DOI: 10.13102/sociobiology.v60i2.174-182Sociobiology 60(2): 174-182 (2013) Arboreal Ant Assemblages Respond Differently to Food Source and Vegetation Physiogno- mies: a Study in the Brazilian Atlantic Rain Forest JJ Resende1, PEC Peixoto1, EN Silva1 JHC Delabie2,3 & GMM Santos1 Introduction The richness and composition of ant assemblages have been related to the different structural aspects or the level of habitat preservation (Schoener 1971, Greenslade & Greenslade 1977, Levings 1983 & Andersen 1995). Ants are frequently chosen for studies that focus on the understand- ing of the effects of repetitive events of man-made habitat or ecosystem simplification on biodiversity (Matos et al. 1994, Majer 1996, Perfecto et al. 1997). Several studies have shown significant correlations between ant assemblages and habitat structural complexity, particularly in the tropics (Andersen & Majer, 2004, Delabie et al. 2006). For example, the richness of ants in the forest leaf litter has a strong correlation with plant diversity (Pereira et al. 2005) and in coffee agrosystems, the diversity of twig dwelling ants increases in habitats with more diverse shade tree cover (Armbrecht et al. 2004). Human activities have often caused the simplification of natural environments, leading to local extinction of popula- Abstract This study aimed to analyze assemblages of arboreal ants in different vegetation physiognomies within the Tropical Moist Forest (Atlantic Rain Forest) domain. The study was carried out at the Michelin Ecological Reserve, State of Bahia, Northeast of Brazil. We used sardine (protein resource) and honey (carbohydrate resource) baits to collect ants foraging in three vegetation types: (1) preserved native forest, (2) forest in regeneration (capoeira) with many invasive plants and (3) a mixed agroystem of rubber and cocoa tree plantation. We recorded 69 ant species at- tracted to the baits, 21 of them exclusive to honey bait and 25 exclusive to the sardine baits. The vegetation physiognomies preserved forest and rubber/cacao agrosystem showed higher species richness in relation to the forest in regenera- tion (capoeira), suggesting that rubber tree plantations can be a good matrix for the maintenance of some ant species typical of the forest matrix. The type of re- source used is important for the structuring of the arboreal ant assemblages. The ants that were attracted to protein resources showed a guild composition that is more differentiated between vegetation types that of ants attracted to glucose resources. Sociobiology An international journal on social insects 1 - Universidade Estadual de Feira de Santana, Feira de Santana, Bahia, Brazil 2 - Universidade Estadual de Santa Cruz / CEPLAC/CEPEC, Ilhéus-Itabuna, Bahia, Brazil 3 - CEPLAC/Centro de Pesquisas do Cacau, Itabuna, Bahia Brazil RESEARCH ARTICLE - ANTS Article History Edited by: Kleber Del-Claro, UFU – Brazil Received 13 May 2013 Initial acceptance 11 June 2013 Final acceptance 17 June 2013 Keywords Habitat preference, resource preference, matrix quality, Formicidae, Community Ecology Corresponding author Janete Jane Resende Departamento de Ciências Biológicas Universidade Estadual de Feira de Santana Av. Transnordestina s/n Novo Horizonte Feira de Santana, Bahia, Brazil 44036-900 E-Mail: antforjane@gmail.com tions and species and, consequently, could negatively impact important ecological processes such as nutrient cycling, seed dispersal and pollination (Thomas 2000, De Marco & Coelho 2004). Simplified environments often harbor a lower richness and diversity of ants, with an ant fauna consisting of gener- alist species (Sobrinho & Schoereder 2006), unlike forested habitats that harbor ant assemblages with higher levels of di- versity, consistent with the characteristics and the complexity of the vegetation (Majer 1996, Pereira et al. 2007). Different sampling methods have been used in surveys of ants and there is no direct means of comparison between different collection procedures (Romero & Jaffe 1989). There are several methods used for sampling ants such as oil sar- dine, carbohydrates, meat and cassava flour baits, Winkler ex- tractor, pitfall or manual collection, each one of them suited to select different classes of ants (soil dwelling, carnivorous, detritivorous, omnivorous) (Bestelmeyer et al. 2000; Freitas et al. 2004). Each of the collection procedures sampled a dif- ferent set of ant species (Romero & Jaffe 1989) and the forag- Sociobiology 60(2): 174-182 (2013) 175 ing activities may reflect or indicate the nutrients that are most limiting to in the respective nesting habitats, some ants prefer honey baits (carbohydrate) and other ants set prefer fish baits (nitrogen- protein) (Hashimoto et al. 2010). Ants provide an ideal system to test how macronutrient availability affects the costs and benefits of competitive dominance (Grover et al., 2007). Considerable evidence suggests that resource competi- tion strongly influences population and community dynamics in ants (Hölldobler & Wilson 1990) This study aimed to analyze arboreal ant assemblages in different vegetation physiognomies within the Atlantic Rain Forest domain. We tested two hypotheses. First, if the agroforestry system constitutes a good matrix for the main- tenance of ant species, richness of ants would be expected to be similar to or even greater than that found in forest like physiognomies. Second, since trophic groups of ants that are glycoside and protein consumers are generalist and special- ist, respectively, the ant assemblages of generalist ants are expected to be more similar while the second ones would be more dissimilar between habitats. Material and Methods Study area The study was conducted at the Michelin Ecological Reserve (headquarters: 13º 50’S, 39º 10’ W) Ituberá, state of Bahia, Brazil. The climate is of the type As according to the Köppen classification, tropical, rainy, hot, characterized by rainfall concentrated in summer and autumn, and average temperatures are never below 20º C. The landscape of the re- gion is characterized by the dominance of cacao agroforestry, mixed rubber/cacao plantations, pastures and forest fragments with natural vegetation comprising primary forest (a small percentage) and different stages of forest regeneration (with a large proportion of secondary forest). We collected ants in six different periods between Oc- tober 2007 and September 2008. Ant assemblies attracted to carbohydrate and protein baits were collected in three vegeta- tion physiognomies: (I) Conserved Atlantic Forest Fragment (PF), a typical Low Land Humid Tropical Forest measur- ing about 550 ha comprising blocks of native forest canopy of 15-25 m in height, with isolated trees reaching 30-40 m; (II) Forest Regeneration Fragment (Capoeira - SF), an early to intermediate stage of secondary succession tract of land, measuring about 10 ha of contiguous forest in regeneration, characterized by the presence of lianas, bromeliads, orchids, rocky ground, shrub vegetation and invasive plants and (III) agrosystem of mixed rubber tree/cacao plantation (AG), with over 20 years of age and 12 ha of area. Sampling methods The sampling consisted of 18 transects distributed among three vegetation types. In each vegetation type six transects of 400 m were established. In each transect there were 20 sampling points spaced at 20 m intervals. Each sam- pling point consisted of a bait rich in proteins and lipids (sar- dine oil) and another bait rich in carbohydrates (honey) in- stalled within the same tree at a height of 2 m and at least 20 cm away from one another. After installation, the baits remained on the plants for about 30 min. They were then collected and the ants present were fixed in alcohol 70%. In total, 120 baits were placed in each vegetation type. The sorting, assembly and morpho- speciation of ant specimens occurred in the Laboratory of Entomology, at Feira de Santana State University (UEFS). For the identification of ant species we used the classification of Bolton (2003), except for the genus Nylanderia (based on Lapolla et al. 2010), and the genera Strumigenys e Basiceros (based on Baroni-Urbani & De Andrade 2007). Vouchers were deposited in the Entomological Col- lection Prof. Johann Becker, Museum of Zoology - UEFS (MZFS) at Feira de Santana and the collection of the Myrme- cology Laboratory of CEPEC/CEPLAC (CPDC) at Ilhéus, state of Bahia. Data analyses The analyses of data took into consideration only presence or absence of the species as usual in ant commu- nity ecology studies (Longino 2000). The observed species richness was calculated using the rarefaction curve (Mao Tau) (Colwell et al. 2004). The total richness was estimated using the 1st order Jackknife estimator based on 50 randomizations (Heltshe & Forrester, 1983), performed with the program Es- timateS, version 7.5.2 (Colwell, 2006). The dissimilarity of ant assemblages between vegeta- tion types was assessed by performing a Principal Coordinate Analysis (PCoA) using the program R (R Development Core Team 2010). This analysis was preceded of the calculation of the Jaccard similarity index for the pair-wise combination of data collected in all transects, considering as a transect the 20 samples collected by vegetation type and sampling date. Therefore, the analyses were based on 18 transects (three veg- etation types times six dates). Results We recorded 69 species of ants attracted to the baits. Out of this total, 21 species were exclusive to honey baits and 25 to sardine baits (Table 1). With respect to vegetation types, 17 species were exclusive to Conserved Forest, 16 to the agrosystem and seven to the “Capoeira”. Regarding the frequency of species, Solenopsis sp.2 was the most frequent on both sardine (31.6%) and honey (20%) baits in rubber/ca- cao tree agroforestry. The second most frequent species in this physiognomy was Camponotus sp.5 which was recorded on 11.6% of sardine baits. In the Conserved Forest physiognomy, the most fre- JJ RESENDE, PEC PEIXOTO, EN SILVA, JHC DELABIE & GMM SANTOS - Vegetation structure and arboreal ant diversity176 quent ants on sardine baits were Crematogaster sp.1 (9.1% of the baits) and Ectatomma tuberculatum (8.3%). The same happened in the “Capoeira” area with Crematogaster sp.1 present on 10% of the baits and E. tuberculatum in 9.1%. For honey baits, Strumigenys sp.9 (5.8%) was the most frequent species in conserved forest while Crematogaster sp.1 (6.6%) was the most abundant in “Capoeira”. Among the most speciose genera, Pheidole (14 spe- cies), Pachycondyla (8), Camponotus (7) and Solenopsis (5) were the most rich. The proportion of species belonging to these genera present in both types of baits remained very close, except for Camponotus which had two and seven spe- cies recorded on honey and sardine baits, respectively. The two axis extracted from the PCoA analyses with data from the two bait types grouped explained 34.3% of the total variation in ant composition among samples (18.6% of variation explained by axis 1 and 15.6% of variation ex- plained by axis 2). According to this analysis, the species col- lected in sardine differed from the species collected in honey baits. Considering only ants collected in sardine baits, there was also a very clear separation between vegetation types, in- dicating that the composition of ant species differs between vegetation types (17.6% of variation explained by axis 1 and 17.1% of variation explained by axis 2). On the other hand, the analysis including only ants collected in honey baits dem- onstrated a low distinction between vegetation types (18.4% of variation explained by axis 1 and 11.7% of variation ex- plained by axis 2), indicating that the assemblage of species that visited the baits rich in carbohydrates were similar among the three habitats (Fig. 1). The greatest observed and expected richness of ants were recorded in the preserved forest fragment and in the agrosystem of mixed rubber/cacao plantation (Fig. 2, Table 1). The observed richness curves (Mao Tau) showed no stabilization in any of phytophysiognomies (Fig. 2). Never- Fig. 1. Ordination by PCoA of species of ants collected in different phy- tophysiognomies, Reserva Ecológica da Michelin, Ituberá e Igrapiúna municipalities Bahia, Brazil. PF – Conserved forest; SF - Capoeira; AG – Agrosystem of mixed rubber tree/cacao plantation. (A) ants collected in honey and sardine baits; (B) ants collected in sardine baits and (C) ants collected in honey baits. A B C Fig. 2. Expected (Jack1) and observed (Mao Tau) richness curves for species of ants collected in a total of 240 samples. Reserva Ecológica da Michelin, Ituberá e Igrapiúna municipalities, Bahia, Brazil. PF – Conserved forest; SF- Capoeira; AG – Agrosystem of mixed rubber tree/cacao plantation. R ic hn es s of s pe ci es Number of samples Sociobiology 60(2): 174-182 (2013) 177 theless, the expected richness according to the species accu- mulation curve (jack1) showed a sharp increase in the number of species as a function of sample size, followed by an asymp- tote in the agrosystem habitat. The shape of the curve sug- gests that in the agrosystem the sampled fauna is more homo- geneous than in “Capoeira” and conserved forest. Therefore the majority of the species is sampled with a lower number of samples. Extrapolation of the curves also suggests that the number of total species in the agrosystem is lower than the two other habitats. Discussion According to our results, the vegetation physiogno- mies of conserved forest and agrosystem presented species richness highest than “Capoeira”, showing that agricultural habitats including the association between cacao and rubber trees forms a suitable matrix for the maintenance of ant spe- cies typical of forest environment. However, the juxtaposition of forest areas close to the agrosystem is an important point allowing to maintain the ant diversity (Delabie et al. 2007). Similarity between bait types and vegetation physiognomies The similarity between the ant assemblages sampled in sardine baits among the studied vegetation physiognomies, is probably a result of the shared occurrence of species of the genera Azteca, Camponotus, Crematogaster, Pheidole and Solenopsis which are considered dominant or subdominant if considering the structure of arboreal ant fauna (Wilson 1976; Majer et al. 1994; Brandão et al. 2009). Some species of these genera can significantly influence the structure of the arthro- pod community, exercising strong predation, especially on larvae of Lepidoptera and Coleoptera (Majer & Delabie 1993, Floren et al. 2002; Philpott & Armbrecht 2006). This same ecological context may apply to explain the great similarity between the ant assemblages attracted to honey and sardine baits within the agrosystem, because al- most the same species were present in both types of baits. Other species that were recorded in this physiognomy such as E. tuberculatum, Pachycondyla venusta and Odontomachus haematodus belong to genera typically considered as those of generalist predator species. Except for P. venusta, these ants may supplement their diet with nectar exudates from plants and honoydew producer insects (Hölldobler & Wilson 1990, Delabie 2001). An additional outcome supporting this expla- nation is the occurrence of the species Tapinoma sp.1 in the agrosystem of mixed rubber/cacao plantation. This species has as main feature the generalist behavior, being undemand- ing in terms of habitat quality and shifting easily from one food source to another. The numerous species of ants visiting sardine baits in each vegetation physiognomy, such as those of the gen- era Hypoponera, Megalomyrmex, Nylanderia, Wasmannia in Conserved Forest, Linepithema in Capoeira and Dorymyrmex in the agroystem, has an important ecological implication. In case where protein-based resources are scarce in these habi- tats, they will be almost exclusively used by specialist species that tend to defend this resource with aggressive behaviors, by exhibiting a rapid recruitment of workers and thus preventing access of other species. The species of the genus Solenop- sis, for instance, have aggressive behavior and are common in disturbed habitats. On the other hand, carbohydrate-based resources are more common and visited by generalist species, which do not have any preference to the types of bait used, such as those of the genera Camponotus, Pheidole and Solen- opsis (Hölldobler & Wilson 1990). Although, in general, species richness correlates posi- tively with habitat complexity, this correlation seems to de- pend on the habitat, because the agrosystem had a higher ob- served richness compared to “Capoeira”, which is considered structurally more complex than the former. Lassau & Hochuli (2004) found similar results with a greater richness in less complex habitats, believing that the movement of ants may be more efficient in terms of energy, in less complex habitats. Gomes et al (2010) demonstrates that ant fauna is more influ- enced by vegetation integrity than by fragment size, distance to edge or forest cover surrounding fragments. Lopes et al (2012) shows that the species that compose the ant assemblies in different phytophysiognomies are a reflex of the environ- ment, especially of the plant species, supporting the hypoth- esis that differences in the vegetational composition result in different position of the ant assembly. The higher number of species in the agrosystem in comparison to “Capoeira” suggests the occurrence of occa- sional species with low occurrence, sometimes called “tour- ists”. These species are treated by the estimators of richness as singletons or doubletons and can boost the expected rich- ness estimation. In spite of this, a greater number of species considered dominant (genera Camponotus, Crematogaster, Solenopsis and Pheidole) was recorded for the agrosystem in comparison to “Capoeira”. It is important to remark that although the species ac- cumulation curve based on Jack1 reached an asymptote in the agrosystem, the estimated richness in the Capoeira is in an as- cending trend and could surpass the estimated richness of the agrosystem if the sampling effort were increased. This leaves the inference on the quality of the rubber/cacao agrosystem as a good matrix open, from the perspective of which habitat is richer. However, looking at the list of species in Table 1, the following results are found. Out of the 69 ant species collect- ed in all habitats, 13 (18.84%) are common to both conserved forest and agrosystem. Six species are protein consumers and seven carbohydrate consumers. Furthermore, the relative fre- quencies of most of these species are low and very similar in forest and agrosystem habitats. Thus, from the point of view of life history, we can infer that the agrosystem is a good qual- JJ RESENDE, PEC PEIXOTO, EN SILVA, JHC DELABIE & GMM SANTOS - Vegetation structure and arboreal ant diversity178 ity matrix for at least some rare species that also occur in for- est vegetation. The choice of food resource (honey or sardine baits) by the ant assemblages, clearly differentiate the three vegeta- tion physiognomies. Both proteins and carbohydrates are of paramount im- portance for maintenance of ant populations. Hashimoto et al. (2010) demonstrated that different populations of ants forage more actively on these resources when they are scarce. On one hand, the proteic resources are fundamental for brood devel- opment and tissue synthesis (Hölldobler & Wilson 1990). On the other hand, carbohydrates are key resources for the main- tenance of the ant activity (Davidson, 1998). Byk and Del- Claro (2011) demonstrated the benefits of the carbohydrate resources to ant populations. The glucidic foods influence ant species composition, abundance, number of individuals per colony, body weight, survivorship, growth, reproduction and interactions at the community level. Bihn et al. (2008) exam- ined bait preferences of litter ants along a successional gradi- ent of forests in the Atlantic Forest of Brazil and observed that ants preferred protein-based baits in secondary forests, and carbohydrate-based baits in old-growth forests. In addition, the ant preference for carbohydrate or protein is subject to change individually according a previous ingestion of one of this food source, extrafloral nectar or insect. This is the theory of ecological stoichiometry, which relates nutrient balance to ecological processes (Sterner & Elser 2002). Such nutri- tional complexity can mediate patterns of ecological interac- tions (Anderson et al. 2004). Therefore, if an ant individual is consuming extrafloral nectar, it is more propable it attacks a herbivore on the host plant aiming to balance the ingestion of protein after too much ingestion of carbohydrate (Grover et al., 2007). Our data support our working hypothesis that the type of resource used by ant species is important in structuring the community. The distribution of ant species that are specialist and consumers of protein-based resources (sardine baits) dif- fered more between the vegetation types than the distribution of ant species that are generalist and consumers of carbohy- drate-based resources. The similarity of species collected in each vegetation physiognomy was higher among ant assem- blages that use carbohydrate-based resources (Fig. 1c) than among ant assemblages that use protein-based resources, in an exclusive manner (Fig. 1b) or together with carbohydrate sources (Fig. 1c). Acknowledgements To CNPq (Project 620021/2008-0 – Edital no 16/2008 and PRONEX FAPESB-CNPq project PNX 011-2009); to Foundation for Research Support de in Bahia state (Fundação de Amparo à Pesquisa do Estado da Bahia - FAPESB (Proj- ect 5577/2009) and the Michelin Ecological Reserve (REM) supported this study. GMMS and JHCD thanks to CNPq for productivity Fellowship. To many colleagues that contribut- ed to this study, specially Allan Rhalff Gomes Teixeira and Emerson Mota da Silva for helped with field sampling. We would like to thank two anonymous reviewers provided their insightful comments on the manuscript. References Andersen, A.N. & Majer, J.D. (2004). Ants show the way Down Under: invertebrates as bioindicators in land manage- ment. Front. Ecol. Environ., 2: 291-298. Andersen, A.N. (1995). A Classification of Australian Ant Communities, Based on Functional Groups Which Parallel Plant life-forms in relation to stress and disturbance. J. Bio- geog., 22:15-29. Anderson, T., Elser, J. J. & Hessen, D. O. (2004). Stoichiom- etry and population dynamics. Ecol. Lett., 7: 884–900. doi: 10.1111/j.1461-0248.2004.00646.x Armbrecht, I., Perfecto, I.& Vandermeer, J. (2004). Enigmatic Biodiversity: Ant Diversity Responds to Diverse Resources. Science 304:284-286. Baroni-Urbani, C. & De Andrade, M.L. (2007). The ant tribe Dacetine: limits and constituent genera, with descriptions of new species. Ann. Mus. Civ. Stor. Nat. Giacomo Doria, 99:1- 191. Bestelmeyer, B.T., Agosti, D., Leeanne, F., Alonso, T., Bran- dão, C.R.F., Brown, W.L., Delabie, J.H.C., & Silvestre, R. (2000). Field techniques for the study of ground-living ants: An Overview, description, and evaluation. In: D. Agosti, J.D. Majer, A. Tennant & T. de Schultz (eds), Ants: standart meth- ods for measuring an monitoring biodiversity. Smithsonian Institution Press, Washington 122-144. Bihn, J.H., Verhaagh, M., Brand, R. (2008). Ecological stoi- chiometry along a gradient of forest succession: Bait prefer- ences of litter ants. Biotropica, 40: 597-599. Bolton, B. (2003). Synopsis and classification of Formicidae. The American Entomological Institute. Gainesville 370p. Brandão, C.R.F., Silva, R.R. & Delabie, J.H.C. (2009). For- migas (Hymenoptera), pp. 323–369. In: A.R.Panizzi & J.R.P. Parra (eds). Bioecologia e Nutrição de Insetos. Base para o Manejo Integrado de Pragas. Brasília, Embrapa Informação Tecnológica, 1263p. Byk, J. & Del-Claro, K. (2011). Ant-plant interaction in the Neotropical savanna: direct beneficial effects of extrafloral nectar on ant colony fitness. Popul. Ecol., 53(2): 327-332. Chazdon, R.L., Colwell, R.K., Denslow, J.S. & Guarigurata, M.R. (1998). Statistical methods for estimating species rich- ness of woody regeneration in primary and secondary rain forests of northeastern Costa Rica. In Forest biodiversity research, monitoring and modelling. (F. Dallmeier & J. A. Sociobiology 60(2): 174-182 (2013) 179 Comiskey, eds). MAB - Man and Biosphere Series, UNES- CO, Paris 285-309. Colwell, R. K. (2006). EstimateS: Statistical estimation of species richness and shared species from samples, version 8.0. Disponível em: http://viceroy.eeb.uconn.edu/EstimateS Colwell, R. K., Mao, C.X. & Chang, J. (2004). Interpolating, extrapolating, and comparing incidence-based species accu- mulation curves. Ecology 85:2717-2727. Davidson D.W. (1998) Resource discovery versus resource domination in ants: a functional mechanism for breaking the trade-off. Ecol. Entomol., 23:484–490 De Marco P. & Coelho, F.M. (2004). Services performed by the ecosystem: Forest remnants influence agricultural cultures’ pollination and production. Biodiv. Conserv., 13: 1245-1255. Delabie, J.H.C. (1993). Formigas exóticas na Bahia. Bahia, Análise Dados. 3:19-22. Delabie, J.H.C. (2001). Trophobiosis between Formicidae and Hemiptera (Sternorrhyncha and Auchenorrhyncha): an overview. Neotrop. Entomol., 30(4): 501-516. Delabie, J.H.C., Jahyny, B., Nascimento, I.C., Mariano, C., Lacau, S., Campiolo, S., Philpott, S.M. & Leponce, M. (2007). Contribution of cocoa plantations to the conservation of native ants (Insecta: Hymenoptera: Formicidae) with a spe- cial emphasis on the Atlantic Forest fauna of southern Bahia, Brazil. Biodiv. Conserv., 16: 2359-2384. Delabie, J.H.C., Fisher, B.L., Majer, J.D. & Wright, I.W. (2000). Sampling Effort and Choice of Methods. In: D. Agosti; Majer, J.D.; Tennant, A.; Schultz, T.R. (Eds). Ants: standard methods for measuring and monitoring biodiversity. Smithso- nian Institution Press, Washington, DC., USA 145-154. Delabie, J.H.C., Paim, V.R.L., Nascimento, I.C., Campiolo, S. & Mariano, C.S.F. (2006). As Formigas como Indicadores Biológicos do Impacto Humano em Manguezais da Costa Su- deste da Bahia. Neotrop. Entomol., 35(5):602-615. Dunn, R.R., Agosti, D., Andersen, A.N., Arnan, X., Bruhl, C.A., Cerdá, X., Ellison, A.M., Fisher, B.L., Fitzpatrick, M.C., Gibb, H., Gotelli, N.J., Gove, A.D., Guenard, B., Janda, M., Kaspari, M., Laurent, E.J., Lessard, J.P., Longino, J.T., Majer, J.D., Menke, S.B., McGlynn, T.P., Parr, C.L., Philpott, S.M., Pfeiffer, M., Retana, J., Suarez, A.V., Vasconcelos, H.L., Wei- ser, M.D. & Sanders, N.J. (2009). Climatic drivers of hemi-Climatic drivers of hemi- spheric asymmetry in global patterns of ant species richness. Ecol. Lett., 12: 324-33. Floren, A., Biun, A.& Linsenmair, K.E. (2002). Arboreal ants as key predators in tropical lowland forest trees. Oecologia, 131: 137-144 Freitas, A.V.L, Francini, R.B. & Brown, K.S. (2004). Insetos como indicadores ambientais. Pp. 125-151. In: Cullen Jr., L.; Rudran, R. & Valladares-Padua, C. (Org.). Métodos de estu-Métodos de estu- dos em biologia da conservação e manejo da vida silvestre. Editora UFPR, Curitiba. 665p. Gomes, J. P., Iannuzzi, L. & Leal, I.R. (2010). Resposta da comunidade de formigas aos atributos dos fragmentos e da vegetação em uma paisagem da Floresta Atlântica nordestina. Neotrop. Entomol., 39(6):898-905 Greenslade, P.J.M. & Greenslade, P. (1977). Some effects of vegetation cover and disturbance on a tropical ant fauna. In- sectes Soc., 24: 163-182. Grover, C.D., Kay, A.D., Monson, J.A., Marsh, T.C., Holway, D.A. (2007). Linking nutrition and behavioural dominance: carbohydrate scarcity limits aggression and activity in Ar- gentine ants. Proc. R. Soc. B, 274: 2951–2957. doi:10.1098/ rspb.2007.1065 Hashimoto Y., Morimoto Y., Widodo, E.S., Maryati Mo- hamed & Fellowes, J.R. (2010). Vertical habitat use and for- aging activities of arboreal and ground ants (Hymenoptera: Formicidae) in a Bornean tropical rainforest. Sociobiology, 56(2): 1-14. Heltshe, J. & Forrester, N.E. (1983). Estimating species rich- ness using the jackknife procedure. Biometrics, 39: 1-11 Hölldobler, B. & Wilson, E.O. (1990). The ants. Harvard Uni- versity Press, Cambridge, 732p. Lapolla, J., Brady, S.G. & Shattuck, S.O. (2010). Phylogeny and taxonomy of the Prenolepis genus-group of ants (Hy- menoptera: Formicidae). System. Entomol., 35: 118-131. Lassau, S.A. & Hochuli, D.F. (2004). Effects of habitat com- plexity on ant assemblages. Ecography, 27: 157-164. Levings, S.C. (1983). Seasonal, annual, and among-site varia- tion in the ground ant community of a deciduous tropical for- est: some causes of patchy species distributions. Ecol. Mono-Ecol. Mono- gr., 53: 435-455. Longino, J.T. 2000. Que fazer com os dados. p.186 - 203. In: Agosti, D.; Majer, J. D.; Alonso, L. E. & Schultz, T. R. (Eds.) Formigas: Métodos padrão para medição e monitoramento da biodiversidade. Smithsonian Institution Press, Washington, DC 280.pp Lopes, J.F.S., Hallack, N.M. dos R., Sales, T.A. de, Brugger, M.S., Ribeiro, L.F., Hastenreiter, I.N. & Camargo, R. da S. (2012). Comparison of the Ant Assemblages in Three Phy-Comparison of the Ant Assemblages in Three Phy- tophysionomies: Rocky Field, Secondary Forest, and Ri- parian Forest—A Case Study in the State Park of Ibitipoca, Brazil, Psyche, vol. 2012, Article ID 928371, 7 pages, 2012. doi:10.1155/2012/928371 Majer, J. D. & Delabie, J.H.C. (1993). An evaluation of Brazilian cocoa farm ants as potential biological control agents, J. Plant Prot. in the Tropics, 10(1): 43-49. Majer, J. D., Delabie, J.H.C. & Smith, M.R.B. (1994). Arbo- JJ RESENDE, PEC PEIXOTO, EN SILVA, JHC DELABIE & GMM SANTOS - Vegetation structure and arboreal ant diversity180 real ant community patterns in brazilian cocoa farms, Biotro- pica, 26(1):73-83. Majer, J.D. (1996). Ant recolonization of rehabilitated bauxite mines at Trombetas, Pará, Brazil. J. Appl. Ecol., 12: 257-273 Matos, J. Z., Yamanaka, C.N., Castellani, T.T. & Lopes, B.C. (1994). Comparação da fauna de formigas de solo em áreas de plantio de Pinus elliottii, com diferentes graus de complexi- dade estrutural (Florianópolis, SC). Biotemas, 7(1-2): 57-64. Pereira, M.P.S., Queiroz, J.M., Souza, G.O. & Mayhé-Nunes, A.J. (2007). Influência da heterogeneidade da serapilheira so- bre as formigas que nidificam em galhos mortos em floresta nativa e plantio de eucalipto. Neotrop. Biol. Conserv. 2(3): 161-164. Pereira, M.P.S., Queiroz, J.M., Valcarcel, R. & Nunes, A.J.M. (2005). Fauna de formigas no biomonitoramento de ambien- tes de área de empréstimo em reabilitação na ilha da madeira, RJ. In: Simpósio Nacional e Congresso 3 Latino-Americano de Recuperação de Áreas Degradadas. Curitiba. Anais Sobra- de, 6: 5-12. Perfecto, I., Vandermeer, J., Hanson, P.& Cartin, V. (1997). Arthropod biodiversity loss and the transformation of a tropi- cal agro-ecosystem. Biodiv. Conserv., 6: 935-945. Philpott, S. M., Perfecto, I. & Vandermeer, J. (2006). Effects of management intensity and season on arboreal ant diversity and abundance in coffee agroecosystems. Biodiv. Conserv. 15: 139–155. Resende, J.J., Santos, G.M.M., Nascimento, I.C., Delabie, J.H.C. & Silva, E.M. (2011). Communities of Ants (Hy-Communities of Ants (Hy- menoptera – Formicidae) in Different Atlantic Rain Forest Phytophysionomies. Sociobiology, 58: 779-799. Romero, H. & Jaffé, K. (1989). A comparison of methods for sampling ants (Hymenoptera, Formicidae) in savannas. Bio- tropica, 21: 348-325. Schoener, T.W. (1971). Theory of feeding strategies, Ann. Rev. Ecol. Syst., 2: 369-404. Sobrinho T.G. & Schoereder, J.H. (2006). Edge and shape ef-Edge and shape ef- fects on ant (Hymenoptera:Formicidae) species richness and composition in forest fragments. Biodiv. Conserv. 16: 1459- 1470. Sterner, R.W. & Elser, J.J. (2002). Ecological stoichiometry. Princeton, NJ: Princeton University Press. 584pp. Thomas, C. D. (2000). Dispersal and extinction in fragmented landscapes. Proc. R. Soc. Lond. B Biol. Sci. 267: 139-145. Wilson, E. O. (1976). Which are the most prevalent ant genera? Studia Ent., 9: 1-4. Sociobiology 60(2): 174-182 (2013) 181 Table 1. Number of records and relative frequency (%) of ants collected in honey and sardine baits in three vegetation physi- ognomies in the Reserva Ecológica da Michelin. Ituberá e Igrapiúna municipalities, Bahia, Brazil PF – Preserved forest; SF - Secondary Forest (Capoeira); AG – Agrosystem of mixed rubber tree/cacao plantation. Food Preferences (C= Primarily Car- bohydrate consumer, P= Primarily Protein consumer, C/P = Both; sensu Brandão et al. 2009). Sampled Species Food Preferences Honey bait Sardine bait PF SF AG PF SF AG AMBLYOPONINAE Prionopelta sp.1 P - 1 (0.8) - - - - ECTATOMMINAE Ectatomma sp.1 P - 2 (1.6) - - - - Ectatomma brunneum F. Smith, 1858 P - 3 (2.5) - - - - Ectatomma tuberculatum F. Smith, 1858 P - 4 (3.3) 8 (6.6) 10 (8.3) 11 (9.1) 7 (5.8) Gnamptogenys sp.3 P - - - 3 (2.5) - - DOLICHODERINAE Azteca sp.1 C/P 2 (1.6) - 3 (2.5) 7 (5.8) 1 (0.8) 6 (5) Azteca sp.28 C/P 4 (3.3) - - 3 (2.5) 2 (1.6) 3 (2.5) Azteca sp.3 C/P - - - - - 2 (1.6) Azteca sp.4 C/P - - 3 (2.5) - - - Dorymyrmex sp.1 C - - - - - 4 (3.3) Linepithema sp.1 C - 3 (2.5) 4 (3.3) - 2 (1.6) - Linepithema sp.3 C - - - - 1 (0.8) - Tapinoma sp.1 1 (0.8) 2 (1.6) 4 (3.3) - - 1 (0.8) FORMICINAE Camponotus sp.2 C - - - - 5 (4.1) - Camponotus sp.3 C 1 (0.8) - - - 4 (3.3) - Camponotus sp.4 C - - - 3 (2.5) - - Camponotus sp.5 C - - 3 (2.5) - - 14 (11.6) Camponotus sp.6 C - - - - - 5 (4.1) Camponotus sp.7 C - - - - - 3 (2.5) Camponotus sp.8 C - - - - - 1 (0.8) Brachymyrmex sp.1 C/P 3 (2.5) - 3 (2.5) - - - Brachymyrmex sp.4 C/P - 3 (2.5) - 1 (0.8) 1 (0.8) - Nylanderia sp.2 C/P - - - 1 (0.8) - - Nylanderia sp.3 C/P - - 1 (0.8) - - - Nylanderia sp.4 C/P - - 1 (0.8) 3 (2.5) - - MYRMICINAE Cephalotes atratus (Linneus, 1758) C - 1 (0.8) - - - - Cephalotes sp.1 C - - - - 3 (2.5) - Cephalotes sp.2 C 1 (0.8) - - - - - Crematogaster sp.1 C - 8(6.6) 3 (2.5) 11 (9.1) 12 (10) 8 (6.6) Crematogaster sp.2 C 1 (0.8) - 2 (1.6) 1 (0.8) 1 (0.8) - Crematogaster sp.3 C - - 3 (2.5) - 1 (0.8) 12 (10) Crematogaster sp.5 C - - - 2 (1.6) 1 (0.8) - Crematogaster sp.6 C - - 10 (8.3) - - - Pheidole sp.1 C/P 2 (1.6) 1 (0.8) - - - - Pheidole sp.2 C/P - - - 1 (0.8) - - JJ RESENDE, PEC PEIXOTO, EN SILVA, JHC DELABIE & GMM SANTOS - Vegetation structure and arboreal ant diversity182 Table 1 (continued) Pheidole sp.5 C/P - - - - - 3 (2.5) Pheidole sp.10 C/P - 1 (0.8) 1 (0.8) - 3 (2.5) 1 (0.8) Pheidole sp.12 C/P - - - 3 (2.5) - - Pheidole sp.14 C/P - 6 (5) - 4 (3.3) - - Pheidole sp.15 C/P 1 (0.8) 1 (0.8) - - - - Pheidole sp.16 C/P - - - 5 (4.1) - 3 (2.5) Pheidole sp.17 C/P 1 (0.8) - 2 (1.6) - - - Pheidole sp.18 C/P 1 (0.8) - 1 (0.8) - - - Pheidole sp.20 C/P 1 (0.8) - - - - 3 (2.5) Pheidole sp.23 C/P 1 (0.8) - - - - 1 (0.8) Pheidole sp.24 C/P 2 (1.6) 1 (0.8) - 1 (0.8) - - Pheidole sp.26 C/P 3 (2.5) - 2 (1.6) - - - Carebara pilosa Fernández, 2004 P 2 (1.6) - - - - - Megalomyrmex sp.1 P - - - 2 (1.6) - - Solenopsis sp.1 P - - - - - 1 (0.8) Solenopsis sp.2 P - 5 (4.1) 24 (20) 2 (1.6) 4 (3.3) 38 (31.6) Solenopsis sp.3 P - - 2 (1.6) - 5 (4.1) - Solenopsis sp.4 P 2 (1.6) 2 (1.6) 4 (3.3) 7 (5.8) 1 (0.8) - Solenopsis sp.5 P - - 2 (1.6) - - 5 (4.1) Strumigenys sp.8 P - - - 1 (0.8) - - Strumigenys sp.9 P 7 (5.8) Wasmannia auropunctata (Roger, 1863) C/P - - - 1 (0.8) - - Basiceros (Octostruma) sp.2 P - - 1 (0.8) - - - PONERINAE Hypoponera sp.1 P - - - 1 (0.8) - - Hypoponera sp.2 P - - - 2 (1.6) - Pachycondyla apicalis (Latreille, 1802) P 2 (1.6) - - - - - Pachycondyla constricta (Mayr, 1884) P - - - - - 1 (1.6) Pachycondyla complexo villosa (Fabricius, 1804) P - - 5 (4.1) - - - Pachycondyla harpax (Fabricius, 1804) P - - - 1 (0.8) - - Pachycondyla venusta (Forel, 1912) P - 2 (1.6) 2 (1.6) - 3 (2.5) 1 (0.8) Pachycondyla villosa (Fabricius, 1804) P - - - - - 2 (1.6) Pachycondyla sp.1 P 1 (0.8) - - - - -