diversidude sao ferramentas importantes que nos ajudam a enteuder as interafres humanas com .... b) Are the same plants used by most individ- uals?
USE OF ECOLOGICAL METHODS IN ETHNOBOTANY: DIVERSITY INDICES 1,2 ALPINA BEGOSSI
Begossi, Alpina (Ndcleo de Estudos e Pesquisas Ambientais, Universidade Estadual de Campinas, CP 6166, Campinas, SP, 13081-970, Brazil). USE OF ECOLOGICALMETHODSIN ETHNOBOTANY:DIVERSITYINDICES.Economic Botany 50(3):280-289, 1996. The application o f ecological concepts to ethnobotanical studies, in particular o f diversity, is analyzed. Diversity indices are important tools that may help in understanding human-environment interactions. Those indices allow comparisons on the use o f plants by different populations in different environments. A review on recent major ethnobotanical journals was carried out, and 10 studies (7from Latin America, 2 from Asia and 1 from Europe) were selected based on available data to calculate diversity indices. The Shannon-Wiener indices and rarefaction curves were obtained. High diversity on plant uses were found f o r studies carried out at Peru, Mexico, Brazil and Thailand. A low diversity was found f o r Tonga, and island biogeography theory is used to discuss these results. Sampling effort is evaluated through rarefaction curves. The estimation o f the diversity o f resources used by native populations may be useful when planning conservation areas and their management. O Uso de Mrtodos Ecol6gicos em Etnobotgmica:Indices de Diversidade. A aplicafdo de conceitos de ecologia em estudos etnobotCmicos, em particular diversidade, ~ analisada. Os (ndices de diversidude sao ferramentas importantes que nos ajudam a enteuder as interafres humanas com o ambiente. Esses (ndices permitem comparar o uso de plantas por populafres diferentes em ambientes diferentes. Foi realizada urea revisao nos principais perirdicos recentes de etnobot~nica e foram selecionados 10 estudos (7 da America Latina, 2 da Asia e I d a Europa), corn base na disponibilidade de dados, para o cdlculo dos indices de diversidude. Indices de ShannonWiener e curvas de rarefafao foram obtidas. Foi encontrada uma alta diversidade de uso de plantas para Peru, Mgxico, Brazil e Tail~udia. Uma haixa diversidade foi encontruda em Tonga, e a biogeografia de ilhas 6 usada para discutir os resultados. 0 esfor9o de amostragem ~ avaliado com base nas curvas de rarefagao. A estimativa da diversidade dos recursos usados p o t populafres nativas pode ser util no planejamento de dreas de conserva96o e em seu manejo.
Key Words: biodiversity; diversity indices; ethnobotany; island biogeography.
Ecological concepts, models and methods have been proven to be useful in the understanding of human-environmental interactions. There is a vast literature on those applications, such as on carrying capacity (Fearnside 1986), on resilience (Begossi 1995; Berkes and Folke 1992), on optimal foraging theory (Begossi 1992; Begossi and Richerson 1992; Hames and Vickers 1982; Hawkes, Kaplan, Hill, and Hurtado 1987; Hill, 1988, among others) and on diversity (Begossi n.d.; Begossi and Richerson 1993; HarReceived 7 September 1995; accepted 21 February 1996. 2This study is dedicated to the memory of my friend and collaborator, the Brazilian botanist and ethnobotanist Hermrgenes de Freitas Leit~o-Filho [19441996].
desty 1975). Although there has been criticism of the use of ecological models in the analysis of human-nature relationships (Balre 1989), such models have provided quantitative tools that increase confidence to the analysis of data, and focus attention on the importance of sample size and sampling effort. In this study I examine the use of ecological diversity indices for the analysis of the diversity of resource use (in this case, plants) by different populations from different areas. For this purpose I draw upon the data from the literature, from different parts of the world. ECOLOGICAL DIVERSITY OR BIODIVERSITY The diversity of species has always been a fundamental area of inquiry in ecology. The
Economic Botany 50(3) pp. 280-289. 1996
9 1996 by The New York Botanical Garden, Bronx, NY 10458 U.S.A.
1996]
BEGOSSI: ECOLOGICAL METHODS IN ETHNOBOTANY
study of diversity is based primarily on comparative and correlative research (Huston 1994:8). Local diversity should be correlated with features of the environment, especially with the diversity of resources (Schluter and Ricklefs 1993). According to Huston (1994:16) the latitudinal gradient was the pattern that first attracted scientific attention to species diversity; plants are well represented in this gradient. Alexander von Humboldt and Aim6 Bonpland, in 1807, were probably the first in comparing the greater diversity found in the tropics to the temperate areas (Latham and Ricklefs 1993). Diversity indices can be used to evaluate the intensity of resources used by human populations, to allow comparisons among different populations in different environments, and to allow evaluations of sampling effort. Diversity indices can also help us to determine a minimum area necessary for a native human population based on data on the resources used. According to May (1975) species-abundance relations form the basis for considerations of species-area relations. Hence, the abundance of individual species used by native populations is of key importance in determining the size of the areas used by those populations, and has consequences for the size of conservation units such as extractive reserves. Very recently, diversity indices have also been used in archaeological studies (Wing and Wing 1995). Many diversity indices have been discussed in the ecological literature (Huston 1994; Pielou 1975; Rosenzweig 1995; Schluter and Ricklefs 1993). The number of species per unit area, richness, depends on sample size; the more samples taken, the more rare species are added to the list (Begon, Harper, and Townsend 1990). Indices of equitability or evenness take into account both the number of species and their relative abundance (Magurran 1988). One of these, the Shannon-Wiener index, employed in this study, is widely used in ecology and is regarded as having a moderate sensivity to sample size (Magurran 1988). Another form of evaluating diversity used here is the rarefaction curve. It is derived from sampling distributions, obtained by plotting the number of species expected to be observed, as a function of the number of individuals in the sample [Sanders (1969) curves--see also Simberloff (1979)]. Many quantitative methods in ethnobotanical studies have been proposed (Johns, Kokwaro, and Kimanani 1990; Moerman 1991), one is the
281
use value of species (Phillips and Gentry, 1993a,b; Phillips et al. 1994; Prance et al. 1987). Carneiro (1978) used a method to evaluate the number of species per area (species/area curve). Diversity indices are not being suggested here as a substitute for the methods proposed above, but as a method that allows comparisons among populations in different areas. Diversity indices are quantitative tools used by ecologists in most studies at the community level, why not apply them to ethnobotanical studies? Diversity indices can help us in answering at least the following questions: a) Does the diversity of plant use represent the diversity of plants available? We should expect a higher diversity of plant uses in the tropics, compared to temperate areas, if uses follow the availability of plants. For this, we need to have data on the plant diversity of the area studied; such data often are not available. Benz and co-workers (1994) found that the floristic abundance of the area studied was related to the floristic composition of plants used, in terms of family composition. Some studies allow comparisons between the richness of available and used species. Such studies include: Boom (1990), which showed that 30% of species were used by the Panare Indians from Venezuela; Glover, Magogo, and Mzee (1969), which reported that 47% of local species were used by the Digo from the Shimba Hills, Kenya; and Prance and co-workers (1987), in Amazonia, which found that 77%, 61%, 49% and 79% of species were used by the Ka'apor, Tembr, Panare and Chficobo, respectively. b) Are the same plants used by most individuals? The calculation of the evenness may be helpful. A low evenness means a high dominance in the use of a few species. c) Are there differences in the diversity of plant uses per category (such as gender/age)? Figueiredo, Leitao-Filho, and Begossi (1993) showed in one community at the Atlantic Forest coast (Brazil), that women and older people tend to cite a higher diversity of plant uses than do men and young individuals. d) Is the sampling effort sufficient? In other words, is the informant sample large enough to cover a high diversity area? Rarefaction curves help to evaluate sampling efforts.
METHODS In order to calculate diversity indices (besides richness), it is necessary to have data on the
282
ECONOMIC BOTANY
number of individuals who cited a plant species. A survey on the major journals of ethnobotany was carried out at the Economic Botany library, Harvard University. All volumes of Advances in Economic Botany (1-9), Conservation Biology (1-9), Journal of Ethnobiology (1-14) and the
Journal of Herbs, Species and Medicinal plants (1-3) were consulted. For Economic Botany and for the Journal of Ethnopharmacology, volumes of the most recent fifteen years were consulted. One study (Amorozo and Grly 1988), published in a Brazilian periodical, was obtained from one of the authors. The calculation of the Shannon-Wiener index was made through the formula H' = - E pi log pi (both base e and 10), where pi is the proportion of individuals of the ith species (here the number of citations or informants per species). Evenness is given by H'/Hmax (Magurran 1988; Pielou 1975). Shannon-Wiener is an information theory index, based on the idea that diversity in a natural system can be measured as information contained in a message. We used the program called Specdiv 1.3 based on Brower and Zar (1977) which calculates many different diversity indices. For the rarefaction method, sub-samples of individuals of equal size are taken at random from the total. By rarefying the samples they are made comparable (Ricklefs 1990). The formula is given in Magurran (1988) as: E(S) = E{ 1 - [(Nn - pi)/(N)]}, where, E(S) = expected number of species in the rarefied sample n = standardized sample size N = total number of individuals recorded in the sample to be rarefied p i = the number of individuals in the ith species in the sample to be rarefied. For the rarefaction curves, T. Lewinsohn program (Unicamp, Brazil) which is based on Shinozaki (1963 in Mueller-Schaerer, Lewinsohn, and Lawton 199l) was used, being pi (x axis) sampling units (in this case, number of informants). We also used the rarefaction program from Krebs (1989), in this case pi (x axis) being the number of individual species cited (number of citations of all informants). These diversity indices are used because they may complement each other and are well
[VOL. 50
known, widely used, and easily found in the ecological literature. The Shannon-Wiener index allows us to compare the diversity of plants cited taking into consideration the relative abundance of citations. For example, two different populations each citing 5 useful species, with a total of 100 citations, have the same diversity apparent. However, looking at the relative abundance of citations, we notice that one population cited the 5 species in the proportion 20:20:20:20:20, and the other in the proportion 80:5:5:5:5, the later showing a lower diversity or, a high dominance on the use of plant species. The rarefaction curves allow us to compare the diversity of plant uses by different populations with different sample sizes (Peet 1974). With these two methods we expect to be able to examine on a broad scale the uses of plants by different cultures. RESULTS AND DISCUSSION A few studies with the number of informants per plant, or per plant use, were found in the literature. Similar observations were made by Etkin (1993) and by Johns, Kokwaro, and Kimanani (1990). The importance of collecting and showing quantitative data is that they allow more confidence in results and, allow the study of other aspects of ethnobotany. Most of the studies found in the literature are from tropical areas, which show a high biodiversity. Seven studies are from Latin America (three from South America), two from Asia and one is from Europe. Some ethnopharmacological studies showed the number of informants per remedy (not per species) and are not part of this study. COMPARATIVE STUDIES AND SITES The following descriptions include general remarks on the area and populations studied, and more particular remarks on the methodology used, in order to explain how the data were applied to calculate the indices. The abbreviations used in figures are shown in brackets. 1) Brazil [Ba]. This study (Kainer and Duryea 1992) was carried out at the Extractive Reserve Cachoeira, located 50 km from Xapurf, in the State of Acre, Brazil. It is a reserve of 29 974 ha, including 68 families. The procedures included interviewing 14 women concerning the use of plants for food, beverages, spices, medicines, animal feed, firewood, construction materials, and other uses. The percent of type of
1996]
BEGOSSI: ECOLOGICAL METHODS IN ETHNOBOTANY
usage per species is shown, but more than one use was mentioned per informant. To calculate the diversity indices, the number of informants per plant species was obtained from the percentages. When more than one use was shown (such as food and medicine) the highest percentage was considered (35 cases out of 145). Naturally, when more than one use was shown, and one of these was 100%, we could be sure that all informants mentioned that species. Data are shown for 145 species (1284 individuals = citations). 2) Brazil [Bp]. This is a study (Amorozo and G61y 1988) on the use of medicinal plants in two caboclo villages of the lower Amazon at Barcarena Municipality (Itupanema and Nova do Piry), with about 80 families in each. Barcarena is located about 30 km from Bel6m (Pardi State), and the villages are located at Maraj6 Bay. Interviews were conducted in 10 houses from both villages (17 informants). Data on species per informant were shown for 34 plant species, cited by more than 3 informants. One informant per species was considered for the other 186 species. The total number of plants species was of 220 (365 citations). 3) Catalonia [Ca]. This study (Bonet, Blanch6, and Xirau 1992) was conducted in an area of 260 km 2, with 18 000 inhabitants in the valley of the river Tenes, near Barcelona. The area is highly variable, allowing for different micro climates and for a rich flora. Interviews were conducted with 28 elderly informants on medicinal plants. The frequency of citation per plant use was available for 30 species. Thirty-three plants were mentioned by two informants and 128 plants were mentioned by one informant. For the calculations, the frequency of uses cited per species was summed (30 species). A total of 191 species (393 citations) were used in the calculations. 4) Guatemala. This study (Gir6n et al. 1991) was conducted at the Municipality of Livingston, Department of Izabal, Atlantic Coast, where 300 caribs were interviewed by a multidisciplinary team. The Municipality of Izabal has 1940 km 2 and 18 811 inhabitants, where different ethnic groups are found, the Caribs being the most frequent. The frequency of citations per use and per plant is shown for 103 plants (1008 citations). As for other studies, when more than one use per plant was found, the total was considered to calculate diversity. For that case, only
283
the Shannon-Wiener index was obtained from the programs. 5) Mexico [Me]. This is a study (Benz et al. 1994) on the Sierra de Manantlan Biosphere Reserve, Jalisco-Colima, Western Mexico, in nine different Mestizo communities (about 100 informants). The area has a high diversity, such as 2500 species of plants and 668 species of vertebrates. Data on local vegetation types are available as well as on floristic composition of the area. Interviews included the use of plants for food, firewood, fence posts, construction, forage, instruments and medicines. Data on 335 plants were shown (2727 citations). The number of reports (citations) per plant species was available. 6) Nicaragua [Na]. Medicinal plants of Nicaragua's Atlantic coast, by Barrett (1994). Five communities (809 households) in the Atlantic coast of Nicaragua were studied: Bluefields, Rama Cay, Kakabila, Orinoco and Karawala. The number of informants per species was available. A total of 149 plants (1659 citations) were used in the study. The rarefaction curve could only be obtained through Krebs (1989) program, because of the high number of informants. 7) Nicaragua [Ni]. Also in the Atlantic coast, this study (Dennis 1988) was carded out at Awastara, close to Puerto Cabezas. Data were obtained from 16 Miskito informants (23 plants, 43 citations). The number of informants per plant is available. 8) Peru [Pe]. This study (Phillips et al. 1994) was conducted at Madre de Dios, a Tambopata reserve of 5500 ha. Data were gathered from seven plots of 1 ha each, by interviewing 29 informants. Plant uses were categorized by food, construction, commerce, medicinal, technology and crafts. This study calculates the use value of species and data on local vegetation is available. The diversity was calculated from a detailed appendix, with 472 plant species (1383 citations). 9) Thailand [Th]. This study (Pake 1987) among Hmong refugees took place at Phanatnikhom refugee camp, 90 km southeast of Bangkok. The informants were nine herbalists. Data were shown in detail for plants mentioned by two herbalists (20 species). Another 130 species were mentioned by one herbalist. Calculations included 150 species (170 citations). 10) Tonga [To]. The Kingdom of Tonga is located in the South Pacific, close to the Fiji and Samoa archipelagoes. Interviews with 50 Tonga
284 T A B L E 1.
ECONOMIC BOTANY
[VOL. 50
DIVERSITY INDICES (RICHNESS AND S H A N N O N - W I E N E R ) ON PLANT USES FROM SELECTED STUD-
IES. USE: M
= MEDICINAL,
G
= GENERAL,
VARIOUSUSES.
Diversity indices Shannon-Wiener Richness
(b.10)~
(b.e)2
Evenness
Local
Use
Informants
Citations
145 220 191 103 335 149 23 472 150 105
2.09 2.20 2.10 1.70 2.24 1.69 1.30 2.58 2.16 1.95
4.80 5.07 4.82 3.90 5.16 3.88 2.99 5.95 4.97 4.49
0.97 0.94 0.92 0.84 0.89 0.78 0.96 0.97 0.99 0.97
Brazil Brazil Catalonia Guatemala Mexico Nicaragua Nicaragua Peru Thailand Tonga
G M M M G M M G M M
14 17 28 300 100 809 16 29 9 50
1284 365 393 1008 2727 1659 43 1383 170 2037
Sources
Kainer and Duryea 1992 Amorozo and G61y 1988 Bonet et al. 1992 Gir6n et al. 1991 Benz et al. 1994 Barrett 1994 Dennis 1988 Phillips et al. 1994 Pake 1987 Whistler 1991
Base 10. 2 Base e.
herbalists were carded out and a list of 105 species (2037 citations) with the number of informants per species is shown (Whistler 1991). In Table 1, the Shannon-Wiener index is shown for the communities mentioned above. Some calculations were made using the number of citations per species, other were based on the number of informants, depending on what kind of data were available. The best procedure would be to calculate the indices with either the number of citations or the number of informants, for all communities. In ethnobotanical studies, results calculated in both ways would be probably highly positively correlated: a plant mentioned by many informants would have more citations, too. Thus, I believe the samples are still comparable. The Shannon-Wiener index, which includes both richness and evenness, is high for communities from Peru, Mexico, Brazil (Pardi State) and Thailand. The communities with a high evenness are Thailand, Peru, Brazil (Acre State) and Tonga (Table 1). These results show that both Peru and Thailand were sites with a high diversity of plant uses, in terms of both richness and evenness. The rarefaction curves (Fig. 1, 2, 3) confirm the high diversity of plants uses at Peru and Thailand, as well as Brazil and Mexico, in spite of the different sample sizes taken from these different populations. In Peru the sample included the use of 472 species by 29 informants and in Thailand, it included the use of 150 species by 9 informants (Table 1). The community in Mexico shows a high
Shannon index and a relatively high rarefaction curve, due to the number of species and of informants (its evenness is moderate--Table 1). For both Mexico and Peru, the authors (Benz et al. 1994; Phillips et al. 1994; respectively) stressed the high diversity of the vegetation in the study sites: 9 Tambopata forest sites at Peru, and 8 vegetation types at Biosphere Reserve in Mexico. Low diversity indices are shown for the populations at Nicaragua, Guatemala (a low evenness), Catalonia, and Tonga (a low rarefaction curve). Catalonia is located in a temperate region. In spite of a local highly diversified flora (Bonet, Blanch6, and Xirau 1992), also expected in Mediterranean shrub communities (Ashton 1992), we can expect a relatively low diversity at Catalonia sites compared to rain forest sites. The expected decrease in species diversity from the low latitude tropics to high latitudes was called the "master gradient of diversity" by Whittaker (1975) and is analyzed in detail by Latham and Ricklefs (1993). The Kingdom of Tonga includes 150 islands (697 km2), of which only 36 are inhabited (Whistler 1991). According to the classical island biogeography theory (MacArthur and Wilson 1967), the diversity of species should increase according to the island area, and decrease according to its distance from the mainland (see Rosenzweig 1 9 9 5 : 2 1 0 - 2 6 3 for a review). Therefore, small or distant islands should have lower diversity than inland communities; moreover, low diversity is expected in both far and
1996]
BEGOSSI: ECOLOGICAL METHODS IN E T H N O B O T A N Y
285
600 r~
500
Pe
ii II /I
400
I
I
r~
Me 1
300
/ /
_+-...... -
_._......
!
i
200 IO0
Bp
I/
..........
, .."'"'"** Ca
/ /
/ , . , " ~ / ,-+
;-/Ba
Th ,1
.........................
TO
I,'/11~ .................. /
O
H-II N l ~ i / l L / i i
1
I N l i l t t l t l l l I I I l l I1111 I l i l l III Iq I I ] l l l l l
10
20
40
30
I1111 II III ]1 III It t l l t I I I I ] 1 ] 1 1 1 1 I I I l l l l l
50
60
70
80
ITqlll
90
1
O0
NUMBER OF INFORMANTS Fig. 1. Rarefaction curves based on the number of informants per plant. Pe = Peru, Th = Thailand, Bp = Brazil (Par~ State), Me = Mexico, Ba = Brazil (Acre State), Ca -- Catalonia, To = Tonga, Ni = Nicaragua (Puerto Cabezas).
600 500
Pe
.....
i~ 4O0 ) 300
/// // /
i
/
/
//
/*
./ /
200 /
100
/
/
Th
+ . ~ " .J ;.....................
.......................
........1
..Bp ~Ca
~.- ~-'~-
Ba
..............
~'"
/
0
,.,--,--,--,--,--,--,
1
5
I
I
10
l
I
t
I
I
15
I
]
I
I
I
I
I
[
20
]
I
25
I
I
I
I
1
30
NUMBER OF INFORMANTS Fig. 2. Rarefaction curves based on the number of informants, up to 30 informants. Pe = Peru, Th = Thailand, Bp = Brazil (Pardi State), Ba = Brazil (Acre State), Ca = Catalonia, Ni = Nicaragua (Puerto Cabezas).
286
[VOL. 50
ECONOMIC BOTANY
4OO
300
r.~ .....j'""
i
200
./ ,/'/'
-- IIIla
/
100
./ ,," J /,p,-
0
-I
I
I
300
600
J
[
I
I
I
I
900 1200 1500 1800 2100 2400
NUMBER OF CITATIONS Fig. 3. Rarefaction curves based on the number of citations per plant. For Mexico and Tonga the curve was calculated through both number of informants and citations (Fig. 1 and 2). Me = Mexico, Na = Nicaragua (five communities in the Atlantic coast), and To = Tonga.
small islands. Begossi, Leit~o-Filho, and Richerson (1993) and Figueiredo, Leit~o-Filho, and Begossi (1993) collected and analyzed data on the use of medicinal plants in two different islands from the Atlantic Forest coast: B6zios and Itacuruq~i islands. Btizios has an area of 7.5 km 2 and is located 24 km off the coast whereas Itacuruq~i has an area of 8.3 km 2 and is 0.8 km off the coast. The Shannon-Wiener index for the use of medicinal plants at Btizios is 1.47 and for Itacumq~i 1.65 (base 10). A statistical method to test the significance of differences between two Shannon index is available in Zar (1984) and Magurran (1988), and the differences were significant (P < 0.05). A study using rarefaction curves to compare the use of plants in 5 Atlantic Forest coastal communities showed also a lower diversity in the island communities (Rossato, Leitao-Filho, and Begossi in prep.). The low diversity of plant uses found at Tonga may be a representation of its low floristic diversity. The rarefaction curves (Fig. 1, 2, 3) allow evaluation in sampling efforts. How many informants should be enough to analyze the diversity of plants used in a population? When more informants are added and the curve does
not continue to increase, it is because the sampiing was sufficient. A very good effort is observed for the data collected at Tonga and Mexico. Thus, the low diversity of the plants used at Tonga is not due to incomplete sampling. On the other hand, the comparatively low effort made in Thailand means that it probably would show a still higher diversity of plants used. Peru, Brazil (Pardi), and Mexico showed a high diversity. We should recall that the plant uses in these areas were for general purposes, whereas in the other areas the focus was only on medicinal plants. The diversity of plant use in Peru and Mexico might be partially explained by more species being added to the list for a variety of different uses, which is not the case for the high diversity found at Thailand. In Thailand, Pake (1987) stressed the use of local plants by people and their tenacity in retaining their indigenous medicine as well as their low utilization of Western medicine. Brazil (Pani) shows a high diversity in spite of a lower sampling effort, compared to the study at Acre (Fig. 1, 2). This diversity is still high if we take into consideration that only medicinal uses were studied at Pani State.
1996]
BEGOSSI: ECOLOGICAL METHODS IN ETHNOBOTANY
These indices may also help us to see the homogeneity/heterogeneity aspects on the use of plants "inside" a population. Less intra-homogeneity means a high diversity of plant uses and a high intra-homogeneity (high dominance or low diversity) mean that just the most important plants are being mentioned by the informants. Thus, a low diversity on the use of plants may signify some loss of local knowledge, assuming that the original diversity was high. Also, an isolated population living in an area with a low floristic diversity may show a higher diversity of plant uses in comparison with a less isolated population that lives in an area of high diversity, but with access to Western medicine. Diversity indices allows us also to compare the use of plants by different categories. For example, Figueiredo, Leit~o-Filho, and Begossi (1993) compared the use of plants among men/ women; fishermen/non-fishermen and old/ young, using the comparative method of Zar (1984). Significant differences in the ShannonWiener index were found for women and older people, who cited comparatively more medicinal plants. Women usually make plant medicines for the children (such as syrups) and the older people usually have a deeper knowledge of plants. Moreover, some loss of knowledge of medicinal plants is occurring in many communities, especially among the young generations as already observed by Anderson (1986) in Thailand and by Begossi, Leitao-Filho, and Richerson (1993) at Btizios Island (Brazil). Finally, data on the diversity of plants used may help in the management of conservation areas. When estimating the diversity of plants used, we can compare it to the available diversity (vegetation or fioristic diversity) and get a minimum idea of the minimum area needed for some native populations. The estimation of minimum living areas are very important in order to guarantee the survivorship of many indigenous populations in both tropical and temperate areas. CONCLUSIONS Diversity indices are useful tools that help us to ask questions and analyze ethnobotanical data, besides allowing comparisons among different communities in different or similar environments. Intra-populational comparisons, such as among different categories, may be also drawn from data. The Shannon-Wiener index gives us an esti-
287
mation of the richness and evenness, whereas the rarefaction curves allow us to compare populations with different sample sizes and to evaluate sampling effort. The results of this study show high indices for high diversity areas, such as Peru, Brazil, Mexico and Thailand. Tonga showed a low diversity, probably reflected by its island environment. The diversity of Thailand could be higher after a stronger sampling effort. Thailand's high diversity may be the result of local resistance to Western medicine, in addition to an expected high local diversity (no data on local biodiversity is available from that study). It is important to have quantitative studies in ethnobotany reporting data on informants, because those data can be very useful for macro scale comparisons. These macro scale studies are particularly important when we observe that both biological and cultural biodiversity are seriously threatened in many parts of the world. ACKNOWLEDGMENTS I am grateful to the grant received from FAPESP-Brazil, which covered all my travel expenses and made possible my stay at Harvard University, Botanical Museum and Herbaria, in June-July 1995, where this study was undertaken. I thank the CNPq-Brazil for a research scholarship. I thank Silvia C. Rossato for helping me with all diversity computer programs and Efraim Rodrigues for a careful reading of this manuscript. I thank Paulo Moutinho for ideas while studying ants (Master's thesis, Unicamp). I am grateful to the Economic Botany Library at Harvard University, for helping with facilities for this study, including computers, and to Donald Pfister, Director of Harvard Herbaria. Finally, I am especially thankful to Prof. Richard E. Schultes, Director Emeritus of the Botanical Museum at Harvard, for this special opportunity of collaboration and also for kindly reviewing this manuscript.
LITERATURE CITED Amorozo, M. C. de Mello, and A. G~ly. 1988. Uso de plantas medicinais por caboclos do Baixo Amazonas, Barcarena, Par~, Brazil. Boletim do Museu Paraense Emilio Goeldi, S6rie Botanica 4:47-130. Anderson, E. F. 1986. Ethnobotany of Hill tribes of northern Thailand. II. Lahu medicinal plants. Economic Botany 40:442-540. Ashton, P. S. 1992. Species richness in plant communities. Pages 4-22 in R L. Fielder and S. K. Janin, eds., Conservation biology. Routledge, Chapman and Hall, Inc., New York. BaiZe, W. 1989. The culture of Amazonian forests. Advances in Economic Botany 7:1-21. Barrett, B. 1994. Medicinal plants of Nicaragua's Atlantic coast. Economic Botany 48:8-20. Begon, M., J. L. Harper, and C. R. Townsend. 1990. Ecology, individuals, populations and communities. Blackwell Scientific Publ., Boston. Begossi, A. 1992. The use of optimal foraging in the understanding of fishing strategies: a case from Se-
288
ECONOMIC BOTANY
petiba Bay (Rio de Janeiro State, Brazil). Human Ecology 20:461-475. 1995. Cultural and ecological resilience among cai~aras of the Atlantic Forest coast and caboclos of the Amazon (Brazil). Presented at the Fifth Annual Common Property Conference, May 24-28, Bodo, Norway, Chapter 2 in forthcoming book: Linking ecological and social systems for resilience, C. Folke and E Berkes, eds. Cambridge University Press. - - . n.d. The application of ecological theory to human behavior: niche, diversity and optimal foraging. Proceedings of the Seventh International Meeting of the Society of Human Ecology. Society for Human Ecology, East Lansing, MI. - - , and P. J. Richerson. 1992. The animal diet of families from Btlzios Island (Brazil): an optimal foraging approach. Journal of Human Ecology 3: 433-458. , and - - - . 1993. Biodiversity, family income and ecological niche: a study on the consumption of animal foods on Btizios Island (Brazil). Ecology of Food and Nutrition 30:51-61. , H. F. Leit~o-Filho, and P. J. Richerson. 1993. Plant uses in a Brazilian fishing community (Btlzios Island). Journal of Ethnobiology 13:233256. Benz, B. E., F. M. Santana, R. L. Pineda, J. E. Cevallos, L. H. Robles, and D. L. Niz. 1994. Characterization of mestizo plant use in the Sierra de Manantlan, Jalisco-Colima, Mexico. Journal of Ethnobiology 14:23-41. Berkes, F., and C. Folke. 1992. Linking social and ecological systems for resilience and sustainability. Beijer Discussion Papers 52. The Royal Swedish Academy of Sciences, Stockholm. Bonet, M. A., C. Blanche, and J. V. Xirau. 1992. Ethnobotanical study in River Tenes valley (Catalonia, Iberian Peninsula). Journal of Ethnopharmacology 37:205-212. Boom, B. M. 1990. Useful plants of the Panare Indians of the Venezuelan Guayana. Advances in Economic Botany 8:57-76. Brower, J. E. and J. H. Zar. 1977. Field and laboratory methods for general ecology. Wm. C. Brown, Dubuque, Iowa. Carneiro, R. L. 1978. The knowledge and use of rain forest trees by the Kuikuru Indians of Central Brazil. Pages 201-216 in R. I. Ford, ed., The nature and status of ethnobotany, Anthropological Papers no. 67. University of Michigan, Ann Arbor. Dennis, P. A. 1988. Herbal medicine among the Miskito of Eastern Nicaragua. Economic Botany 42: 16-28. Etkin, N. L. 1993. Anthropological methods in ethnopharmacology. Journal of Ethnopharmacology 38:93-104. Fearnside, P. M. 1986. Human carrying capacity of
[VOL. 50
the Brazilian rainforest. Columbia U. Press, New York. Figueiredo, G. M., H. F. Leit'5o-Filho, and A. Begossi. 1993. Ethnobotany of Atlantic Forest coastal communities: diversity of plant uses in Gamboa (Itacuruq~i Island, Brazil). Human Ecology 21:420430. Gir6n, L. M., V. Freire, A. Alonzo, and A. C~iceres. 1991. Ethnobotanical survey of the medicinal flora used by the Caribs of Guatemala. Journal of Ethnopharmacology 34:173-187. Glover, P. E., F. C. Magogo, and A. B. H. Mzee. 1969. A list of Digo plant names with their botanical equivalents. Kenya National Parks, Nairobi. Hames, R. B., and W. T. Vickers. 1982. Optimal diet breadth theory as a model to explain variability in Amazonian hunting. American Ethnologist 9: 358-379. Hardesty, D. L. 1975. The niche concept: suggestions for its use in human ecology. Human Ecology 3:71-85. Hawkes, K., H. Kaplan, K. Hill, and A. M. Hurtado. 1987. Ach6 at the settlement: contrasts between farming and foraging. Human Ecology 15:133162. Hill, K. 1988. Macronutrient modifications of optimall foraging theory: an approach using indifference curves applied to some modem foragers. Human Ecology 16:157-197. Huston, M. A. 1994. Biological diversity. Cambridge U. Press, Cambridge, UK. Johns, T., J. O. Kokwaro, and E. K. Kimanani. 1990. Herbal remedies of the Luo of Siaya District, Kenya: establishing quantitative criteria for consensus. Economic Botany 44:369-381. Kainer, K. A., and M. L. Duryea. 1992. Tapping women's knowledge: plant resource use in extractive reserves, Acre, Brazil. Economic Botany 46: 408-425. Krebs, C. J. 1989. Ecological methodology. Harper and Row Pub., New York. Latham, R. E. and R. E. Ricklefs. 1993. Pages 294314 in R. E. Ricklefs and D. Schluter, eds., Species diversity in ecological communities, historical and geographical perspectives. The University of Chicago Press, Chicago. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton U. Press, Princeton. Magurran, A. 1988. Ecological diversity and its measurement. Princeton University Press, Princeton. May, R. M. 1975. Patterns of species abundance and diversity. Pages 81-120 in M. L. Cody and J. M. Diamond (eds.), Ecology and evolution of communities. The Belknap Press of Harvard U. Press, Cambridge. Moerman, D. E. 1991. The medicinal flora of native
1996]
BEGOSSI: ECOLOGICAL METHODS IN ETHNOBOTANY
North America: an analysis. Journal of Ethnopharmacology 31:1-42. Mueiler-Sehaerer, H., T. M. Lewinsohn, and J. H. Lawton. 1991. Searching for weed biocontrol agents--when to move on? Biocontrol Science and Technology 1:271-280. Pake, C. V. 1987. Medicinal ethnobotany of Hmong refugees in Thailand. Journal of Ethnobiology 7: 13-26. Peet, R. K. 1974. Measurement of species diversity. Annual Review of Ecology and Systematics 5:285307. Phillips, O., and A. H. Gentry. 1993a. The useful plants of Tambopata, Peru: I. Statistical hypothesis tests with a new quantitative technique. Economic Botany 47:15-32. - - , and . 1993b. The useful plants of Tambopata, Peru: II. Additional hypothesis testing in quantitative ethnobotany. Economic Botany 47: 33-43. , --, C. Reynel, P. Wilkin, and C. B. Gtiivez-Durand. 1994. Quantitative ethnobotany and Amazonian conservation. Conservation Biology 8:225-248. Pielou, E. C. 1975. Ecological diversity. John Wiley and Sons, New York. Prance, G. T., W. Bal6e, B. M. Boom, and R. L. Carneiro. 1987. Quantitative ethnobotany and the case for conservation in Amazonia. Conservation Biology 1:296-310. Rieklefs, R. E. 1990. Ecology. W. H. Freeman and Company, New York.
289
Rosenzweig, M. L. 1995. Species diversity in space and time. Cambridge U. Press, Cambridge. Sanders, H. L. 1969. Benthic marine diversity and stability-time hypothesis. Pages 71-81 in G. M. Woodwell and H. H. Smith, eds., Diversity and stability in ecological systems. Brookhaven Symposia in Biology no. 22. U.S. Department of Commerce, Springfield. Schluter, D. and R. E. Rieldefs. 1993. Species diversity. Pages 1-10 in R. E. Ricklefs and D. Schluter, eds., Species diversity in ecological communities, historical and geographical perspectives. The University of Chicago Press, Chicago. Simberloff, D. 1979. Rarefaction as a distributionfree method of expressing and estimating diversity. Pages 159-176 in J. E Grassle, G. P. Patil, W. Smith, and C. Taillie, eds., Ecological diversity in theory and practice. International Co. Pub. House, Faifland. Whistler, W. A. 199l. Herbal medicine in the kingdom of Tonga. Journal of Ethnopharmacology 31: 339-372. Whittaker, R. L. 1975. Communities and ecosystems. Macmillan Pub. Co., Inc., New York. Wing, E. S., and S. R. Wing. 1995. Prehistoric ceramic age adaptation to varying diversity of animal resources along the West Indian Archipelago. Journal of Ethnobiology 15(1): 119-148. Zar, J. H. 1984. Biostatistical analysis. Prentice-Hall, Inc., Englewood Cliffs.
BOOK REVIEW Major Weeds of Mongolia. R. K. Malik and D. Tsedev. 1996. Food and Agriculture Organization of the United Nations, Viale delle Terme di Caralla, 00100 Rome, Italy. ix + 101 pp., illus. (paperback). Price not given. In my mind, Mongolia is a mysterious place. Mention of it conjures up visions of yak butter, nomadic herdsman, and the original collection of a dinosaur egg in the Gobi Desert. Botanists and agriculturalists know so little of Mongolia that we are fortunate to have this small, attractively produced book with high quality printing. The plan of the book is simple. For each of the 41 species, there are English and Mongolian common names, a short technical description in both English and Russian, phenology, distribution, related species, and control measures. While several herbicides are noted, I was encouraged to find several non-chemical control measures recommended. Different aspects of the plants are pic-
tured. The quality of the original pictures is not always excellent. A glossary of terms in English and Russian and an index (inexplicably only in English) conclude the book. I could not find callus, lyrate or nut in the glossary but the remainder of the terms are included with clear, simple wording. Most of the weeds will be familiar to North American botanists with such genera as Agropyron, Amaranthus, Arena, Chenopodium, Cirsium, Elymus, Euphorbia. Plantago, Polygonum, Potentilla, and Taraxacum represented. In recent years, FAO has produced helpful guides to the weed flora and control of weeds. This latest is a commendable addition and will be essential to anyone interested in the flora and agriculture of Mongolia. LYTTONJ. MUSSELMAN DFJ'ARTMENTOF BIOLOGICALSCW24CES OLD DOMINIONUNIVERSrrY NORFOLK,VA 23529-0266
