Can apparency affect the use of plants by local people in tropical forests?

Ulysses Paulino de Albuquerque¹ and Reinaldo Farias Paiva de Lucena²

¹Biologist and Doctor in Ethnobotany, Universidade Federal de Pernambuco (UFRPE), Brazil. Professor, Universidade Federal Rural de Pernambuco, Brazil. Address: Departamento de Biologia, UFRPE. Rua Dom Manoel de Medeiros s/n, Dois Irmãos, Recife, Pernambuco, Brazil, 52171-030. e-mail: upa@ufrpe.br

²Biologist and M.Sc. in Ethnobotany, Doctoral student, UFRPE, Brazil.

Summary

    Ethnobotanists generally assume that an increase in a taxon’s abundance implies increased local relative importance. This idea, based on a simple prediction of the apparency hypothesis, suggests that people will tend to use plants that are easy to find. Since this prediction was put forward in the 1990’s, field researchers have often demonstrated relationships (frequently positive) between the structural parameters of a plant community and the manner in which local communities use these resources; but there is no consistent evidence yet to affirm that these relationships are in any way constant, or even exist at all. The importance of examining the apparency hypothesis in tropical regions is discussed in terms of management, conservation, and bioprospection strategies.

Resumen

    Los Etnobotánicos generalmente asumen que un aumento en la abundancia de una planta implica en aumento de su importancia relativa local. Esta idea, basada en una predicción simple de la hipótesis de la apariencia, sugiere que las personas tenderán a usar plantas que son fáciles de encontrar. Desde que esta predicción se formuló en los años 90, investigadores del campo han demostrado a menudo las relaciones (frecuentemente positivas) entre los parámetros estructurales de una comunidad de plantas y la manera en que las comunidades locales usan estos recursos; pero no hay ninguna evidencia consistente para afirmar que estas relaciones son constantes. Se discute la importancia de examinar la hipótesis de la apariencia en las regiones tropicales por lo que se refiere a la conservación, y estrategias del investigación del potencial farmacológico de plantas medicinales.

Resumo

    Etnobotânicos geralmente têm assumido que o aumento na abundância de um táxon levará ao aumento na sua importância relativa local. Essa idéia, baseada numa predição simples da hipótese da aparência, preconiza que as pessoas tendem a usar as plantas que encontram com mais facilidade. Desde as publicações da década de 90, evidências de campo mostram uma relação freqüentemente positiva entre os parâmetros estruturais de uma comunidade vegetal e as formas de aproveitamento de seus recursos por comunidades locais. Todavia, há evidências consistentes de que essas relações podem variar ou não existirem. Discute-se a importância de testar a hipótese da aparência nas regiões tropicais para orientar estratégias de manejo, conservação e bioprospecção.

Key words / Apparency / Ethnobotany / Medicinal plants /

Received: 11/24/2004. Modified: 05/21/2005. Accepted: 06/16/2005.

Introduction

    In the last few decades, science has accumulated a considerable lote of knowledge on the use of plant resources in tropical forests, despite a lack of information about their general use patterns, which limits our ability to make predictions in terms of their sustainable use or conservation. Is the data presently available sufficient to yield reasonable conclusions on use patterns? Perhaps this problem will stimulate the systematization of the available information and help identify patterns and develop theories to interpret and explain some of the observed phenomena.

    Several ethnobotanical studies have emphasized evaluating the degree of importance of plant resources for a given community and have proposed techniques to that end (Phillips, 1996; Silva and Albuquerque, 2004). Phillips and Gentry (1993a, b) put forward the use value to test ethnobotanical hypotheses, a quantitative technique based on informant consensus. An interesting hypothesis brought up by these authors is linked to resource availability and its local relative importance. Does a plant’s availability in nature tend to make it more culturally important? This question derives from a variant of the ecologic apparency hypothesis, also known as the ecologic transparency hypothesis, or evidence hypothesis. This hypothesis, developed from discussions concerning herbivory, postulates that some plants are more easily found by herbivores (‘apparent’ plants), while others are not (‘non-apparent’), and discusses how plant defenses against herbivory distribute themselves ecologically (Feeny, 1976, Rhoades and Cates, 1976). The hypothesis has been tested already in several locations, with different focuses and methodological efforts. In this paper we will try to summarize how a plant’s visibility can affect its local use, and the implications in terms of tropical resource conservation.

The Apparency Hypothesis

    This hypothesis was developed in the 1970’s through the work of Rhoades and Cates (1976) and of Feeny (1976). These authors hypothesized that plants that are ‘apparent’ (visible) would be more susceptible to potential foraging animals. As such, plants could be classified into two basic groups: ‘apparent’, and ‘non-apparent’. ‘Apparent’ plants are usually perennial woody plants or those that are normally dominant in the ecosystem; while the ‘non-apparent’ plants would be represented by the herbaceous species (principally small plants) with a short life cycle, as well as those present in the early stages of ecological succession. This hypothesis can be examined through both ecological and chemical approaches. The ecological aspects involve the facility with which herbivores find the plants, and the chemical aspects deal with the energetic costs involved in plant defenses. Chemical defenses can involve quantitative or qualitative investments. It is presumed that ‘non-apparent’ plants will invest in low-molecular weight compounds that are highly bioactive (qualitative defense), such as alkaloids, which would be produced in relatively low concentrations (Stepp, 2004; Almeida et al., 2005).

    In the transposition of this hypothesis to the realm of ethnobotany it is assumed that people take the role of foragers and demonstrate much the same behavior as herbivores. On the one hand, the most important plants to a particular culture would be the most ‘apparent’ ones from an ecological point of view. On the other hand, and considering principally the category of medicinal plants, the fact that short life-cycle plants predominate within the repertoire of a large number of societies would seem to indicate that these people are mostly attracted to plants that contain strongly bioactive compounds (see Stepp and Moerman, 2001; Stepp, 2004).

    One of the first tests of the ecological apparency hypothesis as applied to ethnobotanical questions was undertaken by Phillips and Gentry (1993a, b). It was based on a simple prediction that related use to abundance: easily found plants would offer more possibilities for local populations to experiment with their uses, thus having a greater probability of being introduced into the local culture (Phillips and Gentry, 1993a) and having its uses perpetuated by folk knowledge (Figure 1). The authors proposed use-value as a measure of the relative importance of a plant. This technique is based on the number of uses attributed to a plant by the totality of informants (Phillips, 1996; Silva and Albuquerque, 2004), and therefore required a relatively large number of interviews. In using this technique, the researcher assumes that a plant is important based entirely on its total number of uses. As such, plants that are infrequently mentioned will be considered as less valued. This technique, however, does not distinguish between ‘real use’ and ‘knowledge’, or ‘real use’ and ‘cognitive use’, as pointed out by Torre-Cuadros and Islebe (2003), as a plant may be frequently cited but little used. Lucena (2005), for example, reported that Schinopsis brasiliensis Engler, a plant from the semiarid region of northeastern Brazil, had the third highest use-value in their survey, although the plant was no longer actively being used.

    Also, this technique does not consider the ‘cultural significance’ of a plant, its frequency of use or exploitation, nor the conduct of the local people in relation to that resource. Prance et al. (1987) and Turner (1988) had proposed other measures of relative importance, classified within a group called ‘subjective allocation’ (Phillips, 1996). These measures were widely criticized due to the fact that the relative importance of a plant would depend, basically, on the vision of the researcher, often ignoring the conduct and opinion of the informant. A technique developed by Turner (1988) and called the ‘cultural significance index’, incorporated elements that were extremely important for determining if a plant was valuable for a given population, such as use-frequency, management of that resource, and current use. In spite of severe criticism, some researchers (Stoffle et al., 1990; Silva et al., 2005) have attempted to improve this index; they have proposed changes in the index that they suggest will make it more objective, not only adding elements of consensus among the informants, but also associating variables such as management, preference, and use-frequency. The idea of cultural importance, for instance, associated with perceived attributes (perceptual salience) of a given plant by the people is related to studies of folk-classification (Berlin, 1972, 1973; Berlin et al., 1966, 1968, 1973; Hunn, 1975, 1976, 1982; Brown, 1977; Dougherty, 1978; Carrington, 1983). In this sense, these authors have pointed out the importance of ‘perception’ of a resource and its ‘salience’ as determinate for its cultural significance.

    The apparency hypothesis has significant methodological and theoretical implications, not only in terms of its importance for the conservation of biodiversity as a measure of the use and exploitation of natural resources, but as well in other areas of ethnobiology. In this paper, we will comment the aplication of this hypothesis within the context of biodiversity and conservation and the use of medicinal plants, considering that these approaches have been used in the great majority of the research papers related to apparency hypothesis. In this sense, we will not consider works that only treat these subjects based on anecdotal observations.

Expected Increase in a Plant’s Relative Importance

Local approaches

    Several researchers have supplied evidence that supports Phillips and Gentry’s (1993a) ecological apparency hypothesis. Galeano (2000) demonstrated that the use value of certain plant families and species in Colombia is directly related to their abundance, and therefore concluded that species and families with high use values (as opposed to abundance) may gradually be eliminated through predatory collection. Relationships between taxon use, abundance, and ecologic importance have been found for trees in the Peruvian Amazon (Phillips and Gentry, 1993a, b), and for lianas of the Ecuadorian Amazon (Paz y Minõ et al., 1991). Results from other studies, although not testing specific predictions, suggest that visibility is a predictor of the relative importance of a given taxon (Mutchnick and McCarthy, 1997; Caniago and Siebert, 1998).

    More recent studies have either found no link between taxon abundance and relative importance, or at best a weak, yet significant, relationship. Torre-Cuadros and Isbele (2003) working with the Mayas in Southwestern Mexico, examined the ecological apparency hypothesis based on the relationship between a species cultural importance, as expressed by its use value, and its availability according to the importance value index. They concluded that this relationship has two implications: i- "not all species are used according to their availability", and ii- "a resource’s real use and cognitive use can cause positive and negative impacts for sustainability". One way of favoring sustainability of plant resources would be by employing a large group of dominant species, allowing resource use according to species abundance (Tacher et al., 2002). It is an efficient conservationist strategy, for it reduces risks to species with lower abundances and that are relatively rare.

    In Brazil, studies undertaken in the Atlantic Forest and the Caatinga suggest that plant visibility did not completely explain its relative importance. Cunha and Albuquerque (2005) verified that all species found in an Atlantic Forest fragment in Northeastern Brazil were locally recognized as useful. The relationship between their visibility (density, dominance, frequency, and importance value index) and their use value was positive, although weak in relation to their relative frequency and importance value index. In the semiarid areas of Northeastern Brazil, the findings of Albuquerque et al. (2005) and Ferraz (2004) suggest a weak or nonexistent relationship between use and visibility. In fact, a careful analysis of each species revealed that many of the more important species tend to be the most locally rare or vulnerable (Albuquerque et al., 2005).

    Previous studies commonly employed use value (as proposed by or modified from Phillips and Gentry, 1993a, b) as a measure of relative importance. This index assumes that a plant’s importance is reflected in the number of uses cited by informants (Silva and Albuquerque, 2004). However, caution must be taken in drawing any conclusions concerning use pressure, as a taxon might have many uses yet the fact be known by only a few individuals.

    In this sense, Lawrence et al. (2005) in a study in Madre de Dios, Peru, proposed that the importance of a given plant species could be more strongly related to its dominance (basal area) than to its abundance. As such, they observed a relatively strong relationship, in the category of construction wood, between a species’ use value and its visibility. However, when the two most important species in this category (Cedrela odorata L. and Swietenia macrophylla) were removed, this statistical relationship became weaker. A positive relationship between a species’ visibility and its use value can result in negative consequences for the more apparent species, as their high use value may well threaten its abundance, due to the constant pressure for domestic or commercial use. Lawrence et al. (2005) suggest, however, that a species’ use value is not absolute, but may vary from year to year according to circumstantial influences.

Regional approaches

    Regional approaches have rarely been undertaken. Silva and Albuquerque (2005) analyzed woody medicinal plants on a regional scale in six areas of Caatinga in Northeastern Brazil. These authors employed the relative importance measurement proposed by Bennett and Prance (2000), which is based on the number of uses and therapeutic properties attributed to a given species. Their study did not involve direct consultations with local communities, and uses were recorded using secondary data. In general, this importance measurement follows the same logic as the use value proposed by Phillips and Gentry (1993a, b). The work was based on compiling floristic inventories and elaborating a data base, and the popular uses attributed to each of the species, based on a wide bibliographic survey, were investigated. The authors found that relative importance is negatively correlated with relative density (r= -0.39; p<0.05) and relative frequency (r= -0.35; p<0.05). Several species were considered to be vulnerable by the authors, as they were very sought after for their medicinal properties and for wood uses by different rural communities in the region. This type of regional approach presents a broader vision of a taxon’s total availability. As such, it would be interesting to carry out other regional studies to gain a better comprehension of the use patterns of tropical resource at greater scales.

Possible Effects of Apparency on the Use and Selection of Medicinal Plants

    The growing interest in the use of plants, especially species with medicinal potential, is linked to many factors, such as therapeutic efficacy and low cost. This growing demand may nevertheless contribute to an increase in deforestation due to excessive collection (Hersch-Martínez, 1995). Medicinal plants are an important aspect of traditional therapy used by local communities (Sequeira, 1994) and ethnobotanical studies have documented the use of medicinal plants in many local communities. This research has revealed an impressive number of species that are used in a traditional manner for a great variety of purposes (Milliken, 1997; Milliken and Albert, 1997) as well as in the industrial manufacture of drugs. Consequently, tropical forests are considered to be an important resource bank, especially for plants of pharmacologic interest.

    Mendelsohn and Balick (1995) suggested that approximately 750000 potential extracts can be obtained from higher plants from the world’s tropical forests, giving force to campaigns seeking to protect these areas. However, some data has suggested that primary forests are not the most important source of medicinal plants, as studies undertaken in different parts of the world have found that local populations identify more medicinal plants as coming from secondary vegetation and anthropogenic zones.

    Stepp and Moerman (2001) suggested the usefulness of the apparency and resource availability theories, initially used in herbivory studies, in explaining the use and selection of medicinal plants by local communities. Analyses of surveys undertaken in several parts of the world have shown that medicinal plant lists are dominated by herbs (Figure 2). According to the prediction of the apparency hypothesis, herbs or plants with short life cycles should concentrate highly active compounds (Feeny, 1976; Coley et al., 1985).

    Almeida et al. (2005) carried out a study in a large semiarid region in Northeastern Brazil in which phytochemical tests were applied to selected native plants comprising the medicinal flora. The authors’ hypothesis considered that the most important local medicinal species should be herbs, and that these herbs should have the highest percentages of compounds considered to be strongly bioactive. Table I summarizes the data obtained. In this case, the measure of relative importance suggested by Bennett and Prance (2000) was also used. The importance of each plant was calculated and then ascribed to the categories listed in the table. Averages were compared using the Kruskal-Wallis test. The conclusion was that a plant’s relative importance is independent of the categories tested.

    The data obtained in the Caatinga does not support the predictions of the hypothesis presented above, nor does it follow the general trends found in tropical and temperate regions. These results may be due in part to the unique characteristics of this dry biome (Almeida et al., 2005). In fact, many studies have shown that herbaceous plants stand out in medicinal floras throughout the world (Stepp and Moerman, 2001; Stepp, 2004), with approaches that validate these ideas (Voeks, 1996). Even though it might seem that there are differences that indicate the great relative importance of herbs, this data must be treated with caution, as such conclusions have been reached with different methodologies and theoretical backgrounds and not all results are comparable. For example, despite the fact that Voeks (1996) and Albuquerque et al. (2005) found that anthropogenic zones have greater relative importance in supplying plant resources, which supports the ideas presented previously, the latter author determined, in a Caatinga region, that in spite of the fact that these zones had a relatively greater number of medicinal plants, people preferred to use native trees. In Voeks’ (1996) study, the idea of preference becomes confounded with that of supply source, which are distinct concepts (Albuquerque et al., 2005).

Conclusions and Future Investigations

    The data presented here becomes especially relevant when forest resource management subsides and conservation proposals are to be contemplated, although there is still much to be discovered concerning the use-patterns of tropical forest diversity. Future results will depend on larger methodological efforts, with quantitative approximations. A qualitative approximation is first necessary to improve our floristic/ethnobotanical database, to be combined then with qualitative efforts that put forward and then test new hypotheses. This implies a discussion about the instruments employed and assumptions made, implicitly or explicitly, to test these hypotheses. In the present case, the data collected in dry forests does not support the ecologic apparency hypothesis, in contrast to similar investigations undertaken in humid forests, such as the work of Galeano (2000) and Phillips and Gentry (1993a, b). These authors were able to show a positive relationship between a plant resource local importance and its availability in the forest. The implications would be extremely interesting if this turns out to be a pattern, although Phillips and Gentry (1993a, b) observed: "Are the most important species more available because of the strength of anthropogenic impact, which favor them in some way? Or are they more important because they are more available?"

    In this paper various examples were selected to illustrate the need for new hypotheses in ethnobotanical research. Nearly all aspects related to the connection between human beings and plants need more in-depth studies, not only because of the imperative of expanding scientific knowledge, but also for practical problem solving issues. What differences are observed in relation to the use of forest diversity by different communities? What factors determine the preferences and the local selection criteria for plant use, especially for medicinal ends? How do chemical, ecological, and cultural factors influence these processes? How does a plant’s availability in the wild influence its use?

    Answering these questions would help better delineate use strategies for natural resources, as well as guide the efforts of prospecting tropical biological diversity. To implement the needed research, there must first be discussions about the critical analysis of the instruments used in data collection, as well as efforts for at least minimal methodological standardization. For example: What are the real limitations of use value as a measure of relative importance in terms of the objectives put forth here? Does the fact that a given plant has many known uses necessarily imply more use pressure on that species?

    The techniques proposed by Phillips and Gentry (1993a, b) to determine use values limit its effectiveness in evaluating the exploitation of natural resources and the pressure these resources may suffer from harvesting. In spite of these obvious limitations, no formal studies have yet been undertaken that critically examine the usefulness of this approach. In truth, the technique seems only to indicate that there exists a relationship between the "apparency’ of a resource and its spectrum of uses, and the number of people who mention it; but it is not possible to speculate on the resource’s real use. Even though the technique allows objectivity and repeatability for comparative purposes, it must be remembered that the use and exploitation of any resource takes place within specific cultural contexts. Thus, considering the importance of testing the apparency hypothesis in terms of the conservation of biodiversity, we suggest that future studies: a) evaluate relative importance using different techniques and different principals with the goal of better defining which aspects of ‘apparency’ influence the use and knowledge of natural resources; b) confront the information obtained with these techniques with field data about the real use and exploitation of natural resources; c) examine ‘apparency’ by other parameters of availability (density, frequency, etc.) considering the totality of species used as well as their individual use categories (medicinal plants, food resources, etc.). Using these methods, it should be possible to compile more information concerning variables that would minimize the ‘apparency’ effect on the importance of a given resource.

Acknowledgements

    The authors thank CNPq/FACEPE for financial support to U.P Albuquerque through the "Primeiros Projetos" program.

References

1. Albuquerque UP, Silva ACO, Andrade LHC (2005) Use of plant resources in a seasonal dry forest (northeastern Brazil). Acta Bot. Brasílica 19: 27-38.        [ Links ]

2. Almeida CFCBR, Silva TCL, Amorim ELC, Maia MBS, Albuquerque UP (2005) Life strategy and chemical composition as predictors of the selection of medicinal plants from the Caatinga (Northeast Brazil). J. Arid Env. 62: 127-142.        [ Links ]

3. Bennett BC, Prance GT (2000) Introduced plants in the indigenous pharmacopoeia of Northern South America. Econ. Bot. 54: 90-102.        [ Links ]

4. Berlin B (1972) Speculations on the growth of ethnobotanical nomenclature. Language in Society 1: 51-86.        [ Links ]

5. Berlin B (1973) Folk systematics in relation to biological classification and nomenclature. Annu. Rev. Ecol. Systemat. 4: 259-271.        [ Links ]

6. Berlin B, Breedlove DE, Raven PH (1966) Folk taxonomics and biological classification. Science 154: 273-275.        [ Links ]

7. Berlin B, Breedlove DE, Raven PH (1968) Covert categories and folk taxonomies. Am. Anthropol. 70: 290-299.        [ Links ]

8. Berlin B, Breedlove DE, Raven PH (1973) General principles of classification and nomenclature in folk biology. Am. Anthropol. 75: 214-242.        [ Links ]

9. Brown CH (1977) Folk botanical life-forms: their universality and growth. Am. Anthropol. 79: 317-336.        [ Links ]

10. Caniago I, Siebert SF (1998) Medicinal plant ecology, knowledge and conservation in Kalimatan, Indonesia. Econ. Bot. 52: 229-250.        [ Links ]

11. Carrington JF (1983) Linguistic pit-falls in upper zairean folk-taxonomy research. J. Agric. Trad. Bot. Appl. 2: 149-158.        [ Links ]

12. Coley PD, Bryant JP, Chapin FS (1985) Resource availability and plant anti-herbivore defense. Science 230: 895-899.        [ Links ]

13. Cunha LVFC, Albuquerque UP (2005) Quantitative ethnobotany in an Atlantic Forest fragment of Northeastern Brazil- implications to conservation. Env. Moni. Assess. In press.        [ Links ]

14. Dougherty JWD (1978) Salience and relativity in classification. Am. Ethnol. 5: 66-80.        [ Links ]

15. Feeny P (1976) Plant apparency and chemical defense. In Wallace JW, Nansel RL (Eds.) Biological Interactions Between Plants and Insects. Recent Advances in Phytochemistry 10. Plenum Press. New York, USA. pp. 1-40.        [ Links ]

16. Ferraz JSF (2004) Uso e Diversidade da Vegetação Lenhosa as Margens do Riacho do Navio Município de Floresta (PE). Thesis. Universidade Federal Rural de Pernambuco. Brazil. 70 pp.         [ Links ]

17. Galeano G (2000) Forest use at the Pacific Coast of Choco, Colombia: a quantitative approach. Economic Botany 54: 358-376.        [ Links ]

18. Hersch-Martínez P (1995) Commercialization of wild medicinal plants from southwest Puebla, Mexico. Econ. Bot. 49: 197-206.        [ Links ]

19. Hunn E (1975) A measure of the degree of correspondence of folk to scientific biological classification. Am. Ethnol. 2: 309-327.        [ Links ]

20. Hunn E (1976) Toward a perceptual model of folk biological classification. Am. Anthropol. 3: 508-552.        [ Links ]

21. Hunn ES (1982) The utilitarian factor in folk biological classification. Am. Anthropol. 4: 830-847.        [ Links ]

22. Lawrence A, Phillips OL, Reategui A, López M, Rose S, Wood D (2005) Local values for harvested forest plants in Madre de Dios, Peru: towards a more contextualized interpretation of quantitative ethnobotanical data. Biodiv. Cons. 14: 45-79.        [ Links ]

23. Lucena RFP (2005) A hipótese da aparência ecológica poderia explicar a importância local de recursos vegetais em uma área de Caatinga? Thesis. Universidade Federal Rural de Pernambuco. Brazil. 92 pp.        [ Links ]

24. Mendelsohn R, Balick MJ (1995) The value of undiscovered pharmaceuticals in tropical forests. Econ. Bot. 49: 223-228.        [ Links ]

25. Milliken W (1997) Traditional anti-malarial medicine in Roraima, Brazil. Econ. Bot. 51: 212-237.        [ Links ]

26. Milliken W, Albert B (1997) The use of medicinal plants by the Yanomami indians of Brazil, Part II. Econ. Bot. 51: 264-278.        [ Links ]

27. Mutchnick PA, McCarthy BC (1997) An ethnobotanical analysis of the tree species common to the subtropical moist forests of the Petén, Guatemala. Econ. Bot. 51: 158-183.        [ Links ]

28. Paz y Minõ G, Balslev H, Valencia R, Mena P (1991) Lianas utilizadas por los indígenas Siona-Secoya de la Amazonía del Ecuador. Reportes Técnicos 1. Ecociencia. Quito, Ecuador. XXX pp.        [ Links ]

29. Phillips O (1996) Some quantitative methods for analyzing ethnobotanical knowledge. In Alexiades MN (Ed.) Selected guidelines for ethnobotanical research: a field manual. New York Botanical Garden. New York, USA. pp. 171-197.        [ Links ]

30. Phillips O, Gentry AH (1993a) The useful plants of Tambopata, Peru: I. Statistical hypothesis tests with a new quantitative technique. Econ. Bot. 47: 15-32.        [ Links ]

31. Phillips O, Gentry AH (1993b) The useful plants of Tambopata, Peru: II. Additional hypothesis testing in quantitative ethnobotany. Econ. Bot. 47: 33-43.        [ Links ]

32. Prance GT, Balée W, Boom BM, Carneiro RL (1987) Quantitative ethnobotany and the case of conservation in Amazonia. Cons. Biol. 1: 296-310.        [ Links ]

33. Rhoades DF, Cates RG (1976) Toward a general theory of plant antiherbivore chemistry. In Wallace JW, Nansel RL (Eds.) Biological Interactions Between Plants and Insects. Recent Advances in Phytochemistry 10. Plenum Press. New York, USA. pp. 169–213.        [ Links ]

34. Sequeira V (1994) Medicinal plants and conservation in São Tomé. Biodiv. Cons. 3: 910-926.        [ Links ]

35. Silva VA, Albuquerque UP (2004) Técnicas para análise de dados etnobotânicos. In Albuquerque UP, Lucena RFP (Org.) Métodos e técnicas na pesquisa etnobotânica. NUPEEA, Recife, Brazil. pp. 63-88.        [ Links ]

36. Silva ACO, Albuquerque UP (2005) Woody medicinal plants of the caatinga in the state of Pernambuco (northeast Brazil). Acta Bot. Brasílica 19: 17-26.        [ Links ]

37. Silva VA, Andrade LHC, Albuquerque UP (2005) Revising the Cultural Significance Index: the case of the Fulni-ô in Northeastern Brazil. Field Methods. In press.        [ Links ]

38. Stepp JR (2004) The role of weeds as sources of pharmaceuticals. J. Ethnopharmacol. 92: 163-166.        [ Links ]

39. Stepp JR, Moerman DE (2001) The importance of weeds in ethnopharmacology. J Ethnopharmacol. 75: 19-23.        [ Links ]

40. Stoffle RW, Halmo DB, Evans MJ, Olmsted JE (1990) Calculating the cultural significance of American indian plants: Paiute and Shoshone ethnobotany at Yucca Mountain, Nevada. Am. Anthropol. 92: 416-432.        [ Links ]

41. Tacher SIL, Rivera RA, Romero MMM, Fernández AD (2002) Caracterización del uso tradicional de la flora espontánea en la comunidad Lacandona da Lacanhá, Chiapas, México. Interciencia 27: 512-520.        [ Links ]

42. Torre-Cuadros MA, Isbele GA (2003) Traditional ecological knowledge and use of vegetation in southeastern Mexico: a case study from Solferino, Quintana Roo. Biodiv. Cons. 12: 2455-2476.        [ Links ]

43. Turner NJ (1988) "The importance of a rose": evaluating the cultural significance of plants in Thompson and Lillooet interior Salish. Am. Anthropol. 90: 272-290.        [ Links ]

44. Voeks RA (1996) Tropical forest healers and habitat preference. Econ. Bot. 50: 381-400.        [ Links ]