site has pro- duced 13 radiocarbon dates with calibrated midpoints ranging from A.D. 160 to 1420. The site deposits are composed of shell, fire-cracked rock, bone, antler, and 128 LEPOFSKY et al. Vol. 23, No. 1 charred and uncharred botanical remains. The artifact assemblage numbers about two dozen items, including five deer ulna awls or knives, bone and wood points, a deer scapula spoon, barbed bone harpoon point, mussel shell blade, and frag- ments of worked bone, wood, and shell. From the rockshelter, 8918 vertebrate remains have been studied. Thirty-two species, nine additional genera, and another nine families of animals have been identified (Moss n.d.; Moss and Losey 2003). During the earliest period of site occupation (A.D. 70 to 270), the faunal remains are dominated by halibut (Hip- poglossus stenolepis Schmidt) and deer (Odocoileus hemionus sitkensis Merriam). After A.D. 600, Pacific cod (Gadus macrocephalus Tilesius) is the most abundant taxon, indicating site occupation during March and April (see Bowers and Moss 2001 for a detailed discussion of Pacific cod). In later periods, harbor seals (Phoca vi- tulina L.) and salmon (Oncorhynchus spp.) become key resources, and offshore resources, such as northern fur seals (Callorhinus ursinus L.), Steller sea lions (Eu- metopias jubatus Schreber), and a variety of seabirds attest to use of oceanic islands, possibly including the Forrester Islands. Both faunal and macrobotanical evidence suggest site use during spring and summer (Lepofsky et al. 2001; Moss and Losey 2003). Plants were collected in the site vicinity, and some plant processing and consumption took place on site. e site occurs within the densely vegetated western hemlock-Sitka spruce (Tsuga heterophylla—Picea sitchensis) rainforest characteristic of the southern half of southeast Alaska (Viereck and Little 1972). Some small shore pine (Pinus contorta) are also found along the beach fringe today. Mountain hemlock (Tsuga mertensiana (Bong.) Carr.), western redcedar (Thuja plicata), and yellow-cedar (Chamaecyparis nootkatensis) occur within the larger region, but were not observed during recon- naissance of southwest Noyes Island. Shrubs and herbs growing in the immediate vicinity of the rockshelter include Sitka alder (Alnus crispa), cow parsnip (Hera- cleum lanatum Michx.), and devil’s club (Oplopanax horridus Smith). A variety of ferns, grasses, and other low-lying plants are found outside the shelter. The shore- line in front of the site is open to the Gulf of Alaska, but nearby headlands and offshore rocks provide some protection from the full force of Pacific swells and storms. Nonetheless, driftwood is common on the beach in front of the site today. The vegetation surrounding the site today likely differs little from that of the last 2000 years. The forest surrounding the site has not been altered significantly by modern human activity, and while the record of glacial advances and mac- rofossils do indicate climatic fluctuations on the Alaska coast in the last 2000 years (e.g., Calkin et al. 2001; Cwynar 1990; Hansen and Engstrom 1996; Mann et al. 1998), there was relatively little change in vegetation in the same period—at least at the coarse level that pollen records can detect (e.g., Gottesfeld et al. 1991; Hebda 1995; Hebda and Whitlock 1997). METHODS As part of a larger paleoethnobotanical analysis of 49-CRG-188, Lyons and Lepofsky identified 90 charred and 30 uncharred wood specimens from five bulk sediment samples and from material collected from the %4-inch screen during the excavation of one deposit (Table 1; Lepofsky et al. 2001). The bulk and %-inch Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 129 TABLE 1.—Paleoethnobotanical samples analysed from Cape Addington Rockshelter. Sample Date volume Unit Stratum Feature (approx.) _ (liters) Notes Unit 1 Layer Ila 18 A.D. 1420 1.4 18 is a hearth with ash lens, hell. Unit 2. Layer IIIc 22 A.D. 660 1.6 22 is a mussel shell-burned rock shell. Large particles of shell, rock; considerable bone Unit 2. Layer IVd N/A A.D. 500 1.9 IV is a dark matrix with less shell than III. Moist compare Unit 7 Stratum B N/A A.D. 1250 1.8 B is a dark matrix with some shell and lots of charred d wood. Unit 7 Stratum H N/A __ A.D. 800 ii H is a dark matrix with bits of shell. Unit 3 Layer IIa N/A A.D. 750 N/A _ botanical remains collected from shell. Large particles of shell, rock; considerable bone. samples were selected to represent the different areas of the excavation as well as the major deposits within the site. Two of the bulk samples were from features and the other three were from charcoal-rich layers within the shell midden. The material from the %-inch screen was from a shell-rich deposit where unique de- positional conditions resulted in the preservation of an abundance of both un- charred and charred botanical remains (Lepofsky et al. 2001). Material collected from the %4-inch screen should be representative of larger-sized botanical remains, but not smaller remains. We processed the sediment and %-inch samples in slightly different ways. The sediment samples were floated using a modified bucket flotation system which collected plant remains >0.425 mm in diameter. These remains were sorted into their constituent parts (charcoal, “seeds,” needles, non-woody tissues) with the aid of a dissecting microscope (maximum magnification 40x). We limited our analysis of the %4-inch screen material to remains >2.0 mm in size, since an un- known amount of smaller material had passed through the screen. We sorted this material directly into three gross categories: charcoal, uncharred wood, and un- charred herbaceous plant. We further subdivided the uncharred wood into two categories based on gross morphology: stem/root and wood. The “wood cate- gory consists of fragments which, based on the curvature of the growth rings, originate from large branches, large roots, or stemwood. The branch /root’ cat- egory consists of specimens whose original diameter (based on ring curvature 130 LEPOFSKY et al. Vol. 23, No. 1 and presence of bark) was <0.5 cm in diameter, and therefore likely originated from either smaller tree roots or small branches. We randomly selected for identification 15 charcoal specimens from each of the five flotation samples and 15 specimens from the charcoal, uncharred wood, and branch/root from the 4-inch screen. Based on previous experience with wood analysis on the North Pacific Coast, 15 specimens is the minimum sample size to represent the abundance of the common taxa. All specimens were >2 mm in size. Charred wood was identified using a reflected light microscope (maximum mag- nification 400X) and uncharred wood was identified with a transmitted light microscope (maximum magnification 400x). All identifications were made by comparing against specimens in the wood reference collection housed in Le- pofsky’s paleoethnobotany laboratory at the Department of Archaeology, Simon Fraser University. RESULTS The Cape Addington Rockshelter flotation and screen samples yielded seeds, needles, buds, wood, and non-woody tissues representing 24 plant taxa (Lepofsky et al. 2001). Of these, uncharred and charred wood taxa dominate the assemblage both in diversity and abundance. The woods are comprised of ten tree species, the majority of which are conifers (Table 2). Despite the relatively small sample size, there is patterning in the distribution of charred and uncharred wood taxa. Of the charred woods, three species, spruce (Picea sitchensis), redcedar (Thuja plicata), and Douglas-fir (Psuedotsuga menziesii) were collected most often and in greatest quantity (Figure 2). The remaining charcoal species were collected either infrequently, and/or were not harvested in abundance. Our sample of uncharred woods is too limited to discern definitive patterns, but the percent abundances of some taxa do indicate distinct formation processes for the uncharred wood and charred wood taxa (e.g., yellow-cedar char- coal versus uncharred branch/root; hemlock charcoal versus uncharred wood; Table 2). This non-random distribution reflects the deliberate collection of particular wood species for specific tasks. We assume that the charred wood recovered was collected for fuel, and/or was scrap generated from other tasks, which was later burned. The preference at the site was clearly for spruce, redcedar, and Douglas- fir fuel; each of these taxa is widely recognized among Northwest Coast peoples as an excellent, all-purpose fuel wood (Turner 1998). The uncharred wood from the %-inch screen sample may have been brought into the site to manufacture artifacts, or may have also been stockpiled for future fuel consumption (e.g., Reger and Campbell 1986). The taxonomic abundance of the different woody taxa reflects both local abundance and cultural preference. In charred and uncharred archaeological sam- ples as well as in the local environment today, conifers are more common than hardwoods. However, while both spruce and western hemlock (Tsuga heterophylla) are common in local forests today, and were likely so in the past as well, only spruce is ubiquitous and abundant in the archaeobotanical assemblage (Table 2; Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 13] TABLE 2.—Wood taxa recovered from Cape Addington Rockshelter. N (% Abundance)! Scientific name UC uC (common name) t W __ B/R _ Distribution Coniferous Trees Abies spp. (true fir) 4(4) — — could be either A. amabilis or A. lasiocarpa; both grow in rare, lo- cal stands in southeast Alaska; not observed on Pha s Island, but may be prese Chamaecyparis nootkatensis D. 4(4) 1(3) 5(17) present in ies Alaska i in Don (yellow-cedar) mixed coniferous forests from Picea cf. sitchensis (Bong.) 27 (30) 6(20) 3(10) abundant in the forests of south- Carr.’ (Sitka spruce) east Alaska from low to mid el- store inl abundant in site vicini- today ty to Pinus sch yarn Dougl.? 39D) = so common in 1 shoreline me pre- (sho sent in site vicinity today. Pinus/Picen 2(2) — — oe Psuedotsuga menziesii 210) — — common component of central (Mirbel) Franco (Douglas- and southern B.C. coastal for- fir) te present in southeast Alaska from low to mid elevations south of Frederick Sound; nea in imme- diate site vicinity tod occur in more x eens habitats on the islan 3(3) 5(17) 2(7) common tree of coastal forests; common in site vicinity today. Thuja plicata Donn (western 11(12) 103) — edcedar Tsuga cf. heterophylla (Raf.) Sarg.” (western hemlock) Tsuga/Chamaecyparis 14(16) 2077) 300) — Unidentified conifer 7(8) — 2(7) — Deciduous Trees ee e Al f. 2 (Regel 3(3) — — common along shore! of south- Rydb. vo bbe) 0 east Alaska and near site today. Pyrus fusca Raf. (Pacific 10) — — grows in mixed and pure thickets crabapple) or as a slow-growing small tree in low to mid elevations of southeast Alaska. None ob- rved in the site vicinity today. i 1 _— several species of willow grow on septs yee oe the outer islands of southeast aes in site vicinity today. Unidentified deciduous 1 (1) ro 1.C = charred, UC W = uncharred stemwood; UC B/R = uncharred peewee aes aun N = identifi i five flotation samples and the one ‘-in 90 (15 identified specimens fom nee “7 sit bates rom each cf the stemwood and branch/root eco’ from %4-inch screen sam * Species identifications of these genera are eet on phytogeography rather than minute anatomy. 132 LEPOFSKY et al. Vol. 23, No. 1 4% # of specimens 1% 4% 3% 3% 3% 1% i «“ dd oP Wood taxa G FIGURE 2.—The number of times different charcoal taxa were identified in each of six samples (15 identifications per sample). For each taxon, each bar represents one sample. e numbers on top of the bars represent the average percent abundance for that taxon. Only secure identifications are illustrated, but unidentifiable specimens are included in the calculations of percent abundance. Figure 2). This may be because unlike spruce, hemlock was not a preferred fuel wood among many Northwest Coast groups (Turner 1998). In contrast, Douglas-fir does not grow locally, yet is common in the archaeo- botanical record. The northern limit of Douglas-fir habitat occurs at about 53° north latitude in central British Columbia, almost 380 km to the south of Noyes Island (Figure 1; Pojar and MacKinnon 1994:17). Distributions of Douglas-fir var- ied in the past, but never extended north of British Columbia’s central coast (Mat- hewes and Rouse 1975; Hebda 1995; Hebda and Whitlock 1997). The Douglas-fir wood in the 49-CRG-188 assemblage was undoubtedly col- lected as drift from Noyes Island beaches. Although it is possible that the Cape Addington Rockshelter occupants travelled south to harvest Douglas-fir or that they received it through trade, both these scenarios seem unlikely, especially given that the archaeological site was not occupied year-round. The Davidson Current, which regularly transports drift from southern California northward along the Alaska coast during the winter (Thomson 1981:231) likely transported the Doug- las-fir wood to beaches in the site vicinity. Other taxa within the archaeobotanical assemblage also may have been col- lected as driftwood, but this is more difficult to demonstrate based on plant ge- ography. In particular, the abundance of redcedar in the archaeobotanical assem- blage is greater than its apparent abundance in the local forest (Table 2). Although we cannot rule out the possibility that site inhabitants went to some effort to transport redcedar back to the rockshelter from elsewhere on the island, or from Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 133 another island, a more plausible scenario is that they harvested the wood as drift from the beach in front of the rockshelter. While we cannot quantify how much of the total wood used at the rockshelter was collected as driftwood, it appears to be significant. Conservatively, based on Douglas-fir alone, driftwood accounts for 8% of the charred and uncharred wood recovered at Cape Addington (n = 120). If redcedar is included in the total, at least 18% of the total assemblage may have originated from drift (Table 2). The actual contribution of driftwood could be much higher. The relative abundance of Douglas-fir and redcedar in the archaeobotanical assemblage suggests that the site inhabitants deliberately selected these woods from among the drift. Although we cannot quantify how common Douglas-fir and redcedar were in the ancient driftwood population, we expect that hemlock was at least as available as the other two species. Yet, whereas hemlock was not collected in abundance as a fuel wood, Douglas-fir and redcedar were clearly sought after. In the case of Douglas-fir, it was only found charred in the deposits (Table 2), suggesting it was selected only for fuel, and not for other purposes. DRIFTWOOD USE ON THE NORTH PACIFIC COAST Even though few North Pacific ethnographers discuss the use of woods much at all, our literature review indicates that groups throughout the region relied on driftwood for fuel and technology (Table 3; Figure 1). For the Makah and Kwak- wa’ka’wakw, for example, driftwood was the most common source of fuel, while only occasional use is mentioned for other groups. For some groups, access to driftwood meant a supply of valued woods that were not locally available or sufficiently abundant, and could otherwise be acquired only through trade. Such was the case of redcedar among the Alutiiq (Wennerens 1985:59) and Makah (Singh 1966:27), and redcedar and yellow-cedar among the Yakutat Tlingit (de Laguna 1972:43; Drucker 1955:61). Several sources mention preferred uses for par- ticular taxa of drift, indicating that driftwood, like wood found in the forest, was selectively harvested for specific uses. S The ethnographic sources are relatively silent as to specifically who collected driftwood. Slaves and other lower class people were often employed to gather firewood in general (Boas 1935 [1969]:7; Jewitt 1974:96; Oberg 1973:79; Ruyle 1973: 611; Singh 1966:54; Suttles 1987:5), which likely included driftwood as well as wood from the forest. Only Drucker (1951:107) had more specific observations about the harvesting of driftwood. He noted, “[bJoth [Nuu-chah-nulth] men and women got wood, although women’s wood gathering consisted chiefly of picking up small driftwood along the beaches. Men got ‘big wood’—big lengths of drift- wood. .. .”” There is some indication that driftwood was a resource that could be claimed and owned. This is clearly illustrated in the case of a large spruce log that drifted ashore in Nitinaht territory in 1930s that was immediately claimed, and then cut up for firewood (Turner et al. 1983:fig. 16). Since slaves belonged to titleholders or to the households of titleholders (Donald 1997), and slaves often collected fire- wood, we surmise that driftwood collected for fuel was considered the private property of an individual or the common property of a household. Some southern 134 LEPOFSKY et al. Vol. 23,. No. 1 TABLE 3.—Aboriginal use of driftwood on the North Pacific Coast.' Group Use of driftwood Reference Puyallup-Nisqually Driftwood piled to make screen to Smith 1940:286 direct heat from cooking fires on bea Twana “Bark was a favored sean’ fuel, but Elmendorf and Kroeber any kind of wood, incl 1992:220 beach drift, was so pe Makah Drift logs were split into boards or Swan 1868:4 made into canoes Quileute and Makah Driftwood was the principal source Singh 1966:23 Nitinaht Nuu-chah-nulth Nuu-chah-nulth Kwakwaka’wakw Kwakwaka’wakw Kwakwaka’wakw Haida Northern and Kai- Yakutat Tlingit Chugach Alutiiq Chugach Alutiiq of firewood; redcedar drift logs were used for canoes, especially and cut up for Drift was collected for fuel Bamboo sometimes drifted ashore Drift logs for firewood Drift logs were most common source of fuel; soft meciageen used as hearth with fire Driftwood was used ay process el- derberries and to singe the hair off harbor seals P. menziesii, Acer, Betula, and other woods sometimes collected as drift Drift logs were lashed together to make rafts to migrate to Alaska Drift logs used for pyre to cremate dead Driftwood used for se on beach to process spruce ro Thuja and noc found as drift ae available at Nuchek only as Sound, used for paddles, quivers; yellow-cedar sought after on Low- er Kenai Peninsula Driftwood, small trees, and broken pieces of wood were preferred e from Tsuga and Thuja drift. eatin trichocarpa rain used to heat homes and ste baths; preferred for part 5 fish Turner et al. 1983:fig. 16 Drucker 1951:107; Arima 1983:62 N.J. Turner? Wolcott 1967:23; Rohner and Rohner 1970:19 Boas 1909:407 Boas 1921:256, 451. N.J. Turner? Blackman 1981:75 Swan 1876:9 Blackman 1980:85 de Laguna 1972:413 Wennerens 1985:59, 78 Russell 1991b: 6, 19, 20, 21 Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 135 TABLE 3.—Continued. Group Use of driftwood Reference Dena’ina Driftwood and windfalls were Kari 1987:15, 39 sought after for firewood; Thuja col- lected as driftwood Dena’ina Fire drills made of Thuja drift Osgood 1966:108 Dena’ina, Aleut Populus trichocarpa drift preferred for Russell 1991a: 9, 12, 13 smoking fish because is ‘clean’, without sap, contains salt, and slow- burning; for steambath switches, whittling Thuja and Chamaecyparis used as building material due to resistance to rot; dugout boats for children. Driftwood used as poles for fish drying racks ‘The aboriginal groups are ordered from south to north along the coast. See Figure 1 for locations of groups * Dr. Nancy J. Turner, ethnobotanist, University of Victoria, email correspondence with D. Lepofsky, 1999. Northwest Coast peoples divided beaches into family-owned territories to claim ownership of beached whales (e.g., Hajda 1990), and it is possible that driftwood was also incorporated into this system of ownership. At the north end of the Pacific Coast, where forest wood is a much scarcer commodity, several groups developed systems of ownership of driftwood. For instance, among the Alutiig, when redcedar drift was found it was cut up and shared (Russell 1991b:19), and among the Koniag of Kodiak Island, ownership of driftwood was often claimed while the logs were still out at sea (Adams 1998). Among the Aleuts of the Aleutian Islands, wars were started when driftwood was taken from another village's territory (Veniaminov 1984, cited in Hoffman 1999:159), and an excavation of an Aleut village suggests that driftwood was shared within a household group (Hoffman 1999:159). The people of Nunivak Island marked ownership of piled driftwood by placing a large log in an upright Position in the pile (Fienup-Riordan 2000:62). DISCUSSION From the time of initial colonization of the North Pacific Coast, driftwood prob- ably has been an important source of wood for people of the region (Ames and Maschner 1999:61). Not only was driftwood a readily accessible and renewable source of wood, but it also provided preferred taxa that would not otherwise have been available except through trade. Both the Cape Addington Rockshelter case study and our review of the ethnographic record indicate that North Pacific Coast peoples se- lectively harvested driftwood for particular purposes. Like living trees or dead wood found in a resource territory, driftwood was integrated into a system of ownership and management typical of other valued resources, by at least some groups. Although we cannot determine the amount of driftwood available on Pacific Coast beaches prior to industrial logging, early accounts indicate that driftwood 136 LEPOFSKY et al. Vol. 23, No. 1 was plentiful. For instance, in 1777, when Captain Cook arrived to Nootka Sound on the west coast of Vancouver Island, he was pleased that many drift logs of all shapes and sizes were available to repair various parts of the Resolution (Gough 1978:10). Similarly, Dorsey (1898:5), travelling in Dixon Entrance in the late 1800s, commented that many beaches were “piled high with drift, often to a height of sixty feet or more.’’ Likewise, early photographs of the coast (e.g., Curtis’s 1912 photograph of “[t]he mouth of the Quinault River’) also show abundant drift- wood on the beaches (Curtis 1913). Despite the overall availability of driftwood on the coast, not all people would have had equal access to this valuable resource. Because of the subtleties of local ocean currents, only certain beaches accumulate drift.* Undoubtedly, North Pacific Coast peoples had knowledge of beaches where drift tended to accumulate and these were valuable locations for collecting wood. Even though we cannot recon- struct the relative abundance of species that were available as driftwood prior to industrial logging, current patterns of deposition of drift on beaches should reflect the overall spatial availability of driftwood in the past. In addition to the availability of drift on local beaches, the role of driftwood to any individual group was undoubtedly influenced by regional variation in forest species composition and overall forest productivity. Importantly, the num- ber of tree species declines northward along the coast (Alaback 1996), and the distribution of highly valued species such as Douglas-fir, redcedar, and yellow- cedar becomes increasingly restricted towards the northern ends of their geo- graphic ranges (Figure 1; Pojar and MacKinnon 1994). Where redcedar and yel- low-cedar do grow in Alaska, they often are small and stunted (Pojar and MacKinnon 1994:17), and thus are unsuitable for canoes or large structures. In the northernmost part of the region (north of 58° north latitude), forests not only contain fewer tree species in general, but less of the landscape is forested, and those forests are less productive overall (Alaback and Pojar 1997). Based on these vegetation patterns, it seems quite clear that driftwood was more important to people on the extreme northern part of the North Pacific Coast. However, the ethnographic data suggest that geographical variation alone does not account for the relative role of driftwood among North Pacific Coast groups (Table 3). A final factor in determining the importance of driftwood was settlement type. Large, long-term, permanent settlements would have placed especially high demands on local forests, both for fuel and technological needs. This is clearly illustrated for the Makah, Nuu-chah-nulth, and Kwakwaka’wakw, who are located in some of the most productive coastal forests, yet according to the ethnographic record, driftwood was their principal source of fuel (Table 3). Driftwood may also have played a relatively more important role in some small and/or low-status settlements since these communities are less likely to be associated with the well- developed regional trade systems that supplied non-local woods. The North Pacific Coast is not unique in northwestern North America with respect to the value of driftwood. Further north in the Western Arctic, where the landscape is sparsely treed or treeless, and the cold winters result in especially heavy demands on fuel, the importance of driftwood permeates the religious, social, and economic systems of most groups (Adams 1998; Barker 1993; Fienup- Riordan 1996, 2000; Oswalt 1957:26). To the east, Plateau groups located their Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 137 villages and camps along floodplains that accumulated drift, and used the wood both for fuel and for special technological tel aa (Miller 1998:258; Rhode 1986; Smith 2000:7.8; Stenholm 1985; Teit 1930:223). Recognizing the value of driftwood has important implications for our un- derstanding of how people of the North Pacific Coast negotiated resource diver- sity across the landscape. Even in a heavily forested region, an archaeologist can- not assume that charcoal represented in an archaeobotanical assemblage repre- sents locally growing tree species. North Pacific Coast peoples clearly identified and understood the different properties of various wood taxa, even in the form of drift. Identifying charred remains to genus or species should become a routine archaeological practice, especially for wood samples from coastal sites submitted for radiocarbon dating. Samples of driftwood may be particularly susceptible to the “old wood,” or inbuilt age problem, and the magnitude of this effect can be species specific (Gavin 2001; Schiffer 1986) NOTES ‘For the purpose of this paper, we define the North Pacific Coast as the region extending from Cook Inlet to the Oregon-California border. * Dr. Curt Ebbesmeyer, Oceanographer, Evans-Hamilton; email correspondence with D. Le- pofsky, 1999. * Dr. Curt Ebbesmeyer, Oceanographer, Evans-Hamilton; email correspondence with D. Le- pofsky, 1999, ACKNOWLEDGMENTS Excavations at 49-CRG-188 were supported by grants to Moss from the National Science Foundation, Tongass National Forest, Craig Ranger District, University of Oregon, and the National Marine Fisheries Service, Auke Bay Laboratory. Moss would like to thank the field team: Jon Erlandson, Rob Losey, Mark Tveskov, Betty Johnston, and Erik Moss Erlandson, all of whom enjoyed driftwood campfires at night. Many thanks to Curt Ebbesmeyer for sharing his extensive knowledge of drift, to Nancy Turner for information on the use of drift by First Nations of the Northwest Coast, and to Karen Adams for sharing her unpublished work on driftwood use in Alaska. We also thank Jesse Morin for his help with the search through ethnographic sources for information on wood use by Northwest Coast peoples. Many thanks to Kathia Guay and Remi Brisson for the translation of the abstract into French and to Andres Valle-Domenech and Marcela Olguin-Algarez (once again) for the translation into Spanish. REFERENCES CITED Adams, Karen vs 1998. The spirit of drift- High-Latitude Rainforests and ‘rcohang wood: Alaskan driftwood use and ar- Ecosystems of the West Coast of the . Minecitirvers Sdipiitilnts Paper pre- ricas, eds. Richard B. Lawford, Paul B sented at the 63rd Annual Meeting of Alaback, i. sregtnil pp. 105-133. er-Verlag, ork i ican Archaeology, Spri ew oma Was v8 rene , e — Paul and Jim Pojar. 1997. Vegeta- Alaback, Paul B. 1996. Biodiversity patterns in relation to climate: the coastal tem- Peishuvets of Home, vee Peter he perate rainforests of North America. In Schoonmaker, Bettina von Hagen, a 138 LEPOFSKY et al. Edward C. Wolf, pe. 69-87. Island Press, Washington, D.C. Ames, Kenneth M., Doria E Raetz, $53 C ilton, and Christi — of Field st 19:275-290. mes, Kenneth M. Herbert D.G. Maschner. 1999. one - the Northwest Coast. Thames and Hudson Ltd., Lon- on. Arima, E.Y. 1983. The West Coast People. The Nootka of Vancouver Island and Cape Flat- tery. British Columbia Provincial Mu- seum, Special Publication, no. 6, Victo- a. Barker, James H. 1993. Always Getting Ready, Upterrlainarluta. Yup’ik Eskimo Subsis- tence in Southwest Alaska. University of Columbia Heritage Trust, Victoria. Woman. University of Washington Press, Seattle. —. 1981. Window on the Past: The Pho- tographic Ethnohistory of the Northern and Kaigani Haida. National Museum of Man Mercury Series, Canadian Ethnol- ogy Service, Paper no. _ National Mu- seums of Canada, Ott Boas, Franz. 1921. hnoloes of the Kwa- kiutl, Based on Data Collected by George Hunt. Bureau of American Eth- nology, Annual Report 35, pts 1-2. In 35th Annual Report of the Bureau of Amer- ican Ethnology for the Years 1913-1914, pp. 43-1481. Washington, D.C. —. 1909 [1975]. The Kwakiutl of Van- couver Island. Publications of the Jesup North Pacific Expedition 5 (2):301-522. Memoirs of the American Museum of Nat- ural History 8 (2). Reprinted, AMS Press, New York. . 1935 [1969]. Kwakiutl Culture as Re- flected in Mythology. Kraus Reprint, New York. Bowers, Peter M. and Madonna L. Moss. 2001. The North Point Wet Site and the People and Wildlife in Northern North America: Essays in Honor of R. Dale Guth- Vol. 23, No. 1 rie, pp. 159-177. British ee Report International Series 9 chaeopress, Oxford. Calkin, Parker E., Gregory C. Wiles, and David J. Barclay. 2001. Holocene coastal oneal of Alaska. Quaternary Science Review. 61. Curtis, Edward S. 1913 [1970]. The North merican Indian: Being a Series of Volumes Picturing and Describing the Indians of the United States, the Dominion of Canada, and Alaska, vol. 19, ed. Frederick W. Hodge. Plimpton Press, Norwood, Massachu- setts. Johnson Reprint, New York. Cwynar, L.C. 1990. A late Quaternary veg- etation history from Lily Lake, Chilkat Peninsula, southeast Alaska. Canadian Journal of Botany 68:1106—1112. de Laguna, Frederica. 1972. Under Mount Saint Elias: The History and Culture of the Yakutat Tlingit, Pt. 1. Smithsonian Contributions to Anthropology, Vol. 7. aie nie sonian Institute Press, Washing- Donald Bee 1997. Aboriginal eres on Northwest Coast of North America. University of Selim Press, Berkeley. Dorsey, George A. 1898. A cruise among Haida and Tlingit villages about Dix- on’s Entrance. Reprinted in Appleton’s Popular Science Monthly from a lecture delivered in the Field Columbian Mu- seum, Nov. 6, 1897. Drucker, Philip. 1951. The Northern and Cen- tral Nootkan Tribes. Smithsonian Insti- tution Bureau of American Ethnology, Bulletin 144. Smithsonian Institution, Washington, D.C. . 1955. Indians of the Northwest Coast. The Natural History Press, Garden City, New York. Elmendorf, William W. and A.L. Kroeber. The Structure of Twana Culture. With Comparative Notes on the Structure of Yurok Culture. Washington State Uni- versity Press, Pullman, Washington. Fienup-Riordan, Ann. 1996. The Living Tra- dition of Yup’ik Masks. University of Washington Press, Seattle. . (editor) 2000. Where the Echo Began and other Oral Traditions from Southwest- orded by Hans Himmelheber. ern Alaska rec University of Alaska Press, Fairbanks Friedman, Janet. 1975. The Prehistoric Uses of Wood at the Ozette Archaeological Site. Ph.D. dissertation, Department of Spring/Summer 2003 aicrnaheen at Pecarenige sn State Uni- versity, Pullm. Gavin, Daniel G. 3001. Estimation of inbuilt age in radiocarbon ages of soil charcoal for fire history studies. Radiocarbon 43: 27. Gottesteld, Alan S., Rolf W. Mathewes, an MJ. Gottesfeld. 1991. Holocene ft : dian Journal of Earth Sciences 28:1583- 1593. Gough, Barry M. 1978. Nootka Sound in James Cook’s Pacific world. In Nu-tka— Captain Cook and the Spanish Explorers on the Coast, eds. Barbara S. Efrat and WJ. Langlois, pp. 1-32. Sound Heritage vol. VIL, no. 1. Province of British Columbia, Victoria, B.C. Hajda, rena 1990. Southwestern es Salish. In The Handbook of North A can Indians, vol. 7, The Northwest ee ed. Wayne Suttles, pp. 503-517. Smith- sonian Institution Press, Washington, A Hansen, B.C.S. and D.R. Engstrom 1996. Vegetation history of Pleasant Island, southeastern Alaska, since 13,000 yr BP. 175. J. 1995. British Columbia vegetation and climate history with fo- cus on 6KA BP. Géographie Physique et a wagon Hebda, Richard J. and Cathy Whitlock. 1997, ironmental history. In Ti Rain paca of Home: Profile % a North American Bioregion, eds. Peter Schoonmaker, Bettina von Hagen an Edward C. Wolf, pp. 227-254. Island Press, Washington, D.C. Hoffman, Brian W. 1999. — Village: household archaeology on Unimak Is- land, Alaska. Journal of Field Archaeology 5. The Faith of a Jewitt, John R. 1974. The Adventures and Suf- ferings of John J. Jewitt, Captive Among the Nootka 1803-1805, ed. Derek G. Smith. McClelland and Stewart Ltd, Toronto. Kari, Priscilla Russell 1987. Tanaina Plant- — Dena’ina K’et’una. National Park rvice, Anchorage, Alaska. paar Dana. In press. The Northwest. JOURNAL OF ETHNOBIOLOGY 139 In Plants and People in Ancient North America, vol. 2, ed. Paul Minnis. Smith- sonian Institution Press, Washington, Dz Lepofsky, Dana, Michael Blake, Douglas own, Sandra Morrison, Nicole Oakes, and Natasha Lyons. 2000. The archae- ology of the Scowlitz site, southwestern British Columbia. a of Field Ar- chaeology 27(4):391—-4 Lepofsky, Dana, ‘nies L. Moss, and atasha Lyons. 2001. The unrealized potential of paleoethnobotany in the ar- chaeology of northwestern North America: Perspectives from Cape Ad- dington, Alaska. Arctic Anthropology 38(1):48-59. Lepofsky, Dana and Brian Pegg. 1996. Ar- chaeological and ethnographic assess- ment. In The Kowesa Watershed As- sessment: Summary Report, pp. 3 Interrain Pacific, Portland, Oregon. Mann, Daniel H., Aron L. Crowell, Thomas D. Hamilton, and Bruce P. Finney. 1998. Holocene geologic and climatic history around the Gulf of Alaska. Arctic An- thropology 35:112-131. Mathewes, Rolf W. andd Glen E. Rouse 1975. Palynology and palaeoecology of postglacial sediments form the lower Fraser River Canyon of British Colum- bia. Canadian Journal of Earth Sciences 12: 5-756. Miller, Jay. 1998. Middle Columbia River Salishan. In The Handbook of North American Indians, vol. 12, The Plateau, ed. Wayne Suttles, pp. 253-270. Smithso- nian Institution Press, Washington, D.C. ae Charles M. and Morley Eldridge 2. Culturally modified trees in the Pacific Northwest. Arctic Anthropology 29:91-110. Moss, Madonna L. n.d. Archaeological In- ged Seacoast on the Outer Prince of Wales Archipelago, Alaska. Calamity of . Anthropological Papers. (in preparation) Moss, psec L. and Robert J. Losey. 2003. Resource Use on the Northwest Case Studies from Southeast aska and Oregon. The Ae of the 17th International Abashiri Sympo- sium, Abashiri, Japan. (in press). Oberg, Kalvero. 1973. The Social Economy of Coast: 140 LEPOFSKY et al. the Tlingit Indians. University of Wash- ington Press, Seattle. Osgood, Cornelius. 1966. The Ethnography of the Tanaina. Yale University Publications in pr tame gy 16. Yale University, New Haven, Connecticut. Oswalt, Wendell H. 1957. A Western Eski- - Reger, Douglas R. and Chris R. Campbell. 1986. Early Historic Use of Sakie Bay Cave, CRG-230. Alaska Archaeological Sur- — State of Alaska, 8 tapheert of ral Resources, Anchora Rhode I David. 1986. Ajctialsobtardaed inves- ons in the Wells Reservoir Archae- snares Project, Eastern Washington. In The Wells Reservoir Project, vol. 1: Sum- ological Survey, Archaeology Report 86- 6, Ellensburg, Washington. Rohner, Ronald P. and Evelyn C. Rohner. 0. The Kwakiutl Indians of British Co- pie Holt, Rinehart and Winston, New York. Russell, Priscilla N. 1991a. Ninilchik Plant- lore. An Ethnobotany of the Ninilchik Dena’ina, Aleut, and Russian Peoples. Ms. on file with author and Pratt Mu- seum, Homer, Alaska. - 1991b. English Bay and Port Graham ch dation, and Alaska Native Plant Society, Anchorage. Ruyle, ey E: °1973. saith oe a and stratification on the west Coast: the ethnoenergetics are an neater ent eng system. Current An- thropolo 603-631. Schiffer, Mehael B. 1986. Radiocarbon dat- in “ the Bas poo problem: the f the Hohokam chronology. Jour- nal of ‘Archiaicloilin Shiote 13:13-30. Singh, Ram Raj 1966. Aboriginal Economic System of the Olympic Peninsula ians, Western ington. The Sacra- mento Anthropological Seeley: Sacra- Vol. 23, No. 1 mento State College, Sacramento, Cali- ornia. Smith, Alan H. 2000. Kalispel ethnography and ethnohistory. In The Calispell Valley Archaeology Final Report, vol. 2. History Andrefsky, G.C. Burtchard, K.M. Pics ler, S.R. Samuels, PH. Sanders, and A. Thoms, pp. 7.1-7.151. Center for North- west Anthropology, Department of An- sie ine Washington State Universi- ullm Smith, Marion W. 1940. The Puyallup-Nis- qually. Columbia University Press, New York ork. Stenholm, Nancy A. 1985. Botanical assem- blage. In Summary of Results, Chief Joseph Dan Cultural Resources Project, Washing- ton, ed. Sarah K. Campbell, pp. 421-453. Office of Public Archaeology, Universi- ty of Washington, Seattle. . 1992. Fort Rock Basin botanical analysis. Paper presented at the North- west Anthropological Conference, Bur- aby, B.C. Stewart, Hilary. 1984. cot Douglas and McIntyre, Vancouv d, Arnoud. 1997. “Culturally Modified Trees of British Columbia. British Colum- try of Forests, Nanaimo, B.C the Meares Island studies. B.C. Studies 99:184-234 Suttles, Wayne. gore Private knowledge, morality, and social class amon e Coast Salish, In aie Salish Essays, ed. Wayne Suttles, pp. 3-14. Talonbooks, Vancouver and University of Washing- ton Press, Seattle. Swan, James G. 1868. Indians of Cape Flat- tery at the Entrance of the Strait of Fuca, Washington Territory. Smithsonian Contribution to Knowledge 16(8):1-106. Washington, D.C. . 1876. The Haidah Indians of Queen Charlottes Islands, British Columbia. igns, etc. Smithsonia Contributions to Knowledge 21(4):1-18. Washington, D.C. Teit, James A. 1930. The Salishan Tribes of the Western Plateaus. In 45th Annual Re- port of the Bureau of American Ethnology for the Years 1927-1928. ed. Franz Boas, pp. 23-396. Washington, D.C. Spring/Summer 2003 Thomson, Richard. 1981. Oceanography of the British Columbia Coast. Canadian Special Publication of Fisheries and Aquatic eae of Fisheries and Oceans, Otta Turner, Nancy J. First Peoples in British Columbia. Royal British Columbia Museum versity of British Columbia Press in col- laboration with the Royal British Co- lumbia Museum, Vancouver. Turner, Nancy J. and Sandra L. Peacock. In press. Solving the perennial paradox: the ethnobotanical evidence for plant resource management on the North- west Coast. In Keeping It Living: na nous Plant Management on the Northwe: Coast, eds. Douglas Deur and Nanc Turner. University of Washington Press, Seattle. 1998, Plant Technology of JOURNAL OF ETHNOBIOLOGY 141 Turner, Nancy J., John Thomas, Barry F Carlson, and Robert T. Ogilvie. 1983. Nitinaht — of and. Occasional Paper the British Columbia Provincial vee um, no. 24, British Columbia Provincial Museum, Victoria Viereck, Leslie A. and Elbert L. Little, Jr. 1972. Alaska Trees and Shrubs. Agricul- tural Handbook No. 410. Forest Service, United States Department of Agricul- hington, D.C. Wennerens, Alex J. 1985. Traditional Plant Chugach Natives Around Prince William Sound and on the pe Kenai Peninsula, Alaska. M.A. Thesis, University of Alaska, Anchorage. Wolcott, Harry FE 1967. A Kunkiutl Village and School. Holt, Rinehart and Winston, New York. ali essa) < ‘ Re NR eieey S, oie sehen) Journal of Ethnobiology 23(1): 143-154 Spring/Summer 2003 WOOD REMAINS FROM ANDEAN ARGENTINA: THE USE OF Prosopis SP. L. IN HUT CONSTRUCTION AYLEN CAPPARELLI,? NORA ZAGORODNY,” and BARBARA BALESTA? “Departamento Cientifico de Arqueologia, Museo de Ciencias Naturales de La Plata, Paseo del Bosque s/n. (1900), La Plata, Buenos Aires, Argentina barcap@way.com.ar "Laboratorio de Anilisis Cerdmico, Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata, Paseo del Bosque s/n(1900), La Plata, Buenos Aires, Argentina norazago@way.com.ar, bbalesta@netverk.com.ar ABSTRACT.—A desiccated fragmented wooden post, part of the structure of a hut, from the archaeological site of Carrizal de Azampay, Department of Belén, Province of Catamarca, Argentina, was identified and interpreted. The material was broken manually for the analysis. Transverse, radial, and tangential sections were examined with a stereoscopic and incident light microscope and SEM. The remains were identified as either Prosopis flexuosa DC or P chilensis (Mol.) L. Stuntz, both of which are common species in the study area. This paper confirms the prehistoric use of this genus in construction and analyzes the function of the post as a part of the hut structure. In addition, recent changes in the utilization of both cultivated and native tree species are discussed. Key words: Andean region, Argentina, archaeological wood, Prosopis. RESUMEN.—E] presente trabajo se basa en la identificacién e interpretacién de un poste de madera, fragmentado y desecado, perteneciente a la estructura de una vivienda, procedente del sitio arqueolégico Carrizal de Azampay localizado en el actual Departamento de Belén, Provincia de Catamarca, Argentina. E] ma- terial fue fracturado manualmente en laboratorio. Se examinaron las secciones transversal, longitudinal tangencial y longitudinal radial del mismo con Micros- copio Estereoscépico, de Luz Incidente y Electrénico de Barrido. Los fragmentos fueron identificados como Prosopis flexuosa DC o bien Prosopis chilensis (Mol.) L. Stuntz, especies comunes en el Area de estudio. El trabajo confirma el uso pre- hist6rico de Prosopis como recurso maderero y analiza la funci6n del poste dentro de la estructura de la vivienda. Asimismo se discuten, a partir de observaciones etnograficas, procesos modernos de cambio en el modo de uso de este recurso a partir de la incorporacién de especies cultivadas. RESUME.—Ce rapport analyse et interpréte un fragment desseché de pieu pro- venant de la structure d’une hutte et trouvé sur le site archéologique de Carrizal de Azampay, département de Belén, province de Catamarca, Argentine. Les mor- ceaux de bois ont été brisés a la main pour I’analyse. Les sections transversales, radiales, et tangentielles ont été examinées aux trois microscopes suivants: sté- réoscopique, 4 lumiére incidente, et éléctronique a balayage. Nous avangons deux hypothéses pour I’identification de l’espéce de bois: il s‘agit ou bien du Prosopis uosa DC. ou bien du P chilensis (Mol.) L. Stuntz. Ces deux espéces d’arbres sont communes dans la région. Cette étude confirme l'utilisation préhistorique de 144 CAPPARELLI et al. Vol. 23, No. 1 ce genre pour la construction et analyse la fonction du pieu dans la structure de la hutte. Cet article aborde également les récents changements survenus dans l'utilisation d’espéces d’arbres cultivées et indigénes. INTRODUCTION The interpretation of wood remains from archaeological sites is important for understanding past plant uses and plant-human evolutionary processes (see Smart and Hoffman 1988), as well as for understanding mechanisms of prehistoric deforestation (Willcox 1974). In Argentina, studies of archaeological wood are rare; however, they are increasing (see for example Garibotti 1998; Heyne 1992; Rodriguez 2000; Roig and Barcena 1997) The main goal of this work is the identification and interpretation of a frag- mented desiccated wooden post from the archaeological site of Carrizal de Azam- pay (27°19’ south latitude and 67°02’ west longitude) and the discussion of the changes resulting from the introduction of some exotic woody species in the area. Regional Setting.—The site of Carrizal de Azampay is located a few kilometers from the modern village of Azampay, in the Department of Belén, Province of Catamarca, Argentina, at an elevation of 2000 m asl. The region is semiarid and the climate subtropical. The mean annual temperature is 18°C, with a range be- tween about 9° and 25°C. The average annual precipitation is about 300 mm, with rainfall occurring mainly in summer. Phytogeographically, the area belongs to the Monte Province, Chaquefio Domain, Neotropical Region (Morlans 1985). The main plant communities are: 1) shrub steppes of Larrea cuneifolia Cav., Plectrocarpa rou- gessii Desc. O’ Don. et Lourt., P tetracantha Gillies ex Hook & Arn, Bulnesia schic- kendantzii Hieron. ex Griseb. and Senna rigida (Hieron.) H. S. Irwin & Barneby, among others, on alluvial terraces and colluvial piedmonts found up to an ap- proximate elevation of 2200 m asl; 2) isolated stretches of open forests of Prosopis flexuosa, Prosopis chilensis and Bulnesia retama (G. ex H) Griseb. along riverbanks and wetter areas; and 3) grasslands above 2200 m asl. Unfortunately, there is no paleoclimatic record available for this area. From data related to periglacial sedimentological features in Patagonia, Mercer (1976) concluded that three glacial advances had occurred during an interval of neogla- cial cooling which spanned the last 5000 years. The first one occurred between 4700 and 4200 B.P., the second, between 2700 and 2000 B.P. and the third one, the Little Ice Age, over the last three centuries. It is important to note that such neo- glacial fluctuation has been confirmed elsewhere in the Andes and in the northern hemisphere (Schubert and Clapperton, in Rabassa and Clapperton 1990). As Pren- tice (1992) noted, however, even when pollen profiles from different areas record the same climatic fluctuation, differences in timing of the event could result from geographic variation (such as elevation and slope aspect), the sensitivity of a par- ticular plant community to climate changes, or variations in the intensity and duration of the climate event. Therefore, it is difficult to hypothesize about Monte vegetation fluctuation from these data. However, the authors agree with D’Antoni (1976), who has proposed for the Gruta del Indio, Province of Mendoza, a warm and dry period between the years 5500 and 3000 B.P., based on pollen analysis. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 145 My ey Mh: pci ee 1m. Speedy Surveying Brunton Compass 1999 Wynvelt F. and L.. Dolout FIGURE 1.—Geographical location and plan of the site at Carrizal de Azampay showing the location of the post remains. During this period the Monte Phytogeographic Province would have reached its maximum size. Then, between the years 3000 and 2000 B.P., the climate became cold and wet, causing the spatial reduction of the Monte to its present distribution. The climate after 2000 B.P. became similar to that of the present day, and human activities began to have an impact on the landscape. The Site and the Archaeological Context.—Ongoing, archaeological research is being carried out at two closely related sites 3 km apart, Carrizal de Azampay and Loma de los Antiguos. Carrizal de Azampay, in particular, has three habitation areas vertically distributed along the south-facing piedmont of the Quebrada El Carrizal. Ancient artificial cultivation terraces are associated with each of these areas. It is thought that this settlement pattern supported extended families of farmers of the Belén culture during the Regional Developing Period (A.D. 1000- 1500) (Balesta and Zagorodny 1999:271; Sempé 1999:250; Zagorodny and Balesta 2000 Habitation Area 1 has two rectangular rooms (R1 and R2) and a communal stone for grinding (R3) (Figure 1). Radiocarbon dating of charcoal from R1 has yielded a date of 310 + 60 “C years B.P—l1o calibrated date range from A.D. 1487 to 1657. Based on the recovered archaeological remains, however, especially the ceramics, it is thought that A.D. 1487 is the date that best fits this site. 146 CAPPARELLI et al. Vol. 23, No. 1 A desiccated preserved wooden post was found in R2 during the fieldwork directed by Dr. M. C. Sempé in 1981. The non-fragmented portion of the post was 35 cm long in all—20 cm above the occupation floor of the room and 15 cm sunk vertically into the occupation floor. The occupation floor was encountered at a depth ranging between 70 and 90 cm under the modern surface. The post was in a standing position directly on the bedrock of the hill, supported by two large granite stones. It was located 4.20 m from the southern wall of the room, 2.85 m from the northern, 2.4 m from the western and 2.0 m from the eastern. Since many fragments of the post were found longitudinally spread along a two-meter line away from its base to the southwest (see Figure 1) at the same depth of the occupation floor, it seems that the post fell down during—or a few years after— occupation, and that it gradually decayed over the past 500 years. This post, estimated to have been at least 2.35 m long, is thought by Zagorodny and Balesta (2000) to have supported the roof. Traditional Use of Construction Resources.—Although in Azampay and surrounding areas the exotic genus Populus sp. L. (dlamo, poplar) is nowadays commonly used for house construction—specifically for roof trusses and beams—the native spe- cies most frequently used are similar to those recorded at the nearby village of El Shincal by Capparelli and Raffino (1997): Prosopis chilensis and P. flexuosa (el arbol), Zuccagnia punctata Cav. (pupo), Porlieria microphylla (Baill.) Desc. O’Don. et. Lourt. (chucupi), Lithraea molleoides (Vell.) Engl. (molle cordoba) and Acacia visco Lor. Ap Gris. (visco). It was expected that the post analyzed here would belong to one of these taxa. Today, traditional houses are constructed in one of two ways. Adobe or stone walls may be built first, and the roof is then supported by them. The roof is generally single pitch, constructed in successive layers as follows: 1) beams, which could be made from tree or cactus trunks, 2) twigs or canes, 3) grass, 4) and a beaten surface made from mud and water. The second type of house involves erecting a wooden structure first, built by means of posts and beams (see Figure 2a). This structure supports the roof. Finally, the walls made of adobe, stones or twigs are built (see Figure 2b, c, and d respectively). MATERIALS AND METHODS The intact portion of the post and the dispersed fragments were recovered manually and were stored for twenty years at the Archaeological Scientific De- partment of the Museum of La Plata. It was evident that even before excavation the post had begun to decay, as seen by intrusive roots in vessels and evidence of fungal attack. After being stored for twenty years in the humid city of La Plata, the post had become highly fragmented and labile, crumbling when cut. The weight of the sample was 177 g, of which 120 g correspond to fragments larger than 1 X 1 3 cm (one 4.5 cm wide X 3.5 cm thick X 7 cm long; twelve about 3 X 2 X 4 cm; twelve about 1.5 x 1.5 X 4 cm; thirty about 1 X 1 X 3 cm) (see Figure 3a), while the other 57 g correspond to a great number of minute frag- ments. Three planes (transverse, radial, and tangential) of the best preserved frag- Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 147 FIGURE 2.—A traditional house at the Hualfin Valley. Different types of construction could be observed at the same habitation area for an extended family: a) wooden structures for shading and drying Prosopis fruits; b) adobe bath room; c) stone room; d) twig kitchen. ments were obtained by manual fracture. The fragments were initially examined under a stereoscopic microscope (Iroscope Mod. M2-14T No. 962329), and for more detail a reflected light microscope (Union ME-3206) was used. One fragment was prepared for SEM examination in a Joel SEM JSM T100. This specimen was mounted on metal stubs with synthetic cement, and then coated with gold. Wood features were photographed at magnifications ranging from 35x to 5000X, ac- cording to the amount of structural detail required. Quantitative and qualitative features were described using the “IAWA List of Microscope Features for Hard- wood Identification’ (AWA 1989). The quantitative values represent an average of 25 measures, except vessels/mm? and rays/mm/? that represent an average of 5 microscope fields at 4X. In all cases the mean is followed of maximum and minimum values between parentheses. The number of annual rings of the largest fragment was recorded, and the diameter of the original trunk was estimated from the circumference of its most external ring, following the methodology proposed by Applequist (1958). Finally, archaeological wood was compared with modern material from the study area in Capparelli’s personal comparative collection. Voucher specimens have been deposited for curation in the herbarium of the Scientific Department of Vascular Plants of the Museum of La Plata (LP). Tangential, transverse and radial planes from 1-cm wide slides of air-dried reference material, without pre- vious treatment, were obtained by the use of a microtome. 148 CAPPARELLI et al. Vol. 23, No. 1 FIGURE 3.—a) General aspect of the best preserved fragments recovered; b) transverse section of the largest fragment where annual rings are evident; c) transverse section dia- gram showing the 13 annual rings counted. RESULTS Identification.—Prosopis sp. (Section Algarobia, Serie Chilenses). Due to geograph- ical distribution of tree species of this genus, it is thought that the post could have been made from either Prosopis flexuosa DC or P. chilensis (Mol.) Stuntz. Voucher Specimens.—Capparelli 7 (Herbarium of the Scientific Department of Vas- cular Plants of the Museum of La Plata: LP) and Capparelli 6 (LP) respectively. Reference Literature—Wood anatomic descriptions of these species were published by Cozzo (1951) and Castro (1994). Wood Description.—Due to the fragmentation and poor preservation of the material it was impossible to decide whether or not it had had some processing before use, such as the sharpening of one of the ends for burying or the removal of its bark. The diameter of the original trunk was estimated to have been equal to or larger than 19 cm. The number of annual rings of the largest fragment was 13 (Figure 3b and c). The color of the wood was brown. Transverse section (Figure 4a). Growth ring boundaries distinct. Wood semi- ring-porous. Vessels mostly solitary (64%) and in radial multiples of two (20%), Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 149 commonly more frequent in early wood. Vessels in radial multiples of three or four scarce (4% and 8% respectively) and more frequent in late wood. Tangential diameter of vessel lumina 87 p (29-194 y), and density 35 vessels/mm? (32-44 vessels /mm?). Thin- to thick-walled polygonal outline fibers (lumina/double wall thickness index = 2.4 1). Diameter of fiber lumina 8.2 yp (4.2-12.7 ».). Mean fiber length 248 p (85.1-364.1 ). Axial parenchyma abundant, confluent paratracheal and in bands (30 to 54 cells wide). Straight rays with a frequency of 6 rays/mm (5-6 rays/mm). Tangential section (Figure 4b). Rays 1- to 8-seriate, more common 5-seriate, and ray width 42.3 yw (16.3-88.4 ). Ray height 24 cells (8-50 cells) and 265.8 p (119-460.6 .). Aggregate rays present occasionally. Vessels extending vertically or partially sinuous. Mean vessel element length 140 p (96.6-185.3 1). Simple fiber pits. Fusiform parenchyma cells present as well as 2 or 3 cells per parenchyma strand. Prismatic crystals in chambered axial parenchyma cells. One crystal per chamber. Radial section (Figure 4c). All ray cells procumbent. Simple perforation plates, oblique in narrower vessels and horizontal in larger vessels. Alternate intervessel pits. Size of pits 5 (3-6 p). Vestured pits and vessel wall (Figure 4d). Finally, Figure 5a—d shows transverse, tangential and radial views of Prosopis flexuosa and a detail of the vessel inner wall for reference, and Figure 5e—h shows the same sections for Prosopis chilensis. DISCUSSION AND CONCLUSIONS As described by Pochettino (1985:206), several findings of pods and seeds of Prosopis from archaeological sites of northwestern Argentina have demonstrated that this genus has been an important food resource from 4000 B.P. to the present, even in cases where agriculture was the main subsistence activity. However, the literature mentioning the importance of Prosopis wood as a construction resource in the past is sparse (see for example the compilation made by Raffino 1990:172- 174). The present work not only confirms its past use as a resource for wooden structures, but also allows us to infer the past presence of this genus in the study area. According to the present-day distribution of plant communities, Prosopis could have been collected either from the riverbank communities of narrow val- leys adjacent to the site or from the open forests surrounding the village of La Ciénaga, 10 km from Carrizal de Azampay. Modern studies of tree rings, growth rates and age-size relationships of Pro- sopis flexuosa in central-west Argentina (Martijena et al. 1988; Perpifial et al. 1995; Villagra et al. 2002) allow us to calculate the values of different related variables, such as diameter of the trunk, age, height, and commercial value of an individual tree. The estimated age for an individual of 19 cm diameter would be, for example, 40 years, while its height would be approximately 4.8 m. Unfortunately, micro- regional climatic and ecological features affect growth rates of this genus. There- fore, extrapolation of data from other regions might not be valid. On the other hand, data coming from the nearby village of El Shincal show that in open forests both Prosopis flexuosa and P. chilensis can reach a height of 5-7 m, while P chilensis of the closed-canopy riverbank forests (less frequent and further away from the 150 CAPPARELLI et al. Vol. 23, No. 1 der of an annual ring (arrow), pon wood (te) fe late wood (ta), semi-ring porosity, axial paratracheal parenchyma confluent to bands, and de Sits in vessels; b) tangential longi- tudinal section (TLS) showing uni- to multiseriate rays and crystals (arrow); c) radial lon- gitudinal section (RLS) showing all cells procumbent in rays; d) detail of the vestured layer of the inner vessel wall and pits. Scale (white bars): a—c, 200 p; d, 5 p site than the open forest) reaches a height of 9-10 m. The most useful length for posts of this species is 2 to 3 m, because at this height the main trunk usually branches into two or more limbs. Posts are usually cut so that a short length of the branched end remains, to better support the beam. In addition, Prosopis is a hard and durable wood (even under water) due to its high tannin content (Castro 1994:13). These advantages are well known to modern local people, and surely were also known to inhabitants in the past. Judging from the location where it was recovered, the post seems to have been one of the supports of a longitudinal beam. A second post may have been located where some big fragments of charcoal were recorded—but unfortunately not recovered—by Sempé during excavation (see Figure 1). It is not known which species was used for beams. It might have been difficult, however, to find a 7-m Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 45) FIGURE 5.—Reference samples: a—d) Prosopis pa: DC; e-h) spun aged (Mol.) Stuntz. essel a) TS; b) TLS showing crystals (arr = Bee f) TLS; g) RLS; — d e-g, 200 p; d and h, 5 n. )P arrow); c) RLS; d) detail of the vestu of the v il of the vestured layer of the vessel wa Scale (white bars): 152 CAPPARELLI et al. Vol. 23, No. 1 long trunk for spanning the entire length of the room, but two posts could have supported two 3.5-m long trunks that would have served as a central beam. The shallow hole in which the post was set (15 cm beneath the occupation floor) and the stone circle at the base of the post that helped to keep it in place indicate that the post was held in place by downward vertical pressure; otherwise it would have fallen down. It is not clear what happened to the roof of this room after its abandonment. It might have been blown off by the strong winds characteristic of the area—mainly the Zonda, a wind originating in the Pacific Ocean that dis- charges all its humidity on the west side of the Andes. (When the Zonda goes down the east side of the Andes and comes to northwest Argentina, it becomes dry and warm, and blows along the valleys in north-south direction at any time of the year, but especially in spring. The maximum mean velocity registered for the Zonda is 28 km/h [Servicio Meteorol6gico Nacional 1969, data collected 1951— 1960]). The post appears to have been fractured at the floor level. There is no evidence it was intentionally broken, so it is thought that insect or rodent attack (for example, by Ctenomys sp.), in combination with water erosion on the ancient surface, could have caused the fracture. Another question that arises from this work is why people of this site used Prosopis for posts and not other species, such as Bulnesia retama, Lithraea molleoides or Acacia visco. Possible answers are that while both B. retama and Lithraea are as hard as Prosopis, the former is a very short tree and the latter is likely to have grown much far away from this site. A. visco is softer than the others and less rot-resistant than Prosopis in moist conditions. There have been changes in the long-term pattern of the management of con- struction resources associated with the introduction of exotic species in Argentina during the period A.D. 1800-1900. Many of them seem to be substitutions rather than changes in construction methods. For example, the exotic dlamo (Populus sp.) replaced native trees for roof beams. Populus has much softer and less durable wood than Prosopis, and today is generally employed for making matches or pack- ing boxes (FAO 1980). In Argentina it is the second most important cultivated tree, between Pinus sp. and Eucalyptus sp. (Politzer 1987). Despite the fact that in Catamarca Populus is widely used as a field and border tree along irrigation ca- nals, its planting for processing in sawmills is minimal (IFONA 1985). It is thought that the modern uses of Populus could have been expanded because it is easy to grow from grafting twigs and can be planted in small plots. Technological factors also favor its cultivation; it is straight, long and easy to work compared with the hard Prosopis. In Europe, throughout prehistory, Populus satisfied wood needs of farmers in regions with scarce wood resources (FAO 1980). Similar cases of sub- stitution have been observed by Johanessen and Hastorf (1990) in the Mantaro Valley. In this case cultivated Eucalyptus sp. replaced cultivated indigenous trees for fuel. The decrease in Prosopis use could be also related to a diminution in the availability of this type of wood. Although there seems to be a decline in the use of native forests for raw material for sawmills (Politzer 1987), modern use of Prosopis for wood fuel, charcoal, and posts is well documented for Catamarca (IFONA 1985). The use of Prosopis could reduce the native forests, a problem Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 153 which could also be intensified by the fragility of arid ecosystems—according to Martijena et al. (1988), P flexuosa forest in Cérdoba take 40 years to regenerate, Further charcoal analysis on material from R1 of Carrizal de Azampay and from R21 of La Loma de Azampay will allow a better understanding of the use not only of wood construction resources, but also of wood resources in general and their prehistoric pattern changes. ACKNOWLEDGMENTS Authors are indebted to the Comisién Nacional de Investigaciones Cientificas y Téc- nicas (CONICET), which has provided institutional support for this work. The study of prehistoric wood management at the Hualfin Valley is also supported by both the Inter- national Foundation for Science, Stockholm, Sweden and Fundacién Antorchas, through grants to Aylen Capparelli. The authors would like to thank Dr. Carlota Sempé, who re- covered the material; Dr. Rodolfo Raffino (Museo de Ciencias Naturales de La Plata), Lic. Stella M. Rivera, Lic. Silvia Monteoliva (Facultad de Ciencias Agrarias y Forestales, Uni- versidad Nacional de La Plata), Dr. Ian Farrington (Australian National University, Can- berra), and Dr. Naomi FE Miller for the revision of this manuscript and their critical sug- gestions; and to Arch. Aldo Capparelli for his valuable architectonic interpretation. Thanks also to Lic. Rafael Urrejola and Lic. Diego Gobbo for their kind assistance with the SEM and graphical designs in the PC respectively. REFERENCES CITED Universidad Nacional de La Plata, La Applequist, M.B. 1958. A simple pith lo- cator for use with off center increment cores. Journal of Forestry 56(2):141. Balesta, Barbara and Nora Zagorodny. 1999. La Loma de los Antiguos. Azam- pay. (Depto. de Belén, Catamarca). In Actas del 12° Congreso de Arqueologia Ar- gentina (La Plata, 1997), T 3, ed. C. Diez Marin, pp. 271-281. Editorial Universi- taria, La Plata. Capparelli, Aylen and Rodolfo Raffino. 1997. La etnobotdnica de ‘El Shincal’’ (Catamarca) y su importancia para la arqueologia I: recursos combustibles y madereros. Parodiana 10(1-2):181-188. Castro, M.A. 1994. Maderas Argentinas de Prosopis. Aflas anatémico. Secretaria General Presidencia de la Nacién, Bue- nos Aires, Arg : Cozzo, D. 1951. Anatomia del lefio secun- dario de las le inosas mimosoideas de Ciencias Naturales “Bernardino Rivada- via” 1:363—-405. D’Antoni, Héctor. 1976. Paleoecologia de la Gruta del Indio (San Rafael, za) sus Correlaciones. Tesis Doctoral. Facul- tad de Ciencias Naturales y Museo, Plata. FAO. 1980. Los dlamos y los sauces en la pro- duccién de madera y la utilizacién de las t ci6 s para la Agricultura y la Alimentacién, R oma. Garibotti, I. 1998. Andlisis de la estructura anat6mica de carbones arqueolégicos de sitios incaicos (ca. 1480-1530) del Va- lle de Uspallata (Mendoza, Argentina). Boletin de la Sociedad Argentina de Botan- ica 33(3-4):195—205. Heyne, C. 1992. Identifying Prehispanic Wood from Archaeological Contexts in Andean Argentina. B.A. Thesis. De- partment of Anthropology, University of Minnesota, Minneapolis. IAWA Committee. 1989. [AWA list of mi croscopic features for hardwood iden tification. International Association of Wood Anatomists, Bulletin n.s. 10:219- 332. IFONA. 1985. Anuario de estadfstica forestal. Ministerio de Economia, Secretaria de Agricultura y Pesca, Instituto Forestal Nacional, Buenos Aires. Johannesen, S. and C. Hastorf. 1990. A his- 154 CAPPARELLI et al. tory of fuel management in the Man taro ae Peru. Journal of Ethaohilaey 10: piers N R. Coirini and L. Pietrarelli. 1988. Estimacién de la produccién eco- némica de un bosque de algarrobo. In Actas del 6° Congreso Forestal Argentino, T. 1, pp. 172-176. Santiago del Estero. Mercer, J.H. 1976. Glacial history of south- ernmost South America. Quaternary Re- search 6:125-166. Morlans, C. 1985. Regiones naturales de Ca- tamarca: provincias geolégicas y provincias ai 6 a Universidad de Catamar- a, Catam. Perpifal, E. M.. M. Balzarini, L. Catalan, L Pietrarelli, get U. Karlin. 1995. Edad de culminacién del crecimiento en Prosopis flexuosa DC en el Chaco Arido Argen- tino. Investigacién agronémica: sistema de recursos forestales 4(1):45-74. aria Lelia. 1985. Diseminulos Politzer, A. 1987. E tudio de mercado de roductos forestaies de las principales especies. Convenio de Cooperaci6n Ins- tituto Forestal Nacional (IFONA)—Con- sejo General de Inversiones (CFI). Bue- nos Aires, Argentina. Final report un- lished. Prentice, I.-C. 1992. Climate change and long-term vegetation dynamics. In Plant Succession: Theory and piesa eds. D.C. Glenn-Lewin, R.K. Peet, and TT. Veblen, pp. 293-339. Gaps and Hall, London. a, ite and Chalmers M. Cla 0. Quaternary glaciations of the Andes. Quaternary Science Re- views 9: 153-174, Raffino, R.A. 1990. Poblaciones Indigenas Ar- gentinas. Tipoerdties Editora Argentina. Buenos Aire Vol. 23, No. 1 Rodriguez, M.F. 2000. Woody plant specie? d tuales y detain arqueolégicos del area del Tambo Incaico de Tambillos (Uspallata), preset Argentina. Paro- diana 10(1-2):91-112. rvicio Meteorolégico Nacional. 1969. Es- tadisticas climatolégicas. 3rd Edition. Pu- blicaci6n B, N° 6. Published by the E j ; logia ah rh (La en 1997), T2, ed. #8 in, pp. 250-258. Editorial Se ena ta Plata Smart. LL. and EC. rina, 1988. Envi- Analytical Methods and Cultural Interpre- tation of bth iee ee Plant Remains, a Christine A. Hastorf and Virginia S. Popper, ie nema University of Chicago Press, Chica Villagra, Pablo, Mariano g “Morales, Ricar- do Vi a, and José Boninsegna. 2002. Dendrocronology of Prosopis wood- lands in the argentine arid zone. In [A- NIGLA, 30 Years of Basic and Applied Re- search on Environmental Sciences, eds. D. Trombotto and R. Villalba, pp. 53-57. Zeta Editores, Mendoza entina. Willcox, G. 1974. A history of deforestation as indicated by charcoal analysis of four sites in eastern Anatolia. Anatolian Stud- ora and Barbara Balesta. asado y presente de una pobla- versidad en América latina: conflictos y coincidencias, T1, pp. 181-95. Universi- dad Catdlica Argentina, Buenos Aires journal of Ethnobiology 23(1): 155-166 | BOOK REVIEWS Spring/Summer 2003 Biodiversity and Native America. Paul E. Minnis and Wayne J. Elisens (eds.). University of Oklahoma Press, Norman. 2000. Pp. 310. $34.95 (cloth) ISBN 0- 8061-3232-9, $14.95 (paper) ISBN 0-8061-3345-7. With contributions from a wide range of respected scholars, Biodiversity and Native America is an important contribution to the ethnobotanical literature of North America. Yet this volume is more than a collection of studies about the relationship between Native North Americans and biodiversity; the strong theo- retical orientation greatly advances the conceptualization of human-environmen- tal interactions globally. In the introduction to the volume, the editors assert, “Native peoples have been neither passive consumers of nature's economy nor primitive rapists of pristine environments” (p.3), thereby challenging the validity of the ‘noble savage’ versus ‘Homo devastandus’ dichotomy. Instead, the editors suggest that aboriginal peoples have actively shaped North American landscapes and patterns of biodiversity for millennia, and have accumulated important eco- logical knowledge in the process. These two themes—active indigenous manage- ment of the environment and the utilitarian value of Native American knowl- edge—are explored throughout the book. In doing so, the contributors challenge common biases that have left ethnobiological investigations of indigenous North American peoples under-researched in comparison to the growing body of liter- ature on tropical regions. The editors rightfully address the irony that although concern for biodiversity is strong in developed countries, temperate regions are often considered less rel- evant because they are presumed less diverse biologically and occupied by more acculturated indigenous peoples. Collectively, the authors show just how wrong these perceptions are. Minnis and Elisens convincingly establish the importance of ethnobiological research in indigenous North America, given the unique cul- tural and biological diversity of the region, the significant economic and political resources available for environmental research, and the fact that, despite historical disruptions, ethnobiologically rich cultural traditions still persist. ; The editors argue that recognition of indigenous environmental experiences serve multiple purposes, and these purposes fall into two broad categories. First, indigenous environmental experience allows a better understanding of ecosystem dynamics and environmental history, which can help establish ecological base- lines’ and lead to better biodiversity restoration, management, and conservation programs. Second, the text recognizes the essential value of Native American medicines, crop diversity, and past and present farming strategies, which may benefit contemporary indigenous peoples and the broader population. The editors define the concept of biodiversity broadly to include the cultural context of human ecology among Native Americans. Native America is defined as encompassing contemporary Mexico, the United States of America, and Canada. The text's first section, ‘Issues and Overviews,” explores interactions between Native Americans and knowledge. Helping clarify the debate over whether indig- nous management practices positively impacted biodiversity, Gary Nabhan pro- poses longitudinal studies based on careful analysis of case studies within a par- 156 BOOK REVIEWS Vol. 23, No. 1 ticular biome—the Sonoran Desert region. Nabhan provides a much-needed cri- tique of how narrow definitions of “species richness” and “language richness” overlook important species interactions of which Native Americans are a part. He likewise criticizes ‘salvage’ approaches to ethnobiology and purely descriptive surveys that provide little information about indigenous perceptions and man- agement of the natural world. In order to overcome these methodological and philosophical shortcomings, Nabhan proposes a focus on indigenous ecological knowledge of plant-animal interactions, the component of biodiversity most likely to be lost. In ‘Relationships between Mexican Ethnobotanical Diversity and In- digenous Peoples,” Robert Bye and Edelmira Linares’ overview includes a helpful discussion of the evolutionary, morphophysiological, and ecological impacts of different management intensities. The authors also discuss how the commonly assumed correlation between biological and cultural diversity is inexact and may undervalue entire cultural and ecological regions of importance. Walter Lewis’ “Ethnopharmacology and the Search for New Therapeutics” focuses on the drug discovery process in North America and develops a model procedure for the development of future pharmaceuticals based on experiences in Peru. He suggests that bioprospectors should target plants utilized by early settlers and indigenous peoples, highlighting the difficulty of defining and compensating collaborators due to the shared nature of ethnobiological knowledge. The second section, ‘Ethnographic Case Studies,” presents overviews of in- digenous knowledge, management practices, and belief systems in the Great Ba- sin, British Columbia, and Chihuahua, Mexico. Catherine Fowler’s chapter ex- plores the dual meaning of utilitarian and spiritual value of plants among the former broad-spectrum hunters and gatherers of the Great Basin. Fowler docu- ments how the Basin’s Timbisha Shoshone have “domesticated” their environ- ments in ways that fit their standards of a managed landscape, leaving open for debate whether such changes are ultimately positive or negative. In an exception- ally well-organized and well-written chapter, Sandra Peacock and Nancy Turner discuss how the foraging populations of British Columbia actively managed their environment to increase the local abundance of plants, and argue that the ces- sation of such management practices has actually led to the deterioration of cer- tain habitats and resources. Enrique Salmén’s “Iwigara, A Rardmuri Cognitive Model of Biodiversity and Its Effects on Landscape Management,” rounds out this section by discussing how land management practices are often embedded in cultural values and belief systems. In the final section of the book, “Prehistory and Biodiversity,”’ Ford, Fritz, and Hammett demonstrate how knowledge of environmental interactions and impacts can be extended into the past, with important implications for under- standing environmental history and contemporary ecosystem patterns. Richard Ford, in his chapter, “Human Disturbance and Biodiversity: A Case Study from Northern New Mexico,” introduces the working concept of “‘prehistorical cultural topography,” or the cultural transformation of place due to intentional interaction with the natural world. Ford argues that prehistoric landscape alterations, such as water control devices, create new habitats, which can have both short term and long term impacts on the distribution of plants and animals. In her chapter, ‘“Lev- els of Biodiversity in Eastern North America,” Gayle Fritz discusses the need to Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 157 ground information in historical and geographical contexts rather than making sweeping generalizations about all indigenous peoples. She accomplishes this through a well-documented discussion of the environmental impacts of the mixed foraging and farming systems prevalent among eastern indigenous peoples. Her chapter recognizes Eastern North America as an important independent center of plant domestication, and discusses the historical effects of diversified foraging and farming strategies on species diversity, community level diversity, and eco- system diversity. However, Fritz cautions against the current reactionary trend to overestimate the impact of Native Americans, and warns that such arguments may be appropriated by profit-oriented industries to justify currently destructive practices. Julia Hammett’s final chapter utilizes ethnohistorical evidence to de- scribe the initiation and maintenance of a shifting mosaic of environmental “patches” in the Southeast prior to European contact. By providing a valuable assessment of how to use documentary records as data sources, Hammett broad- ens the scope of methodological possibilities for exploring historical human-en- vironmental interactions. Collectively, the contributors to this volume provide an insightful and well balanced understanding of the relationship between biodiversity and Native Americans, past and present, as well as a much needed commentary about the utility of this knowledge. Although the theme of indigenous empowerment, in- tellectual property rights, and involvement in management projects could have been developed further, this book clearly demonstrates that issues of native bio- diversity have more than mere academic relevance. As a result, this book is useful for students, scholars, and environmental managers of North American biodiver- sity, as well as persons interested in human-environmental relations throughout the world. Meredith Dudley Tulane University New Orleans, Louisiana Participatory Approaches to the Conservation and Use of Plant Genetic Re- sources. Esbern Friis-Hansen and Bhuwon Sthapit (eds.). International Plant Genetic Resources Institute Publications, Rome. 2000. Pp. 214. $32.00 (paper). ISBN 92-9043-444-9. Several sections of the Convention on Biological Diversity (hereafter, the Con- vention) highlight the relationship between indigenous and peasant peoples and the sustainable use of our world’s biodiversity. Signatory countries to the Con- vention have agreed to recognize this relationship and respect the ecologically sound traditional knowledge that indigenous and peasant communities employ in managing their resources. The Convention signatories also have agreed to in- volve and equitably share benefits with these local communities In development interventions undertaken by their governments and their partners. One of the international organizations that appears to have taken these sensible and ethical prescriptions to heart is the International Plant Genetics Resources Institute 158 BOOK REVIEWS Vol. 23, No. 1 (IPGRI), based just outside of Rome, Italy. IPGRI is the premier NGO that un- dertakes, facilitates and coordinates research, conservation and educational pro- grams on plant genetic resources (PGR) worldwide, especially in less-developed countries. While engaged in a broad range of activities linked to both wild and domesticated plants, the greater part of IPGRI’s efforts focus on agrobiodiversity. With funding from Denmark’s Center for Development Research, IPGRI published this most welcome addition to a growing collection of texts on participatory agro- biodiversity conservation. This body of work, and specifically this book, is an important mechanism for helping agricultural scientists, development specialists, and NGOs understand how they can better involve and assist indigenous and peasant peoples in sustainable agricultural development. Composed of a preface, an introduction, 25 chapters distributed over five sections, and two appendices (a list of acronyms and a glossary), this edited volume provides a methodological initiation to participatory approaches in the conservation of PGRs, and a compilation of case studies describing the successful application of these approaches in developing countries. A “participatory ap- proach” (or “informal model’’ approach) is the agrobiodiversity community’s way of labeling what occurs when indigenous or peasant farmers are invited (and they accept) to become fully involved in defining problems and seeking solutions with- in the context of PGR conservation and sustainable use programs. In other words, traditional peoples and their ecological knowledge are integrated into a range of agricultural developmental undertakings from planning, to research and conser- vation, to production system change, to public awareness and policy revision. Many anthropologists, and especially ethnobiologists, will not be surprised by the positive results attributed to this approach, which is described on p. 12 of the book as a “dramatic change’ from existing approaches. Ethnobiology as a field took the lead in promoting this change, and it is a pleasure to read how this paradigm shift is now bearing fruit in so many agrobiodiversity conservation and agricultural development projects around the world. The relative lack of success with the “top-down” (or “formal model’’) development approach, especially when compared to results achieved through participatory approaches, is also de- scribed in several chapters. This conclusion is especially relevant because the ag- ricultural development specialists from the less-developed countries reached it themselves, indicating the extent to which the participatory approach is now ac- cepted and being applied at local levels throughout the world. Similarly encour- aging are the calls throughout the chapters for greater involvement of social sci- entists in agricultural development programs. The editors have grouped the contributions into five thematic sections: par- ticipatory approach methods and gender issues; descriptions of how gene banks now serve local communities and not just formal sector scientists; participatory approaches in crop selection; public awareness raising; and policy changes asso- ciated with plant genetic resource conservation and use. Several weaknesses in the presentation detract from the book’s readability. It is perhaps inevitable in a book with 25 short chapters that many of the introductory statements repeat the same theoretical justifications and methodological prescriptions. But after reading similar opening paragraphs five or six times, I wished the editors had exerted more control over introductory statements by the chapter authors. The strength Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 159 of the case studies resides in how and why participatory approaches succeeded in location- and culture-specific contexts. In this regard, the chapters providing background on ethnicity, history, and geography are the ones that succeeded best. Though the editors tried to overcome jargon and institutional naming with the addition of an acronym appendix and a glossary, overuse of acronyms and jargon in some chapters also weakened the book’s readability. These relatively minor criticisms aside, this is a useful book for ethnobiolo- gists and other experts specializing in the area of plant genetic resources, and especially for those people who are working or planning to work in the area of agricultural development in less developed countries. The book confirms that many of the theoretical, ethical and methodological positions fashioned by eth- nobiologists over the last 20 years are beginning to influence professionals in other disciplines. At the same time it also provides many useful tips for field-oriented PGR conservation and agricultural development specialists as they plan and im- plement their own interventions. Brien A. Meilleur Muséum National d'Histoire Naturelle Paris, France Mayo Ethnobotany: Land, History, and Traditional Knowledge in Northwest Mexico. David Yetman and Thomas R. Van Devender. University of California Press, Berkeley. 2002. Pp. 359. Illus.; scientific, Mayo and Spanish name lists; index. $48.00. ISBN 0-520-22721-2. A valuable resource for ethnobotanists in northwestern Mexico, this book pro- vides a detailed account of plants used by the Mayo of Sinaloa and Sonora, Mex- ico. According to the authors, this arid region is losing both its biological diversity and traditional plant knowledge. Their book does a great service by increasing the reader’s awareness of indigenous plant knowledge in general and the sophis- tication of Mayo plant knowledge in particular. The first third of the book is dedicated to Mayo history, culture, and biophysical landscapes, and the remainder lists the ethnobotanical information compiled by the authors over a ten-year pe- riod. The chapters on Mayo ethnography and history are interesting and well writ- ten, but leave the reader with only a taste of the complex political and cultural ecologies that characterize landscape and knowledge change in northern Mexico. The title of the book led me to expect more historical ecology and biocultural synthesis, but this book is really a text of economic botany with an ethnographic introduction to the Mayo region. The authors do deliver discussions on land (veg- etation and ecotypes), history (political and economic), and traditional knowledge (botanical), but these are presented as separate entities. All the elements are there but, for the most part, the connections amongst them are left for the reader to sro few glimpses of biocultural synthesis that are provided do keep the read- er interested. For example, the discussion of landscape transformations from pro- 160 BOOK REVIEWS Vol. 23, No. 1 ductive pitahaya cactus (Hylocereus polyrhizus Britton & Rose) fields to bulldozed buffelgrass (Pennisetum ciliare (L.) Link) plantations (and associated erosion of traditional plant knowledge) is an excellent illustration of how economic and po- litico-historical forces propel change in biophysical and sociocultural spheres. The reader comes away with a good understanding of the native ecotypes of the Mayo region (coastal vegetation, coastal and foothills thorn scrub, tropical deciduous forest, oak woodlands and riparian zones), but much of this information, includ- ing some of the photographs, has already been published in Martin et al. (1998). The brief discussion of Mayo plant classification is disappointing and would have been better left out. The authors failed to discern a taxonomic system during their research, and go so far as to say, ‘[w]Je now believe that such a system is not to be found among Mayos” (p.129). The methods they mention having used, however (translating names into Spanish, and noting comments regarding relatedness), are not enough to get at any deeper structure. Pile sorts and triad tests would surely have shed some light on this issue. The chapters “Historical and Contemporary Mayos” and “Eight Plants that Make Mayos Mayos” are stimulating and evocative. They allow the reader to approach an understanding of the links and relationships of human—plant sys- tems of the Mayo, but the book as a whole falls short of providing comprehensive ecological understanding of these relationships. That said, the book is a fun read, accessible to students and aficionados, and an important resource for ethnobot- anists, anthropologists, and ecologists interested in arid and coastal lands. The last and longest chapter, ‘An Annotated List of Plants,” is the heart of the book, providing plant descriptions, uses (in the categories medicine, construc- tion, artifacts, livestock, industry) and names in Latin, Mayo and Spanish, in order of plant family. The bare facts are enriched with anecdotes, observations and comments made by the authors or their Mayo consultants, which makes this list an exceptionally readable, interesting and useful reference tool. One frustrating aspect of the naming of plants here and throughout the book is a lack of differ- entiation in type style or markings (or inconsistent differentiation) between Span- ish, Mayo and English words. Luckily, there is an appendix with thorough and cross-referenced lists of plants in each language, plus a glossary. The text also considers intellectual property and indigenous knowledge. The authors do a good job of giving credit to their Mayo teachers and consultants, with references throughout and an appendix listing the names of consultants along with brief biographies. It is unclear, however, what kind of consultation was made with individuals or communities regarding the future publication of names of consultants and names and uses of Mayo plants, some of which could be utilized by non-local commercial interests. This places Mayo Ethnobotany in that zone of ethical ambiguity that so many of us are trying to navigate through: weighing the value against the dangers to the Mayo community and the scientific community at large of recording and publishing indigenous plant knowledge. If, as the well-presented case is made in this book, Mayo landscapes and traditional knowledge are endangered, this publication is important as salvage ethnobotany. But to be truly useful to future generations of Mayo residents of the thorn-scrub coast, this compilation of plant knowledge should be translated into Spanish—or better still, Mayo—and made available in local communities. Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 161 Mayo Ethnobotany is as an important contribution to the growing body of research on arid lands ethnobotany and I recommend it for people working in northwest Mexico who want to learn about its plants and people. The text would also be useful as a preliminary case study for human-plant systems undergoing major ecological transitions. Felice S. Wyndham University of Georgia Athens, Georgia REFERENCE CITED Martin, P.S., D. Yetman, M. Fishbein, P. Jenkins, T.R. Van Devender, and R.K. Wilson, eds. 1998. Gentrys Rio Mayo Plants: The Tropical Deciduous Forest and Environs of Northwest Mexico. University of Arizona Press, Tucson. Fire, Native Peoples, and the Natural Landscape. Thomas Vale (ed.). Island Press, Washington, D.C. 2002. Pp. 238, maps. $50.00 (cloth) ISBN 1-55963-888-5; $25.00 (paper) ISBN 1-55963-889-3. With a nod towards environmental determinism, the geographers who have contributed chapters to this book set out, in the word chosen by the editor, to “demythologize’” the variably emphasized claims that hunting-gathering societies made significant impacts in using fire in much of the American West. The stated aim of the book as presented by Vale in the first chapter is reasonable enough: to achieve a ‘‘middle ground” between the polar positions that North America was an “‘untouched wilderness’ and the counter claims that it was a ‘“humanized landscape.” The outcome, however, is something quite different from a middle ground, and worse, the overall analysis is flawed. The authors conclude that except for the very limited areas of agriculture in the Southwest, indigenous people made little impact on the ‘natural landscape.” As Vale states in the final sentence of the concluding chapter, it was “an American wilderness—a natural landscape—{that] greeted the first Europeans.” This is quite different from what Thomas Bonnick- sen, an acknowledged authority on forest ecology, concluded in his recently pub- lished book, Americas Ancient Forests: From the Ice Age to the Age of Discovery (2000: 259): ‘Native Americans helped to create and sustain the ancient forests that Europeans found beautiful enough to set aside in national parks.” Vale, in making his claim against what he calls the “arm-waving, careless generalizations” made by “anthropologically minded observers,” is apparently unaware that few anthropologists know about, much less would support, the idea that hunter-gatherers increased the abundance and influenced the distribution of natural resources. As taught to anthropology students, environmental manipu- lations do not occur until people take up farming. Indeed, the seminal thinking about hunter-gatherer uses of fire comes out of Vale's own discipline, and only later influenced anthropologists like Omer Stewart. Stewart took seminars from the geographer, Carl Sauer, during the 1930s while 162 BOOK REVIEWS Vol. 23, No. 1 a graduate student at the University of California, Berkeley, and on the subject of foragers and fires, went on to become a maverick in anthropology. In this respect, it was geography-minded observers who initiated the ‘‘arm-waving, careless gen- eralizations’” that concern Vale and his colleagues. Carl Sauer wrote extensively about the importance of human uses of fire; but his works are scarcely mentioned by Vale and his colleagues. Also, and continuing on that aspect of Sauer’s work, the geographer William Denevan has written a great deal about the significance of native people and their uses of fire in both North and South America; and neither he nor Sauer is all that “anthropologically minded.” It may well be that Denevan’s important 1992 article, ‘“The Pristine Myth: the Landscape of the Americas in 1492,” was the motivation behind Vale's charge to ‘‘demythologize” the opposing view. Thus, instead of taking the high middle ground, Vale attempts to resurrect and spruce up the alternative myth that Native Americans lived in essentially passive relationship to nature, seemingly content to wait for natural fires to do for them what they somehow couldn't figure out and do for themselves. Contrary to this view, Stephen Pyne (1982:71) points out that without the knowledge and the will to use fire in ways significantly different from natural fire regimes, ‘“most Indian economies would have collapsed.’ Collapse they certainly would, had they depended upon disruptive natural fires. Fortunately, they knew how to use fire to help manage local habitats, and in some cases whole regions (e.g., the North American Plains), at preferred stages of ecological succession. Thus, more appropriate than merely asking why hunter-gatherers set fires, and knowing what we now know about the ecology of fire, it makes more sense to ask why not? Why on earth wouldn’t hunter-gatherers have employed such a readily available and easily understood tool, given the resources they sought and their understandings of the complex environments in which they lived? Though we wouldn't be here if our hunter-gatherer ancestors had not been successful foragers, most of us haven't the slightest idea of how to live off the land. But that doesn’t stop some of us from claiming to know all about the ecology of hunting and gathering adaptations. In his opening discussions, Vale makes the claim that the interpretations (two chapters of which are his own) derive from the ‘wisdom that guide[s] the final words” of the authors. This aggrandizing would be off-putting enough, even if the authors had supported their claims with new evidence, either archival or from field studies. Instead, they criticize and selectively use the work of others (e.g., opting for the lowest population estimates for Native Americans); ignore or over- look a considerable number of relevant publications that do not fit their assump- tions; and, worst of all, they fail to include (or even note) any comparisons from what has been written about the uses of fire by indigenous people elsewhere in the world—South America, Asia, Africa, and (most studied of all) Australia. Per- haps most egregious is their omission of any reference to William Cronon’s (1983) major work on Native American uses of fire in New England, an area inhabited by people who fall outside their limited view of ‘‘native.’”” As someone who has researched and written a fair amount on the topic of anthropogenic fires, it is disappointing to have so few of my publications cited, the materials selectively used, and the conclusions heavily criticized. Though Aus- Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 163 tralian Aborigines are not within the book’s narrow purview of native people as being only those in western North America, in Lewis (1973) and other venues, I have published several items that go to the heart of what some of the authors argue (for example, Lewis 1977, 1980, 1991). The most important not included in their criticisms (‘How to Burn a Boreal Forest: Yards, Corridors and Mosaics,” Lewis and Ferguson 1988) is applicable to a number of chapters in the book, and could have actually been used to strengthen their arguments, particularly Vale's conclusions that “mosaics” are a feature of the “‘natural’’ landscape. Expressing a typical criticism, one author dismisses the value of historical and ethnographic evidence because of “biases inherent in oral and written accounts.” Without question, however, the authors accept the quantitative evidence presented in “‘tree-ring-based fire history studies” as if there are no biases in the way fire scars are measured and interpreted, or in the way researchers extrapolate data to create whole regional histories of fire. Were it only a question of evaluating the advantages and disadvantages of quantitative vs. qualitative research or, better yet, how such approaches might complement each other, the book could serve as a starting point for a real debate of the topic and not simply a one-sided polemic. In no way does this book represent a definitive argument about the importance of ‘Native Peoples” and their uses of fire. Admittedly, historic and ethnographic data are rarely useful for quantitative analysis. Nevertheless, anecdotal evidence is frequently consistent with ecological paradigms, thereby helping to validate such data. In this respect, the ecological effect of hunter-gatherer uses of fire has been widely “tested” by comparing the practices of culturally distinct groups within regions and also of those living in widely separated regions or even on different continents. Most important for sci- entific verification, however, are the comparisons that can be made with conclu- sions from ecology, which takes historical and ethnographic accounts (“anecdotal evidence’) and converts them into scientific data. Cross-cultural, particularly in- tercontinental, comparison is anthropology’s main claim to scientific credibility; it is our way of getting beyond more parochial kinds of generalization derived from particular sites or even regions—as Vale and colleagues argue for the Amer- ican West. is History is a messy business, full of “oral and written accounts’; whether social or natural, historical evidence is properly considered, corrected, accepted, or rejected in terms of its overall fit and coherence within larger contexts. Some geographers, like Sauer and Denevan, express this clearly in their writings. Un- fortunately, that perspective is largely missing from this book. As I wrote three decades ago in my first publication on indigenous uses of fire (‘Patterns of Indian Burning in California,” 1973:49-50), “It is not the individual facts nor the total number of such facts that is significant but, rather, how the information fits an ecological system of knowledge that has been gained from the actual study of fire in field situations.” T. Lewis H University of Alberta Edmonton, Alberta 164 BOOK REVIEWS Vol. 23, No. 1 REFERENCES CITED Bonnicksen, T.M. 2000. Americas Ancient Forests: From the Ice Age to the Age of Discovery. John Wiley and Sons, New York. Cronon, W. 1983. seh in the Land: Indians, Colonists, and the Ecology of New England. Hill and Wang, New Y Denevan, W.M. 1992. The Pristine Myth: The penercaee of the Americas in 1492. Annals of the Association of American Geographers 82:369- Lewis, H.T. 1973. Patterns of Indian Burning in California ea and Ethnohistory. Ballena Press Anthropological Papers no. 1. Ramona, Californi . 1977. Maskuta: The Ecology of Indian Fires in Nithaer Alberta, Western Canadian journal of Anthropology 7:15-52. . 1980. Indian Fires of Spring. Natural History 89:76-83. . 1991. A Parable of Fire: Hunter-Gatherers in Canada and Australia, In Traditional Ecological Knowledge: A Collection of Essays, ed. R.E. Johannes, pp. 9-16. World Conser- ), vation Union and, Switzerland. Lewis, H.T. and T.M. Ferguson. 1988. Yards, Corridors and Mosaics: How to Burn a Boreal Forest. Human Ecology 16:57-77. Pyne, S.J. 1982. Fire in ait A Cultural History of Wildland and Rural Fire. Princeton University Press, Princeto Trekking Through History: The Huaorani of Amazonian Ecuador. Laura M. Rival. Columbia University Press, New York. 2002. Pp. xiii + 257. Illus., tables, index. $60.00 (cloth). ISBN 0-231-11844. Rival’s Trekking Through History is an ethnographic account of the Huaorani of Ecuador contextualized through a historical ecological approach. Although the Huaorani use domesticated plants, she describes them as primarily nomadic trek- kers, engaged in a hunting and gathering mode of subsistence. In part, this book is a critique of William Balée’s agricultural regression model of Amazonian for- aging, which posits that foraging in some Amazonian groups likely arose as a postcontact adaptation, with loss of knowledge of how to cultivate being one consequence of colonization (e.g., Balée 1994). Rival argues that Huaorani nomadic trekking is not necessarily a postcolonial adaptation, but instead is more likely a long-standing tradition that represents a political choice to trek rather than to cultivate. Although the two views would seem opposed at first glance, I believe they are complementary, for they are addressing different questions. Balée’s work places more emphasis on ecological change, critical historical events, and the loss of indigenous cultural knowledge, while Rival’s work places more emphasis on environmental perception, historicity, and ontology. Balée has addressed the question of why the Guaja foragers of eastern Bra- zilian Amazonia do not know how to propagate domesticated plants. Based on the presence of linguistic artifacts of domesticates in their language, ethnohisto- rical evidence, and their adaptation to anthropogenic forests, Balée’s work has provided a convincing case that the Guaja were formerly a horticultural people. He places the likely time for their loss of indigenous horticultural knowledge subsequent to the devastating and chaotic circumstances following European col- onization. Guajé foraging involves exploiting dominant palm colonizers of old fallow fields, the fruits of which serve as a caloric staple. Further, his model pro- Spring/Summer 2003 JOURNAL OF ETHNOBIOLOGY 165 vides an explanation of the process through which domesticates may be sequen- tially lost, with high investment bitter manioc disappearing before low-start up cost maize. In Trekking Through History, Rival is addressing a related, but rather different issue: why the Huaorani, who do know how to grow domesticated plants, prefer and emphasize nomadic foraging over sedentary horticulture. Taking inspiration from Sahlins that history is organized through structures of meaning, she argues that Huaorani nomadic foraging is not necessarily postcolonial, but likely a man- ifestation of long-standing cultural and political forces that were in place prior to European colonization. Rival reaches her conclusions by analyzing an essential division in Huaorani thought between true people—the Huaorani—and all non-Huaorani others. She also describes the essential relationship between the Huaorani and non-Huaorani as being based on predator/prey dynamics: ‘others’ are always predators and the Huaorani are always prey. This holds true whether the other is a supernatural being, a non-Huaorani Amerindian, or a European colonist. These ways of being and relations among beings are expressed repeatedly in myth, in cosmology, as well as in interpretations of more recent historical events. The construction tran- scends specific history for it is through the structural relationship between pred- ator and prey which the mythic, the sacred, and all historical events are under- stood. One of the most intriguing discussions in Rival’s work is that it is also the structure through which Huaorani homicide of other Huaorani is framed. Simply put, the death of a kinsman is viewed to appropriately invoke rage. A Huaorani, in this state of necessary rage, temporarily becomes “‘other’’ and therefore pred- atory. Rival describes Huaorani foraging as one manifestation of the relationship between predator and prey. Foraging is a tangible expression of Huaorani em- bodiment as prey, fleeing from predatory others, whoever those specific others may be. As such, the European colonist is no different from any other non-Huaor- ani, even if the individual is a Huaorani temporarily taking the position of a predatory, non-Huaorani other. However, foraging is not merely ontological flight. Rival describes food procurement as an encounter with the past. The Huaorani recognize the forest as a patchwork of old fallows, light gaps, and old growth forest with differential proliferation of plants and animals in these zones. Forest sites and cleared fields are remembered not only for their characteristic species, but are historical mnemonics for the activities of specific individuals and more generalized ancestral areas. Foraging includes not only collecting nondomesticat- ed plants, but remnant cultigens from recently abandoned fallows. Anthropogenic growth itself is amplified through planting chonta palm seedlings in cleared fields resulting from past activities. Rival stresses that it is also important to understand that foraging and flight are viewed as temporary conditions. Huaorani identity is centered in the long- house to which one returns. Here, kinship and community are created, reinforced, and magnified through day to day interactions. Consanguinity is described as far more dependent on lived relationships than genealogical links. Further, co-resi- dents become consanguinealized through sharing of “substance (food, semen, breast milk, illness, parasites, space, etc.). Thus, one risks becoming ‘‘other if 166 BOOK REVIEWS Vol. 23, No. 1 foraging and flight are not predicated upon and actualized by the eventual return to the longhouse. Rival’s fascinating ethnography demonstrates that ecological adaptation can- not be understood as resource extraction alone, it is deeply embedded in Huaorani identity, sociality, symbolism, and historicity. As she states, “I have analyzed trek- king, not as a mundane activity relating to the pragmatics of subsistence and to environmental or historical adaptation, but rather, as a fundamental way of re- producing society through time (p. 178).” Trekking Through History is the fourth book in Columbia University Press’s Historical Ecology Series, edited by William Balée and Carole L. Crumley, and Rival’s work represents an important contri- bution to this developing approach. In my mind, the most important question her book raises for historical ecology is whether or not Amazonian foraging is a meaningful category of analysis amenable at all to a general model. The Maku of the Northwest Amazon are foragers, yet are involved in direct trade relations with Tukanoan food producers. The Guaja have probably lost indigenous horti- cultural knowledge, but live independently from food producers through exploi- tation of anthropogenic forests. The Huaorani know how to grow plants, yet pre- fer foraging to horticulture. Given these particular cases: Is it possible to integrate continuity and change, history and historicity, epistemology and ontology among Amazonian foragers? Or, are the historical, cultural, and ecological conditions (pre- or post-contact) under which foraging is practiced in specific groups partic- ular enough that generalization is futile? Loretta Cormier University of Alabama, Birmingham Birmingham, Alabama REFERENCE CITED Balée, W. 1994. Footprints of the Forest: Ka’apor Ethnobotany—The Historical pony of Plant Utilization by an Amazonian People. Columbia University Press, New Yor IT NOTICETO AUTHORS __% 1/99 90911 3237_ The Journal of grein s savior “Guidelines for Authors” appears in volume 22, number 1. It is also posted on our we al/guidelines/. Careful scrutiny of recent issues of the ea ani provide models of ioral ont style for any manuscript you may wish to submit. | 3 Those submitting manuscripts for coreideration sae SS in the Journal should send three hard copies. Note that sagas ilustrati tions e sent as high-quality p or print-outs. Send diskettes and original a ) after consultation with the Editor. Manuscripts submitted in inappropriate Wie ft format will be returned. Manuscripts should be sent to: NAOMI EF. MILLER Editor, Journal of Ethnobiology University re Pennsylvania Museum—MASCA 3rd and Spruce Streets poten ts PA 19104 (nmiller0@sas.upenn.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: DARRON COLLINS Regional Forest Coordinator Latin American and Caribbean Secretariat World Wildlife Fund 1250 24th St., NW Washington, DC 20037-1193 (darron.collins@wwfus.org) www.ethnobiology.org L: For information about the Society of Ethnobiology, its activities, the Journal, and le scription forms, consult our website. For comments and suggestions about our website, cee cophect DOUGLAS TRAINOR (trainor@ethnobiology.org) SUBSCRIPTIONS Other correspondence to the Society and subscription inquiries for the Journal o Ethnobiology should be addressed to the Secretary / Treasurer, C. Margaret Scarry, Department of Anthropology, ii Bldg., CB# 3115, University of North Carolina at Chapel Hill, Chapel — NC 27599-3115 (scarry@email.unc.edu Annual subscription rates (subject to change; consult our website): For institutions in the U.S., Canada, and Mexico—$80; for institutions elsewhere—$90; for individu- U.S. and Canada—$35; for Sphage in Mexico—$25; for individuals in Central and South recent $35; for i a, Asia, Australia, and Europe—$45; for students in the U.S., Canada and | Mexico—$25; for cere soertls 5. Checks made out to the Society of Ethnobiology should be payable in U.S. dollars; we cannot accept es. Defective or - a of the daha will be i mee if a written request is Fo ct Paul E. Minnis, OK 73019; (405) 325-2519 other currenci received within one year of iss Department of Anthropology, Cintventy of Oklahoma, Norman, (minnis@ou.edu). CONTENTS ETHNOBIOTICA ALUNE ARACHNOPHAGY AND APPROACHES TO SPIDERS AMONG AN EASTERN INDONESIAN PEOPLE Christopher Healey and Margaret Florey “HEAD OF SNAKE, WINGS OF BUTTERFLY, AND BODY OF CICADA”: IMPRESSIONS OF THE LANTERN-FLY (HEMIPTERA: FULGORIDAE) IN THE VILLAGE OF PEDRA BRANCA, BAHIA STATE, BRAZIL Eraldo Medeiros Costa-Neto, and Josué Marques Pacheco MARKETS AND THE USE OF WILD ANIMALS FOR TRADITIONAL MEDICINE: A CASE STUDY AMONG THE TSIMANE’ AMERINDIANS OF THE BOLIVIAN RAIN FOREST Lilian Apaza, Ricardo Godoy, David Wilkie, Elizabeth Byron, Tomas Huanca, William R. Leonard, Eddy Peréz, Victoria Reyes-Garcia, and Vincent Vadez .............. THE GRAMMAR OF SNCHITSU’UMSHTSN (COEUR D’ALENE) PLANT NAMES Gary B. Palmer, M. Dale Kinkade, and Nancy J. Turner SHELLFISH ASSEMBLAGES FROM TWO LIMESTONE QUARRIES IN THE PALAU ISLANDS Scott M. Fitzpatrick THE USE OF DRIFTWOOD ON THE NORTH PACIFIC COAST: AN EXAMPLE FROM SOUTHEAST ALASKA Dana Lepofsky, Natasha Lyons, and Madonna L. Moss WOOD REMAINS FROM ANDEAN ARGENTINA: THE USE OF PROSOPIS SP. L. INHUT CONSTRUCTION Aylen Capparelli, Nora Zagorodny, and Barbara Balesta BOOK REVIEWS 2 47 125 143 155