Introduction

Rural livelihoods are based on a diversified production strategy (Ellis 1998). In rural Mexico small-scale farmers have adopted different means of diversification such as rural self-employment, salaried jobs, manufacturing of handmade crafts, and national or international migration, which combine with other subsistence activities including agriculture, forestry, hunting and homegardens (González-Jácome 2003). Family homegardens in the tropics are socio-ecological production systems that involve a diversified management of plants and animals for farmer subsistence (Kumar and Nair 2004). The management of homegardens contributes to the maintenance of a high biodiversity and acts as a local diversity reservoir (Aguilar-Støen et al. 2009; Castiñeiras et al. 2001; Jarvis et al. 2007). Homegardens supply farmers with essential goods such as food, medicinal herbs and timber, as well as occasional income from selling homegarden products. The homegardens' contribution to the farmers' livelihoods is critical for marginalized communities (Fernandes and Nair 1986; Montagnini 2006). The benefits of tropical homegardens are influenced by biophysical and social factors where the traditional ecological knowledge of different ethnic groups plays a key role in the botanic composition, plant diversity and structure of the homegardens (Caballero 1992; Netting 1993; Vogl et al. 2002). Among these groups, the Maya in the lowlands of the Yucatan Peninsula maintain a high diversity of plant species in their homegardens within a complex vertical and horizontal structure that mimics the natural ecosystem and often contains 300 to 500 species of plants and animals (De Clerck and Negreros-Castillo 2000; Rico-Gray et al. 1990; Toledo et al. 2008). Based on the ecological knowledge of the local environment and the specific needs of plants and animals, farmers combine practices for soil management, nutrient conservation, and water and crop protection (Barrera-Bassols and Toledo 2005; Netting 1993). However, over the last decades due to global socioeconomic changes and migration, the composition and diversity of plants found in Mayan homegardens has changed, resulting in a decrease in species cultivated for food and medicinal purposes and an increase in ornamentals (Hernández 2010). Diversity is lost as migrating families lose interest in maintaining a variety of plant species. Also, families permanently adjust their mode of life in response to social, economic, and ecological changes, which may lead to the decline of the family homegarden (Corzo-Márquez and Schwartz 2008).

Global socioeconomic changes as well as structural reforms in land tenure made by the Mexican government at the end of the 1970s and early 1980s forced the migration of indigenous and non-indigenous (Mestizo) farmers to what were then almost uninhabited forests in southeastern Mexico (Kelly 2001; Redclift 1983). Mestizo refers to a person of mixed racial ancestry, especially mixed European and Native American ancestry (Larousse 2004). In the present study the term Mestizo is used for migrant farmers from the Mexican states of Veracruz and Tabasco. The migratory movements to the southern Yucatan Peninsula led to a concentration of diverse ethnic groups encompassing a wide bio-cultural diversity in the region. In the last decade, immigrants represented 20% to 40% of the total population, with 16% being of Maya, Chol, Kanjobal, Tzeltal, Nahuatl, and Tzotzil background. The Mestizo population, representing 26% of immigrants, came mainly from the Mexican states of Tabasco, Chiapas, Veracruz, and Michoacán (López-Villar 2005). With the immigration and settlement in the forest, families adapted to the unfamiliar environment by implementing various subsistence strategies that relied on the intensive use of natural resources, notably through homegardens (Guerrero-Peñuelas 2007; López-Villar 2005; Wilk 1997). Immigrants tested different strategies, growing native plant species or those brought from their homeland, to guarantee their food supply and economic income. As a result, the homegardens of immigrants were more diverse and productive than those of local people (Guerrero-Peñuelas 2007; Shrivastava and Heinen 2005). Still, in some cases (e.g., Tzotzil Maya in Chiapas, Mexico, and migrant families in Petén, Guatemala), immigration caused the loss of species, abandonment of traditional practices and less use of homegardens (Atran et al. 2002; Corzo-Márquez and Schwartz 2008; Lipton 1980). Similarly, Guerrero-Peñuelas (2007) found that migration in the state of Mexico led to a general decrease of homegarden plant diversity, in particular of food plants. Our research explores and compares the homegardens of Maya and Mestizo migrants to the region of Calakmul, Campeche. We propose that because of their different cultural and geographic backgrounds and ethnobotanical knowledge, Maya and Mestizo smallholders would show differences in the composition and plant diversity in family homegardens.

Study Area

The research was conducted in the state of Campeche, Mexico, located in the southern Mexican frontier zone (Figure 1). The sparsely populated state (14 inhabitants/km²) has 174,853 inhabitants; 23.1% of the total population belongs to indigenous groups (Navarrete-Linares 2008). The Calakmul municipality encompasses 13,839 km² and has a population of 26,882 (INEGI 2011). The Calakmul Biosphere Reserve (CBR) covers the majority of this municipality (7231.85 km²) and is divided in two nuclear zones and one buffer zone (CONANP 2008).

Figure 1.

Location of the study villages in the Calakmul Biosphere Reserve (CBR), State of Campeche, Mexico.

The average annual temperature of the humid subtropical climate is 24.6° C and the annual average precipitation in the municipality of Calakmul ranges significantly from 552 mm in the north to 1,634 mm in the south (INEGI 2011; Martínez and Galindo-Leal 2002). The rainy season lasts from June to November followed by the dry season from December to May. Trade winds and hurricanes occur mainly at the end of the rainy season. The karstic parent material of the Yucatan Peninsula allows only few permanent sources of surface water. A high level of evapotranspiration, in addition to ground water levels that reach depths of 150 m to 200 m are factors that influence acute seasonal water scarcity and hence make water a limiting factor for the region (Morales-Rosas 1999). The undulating landscape is part of a central plateau (Meseta de Zoh-Laguna) and reaches elevations of up to 350 m above sea level in the south (Gates et al. 1999; Martínez et al. 2001; Morales-Rosas 1999).

The CBR, together with the neighboring protected zones in Chiapas, the Petén of Guatemala and Belize, make up the second largest adherent tropical forest in the Americas (CONANP 2008). The seasonal tropical forest of Calakmul forms an ecocline between the dry northern Peninsula and the humid Petén of Guatemala in the south, constituting the Mesoamerican Biological Corridor. This forest shows spatial heterogeneity in the vegetation, with no species or vegetation type being dominant (Martínez and Galindo-Leal 2002; Turner II 2010; Vester et al. 2007).

In the zone of Calakmul, impressive architectural relics from the Mayan culture remain, dating back to the Late Pre-Classic (200 B.C.) and the Late Classic (A.D. 800) periods when the region was probably densely populated (Gates et al. 1999). Ancient Maya agricultural systems modified the natural landscape of the tropical forest (Zarate Hoyos 1998). The decline of the Mayan culture was followed by centuries of human abandonment of this region, until the end of the 1970s, when a second wave of settlement started in the Calakmul region (García-Gil and Pat-Fernández 2000). People migrated to this region due to scarcity of land, poverty and unemployment as well as social, cultural, political or economic factors that led to conflicts in other regions in Mexico (López-Villar 2005). The settlers emigrated from 23 different Mexican states bringing various languages, cultural identities and geographic backgrounds to the Calakmul region (Alayón-Gamboa and Gurri 2007). Because of its biological and cultural diversity, this region is considered to be a hot spot of bio-cultural diversity (Stepp et al. 2008). Indigenous languages here include Yucatec Maya (80.9%), Maya Ch'ol (9.4%), Kanjobal (2.0%) and Tzeltal (1.8%), among others (CNDI-PNUD 2000; López-Villar 2005). The last 40 years were also characterized by significant human disturbances to the local ecosystem such as increased settlement, the expansion of livestock farming and the intensification of agriculture, all leading to high rates of deforestation and thus posing a threat to biodiversity (Turner II 2010; Vester et al. 2007).

This study was conducted in four communities that exemplify the region's different cultural backgrounds. Cristóbal Colón (CC) and Narciso Mendoza (NM) are villages with Mestizo inhabitants from Acayucan, Xalapa, Puerto de Veracruz and San Andrés Tuxtla, Veracruz (CC), and Comalcalco, Tabasco (NM) (Table 1). Coming from a humid tropical region, these Mestizos have a broad knowledge of the biological diversity commonly found in tropical ecosystems. El Carmen II (EC) and Niños Héroes (NH) are Mayan villages whose inhabitants came from Salto de Agua and Tila (EC) and Salto de Agua and Tumbalá (NH) in the state of Chiapas where the climate is temperate and the biological resources different than those found in humid subtropical Calakmul. Maya Ch'ol is the native language used amongst these villagers, and Spanish is primarily used by men in communication with Mestizos. For cultural reasons, women communicate little with people from outside and rarely leave their communities. Therefore, their Spanish skills and mobility are more limited than men's.

Table 1.

Geographic and demographic data for the villages studied in Calakmul, Campeche.

All four villages are located along a principal road 60 km south of the district capital Xpujil and lie within the buffer zone of the CBR (Figure 1). The settlements are organized as ejidos, which are communally managed land units granted by the Mexican government to farmers. This land is allocated for usufruct activities and is distributed individually to families who hold the rights to use these plots to sustain their families' livelihoods (Hinojosa 1983). All families own approximately 40 ha of farmland outside the village, where they practice slash and burn agriculture. The main crops cultivated are maize (Zea mays L.), beans (Phaseolus lunatus L., Phaseolus vulgaris L.), pumpkins (Cucurbita mixta Pangalo, Cucurbita moschata Duchesne, Cucurbita pepo L.) and sweet potatoes (Ipomoea batatas (L.) Lam.). The cultivation of jalapeño pepper (Capsicum annum L.) for commercial sale is also common.

Methodology

The fieldwork was conducted in 2008 from the end of May to the beginning of September. Altogether 20 homegardens, 5 in each village, were selected using the snowball sampling method (Bernard 2006) and each homegarden was visited at least twice. The area of the homegarden was defined as the parcel of land in the village where farmers established their homestead. All the plants and constructional infrastructure inside the limited plot were considered part of the homegarden area. Structured, semi-structured and unstructured interviews were conducted mainly with women, who claimed to be responsible for the management of the homegardens. Interviews collected information on the socioeconomic background of the families, the characteristics and uses of the homegardens, and the preparation of products from the homegardens. Ethnobotanical information was gathered through structured interviews as well as participant and non-participant observation in the homegardens. Only plants with a specific purpose for the farmers were documented and categorized, and their ethnobotanical uses were determined while walking through each homegarden with the interview partner. On site, each plant species was pointed out and its use was requested with the same set of pre-tested questions (Martin 1995; Vogl et al. 2004). In addition, a ranking method, using a set of cards with images of possible functions of the homegardens was employed, where female and male heads of the household evaluated the importance of different functions of the homegardens. Inventories were conducted to record the size, botanical composition and infrastructure of the homegardens. Four voucher specimens of each plant species that could not be identified on site were collected. These samples were identified in collaboration with botanists from the Autonomous University of Campeche (AUC). Modified from Vogl-Lukasser (1998) and based on local characteristics, the authors defined four layers according to the vertical height of plants and into five different horizontal areas (areas close to the house, for ornamental plants, herbs and vegetables, fruit trees and secondary vegetation, Figure 2). Chi-square-tests, analysis of variance, lineal regression, cluster analysis and ranking correlations were carried out using Statistic Package for Social Sciences (Ver. 13.0, SPSS Inc.; Dytham 2001).

Figure 2.

Schematic drawing of horizontal zonal plant distribution in the homegarden: (i) living areas close to the house; (ii) areas for ornamental plants; (iii) areas for herbs and vegetables; (iv) areas for fruit trees; (v) secondary vegetation (Modified from Neulinger 2013).

As the research area lies within the buffer zone of the CBR, a permit for plant collection was obtained from the Ministry of Environment and Natural Resources (SEMARNAT) of the Mexican government. Also, prior consent was obtained from the heads of households –male or female– for their participation in interviews and the use of the ranking method, and for the collection of botanical samples.

Results

Botanical Composition in Maya and Mestizo Homegardens

A total of 310 plant species that possessed one or more uses for farmers were recorded in the 20 homegardens, 275 in the Mestizo communities and 185 in the Maya communities, representing a significant difference (Pα _0.0.5  =  0.02). Of these, 125 plant species appeared exclusively in Mestizo homegardens, with 28 of these found in both Mestizo communities. In comparison, 35 species were recorded exclusively in Maya homegardens, with 5 of these found in both Maya communities. The highest species diversity was found in the Mestizo community of Narciso Mendoza (203 species), and the lowest in the Maya community Niños Héroes (111 species). The variety of species in a single homegarden varied from 32 to 141 species; in Maya homegardens an average of 57.6 species was found whereas in Mestizo homegardens an average of 82.7 species was observed.

Of the 310 plant species found in homegardens, 257 were cultivated and managed consciously by the farmers. The rest were spontaneously growing wild species that were protected or tolerated. The most abundant phenotypes were annual herbs and herbaceous perennials (119 species), followed by trees (93 species) and shrubs (57 species). The rest were succulents, palms, bananas and vines. When comparing the 20 most abundant plant species across Mestizo and Maya villages, it is notable that in the homegardens of Mestizo villages (Table 2) 6 ornamental plants (Aloe vera L., Zephyranthes carinata Herb., Rosa sp., Hibiscus rosa-sinensis, Crinum amabile Donn, and Acanthocereus subinermis) were found, whereas in Maya homegardens (Table 3) 2 ornamental plant species (Catharanthus roseus (L.) G. Don and Yucca elephantipes) were present.

Table 2.

Most ubiquitous plant species in Mestizo homegardens in Calakmul and their uses.

Table 3.

Most ubiquitous plant species in Maya homegardens in Calakmul and their uses.

Hierarchical cluster analysis shows a clear differentiation between Mestizo and Maya homegardens (Figure 3). Their plant compositions form two groups identical to their cultural affiliation, with the exception of garden 2, a Mestizo homegarden more similar to Maya ones. Figure 3 also shows a greater similarity in species composition among Maya homegardens than Mestizo homegardens. However, the functional horizontal zonal division was similar between Maya and Mestizo homegardens. In addition, the age of the sampled homegardens ranged between 1 and 35 years (23.3 years on average), but age was not related to the number of different plant species found (r² _Pearson =  0.038).

Figure 3.

Hierarchical cluster analysis (Ward method) of the floristic composition in the homegardens. The numbers indicate the identity of the homegarden (Mestizo 1–10; Maya 11–20).

Uses of Plant Species

Although the number of plant species per homegarden differed between the cultural groups, the functional distribution of plants within Maya and Mestizo homegardens was similar (Figure 4, χ² P_α0.05  =  0.55). Also, farmers' personal evaluation of the functions of their homegardens did not show any significant variation between cultures or between women and men. Thus the farmers' personal evaluation of the functions of the homegardens is congruent with the observed functional distribution of plant species.

Figure 4.

Functional distribution of the richness of plant species in the homegardens of the villages in Calakmul, Campeche.

Most of the 310 plant species cultivated in homegardens were used for family needs, and only 5.2% were occasionally sold in the village or at local events. Ornamental plants were most abundant (123 species), followed by plants grown for food (114 species), and plants grown for medicinal use (94 species) (Table 4). Farmers often used multiple parts of the plants (leaf, fruit, root tuber, root, bark, and trunk), and most plant species had more than one use. For example, medicinal plants such as papaya (Carica papaya L.), pitahaya (Hylocereus undatus (Haw.) Britton & Rose), and jobo (Spondias mombin), and ornamental plants such as canna (Canna indica L.) and flor de mayo (Plumeria rubra L.) can also be used for food.

Table 4.

Plant species richness by functional category in all sampled homegardens and those found exclusively in Mestizo and Maya communities.

Ornamental plants tended to be placed in similar locations in all homegardens. Farmers often planted patches of ornamentals close to the house and on the side of the homegarden that faced the street, and arranged ornamentals in pots along boundaries. In the Mestizo villages, significantly more (χ² P_α0.05  =  0.001) and different species of ornamental plants were recorded than in Maya villages (108 vs. 65 species respectively); 58 ornamental species were found only in Mestizo homegardens, whereas 15 species were found exclusively in Maya homegardens (Table 4).

Maya and Mestizo families used 114 plant species for sustenance, and 34 food plants appeared exclusively in Mestizo homegardens whereas 8 species were recorded exclusively in Maya homegardens (Table 4). The traditional cuisine of Maya and Mestizo farmers of Calakmul consists of a variety of dishes, often using products from the homegardens like cassava (Manihot esculenta Crantz), ñame (Manihot aesculifolia (Kunth) Pohl), sweet potato (Ipomoea batatas), makal (Xanthosoma yucatanense Engl.), and two species of cactus (Opuntia sp. and Acanthocereus subinermis). Also, spontaneously growing wild plant species, such as verdolaga (Portulaca oleraceae L.), black nightshade (Solanum nigrum L.), and cultivated mustard greens (Brassica juncea (L.) Czern.) were fried and eaten with corn tortillas. The seeds of ramon (Brosimum alicastrum Sw.) were consumed toasted. Other fruit harvested in homegardens (52 species) was eaten raw (e.g., Mangifera indica L., Tamarindus indica L., Citrus grandis, Carica papaya, Ananas comosus (L.) Merr.) or prepared as a drink with water or milk (e.g., Citrus grandis, Pouteria sapota (Jacq.) H.E. Moore & Stearn, Manilkara zapota (L.) P. Royen, Morinda citrifolia L.). Thirty-eight species of vegetables were grown. Sixteen plant species were used as spices, the most common being achiote (Bixa orellana L.), basil (Ocimum basilicum L.), peppermint (Mentha sativa L.), epazote (Chenopodium ambrosioides), Mexican coriander (Eryngium foetidum L.), and Chinese parsley (Coriandrum sativum L.). The flower petals of eight species were used for food and to prepare drinks (Yucca elephantipes, Cucurbita moschata, Cucurbita mixta, Cucurbita pepo, Plumeria rubra, Hexopetion mexicanum (Liebm.) Burret, Hibiscus sabdariffa L. and Sambucus mexicana).

A total of 94 plant species were used as medicine with Citrus grandis, Chenopodium ambrosioides and Spondias mombin being the most common. Of these, 18 species were found exclusively in Mestizo homegardens and 10 species in Maya homegardens (Table 4). Medicinal plants were used to treat 64 physical illnesses and symptoms, as well as spiritual disorders. The most common symptoms and illnesses treated were malaria and dengue fever, gastrointestinal parasites, cough, skin irritations, and insect bites. As malaria and dengue commonly occur in the rainy season, farmers used remedies such as teas and compresses to lower the high body temperatures caused by these illnesses. Also, remedies against spiritual disorders like “bad air,” “bad glance” or “fright in the blood” could be obtained from products of homegardens (e.g., Nicotiana tabacum L., Ocimum basilicum, Petiveria alliacea L.).

To treat symptoms of gastrointestinal parasites, farmers prepared teas, for example from the roots of Amaranthus spinosus L., the leaves of Coccoloba uvifera (L.) L., Hibiscus rosa-sinensis, Chenopodium ambrosioides, Bursera simaruba (L.) Sarg., and Tradescantia zebrina Heynh. ex Bosse and the bark of Citrus grandis. For coughs, farmers prepared teas, cold drinks or baths from 12 plants (e.g., Manilkara zapota, Tagetes erecta L., Saccharum officinarum L.). Morinda citrifolia leaves and fruits were used for skin burns, diabetes and to improve blood circulation while young Cecropia peltata L. leaves were reported to treat diabetes.

Woody plants provided farmers with materials for house construction and tools. In all households, women cooked on an open fire, hence farmers required firewood year round. Timber and firewood were extracted from the surrounding forests, but were also occasionally obtained from the 38 species of trees found in homegardens. For house construction, farmers used hardwoods such as Metopium browneii (Jacq.) Urb., Ehrethia tinifolia L., Cordia dodecandra, Cedrela odorata, Swietenia macrophylla, Brosimum alicastrum and Guaiacum sanctum L. Seventeen plant species were used for practical purposes like roofing (Orbigyna cohune, Sabal mauritiiformis, Sabal mexicana) or for the construction of ladders (Cecropia peltata, Bursera simaruba). The dry leaf of Cordia dodecandra has a very coarse surface and was used to polish wood. The hard wood of Muntingia calabura L. and Cordia dodecandra was used to make axe handles and wooden frames that were harnessed to horses for transporting crops. Leaves of Heliconia sp. and Musa paradisiaca L. were used in cooking (e.g., tamales) and to wrap goods. Farmers used the dried fruit of Luffa cylindrica M. Roem. as a sponge to clean plates or for body care and Sapindus saponaria L. seeds as a detergent for the laundry. For handcrafted adornments like necklaces or bracelets, farmers used the seeds of Erythrina americana Mill. and Chamaedorea seifrizii Burret.

Discussion

The recent migration of groups from diverse cultural and geographic origins characterizes the sociocultural background of the inhabitants of Calakmul. The highly diverse natural environment of this zone combined with its multicultural inhabitants justifies the designation of the region as an area of high biocultural diversity (Stepp et al. 2008). This outstanding biocultural diversity becomes evident through the cultivation and diversified use of the vegetation of homegardens by the different ethnic groups observed in the present work. The high species richness of cultivated and tolerated plants in the homegardens of both cultural groups underlies an ideal structural and spatial arrangement, which allows for the use of the full capacity of both horizontal and vertical zones (Ruonavaara 1996). A complex vertical and horizontal structure allows an optimal exploitation of light and space, augments productivity, and is akin to the structure of the natural forest (De Clerck and Negreros-Castillo 2000; Kumar and Nair 2006; Montagnini 2006). The presence of differentiated zones of plants in the studied homegardens (e.g., zones for ornamental plants, herbs, vegetables, and secondary vegetation), combined with structures (e.g., buildings, pens for chickens or pigs), demonstrate the complex and diversified arrangement in homegardens that has also been reported for Maya and Mestizo homegardens in Mexico and Guatemala (Jimenez-Osornio et al. 1999; Ruonavaara 1996; Vogl-Lukasser 1998). Those complex systems involve people, animals, and plants with multiple ecological interactions promoting high diversity (Atran et al. 2002).

The high species richness of plants reported in this study confirms previous findings for homegardens in Veracruz (Lazos-Chavero and Alvarez-Buylla 1988), Yucatán (De Clerck and Negreros-Castillo 2000; Jimenez-Osornio et al. 1999; Osorio-Hernández 1997; Rico-Gray et al. 1990), Campeche (Chi-Quej 2009) and Chiapas (Gasco 2008; Vogl-Lukasser 1998). Most of these studies recorded food and medicinal plants as the most common; however in our study ornamental plants predominated. Chi-Quej's (2009) research on Campeche homegardens and Ruonavaara's (1996) study on household gardens in El Petén, Guatemala found similar results. The large number of plant species found in the homegardens has been associated with family knowledge of the local natural resources (Rico-Gray et al. 1990; Toledo et al. 2008), family preferences, and the age of homegardens (Coomes and Ban 2004). However, in the present study, the age of the homegardens was not related to species richness, which contrasts with other studies in Calakmul (Perea-Mercado 2011). The species richness and composition of the homegardens are influenced by the cultural background of the farmers (Kumar and Nair 2004; Lazos-Chavero and Alvarez-Buylla 1988; Rico-Gray et al 1990). In the present study, a significant difference in plant species richness was detected between Maya and Mestizo homegardens. Also, cluster analysis verified the relationship between the floristic composition of the homegardens and the cultural identity. Plant species composition within the Maya homegardens was less diverse, while Mestizo homegardens showed greater variation of plant species, suggesting an extensive utilization of the floristic resources of the Mestizo homegardens (Ruonavaara 1996).

In addition to differences in floristic diversity, plant species varied between Maya and Mestizo homegardens with some found only among one group or the other. Ornamental plants in particular contributed to higher plant species richness in Mestizo homegardens, providing aesthetic value and prestige to families (Heckler 2004). Ornamental plants are selected for different reasons, including attractiveness, (e.g., Allium aflatunense B. Fedtsch, Hibiscus rosa-sinensis), pleasant aromas (e.g., Murraya paniculata (L.) Jacq., Cestrum nocturnum L.) or leaves with different colors and shapes (e.g., Begonia rex Putz, Sansevieria trifasciata). Many ornamental plants were obtained from the surrounding forests (e.g., Chamaedorea seifrizii, Aristolochia maxima Jacq., Monstera deliciosa Liebm.); other non-natives were bought (e.g., Lagerstroemia indica L., Rosa sp., Datura stramonium L.). We observed in Mestizo homegardens that new ornamental plants are constantly incorporated into homegardens and that families also like to integrate non-ornamental plants with other purposes like food (e.g., Coffea Arabica L.) or medicine (Zingiber officinale Roscoe).

The lower species richness that appeared in Maya homegardens of Niños Héroes could be influenced, in part, by the economic condition of these households. In contrast to Mestizo communities, all houses in Niños Héroes contained modern construction material, as well as commodities like refrigerators and satellite televisions. These reflect higher income, from money sent by male family members who emigrated to the United States (60% of the households interviewed in Niños Héroes had a family member in the U.S.A.). It is possible that having the additional income reduced the need for different plant species in the homegardens, as families had money to buy more products for their basic needs (Cuanalo de la Cerda and Mukul 2008). Another reason for fewer species in the Niños Héroes homegardens could be a preference for growing plant species that possess market value which changed the aim of the homegardens from subsistence to commercial farming (see Abdoehllah et al. 2006; Shrivastava and Heinen 2005). In Niños Héroes farmers grow much citrus and maize, both of which are sold occasionally. Hence, the plants grown for multiple functions in the homegardens, such as food, medicine, fodder for animals, wood and utensils are replaced by species that are sold for cash, as Alayón-Gamboa and Gurri-García (2008) also found in homegardens of households that practice commercial agriculture. A similar situation has been observed in homegardens under constant process of adaptation as a consequence of modernization (Caballero 1992; Corzo-Márquez and Schwartz 2008; Vogl-Lukasser 1998).

Differences in Maya and Mestizo homegardens also probably reflect that these Maya migrants originated from a distinct environment from Calakmul and gaining local ecological knowledge may have been limited by the principal use of their indigenous language. The dominant use of the Ch'ol language could limit their communication and acquisition of local knowledge on plant management and use, especially among women with restricted mobility and Spanish skills. Similar findings were reported for Maya Q'eqchi immigrants in El Petén, Guatemala, who also migrated from a temperate climate to tropical lowlands (Atran et al. 2002).

The cultural differences in plant species richness and exclusive species in Maya or Mestizo homegardens were also reflected in how the families used the plants. In both cultural groups, the most numerous use category was ornamental plants. Of the 123 ornamental plants reported from the 20 homegardens, 46.4% appeared exclusively in Mestizo homegardens. Apart from the ornamental uses of plants, Mestizo homegardens presented more plant species in all functional categories (e.g., food, delimitation, medicine, timber or for animal fodder) than Maya homegardens (Figure 4). The authors observed a constant incorporation of new species, particularly in Mestizo homegardens, obtained locally through trade with neighbors, extracted from the surrounding forests, or brought by the migrant farmers from their native land as also reported by Perea-Mercado (2011). Within the sampled families we observed that Mestizo women had greater mobility and a more active social life than Maya women, giving Mestizo women more opportunities to acquire a diversity of plant material. The higher plant species richness provided a wider array of resources from the Mestizo homegardens; these provisions might make these families less vulnerable to the poverty and seasonal food shortages which characterize rural Calakmul and have negative impacts on living conditions and health (Alayón-Gamboa and Gurri-Garcia 2007). The availability of medicinal plants for remedies of common illnesses (e.g., gastrointestinal parasites, malaria and dengue fever) assists families considering the limited government support programs for healthcare in Calakmul. The supply of medicine and additional food from homegrown plants is more limited within the Maya families, making them more dependent on supplemental income from emigrated family members for their families' livelihoods.

Conclusion

The homegardens managed by the farmers of Calakmul present high species richness and structural diversity, generated by an optimal spatial arrangement and a management of multiple layers, which create a complex agroforestry-system analogous to the natural ecosystem. These plant species and their diverse uses fill a broad range of family needs and are strongly related to the socioeconomic conditions and cultural background of the farmers. Mestizo families with low income integrate more useful plant species in their homegardens than Maya families, making them less vulnerable to difficult conditions related to the scarcity of food and healthcare that is characteristic of the highly marginalized and impoverished region of Calakmul. Maya families benefit from higher income and manage the natural resources obtained from their homegardens in a very similar way. Nevertheless, with less species richness, Maya families are more dependent on cash incomes and less self-sustainable in terms of providing resources for their livelihoods.

The higher number of species reported in the Mestizo homegardens did not correlate with a preference for a particular functional category of plants (e.g. medicinal plants, food plants, or plants used for timber). However, ornamentals contributed considerably to the significantly higher species richness of Mestizo homegardens. This use category has received little attention in previous research and could be an important indicator of the culture-related species diversity of homegardens, especially in the context of women's preferences and cultural behavior. In this context, the aesthetic of the homegarden confers prestige among women and, for those with an active social web, provides more opportunities to acquire a diversity of plant species.