Ant diversity and its relationship with vegetation and soil factors in an alluvial fan of the Tehuacán Valley, Mexico

AC TA O E C O L O G I CA

29

(2006) 316–323

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/actoec

Original article

Ant diversity and its relationship with vegetation and soil factors in an alluvial fan of the Tehuacán Valley, Mexico Leticia Ríos-Casanova a,*, Alfonso Valiente-Banuet a, Víctor Rico-Gray b a Departamento de Ecología de la Biodiversidad. Instituto de Ecología, Universidad Nacional Autónoma de México. Ap. Postal 70275 Coyoacán, 04510 Mexico, D.F. Mexico b Departamento de Ecología Vegetal. Instituto de Ecología A. C. Ap. Postal 63 Xalapa, 91070 Veracruz, Mexico

A R T I C L E

I N F O

Article history: Received 25 October 2004 Accepted 6 December 2005 Available online 27 January 2006 Keywords: Alluvial fan Ant community Atta mexicana Pogonomyrmex barbatus Solenopsis xyloni Soil Tehuacán Valley Mexico Vegetation structure

A B S T R A C T

In this study, we analyze the ant community found along an alluvial fan located in the Tehuacán Valley, central Mexico. Considering that this fan is composed of four terraces with different soils and vegetation structures, our main goal was to determine whether there are significant differences in ant diversity among terraces. To accomplish this goal, we determine species richness and abundance in order to calculate diversity and evenness indices. In addition, we classify species in different feeding guilds to evaluate whether differences among terraces exist. We expected higher ant diversity and variety of food guilds in terraces with sandy soils and complex vegetation structures than in terraces with argillic and calcic horizons. Correlations between several diversity parameters, and soil percent-sand and vegetation structure were also conducted. A total of 26 ant species were recorded along the fan. Species richness was not different among terraces whereas abundance was higher in sandy soils and on terraces with complex vegetation structure. Particularly, the abundance of the harvester ant Pogonomyrmex barbatus was higher in these terraces decreasing total ant diversity and evenness. Species richness within feeding guilds was similar among terraces with the generalized foragers as the most common. Our work suggests that percentage of sand in the soil and complexity of vegetation structure of the alluvial fan studied might be influencing ant distribution and favoring the abundance of numerically dominant species which could be affecting the diversity patterns of the whole community. © 2006 Elsevier SAS. All rights reserved.

1.

Introduction

Several physical and biological factors could affect species richness and abundance of ant communities inhabiting particular environments. Because ants are small sized insects with high surface to volume ratios that make them prone to desiccation (Edney, 1977), physical factors such as solar radiation, temperature, and water, could play an important role

Corresponding author. Fax: +52 55 5622 9010. E-mail address: [email protected] (L. Ríos-Casanova). *

in determining ant diversity (MacKay, 1991; Bestelmeyer, 1997). Arid and semiarid lands have harsh environmental conditions which are characterized by frequent incidence of high solar radiation, wide fluctuations of temperature, and low precipitation regimes (Whitford, 2002). These environmental conditions impose severe restrictions that could limit the distribution and abundance of ants. Some ant species inhabiting these ecosystems have wide physiological tolerances and feeding habits (Hölldobler and Wilson, 1990; MacKay, 1991). However, other species with low physiological tolerances and restricted feeding habits are also found in these

1146-609X/$ - see front matter © 2006 Elsevier SAS. All rights reserved. doi:10.1016/j.actao.2005.12.001

ACTA OECOLOGICA

29

(2006) 316–323

317

Fig. 1 – Terraces of the San Rafael Coxcatlán alluvial fan showing main vegetation and soil traits (from Fernández, 1999 and Medina, 2000).

habitats (Bestelmeyer, 1997; Bestelmeyer and Schooley, 1999). The presence of species with low physiological tolerances strongly depends on the modification of physical factors which in turn, provide suitable conditions for their activity (Whitford and Ettershank, 1975; Kay and Whitford, 1978; Bestelmeyer, 1997). Several studies have suggested that soil traits and vegetation structure could modify the environmental conditions of arid and semiarid lands, affecting ant diversity and distribution (Johnson, 1992; Bestelmeyer and Wiens, 1996; Bestelmeyer and Schooley, 1999; Boulton and Ward, 2002; Johnson and Ward, 2002). Soil traits related to age of formation such as soil percent-sand and cemented calcic horizons are able to maintain moisture and low temperatures inside nests and in foraging locations (Whitford et al., 1976; Johnson, 1992, 2000; Dugas, 2001; Boulton et al., 2005). Likewise vegetation structures also affect the incidence of direct solar radiation, soil temperature, and water evaporation. Sites with complex vegetation structures provide better conditions for ant activity than sites with simple structures (Wisdom and Whitford, 1981; Bestelmeyer and Schooley, 1999; Retana and Cerdá, 2000). Differences in vegetation structure can also influence the distribution of the most preferred ant food, namely plants, which might also affect the distribution of ant species (Johnson, 2000). Alluvial fans are one of the most common landforms in arid and semiarid lands of North America (Whitford, 2002). These landforms have highly heterogeneous mosaics of soils that differ in their degree of development and age of formation (McAuliffe, 1994; Parker, 1995). Clay-enriched argillic and cemented calcic horizons are common in old alluvial deposits, which have important negative effects on hydric soil properties (Hennesy et al., 1983; Hamerlynck et al., 2000). Soils in young alluvial deposits are characterized by poorly developed coarse-textured horizons, which enhance water

infiltration and soil water availability for plants (Burges, 1995; Hamerlynck et al., 2000). This enhanced water availability leads to large vegetation differences among alluvial terraces (McAuliffe, 1994; Parker, 1995). In this study, we analyze the ant community found along an alluvial fan located in the Tehuacán Valley, a semiarid land in central Mexico. This fan is composed of four terraces with different soils, according to their degree of pedogenic development, and vegetation structures (Fig. 1). Soils in some terraces are predominantly sandy, whereas in other terraces, they have argillic and calcic horizons with low proportions of sand (Medina, 2000). Likewise vegetation is diverse and composed of herbs, trees, and shrubs in terraces with sandy soils, whereas it is relatively simple and mainly composed of trees in terraces with argillic and calcic horizons (Fernández, 1999). Considering the soil and vegetation differences in this alluvial fan, our main goal is to determine whether there are significant differences in ant diversity among terraces. To accomplish this goal, we determine species richness and abundance in order to calculate diversity and evenness indices. In addition, we classify species in different feeding guilds to evaluate whether ant community structure differs. We expect higher ant diversity and greater variety of food guilds in terraces with sandy soils and complex vegetation structures than in terraces with argillic and calcic horizons. Correlations between several diversity parameters (species richness, abundance, diversity indices), and soil percent-sand and vegetation structure were also conducted in order to assess ant species responses to environmental differences among terraces. However it is important to note that results from these correlations are limited, since data were obtained from contiguous terraces of one alluvial fan and no replicates were conducted.

318

ACTA OECOLOGICA

2.

Materials and methods

2.1.

Study site

This study was conducted in an alluvial fan of Barranca Muchil that is located in San Rafael Coxcatlán, in the Southeastern portion of the Tehuacán Valley, Puebla, Mexico (18°12′– 18°14′N, 97°07′–97°09′W). At an elevation of 1000 m, it has a mean annual temperature of 25 °C and a mean annual rainfall of 395 mm with a rainy (June to October) and a dry (November to May) season (Valiente, 1991). This alluvial fan is composed by four terraces differing in age, soil traits, and dominant plant species (Fernández, 1999; Medina, 2000). The first terrace (hereafter Fouquerial) is the oldest one as evidenced by its well-developed soils with cemented calcic horizons at 48 cm deep, which are dominated by the tree Fouqueria formosa Kunth. Other less abundant plant species are Bursera aptera Ramírez, Ceiba parvifolia Rose, Mimosa polyantha Benth., and Sanvitalia fruticosa Hemsley. The second terrace (hereafter Cuajiotal) is also an old terrace because it presents cemented calcic horizons at a depth of 32 cm. It is dominated by the plant species Bursera morelensis Ramírez, Mimosa polyantha Benth., Melochia tomentosa L., and Opuntia pilifera Weber. The third terrace (hereafter Chiotillal) is a young terrace due to the incipient development of its soil which presents few carbonates in its deepest layers. The dominant plant species are Escontria chiotilla (Weber) Rose, Stenocereus stellatus (Pfeiffer) Riccob., Viguiera gramatoglossa DC., and O. pilifera. The fourth terrace (hereafter Cardonal) is the youngest terrace since it has soils characterized by scarce modification of the parental material and high proportions of sand (Medina, 2000). Pachycereus weberi (Coulter) Backeb. and Senna wislizenii (A. Gray) Irwin and Barneby are the dominant plants, along with M. polyantha and Viguiera dentata (Cav.) Sprengel (Fernández, 1999; Medina, 2000; Fig. 1).

2.2.

29

(2006) 316–323

dance with percentage of occurrence, minimizing the effect of nearest nests or trunk trails to traps. Chi-square tests were performed to know whether there were differences in species richness and adjusted abundances among terraces. To determine species numerical dominance (Davidson 1988) on each terrace, adjusted abundance was natural log transformed and ranked from highest to lowest (Samways, 1983). Because estimation of diversity and evenness could be strongly affected by the sampling effort made during the fieldwork, we analyzed species accumulative curves of each terrace to determine the proportion of species known from an expected total species richness. This analysis use individual pitfall trap abundance data to perform 100 iterations and estimate the expected species richness. Iterations were conducted using the first order jackknife method with the program EstimateS 5.0 (Colwell, 1997; Chazdon et al., 1998). Diversity for each terrace was estimated using the Shannon–Wiener index: H′ = – Σ (pi ln pi), where pi = proportional abundance of the ith species (Magurran, 1988). Evenness was calculated as E = H′/ ln S, where S = species richness (Magurran, 1988). Diversity was compared between pairs of terraces with t tests for diversity indices (Magurran, 1988) and Bonferroni corrections (α/k, where α = 0.05 and k = No. of comparisons; Sokal and Rohlf, 1995).

2.3.

Feeding guilds

In order to determine whether there were significant differences in trophic guilds among terraces, trapped ants were classified as generalized foragers, harvesters, fungivores, and predators based on field observations and classification by Brown (2000). The number of species for each feeding guild and terrace was compared with a 4 × 4 contingency table (Zar, 1988).

2.4.

Relationship between ant diversity, vegetation, and soil factors

Ant diversity

In order to analyze diversity patterns, ants were captured in each alluvial terrace by using 20 pitfall traps arranged in a grid of 5 × 4 and separated by 10 m. This arrangement was repeated five times from 1999 to 2001 (November 1999, April, July and October 2000, and January 2001), for a total sampling effort of 400 traps (20 traps × 4 terraces × 5 captures). Pitfall traps were plastic containers of 6 cm diameter and 10 cm deep, containing ethanol (70%), water (27%) and glycerol (3%) (Samways, 1983; Andersen, 1991; Lindsey and Skinner 2001). Traps were buried 24 h before sampling to reduce digging effect. After this period, traps were simultaneously opened in all terraces during 48 h. Ants were sorted in the laboratory to species level (Bolton, 1995). Species richness for each terrace was calculated as the total number of species captured in 100 traps (20 traps × 5 captures). Abundance of each species was estimated using adjusted abundance (AA) through the following formula: AA = (A) * (O/100), where A = abundance or the total number of individuals of each species, and O = occurrence or percentage of traps in which a species occurred (Lindsey and Skinner, 2001). This measure was used because it combines abun-

Vegetation structure and soil percent-sand were measured in each terrace to explore their possible relationship with ant diversity. Vegetation structure was determined through foliage height profiles conducted in 20 randomly chosen points along two 50 m transects. At each point, a 7 m rod was placed to record the number of foliage contacts and their height. Foliage height profiles were conducted in October 2000, at the end of the rainy season. The number of foliage contacts was grouped into three categories according to their height: < 0.5 m (herbs), 0.5–2 m (shrubs), and > 2 m (trees). Data were analyzed with a 4 × 3 contingency Table to determine whether there were differences in the number of foliage contacts among terraces (Everitt, 1977). Soil percent-sand was obtained from Medina (2000), who analyzed soil profiles and their hydric properties in each terrace. Spearman rank correlations between diversity parameters (species richness, abundance, diversity indices, and evenness) as dependent variables, and vegetation structure and soil percent-sand as independent variables were conducted through randomization tests. These tests determine whether the observed correlation pattern might arise by chance, through the comparison of the original data set to a new dis-

ACTA OECOLOGICA

tribution generated by randomly re-sampling and reordering data (Manly, 1991). For each correlation analysis, traps from all terraces were considered as replicates, and 10 000 shuffles of these data were made to test the null hypothesis that correlation coefficient did not differ from zero (r = 0). Randomization tests were performed with the program RT version 2.0, and using the regression residual method (Manly, 1991, 1996). Correlations between abundance of the most dominant species in all terraces, and vegetation structure and soil percent-sand were also conducted to determine whether there were significant relationship. These correlations were conducted using the previously described methods.

3.

Results

3.1.

Ant diversity

A total of 26 species belonging to four subfamilies were captured on the four terraces (Table 1). The highest number of species was recorded in Cardonal (22) followed by Chiotillal (21), Cuajiotal (19) and Fouquerial (19) but there were no significant differences among the sites (X2 = 0.33, P = 0.95, d.f. = 3). Of all species recorded in the alluvial fan, 16 species were found in all terraces, whereas some species were exclusive to one of them. Strumigenys sp. and Dorymyrmex sp. 1

29

319

(2006) 316–323

Table 2 – Species richness (S), Adjusted abundance (AA), evenness (E), and Shannon–Wiener Index (H′) of the ant community in four terraces Terrace

S

Cardonal Chiotillal Cuajiotal Fouquerial

22 21 19 19

a a a a

AA

E

H′

1517.4 a 1898.6 b 279.8 c 242.3 c

0.45 0.42 0.77 0.77

1.39 1.29 2.26 2.27

a a b b

In S, AA and H′, different characters mean significant differences (P < 0.001).

were only found in Cardonal, Pheidole sp. 3 in Chiotillal, Pheidole sp. four in Cuajiotal, and Tetramorium spinosum in Fouquerial (Table 2). The analysis of AA (adjusted abundance) showed significant differences among terraces (X2 = 2201.1, P < 0.0001, d.f = 3), but not between terraces Cuajiotal and Fouquerial (X2 = 2.7, P = 0.1, d.f. = 1). Pogonomyrmex barbatus was the most dominant species in Cardonal, Chiotillal and Cuajiotal. Although in this last terrace, there were other species with high numerical dominance such as Brachymyrmex sp., Pheidole sp. 5, and Forelius mccooki. Dominant species in Fouquerial were Solenopsis xyloni followed by Atta mexicana. Camponotus festinatus, Dorymyrmex flavus and Strumygenys sp. were some of the species with lowest numerical dominance (Fig. 2). The analysis of species accumulation curves of the four terraces together, through the first order jackknife method, estimated a total of 29 species. Considering that our sam-

Table 1 – Ants of San Rafael Coxcatlán, their adjusted abundance and feeding guild Subfamily

Cardonal

Terraces Chiotillal Cuajiotal

Fouquerial

Total

Feeding guild

< 0.1

< 0.1

< 0.1

0.2

0.2

P

1426.3 < 0.1 0.8 0.2 < 0.1 – – 60.5 4.2 < 0.1 0.7 < 0.1 – < 0.1

1844.3 < 0.1 – 0.6 0.2 < 0.1 – 15.7 1.7 < 0.1 < 0.1 < 0.1 – –

95.7 < 0.1 0.7 6.1 0.5 – < 0.1 39.9 8.3 <0.1 0.5 < 0.1 – –

20.1 42.3 114.2 – 14.7 – – 3.5 < 0.1 0.1 0.2 < 0.1 0.1 –

3386.4 42.3 115.8 6.9 15.5 < 0.1 < 0.1 119.6 14.2 < 0.1 1.5 < 0.1 0.1 –

H Fun GF H H H H H H GF GF GF GF P

< 0.1 0.3 < 0.1 14.7 < 0.1

< 0.1 0.6 – 7.0 11.5

< 0.1 2.0 – 24.6 28.3

< 0.1 13.6 – 2.2 13.8

< 0.1 16.6 < 0.1 48.5 53.7

GF GF GF GF GF

6.5 2.9 < 0.1 – < 0.1

16.1 0.4 < 0.1 < 0.1 < 0.1

71.2 1.5 0.3 – –

15.8 1.6 < 0.1 – –

109.7 6.5 0.3 < 0.1 –

GF GF GF GF GF

< 0.1 1517.4

< 0.1 1898.4

– 279.8

< 0.1 242.3

< 0.1 3938.1

GF

Species

Subfamiy Ponerinae Odontomachus clarus Roger Subfamily Myrmicinae Pogonomyrmex barbatus (Smith) Atta mexicana (Smith) Solenopsis xyloni McCook Pheidole sp. 1 Pheidole sp. 2 Pheidole sp. 3 Pheidole sp. 4 Pheidole sp. 5 Pheidole sp. 6 Leptothorax (Macromischa) sp. Leptothorax (Myrafant) sp. Leptothorax sp. 1 Tetramorium spinosum (Pergande) Strumigenys sp. Subfamily Dolichoderinae Dorymyrmex cf. flavus McCook Dorymyrmex cf. insanus (Buckley) Dorymyrmex sp.1 Forelius sp. pruinosum group. Forelius mccooki McCook Subfamily Formicinae Brachymyrmex sp. Camponotus (Myrmobrachys) rubrithorax Forel Camponotus (Myrmothix) atriceps (Smith) Camponotus (Tanaemyrmex) flavopubens Emery Camponotus (Tanaemyrmex) festinatus Buckley

Paratrechina mexicana Forel Total

Fun: fungivores; GF: generalized foragers; H: harvesters; and P: predators.

320

ACTA OECOLOGICA

29

(2006) 316–323

Fig. 2 – Ranking of species according to their adjusted abundance in the four terraces showing the most and less abundant species. Pogonomyrmex barbatus (Pb), Pheidole sp. 5 (P5), Forelius pruinosus (Fp), Brachymyrmex sp. (B), Pheidole sp. 6 (P6), Dorymyrmex flavus (Df), Dorymyrmex. sp. 1 (D1), F. mccooki (Fm), Camponotus festinatus (Cfe), Leptothorax sp. 1 (L1), Leptothorax (Macromischa) sp. (LMa), Solenopsis xyloni (Sx), Atta mexicana (Am), Pheidole sp. 2 (P2), Tetramorium spinosum (Ts).

Fig. 3 – Species accumulation curves for the four terraces in the San Rafael Coxcatlán alluvial fan. Observed (doted line) and calculated (continuous line).

pling effort allowed us to capture 26 species, this means that we captured 89.7% of the total estimated species richness. The number of predicted species when each terrace was considered alone was 25. These results indicate that 76% of the species richness was captured in Fouquerial (19) and Cuajiotal (19), 84% in Chiotillal (21), and 88% in Cardonal (22) (Fig. 3). Ant diversity index was higher in Cuajiotal (2.26) and Fouquerial (2.27) than in Chiotillal (1.29) and Cardonal (1.39). These differences were statistically significant (Cuajiotal vs. Cardonal: t = 8.4, d.f. = 3763; vs. Chiotillal: t = 9.2, d.f. = 4118; Fouquerial vs. Cardonal: t = 8.5, d.f. = 3782; vs. Chiotillal:

t = 9.2, d.f. = 4138; in all cases p < 0.0001 after Bonferroni correction) (Table 2). Similarly, evenness was higher in Cuajiotal (0.77) and Fouquerial (0.77) than in Cardonal (0.45) and Chiotillal (0.42; Table 2).

3.2.

Feeding guilds

Generalized foragers had the highest species richness in all terraces followed by harvester ants. Predator and fungivore species were scarce since they were only represented by Odontomachus clarus and Strumigenys sp., and Atta mexicana,

ACTA OECOLOGICA

29

321

(2006) 316–323

Table 3 – Species richness in each feeding guild of four terraces from the San Rafael Coxcatlan alluvial fan

Terraces

Cardonal Chiotillal Cuajiotal Fouquerial

Predators

Harvesters

2 1 1 1

5 6 6 4

respectively. Odontomachus clarus and A. mexicana were found in the four terraces whereas Strumigenys sp. was only found in Cardonal (Table 3). There were no significant differences in the number of species among feeding guilds for the different terraces (G = 12.1, d. f. = 9, P = 11.8).

3.3.

Relationship between ant diversity, vegetation, and soil factors

Foliage height profiles varied across terraces (X2 = 26.86, d.f. = 6, P < 0.001), being Cardonal and Chiotillal the sites with more number of contacts with vegetation (134 and 97, respectively) in comparison to Cuajiotal and Fouquerial (55 and 62, respectively). The number of contacts for herbs, shrubs, and trees in Cardonal and Chiotillal did not differ from that expected by chance, except for herbs which had a higher number of contacts in Cardonal. On the contrary, the number of contacts for shrubs and trees in Cuajiotal and Fouquerial was significantly higher than that expected by chance. The frequency of herbs tended to be low (Fig. 4). Soils in Cardonal and Chiotillal had higher proportions of sand (75–79%) than soils in Cuajiotal and Fouquerial (63– 67%). Soils in these last terraces had high proportions of clay (16–23%; Medina 2000; Fig. 1). There were no significant relationships between soil percent-sand or vegetation structure and diversity parameters (species richness, abundance, diversity indices, and evenness). After 10,000 randomizations, none of the r values differed from zero (Table 4). Correlations between species abundance and vegetation structure and soil percent-sand were conducted only for P. barbatus, since it was the only species with high numerical dominance in all terraces. Results showed that there were significantly positive relationship with soil percent-sand (rs = 0.42, P < 0.001) and vegetation structure (rs = 0.35, P < 0.001).

Fig. 4 – Vegetation structure (foliage height profile) plants < 0.5 m, 0.5–1.0 m, , > 1 m. (+) frequency higher or (–) lower than randomly expected according to adjusted residual.

Feeding guild Generalized foragers

Fungivores

14 13 11 13

1 1 1 1

Table 4 – Correlation between soil percent sand and vegetation structure with some parameters of the ant community in the alluvial fan of San Rafael Coxcatlan

|

Vegetation Soil texture structure (Sand %) (Foliage contacts) P rs P rs

Species richness Adjusted abundance Diversity Evenness

1.0 0.176 0.082 –0.116

0.56 0.15 0.27 0.20

0.078 0.279 –0.04 –0.098

0.13 0.23 0.33 0.21

(P values after 10 000 randomizations).

4.

Discussion

The main goal of this study was to determine differences in ant diversity and feeding guilds of ants in four terraces of one alluvial fan in central Mexico. Results showed that ant diversity was lower in Cardonal and Chiotillal than in Cuajiotal and Fouquerial. However, no significant differences were found in the feeding guilds of ants inhabiting these terraces. The analysis of ant diversity showed that species richness was similar in all terraces (19–22 species), because 62% of the total species were shared among them. Therefore, lower ant diversity in Cardonal and Chiotillal could be the result of the high numerical dominance of particular ant species. The harvester ant Pogonomyrmex barbatus was the most abundant species in almost all terraces, which represented 65% of all captured workers. Particularly, it was the species with the highest abundance in Cardonal and Chiotillal and was negatively associated with ant diversity of these terraces. This idea is supported by the low evenness calculated for Cardonal and Chiotillal compared to Cuajiotal and Fouquerial. Pogonomyrmex barbatus has been considered a species with relatively lower tolerance for high temperatures compared to other Pogonomyrmex species (Johnson, 2000). Therefore the high abundance of this species in Cardonal and Chiotillal could be related to a high soil-percent sand and complex vegetation structure, since it has been suggested that both traits tend to decrease solar radiation, temperature, and water evaporation (Burges, 1995; Perfecto and Snelling, 1995; Dugas 2001). The significantly positive correlation found between P. barbatus abundance, and soil percent-sand and vegetation structure supports this finding. For other ant communities, such as those inhabiting grasslands of California, the overall ant richness and abundance were positively associated with high percent sand content across different soil types. Similarly to our results, the abundance of the dominant Pheidole californica was positively related to percent sand content (Boulton et al., 2005).

322

ACTA OECOLOGICA

Other studies conducted in different Mediterranean sites of Spain have reported that the relative abundance of the most common species, which have low thermal tolerance, tended to increase with vegetation cover resulting in a significant decrease of evenness (Retana and Cerdá, 2000). Dominant species in the Mediterranean region and Australian tropics such as Camponotus sylvaticus (Cerdá et al., 1998), Iridomyrmex spp. and Oecophylla smaragdina (Andersen, 1992) could also reduce diversity and evenness of these ant communities. In agreement with results previously reported by other studies, our study supports the idea that soil texture and vegetation have important effects on ant richness and abundance (Bestelmeyer and Wiens, 2001; Boulton et al., 2005). The number of ant species found in feeding guilds was similar among all terraces, contradicting our original hypothesis. This finding could be related to the presence of a high number of generalized foragers (16) in relation to harvesters (7), fungivores (1), and predators (2). Generalized foragers are the most common ant species in arid and semiarid lands, because they use a broad range of food items such as feces, dead arthropods, and honeydew which are available in almost any place and time (Whitford, 1978; Rojas and Fragoso, 2000). Our results suggest that differences among terraces in vegetation structure and possibly food resources did not affect the presence of generalized foragers. Correlations did not show significant relationship between ant diversity parameters (species richness, abundance, diversity indices, and evenness) and vegetation structure, and soil percent-sand. These results contrast to those reported by Majer (1985) and Perfecto and Snelling (1995) who found highest ant diversity in sites with complex vegetation structures or high vegetation cover. Our results also differed from other studies in which distribution of some ant species such as Linepithema humile (Way et al., 1997), P. rugosus (Dugas, 2001), P. barbatus, and Messor pergandei (Johnson, 1992; 2000) was mainly determined by moisture content of different soil types. At this point it is important to note that our analyses were conducted using four contiguous non-replicated terraces. In this respect, our results should be interpreted with caution since they are valid only for the alluvial fan studied. Several studies analyzing diversity of ant communities in tropical (Torres, 1984), semiarid (Bestelmeyer and Wiens, 1996) and arid habitats (Bestelmeyer and Schooley, 1999), have shown that different physical factors could affect species richness and abundance of ants. The alleviation of the environmental conditions prevailing in particular sites may enhance the distribution of some ant species, particularly those species with low thermal tolerances (Cerdá et al., 1998; Bestelmeyer, 1997; Johnson, 2000). This situation could be especially important in sites such as arid and semiarid lands, where high levels of direct solar radiation, temperature, and water loss through evaporation exist. Results of this study suggest that physical factors could play an important role in determining the presence of certain species in the ant community of the Tehuacán Valley. Due to the rather limited sampling of our data, it is necessary to conduct field research in other alluvial fans of the Tehuacán Valley, and elsewhere. At the same time, it is also necessary to evaluate

29

(2006) 316–323

the relative importance of other factors such as positive and negative interactions among ant species and among ants and other species of animals and plants. To summarize our results suggest that, at least in the alluvial fan analyzed in this study, ant diversity varied between terraces differing in soil traits and vegetation structure. These diversity differences arise from the high abundance of the harvester ant P. barbatus in terraces with high soil percent-sand and complex vegetation structure, and were not related with changes in species richness. On the other hand, differences in soil traits and vegetation structure along this alluvial fan do not have a significant effect on the number of species found in different feeding guilds. These similarities could be related to the high number of generalized foragers found in all terraces.

Acknowledgments We thank Héctor Godínez-Alvarez for his invaluable help during different stages of this work, and Dr. Philip Ward (University of California at Davis) and Roy Snelling (Natural History Museum of L.A. County), for ant identification. CONACyT, DGEP and Institute for Ecology gave a scholarship to L. R-C. DGAPA project IN-208301 gave funding to A. V-B., and IDEA WILD provided economic support for fieldwork. We also thank April Boulton and an anonymous reviewer for their comments which greatly improve the manuscript.

R E F E R E N C E S

Andersen, A.N., 1991. Sampling communities of ground-foraging ants: Pitfall catches compared with quadrat counts in an Australian tropical savanna. Aust. J. Ecol. 16, 273–279. Andersen, A.N., 1992. Regulation of "momentary" diversity by dominant species in exceptionally rich ant communities of the Australian seasonal tropics. Am. Nat. 140, 401–420. Bestelmeyer, B., 1997. Stress tolerance in some Chacoan dolichoderine ants: implications for community organization and distribution. J. Arid. Environ. 35, 297–310. Bestelmeyer, B., Wiens, J., 1996. The effects of land use on the structure of ground foraging ant communities in the Argentine Chaco. Ecol. Appl. 6, 1225–1240. Bestelmeyer, B., Schooley, R., 1999. The ants of the southern Sonoran desert: community structure and the role of trees. Biodiversity Conserv. 8, 643–657. Bestelmeyer, B., Wiens, J., 2001. Ant biodiversity in semiarid landscape mosaics: The consequences of grazing vs. natural heterogeneity. Ecol. Appl. 11, 1123–1140. Bolton, B., 1995. A New General Catalogue of the Ants of the World. Harvard University Press, Cambridge. Boulton, A.M., Ward, P.S., 2002. The ants. In: Case, T.J., Cody, M.L., Ezcurra, E. (Eds.), A New Island Biogeography of the Sea of Cortés. Oxford University Press, London, pp. 112– 128. Boulton, A.M., Davies, K.F., Ward, P.S., 2005. Species richness, abundance, and composition of ground-dwelling ants in Northern California grasslands: role of plants, soil, and grazing. Environ. Entomol. 34, 96–104. Brown, W., 2000. Diversity of ants. In: Agosti, D., Majer, J., Alonso, L., Schultz, T. (Eds.), Ants. Standard methods for

ACTA OECOLOGICA

measuring and monitoring biodiversity. Smithsonian Institution Press, Washington, pp. 45–79. Burges, T., 1995. Desert grassland, mixed shrub savanna, shrub steppe, or semidesert scrub? The dilemma of coexisting growth forms. In: McClaran, M.P., Van Devender, T.R. (Eds.), The Desert Grassland. The University of Arizona Press, Tucson, pp. 31–67. Cerdá, X., Retana, J., Manzaneda, A., 1998. The role of competition by dominants and temperature in the foraging of subordinate species in Mediterranean ant communities. Oecologia 117, 404–412. Chazdon, R.L., Colwell, R.K., Denslow, J.S., Guariguata, M.R., 1998. Statistical methods for estimating species richness of woody regeneration in primary and secondary rain forest of NE Costa Rica. In: Dallmeier, F., Comiskey, J.A. (Eds.), Forest Biodiversity Research, Monitoring and Modeling: Conceptual Background and Old World Case Studies. Parthenon Publishing, Paris, pp. 285–309. Colwell, R.K., 1997. EstimateS: Statistical estimation of species richness and shared species from samples, Version 5, User’s Guide and application, published at: http://viceroy.eeb.unconn.edu/estimates. Davidson, D.W., 1988. Resource discovery versus resource domination in ants: a functional mechanism for breaking the trade-off. Ecol. Entomol. 23, 484–490. Dugas, D.P., 2001. The influence of arroyo edges on nest distribution in the Chihuahuan Desert, New Mexico. J. Arid. Environ. 47, 153–159. Edney, E.B., 1977. Water balance in land arthropods. Springer Berlin, New York. Everitt, B.S., 1977. The analysis of contingency tables. Chapman and Hall, London. Fernández, N., 1999. Análisis de la dinámica de comunidades vegetales con relación a la evolución del paisaje en la zona semiárida de Coxcatlán, Puebla. Caso: Abanico aluvial de la Barranca del Muchil. MSc thesis, Facultad de Ciencias, Universidad Nacional Autónoma de México, México, D.F. Hamerlynck, E.P., McAuliffe, R.J., Smith, D.S., 2000. Effects of surface and sub-surface soil horizons on the seasonal performance of Larrea tridentata (creosotebush). Funct. Ecol. 14, 596–606. Hennesy, J.T., Gibbens, R.P., Tromble, J.M., Cardenas, M., 1983. Water properties of caliche. J. Rang. Man. 36, 723–726. Hölldobler, B., Wilson, E., 1990. The Ants. The Belknap Press of Harvard University Press, Cambridge. Johnson, R.A., 1992. Soil texture as an influence on the distribution of the desert seed-harvester ants Pogonomyrmex rugosus and Messor pergandei. Oecologia 89, 118–124. Johnson, R.A., 2000. Habitat segregation based on soil texture and body size in the seed-harvester ants Pogonomyrmex rugosus and P. barbatus. Ecol. Entomol. 25, 403–412. Johnson, R.A., Ward, P.S., 2002. Biogeography and endemism of ants (Hymenoptera: Formicidae) in Baja California, Mexico: a first overview. J. Biogeogr. 29, 1009–1026. Kay, C.A.R., Whitford, W.G., 1978. Critical thermal limits of desert honey ants: possible ecological explanations. Physiol. Zool. 51, 206–213. Lindsey, P.A., Skinner, J.D., 2001. Ant composition and activity patterns as determined by pitfall trapping and other methods in three habitats in the semi-arid Karoo. J. Arid. Environ. 48, 551–568. MacKay, W.P., 1991. The role of ants and termites in desert communities. In: Polis, G.A. (Ed.), The ecology of desert communities. The University of Arizona Press, Tucson. Magurran, A.E., 1988. Ecological diversity and its measurement. Princeton University Press, Princeton, New Jersey.

29

(2006) 316–323

323

Majer, J.D., 1985. Recolonization by ants of rehabilitated mineral sand mines on North Stradbroke Island, Queensland, with particular reference to seed removal. Aust. J. Ecol. 10, 31–48. Manly, B.F.J., 1991. Randomization, bootstrap and Monte Carlo methods in biology. Chapman & Hall/CRC, London. Manly, B.F.J., 1996. RT, A program for randomization testing, Version 2.0. The Centre for Applications of Statistics and Mathematics, University of Otago, New Zealand. McAuliffe, J.R., 1994. Landscape evolution, soil formation, and ecological patterns and processes in Sonoran Desert bajadas. Ecol. Monogr. 64, 111–148. Medina, J., 2000. Determinación del vigor y el estado reproductivo de Stenocereus stellatus (Cactaceae) a lo largo de una cronosecuencia edáfica en un abanico aluvial en Coxcatlán, Valle de Tehuacán. Bch. Thesis. Escuela Nacional de Estudios Profesionales Campus Iztacala. Universidad Nacional Autónoma de México, México, D.F. Parker, K.C., 1995. Effects of complex geomorphic history on soil and vegetation patterns on arid alluvial fans. J. Arid. Environ. 30, 19–39. Perfecto, I., Snelling, R., 1995. Biodiversity and the transformation of a tropical agroecosystem: ants in coffee plantations. Ecol. Appl. 5, 1084–1097. Retana, J., Cerdá, X., 2000. Patterns of diversity and composition of Mediterranean ground ant communities tracking spatial and temporal variability in the thermal environment. Oecologia 123, 436–444. Rojas, P., Fragoso, F., 2000. Composition, diversity, and distribution of Chihuahuan Desert ant community (Mapimí, Mexico). J. Arid. Environ. 44, 213–227. Samways, M.J., 1983. Community structure of ants (Hymenoptera: Formicidae) in a series of habitats associated with citrus. J. App. Ecol. 20, 833–847. Sokal, R.R., Rohlf, F.J., 1995. Biometry. The principles and practice of statistics in biological research. W.H. Freeman and Company, New York. Torres, J.A., 1984. Diversity and distribution of ant communities in Puerto Rico. Biotropica 16, 296–303. Valiente, B.L., 1991. Patrones de precipitación en el Valle semiárido de Tehuacán, Puebla, México. B. Sc. thesis, Facultad de Ciencias, Universidad Nacional Autónoma de México, México, D.F. Way, M.J., Cammell, M.E., Paiva, M.R., Collinwood, C.A., 1997. Distribution and dynamics of the Argentine ant Linepithema (Iridomyrmex) humile (Mayr) in relation to vegetation, soil conditions, topography and native competitor ants in Portugal. Insectes Sociaux 44, 415–433. Whitford, W.G., 1978. Structure and seasonal activity of Chihuahuan desert ant communities. Insectes Sociaux 25, 79–88. Whitford, W.G., 2002. Ecology of desert systems. Academic Press, San Diego. Whitford, W.G., Ettershank, G., 1975. Factors affecting foraging activity in Chihuahuan desert harvester ants. Environ. Entomol. 4, 689–696. Whitford, W.G., Johnson, P., Ramirez, J., 1976. Comparative ecology of the harvester ants Pogonomyrmex barbatus (F. Smith) and Pogonomyrmex rugosus (Emery). Insectes Sociaux 23, 117–132. Wisdom, W.A., Whitford, W.G., 1981. Effects of vegetation change on ant communities of arid rangelands. Environ. Entomol. 10, 893–897. Zar, J.H., 1988. Biostatistical Analysis. Prentice Hall, Englewood Cliffs, New Jersey.