Open access journal: http://periodicos.uefs.br/index.php/sociobiology ISSN: 0361-6525 DOI: 10.13102/sociobiology.v69i3.8261Sociobiology 69(3): e8261 (September, 2022) Introduction Rainforest fragmentation is been a prominent topic of discussion between members of the scientific community for years (Skole & Tucker, 1993; Laurance et al., 200a; Vedovato et al., 2016). In general, historic forest fragmentation has occurred due to the subdivision of large areas of continuous Abstract Rainforest fragmentation drastically affects biodiversity and species composition, mainly due to habitat loss. Several studies have already shown the effects of forest fragmentation on plant and ant communities. To date, however, there is limited empirical knowledge of how forest fragmentation affects ant-plant interaction in networks. We investigated the effects of the configuration of rainforest fragments on the structure of ant-plant interaction networks mediated by extrafloral nectaries (EFNs). We carried out this study in ten forest fragments, ranging in size from approximately 5 to 3,000 ha, located in the Brazilian Amazon. In each fragment we established a plot of 6,250 m2, in which all ant-plant interactions were recorded, and calculated the following network descriptors: number of interactions, network size, network specialization, diversity of interactions, and nestedness. We used four explanatory variables to investigate the effects of forest fragmentation on these network descriptors: three metrics of the configuration of fragments (i.e., fragment area, edge irregularity, and connectivity) and the forest structure within each fragment, represented by canopy cover. We did not detect any effect of the explanatory variables on the network descriptors. The structural stability of the networks sampled in forest fragments with different configurations is possibly related to the observed constancy of ant species in the central core of highly interacting species. Our results corroborate other studies highlighting the structural stability of these facultative ant-plant networks mediated by EFNs in different spatial and temporal gradients. Nonetheless, the low constancy of plant species in the generalist core should be understood as a warning, mainly because the functionality of this protective mutualism (i.e., food secretions in exchange for protection against herbivory) remains unknown. Sociobiology An international journal on social insects Patricia N. Miranda1,2, José Eduardo L.S. Ribeiro2, Erick J. Corro3,4, Izaias Brasil5, Jacques H.C. Delabie6,7, Wesley Dáttilo3 Article History Edited by Kleber Del-Claro, UFU, Brazil Received 24 May 2022 Initial acceptance 25 May 2022 Final acceptance 08 August 2022 Publication date 29 September 2022 Keywords Community ecology; ecological networks; landscape ecology; plant-animal interactions; tropical rainforest. Corresponding authors Patricia Nakayama Miranda Instituto Federal de Educação, Ciência e Tecnologia do Acre Avenida Brasil, 920, Xavier Maia CEP: 69903-068, Rio Branco-AC, Brasil. E-Mail: patricia.miranda@ifac.edu.br Wesley Dáttilo Red de Ecoetología, Instituto de Ecología AC Carretera Antigua a Coatepec 351, El Haya, Xalapa, Veracruz, C.P.: 91070, Mexico. E-mail: wesley.dattilo@inecol.mx forest generating an altered landscape composed of forest remnants with different size, shape, and connectivity level (Klingbeil & Willig, 2009). This process is usually accompanied by habitat loss which involves the outright removal of habitat patches (Bartlett et al., 2016), thus making forest fragmentation one of the main drivers of biodiversity loss (Newbold et al., 2015). In the Brazilian Amazon, productive 1 - Instituto Federal de Educação, Ciência e Tecnologia do Acre, Campus Rio Branco, Rio Branco-AC, Brazil 2 - Centro de Ciências Biológicas, Universidade Estadual de Londrina, Londrina-PR, Brazil 3 - Red de Ecoetología, Instituto de Ecología AC, Xalapa, Veracruz, Mexico 4 - Facultad de Ciencias Biológicas y Agropecuarias, Universidad Veracruzana, Peñuela, Veracruz, Mexico 5 - Programa de Pós-Graduação em Biodiversidade e Biotecnologia da Rede BIONORTE, Universidade Federal do Acre, Rio Branco-AC, Brazil 6 - Comissão Executiva do Plano da Lavoura Cacaueira, Centro de Pesquisas do Cacau, Laboratório de Mirmecologia, Ilhéus-BA, Brazil 7 - Universidade Estadual de Santa Cruz, Ilhéus-BA, Brazil RESEARCH ARTICLE - ANTS Structural Stability of Ant-plant Mutualistic Networks Mediated by Extrafloral Nectaries: Looking at the Effects of Forest Fragmentation in the Brazilian Amazon mailto:patricia.miranda@ifac.edu.br Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes2 activities such as logging, cattle ranching, and small-scale farming are the main types of land use responsible for this process of landscape change (Laurance et al., 2001a; Laurance et al., 2011). Changes in landscape configuration (i.e., the spatial arrangement of habitat in a landscape), such as size, isolation and edge density (Haan et al., 2020), can lead to changes in plant and animal communities (Fischer & Lindenmayer, 2007; Laurance et al., 2011). For instance, forest area reduction may directly affect the rate of local species losses, with small forest remnants tending to lose species more quickly (Stouffer et al., 2008), due to their inability to support viable populations of certain groups (Didham et al., 1998). Connectivity reduction between forest fragments and/or continuous forests also tends to negatively affect the maintenance of some populations, since it changes resource availability spatially and, in consequence, the foraging behavior or pollination success of these populations (Haddad et al., 1999; Martensen et al., 2008; Hadley & Betts, 2011). Finally, the shape of forest fragments might also cause disruptions in community structure (DeSouza et al., 2001). For instance, more irregular forest fragments tend to present a higher proportion of edge in relation to the total area of the forest remnant, which can result in high proportions of pioneer species in regenerating and light demanding (Hill et al., 2003). Ants are frequently studied in forest fragments (Vasconcelos et al., 2006; Leal et al., 2012). In general, the population density, number and diversity of ant species decline with reductions in the area of forest fragments (Bruhl et al., 2003; Vasconcelos et al., 2006; Hill et al., 2011). The ant species composition may also change due to habitat loss and fragmentation (Vasconcelos et al., 2006; Leal et al., 2012; Ahuatzin et al., 2019), mainly due to local extinctions, or generalist or even invasive species colonization (i.e., biotic homogenization) (Carvalho & Vasconcelos, 1999; Bruhl et al., 2003; Holwaya & Suarez, 2006; Solar et al., 2015). However, the species richness and species composition do not seem to vary with shape complexity of fragments (Sobrinho & Schoereder, 2007). Many plant species, mainly in the tropics, secrete a liquid rich in carbohydrates and amino acids through specialized and non-floral glands called as a whole extrafloral nectaries (EFNs), which attracts different ant species (Koptur et al., 1998; Marques et al., 2015). In exchange for the food provided by the plant, some ants can protect their host plants against herbivores (Rico-Gray & Oliveira, 2007). At the community level, these mutualistic associations are usually evaluated using a network approach, in which different ant and plant species are depicted as nodes and their interactions as links (Dáttilo et al., 2013a). This network approach has provided important information about the organization of these ant- plant interactions (Lange et al., 2013; Dáttilo et al., 2013a, b; Dáttilo et al., 2014a). For instance, some studies have recently shown that ant-plant networks are highly nested, indicating that species with few links interact with a subset of interactive species with several interactions (Del-Claro et al., 2018). Biologically, nestedness describes the organization of the niche breadth of an interactive community, in which more nested networks tend to have the highest niche overlap (Blüthgen, 2010). In general, the structure of these ant-plant networks mediated by EFNs is relatively stable in terms of connectance, network specialization and nestedness, at different spatial (Dáttilo et al., 2013a) and temporal gradients (Díaz-Castelazo et al., 2013; Lange et al., 2013; Dáttilo et al., 2014a), or even after perturbations caused by tropical hurricanes (Sánchez- Galván et al., 2012). The structural stability of these networks is related to the facultative character of these ant-plant interactions, due the low fidelity of ant species when foraging on EFN-bearing plants (Schoereder et al., 2010). Furthermore, the ant species constituting the core of these networks (i.e., ant species that interact with several partner species at high frequencies) are mainly competitively superior species, which gives the network core certain spatial and temporal stability (Dáttilo et al., 2014b). More specialized ant-plant interactions, such as the interactions mediated by domatia, tend to present networks structurally stable in the context of fragmentation (Passmore et al., 2012), since the plant presence determines the ant presence. On the other hand, it is possible that ant-plant networks mediated by EFNs, even being stable in different natural environments, have their structure altered by forest fragmentation, since the associated species can be differently affected by this process. So we wonder if the structure of these ant-plant networks mediated by EFNs responds to a disturbance gradient generated by habitat loss and forest fragmentation. It is known that the structure of these ant-plant networks, for example, is influenced by the vegetation structure (Dáttilo & Dyer, 2014), which tends to be quite modified due to fragmentation (Ewers & Banks-Leite, 2013). Dáttilo and Dyer (2014) found that the diversity of ant-plant interactions may also be positively affected by canopy cover reduction, since some plant species secrete larger amounts of nectar in environments with higher light availability (Szabo, 1980; Kersch & Fonseca, 2005). Therefore, smaller, more irregular and less connected forest fragments, would present a greater number of interactions, with consequent increase in the overlap of pairwise interactions and nestedness. This is because forest fragments with a higher proportion of edges tend to present a reduction in canopy cover, with a consequent increase in light availability (Ewers & Banks-Leite, 2013). Sugiura (2010), in a study conducted on Japanese islands, showed that the connectance and nestedness values of ant- plant interaction networks mediated by EFNs tend to increase with decreasing island size. These results, although based on subtropical forest islands (Government of Japan, 2010), that present a matrix impossible to be used by ants and plants, at least suggested investigations on the effect of rainforest fragmentation on these ant-plant networks. Sociobiology 69(3): e8261 (September, 2022) 3 The aim of this study was therefore to investigate the effects of configurations of rainforest fragments on the structure of ant-plant interaction networks mediated by EFNs in the Brazilian Amazon. We based our research questions on hypotheses related to the configuration of forest remnants and the forest structure within the fragments, here represented by canopy cover. Specifically, we postulated that smaller, less connected, and more irregular forest fragments with lower canopy cover would present: (i) smaller network size, since less-preserved forest fragments tend to support fewer number of species (Ahuatzin et al., 2019); (ii) higher number of interactions, since extrafloral nectaries are more active in higher light environments (Radhika et al., 2010); (iii) less network specialization, since dominant ant species do not need to monopolize the resource (extrafloral nectar) when it is available in large amounts (Dáttilo et al., 2014b); (iv) lower diversity of interactions, due to the lower number of species expected for these fragments (Dátillo & Dyer, 2014); and (v) greater nestedness, due to the greater overlap of pairwise interactions arising from the greater number of interactions expected (Dáttilo et al., 2013b). Materials and methods Study area This study involved ten forest fragments, ranging in size from approximately 5 to 3,000 ha (Fig 1 and Supplementary Material 1), situated in the west of the state of Acre, in the Brazilian Amazon. All forest fragments result from human activities since the 1980s. Most of the primary type of vegetation was converted to croplands and pasture from this decade onwards (Acre, 2011). The matrix types (i.e., pasture, urban, agriculture) surrounding the forest remnants are decisive for the biodiversity in the landscapes (Prugh et al., 2008). The matrices will be more efficient from the point of view of functional connectivity if they are more similar in structure to the habitat patches (Prevedello & Vieira, 2010). For this reason, as an attempt to control the effect of matrix type, we selected only forest fragments surrounded by pasture. According to the Köppen (1936) climate classification, the region is classified as monsoon climate (Am), with an average rainfall of 1,450 mm per year (Macêdo et al., 2013) and marked seasonality, with most rainfall between November and March (Acre, 2006). The average annual temperature is 24ºC (INMET, 2016) with a daily thermal amplitude of around 9ºC (Acre, 2006). The region varies between 110 and 270 m.a.s.l. (Acre, 2006). The predominant vegetation type is open tropical rainforest (i.e., open ombrophilous forest) dominated by native species of bamboo and/or palm trees (Acre, 2006). Sampling ant-plant interactions We established a plot of 6,250 m2 (250 x 25 m) in each forest fragment, at a minimum distance of 100 meters from the edge, except for the smallest forest fragment, which was Fig 1. Map of the study area showing all forest fragments sampled. Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes4 a plot beginning at approximately 20 meters from the edge. We decided not to establish the plots in regions under direct effects of edges, because in these regions the microclimatic conditions, and community composition, tend to be highly altered in any forest fragment (Laurance et al., 2002), which would make it difficult to detect structure changes in ant-plant networks resulting from a disturbance gradient caused mainly due to fragmentation. We recorded all interactions between ants and EFN-bearing plants with a height accessible to the collector (ranging from 0.5 to 3 m) between 09:00 am and 03:00 pm for each plot (Dáttilo & Dyer, 2014). All EFN-bearing plants were observed for five minutes. Ants were considered to be feeding on nectar when they were immobile, with mouthparts in contact with nectar secreting tissues during the observation period (Falcão et al., 2016). When the interaction was confirmed, we collected all foraging ants and a plant sample for further identification (see below). Plants in which we did not detect the presence of EFNs, but which had immobile ants with mouthparts in contact with plant tissues, were also collected for later confirmation of the presence of EFNs through stereoscope observations and literature review. Ants were manually collected with the aid of an entomological umbrella (Bestelmeyer et al., 2000) so as to record those that drop from the plant at the slightest sign of disturbance (Dáttilo & Dyer, 2014). Considering the temporal variation in ant-plant networks mediated by EFNs (i.e., phenological variation) (Lange et al., 2013; Falcão et al., 2016), we sampled each plot twice (June 2016 and February 2017). Ants were identified by species or morphospecies using identification keys (Fernández, 2003; Wilson, 2003; Bolton et al., 2006) and by comparison with specimens deposited at the Laboratório de Mirmecologia of the Centro de Pesquisas do Cacau, Brazil (CPDC Collection), where voucher specimens were deposited. Plants were identified at species or morpho- species level using identification guides (Ribeiro et al., 1999; Pennington et al., 2004) and by comparison with specimens deposited at the Herbário do Parque Zoobotânico (HPZ) of the Universidade Federal do Acre, Brazil. Voucher specimens of plants were deposited at the Herbário (FUEL) of the Universidade Estadual de Londrina, Brazil. Canopy cover To measure the canopy cover in the plots where the ant- plant interactions were sampled, we established a 250 m long transect within each plot, located along the longitudinal direction of the plot, and maintained the same distance from the sides (12.5 m). In each transect, we installed ten subplots of 100 m2 (10 x 10 m) at intervals of 15 m. The first and last 100 subplots were installed 7.5 m from the beginning and end of transects, respectively. We estimated canopy cover using a spherical convex densiometer. The measurements were taken at the center and at the four edges of the 100 m2 subplots. The mean of these five measurements was used to calculate the average canopy cover per sampling point. We used the average of all sampling points to characterize each fragment for canopy cover. Landscape descriptors We used LANDSAT8 satellite images (path/row: 002/ 067) with a 30 m resolution, acquired on 07/23/2016, in the QGIS Software 2.18.10 (QGis Development Team, 2016) to describe the configuration of the forest fragments. The images were classified into two categories: forested areas (primary and secondary forests) and non-forested areas (water bodies, built areas, and pastures). We described the configuration of each forest fragment through metrics: forest fragment area (ha), edge irregularity index, and connectivity index. The forest fragment area was calculated based on its polygon shape. Patton’s diversity index (DI) (Patton, 1975) was adopted as the edge irregularity index, which evaluates the forest fragment regularity, defined as DI = P/2(√πA), where P = perimeter (m) and A = Area (m2). To measure the isolation and fragmentation degree around the forest focal fragments, we used the index of connectivity named proximity (PROX) (Gustafson & Parker, 1992), calculated by the following equation: In this equation, aijs represents the area (m 2) of fragment ijs within specified neighborhood (m) of the focal fragment; and hijs represents the distance (m) between fragment ijs and the focal fragment based on fragment edge- to-edge distance. The connectivity index is the inverse of the isolation and fragmentation degree (Bender et al., 2003). We define buffers of 500 m around the focal fragment, established from the respective edge, to define the neighboring fragments considered for the calculation of the index. We used this buffer size for two main reasons: i) this distance explains the greater variation in the patterns of richness and diversity of ant species at the community level (Spiesman & Cumming, 2008); and ii) it is equivalent to the greater observed distance of an ant gyne’s flight (Hölldobler & Wilson, 1990). All forest fragment metrics were calculated using QGIS Software 2.18.10 (QGis Development Team, 2016). Ant-plant interaction networks Initially, we accumulated the data of the two samples per plot on different dates, in order to maximize the number of EFN-bearing plants species in our samples, considering the annual variation in the extrafloral nectar production by different plant species (Falcão et al., 2016). To verify whether our sampling was adequate to describe the ant-plant networks of each plot studied, we then generated accumulation curves with the number of ant and plant species and the number of distinct pairwise of interactions, as a function of the number of plants sampled (Falcão et al., 2016). We used the non- parametric bootstrapping estimator with 1000 replicates (Gotelli & Colwell, 2001) for all accumulation curves, to estimate the expected number of plant and ant species and interactions for each network. ∑ = = n s ijs ijs h a PROX 1 2 Sociobiology 69(3): e8261 (September, 2022) 5 The ant-plant interaction patterns were examined using an ecological network approach. The data of each plot was organized in a quantitative matrix in which the elements (aij) represented the frequency (number of times) that plant species i interacted with ant species j inside the plot (Bascompte et al., 2003). To evaluate the constancy of the central core of highly interacting species among the different forest fragments sampled, we categorized ant and plant species according Dáttilo et al. (2013a): Gc = (ki - kmean)/σk, where ki = mean number of links for a given plant/ant species, kmean = mean number of links for all plant/ant species in the network, and σk = standard deviation of the number of links for plant/ant species. When Gc ≥ 1, the species presents a large number of interactions in relation to other species of the same trophic level, as a species constituting the generalist core. When Gc < 1, the species presents a lower number of interactions in relation to other species of the same trophic level, such as species constituting the periphery of networks. We used three network descriptors established from a quantitative matrix (number of interactions, network specialization, and diversity of interactions), and two calculated from a binary matrix (network size and nestedness), to describe the patterns of ant-plant interactions in the fragment. We adopted binary matrices for nestedness to facilitate the discussion of our results compared to other published studies, since this data category has been widely used to assess the structure of ant-plant networks mediated by EFNs (Rico-gray et al., 2012; Díaz-Castelazo et al., 2013; Lange et al., 2013; Falcão et al., 2016; Dáttilo et al., 2014a; Díaz-Castelazo et al., 2020). Furthermore, the nestedness values calculated with binary and quantitative data have been shown to be highly correlated (Corso et al., 2015; Miranda et al., 2019). In frequency matrices, as mentioned above, the elements represent the number of times in which a plant species interacted with an ant species within the plot, whereas in binary matrices the elements (1 or 0) represent the presence or absence of an interaction between a plant and an ant species within the plot. These binary-weighted descriptors are the most commonly used in studies dealing with ant-plant networks, and they cover a wide range of possible structures with complementary biological meanings (Del-Claro et al., 2016). Network size was calculated by multiplying the number of plant species by the number of ant species. The number of interactions represents the total interactions observed in the network, considering all ant and plant species. The diversity of interactions was calculated using an index (H’) based on the Shannon’s diversity index, which ranges from zero to infinity (Bersier et al., 2002). We estimated specialization using the H2’ index, which describes how species restrict their interactions from those randomly expected based on a partner’s availability (Blüthgen et al., 2006). In this index, the low specialization of an ecological network represents values close to H2’ = 0, and total specialization is H2’ = 1. Nestedness, which evaluates whether selective ant species only visit a subset of plant individuals visited by the generalist ant species (Dáttilo et al., 2014c), was calculated using the NODF-metric (Nestedness Based on Overlap and Decreasing Fill) (Almeida- Neto et al., 2008) in the ANINHADO software (Guimarães & Guimarães, 2006), using binary matrices. These index values range from 0 (non-nested) to 100 (perfectly nested). The network specialization and diversity of interaction were calculated using the bipartite package (Dormann et al., 2017) in R 3.2.3 (R Core Team, 2016). Data analysis The NODF significance was estimated using p-values based on the Null Model II (1000 randomizations). In this null model, the probability of an interaction occurring is proportional to the observed number of interactions of both plant and ant species (Bascompte et al., 2003). To evaluate the effect of rainforest fragmentation on ant-plant networks, we related each network descriptor (network size, number of interactions, network specialization - H2’, diversity of interactions - H’ and, nestedness - NODF) to the explanatory metrics of the configuration of forest fragments (fragment area, edge irregularity index, and connectivity index), and to canopy cover. Initially, we verified multicolinearity among all explanatory variables (fragment area, edge irregularity index, connectivity index, and canopy cover) using the variance inflation factor (VIF) value (Dormann et al., 2013). The VIF indicates the degree to which each explanatory variable is explained by another exploratory variable in the model. VIF values greater than 10 indicate high multicolinearity. As none of our exploratory variables showed collinearity between them, we included all four in the model selection. We used generalized linear models for all response variables (i.e., network descriptors). We used a negative binomial distribution for number of interactions and network size, due to an overdispersion of residues from our data, which occurs when the deviance of the response is greater than expected by the chosen distribution (Hinde & Demétrio, 1998). We used a Gaussian distribution for network specialization - H2’, diversity of interaction - H’, and nestedness - NODF. For all response variables, candidate models were classified using the corrected Akaike Information Criterion for small samples (AICc) (Akaike, 1974). For each response variable, 16 competing models were used to explain the patterns, including a null model representing the lack of effect. We adopted balanced model sets (i.e., all explanatory variables were present in the same number of models) without considering the interaction effects between them. Subsequently, the corrected Akaike Information criterion for small samples was estimated, the ΔAICc (difference between the AICc of each model in relation to the best model) and the Akaike weight (wAIC) (the probability of a given model being the best among a set of competing models) (Johnson & Omland, 2004). Models with ΔAICc <2.0 and wAIC> 0.1 were considered equally plausible to explain the patterns observed Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes6 (Burnham & Anderson, 2003). Model selections were made according to Akaike’s Information Criterion (AIC) using the MuMIn package (Bartoń, 2016) and the model diagnostics were run with the RT4Bio package (Reis Jr. et al., 2013) in R 3.2.3 (R Core Team, 2016). Results We observed 930 pairwise interactions between ants and EFN-bearing plants on the ten plots studied. We recorded a total of 56 ant species, distributed in 19 genera and seven subfamilies (Supplementary Material 2). The subfamily Myrmicinae comprised 48.2% of the total ant species (n = 27 ant species), followed by Formicinae (26.8%, n = 15) and Dolichoderinae (12.5%, n = 7). Myrmicinae was also the subfamily that showed higher interaction frequency (64.6% of the total interactions, n = 601 interactions), followed by Ectatomminae (13.6%, n = 126) and Formicinae (8.9%, n = 83). We recorded a total of 148 plant species, distributed in 44 genera and 25 families, where liana was the plant habit with the highest number of species (60.1% of the plant species, n = 89 species) followed by trees (34.5%, n = 51) (Supplementary Material 3). The Fabaceae family comprised 37.8% of the total plant species (n = 56 plant species), followed by Bignoniaceae (25.7%, n = 38) and Malpighiaceae (6.8%, n = 10). Fabaceae was also the family that had a higher interaction frequency (41% of the total interaction, n = 381), followed by Bignoniaceae (16.7%, n = 155) and Malpighiaceae (14.0%, n = 130). The generalist core of ant species in the ant-plant networks tends to be more constant than the generalist core of plant species, between forest fragments. We observed that the generalist core of ant-plant networks ranged from one to five ant species, and from two to five plant species per forest fragment. Seven ant species were present in the generalist cores, taking into account all ten ant-plant networks evaluated, and the most common species was Crematogaster brasiliensis (present in the generalist core in 70% of the networks, n = 7 forest fragments), followed by Crematogaster carinata (60%, n = 6) and Ectatomma tuberculatum (50%, n = 5) (Table 1). Twenty-one plant species were present in the generalist cores of the ant-plant networks evaluated, and the most common species were Bauhinia sp.1 (90%, n = 9) followed by Polygonaceae sp.1 (50%, n = 5) and Senegalia sp.2 (50%, n = 5) (Table 1). Generalist core plant species Occurrence frequency Generalist core ant species Frequency Bauhinia sp.1 90% (n = 9) Crematogaster brasiliensis 70% (n = 7) Polygonaceae sp1 50% (n = 5) Crematogaster carinata 60% (n = 6) Senegalia sp.2 50% (n = 5) Ectatomma tuberculatum 50% (n = 5) Fridericia sp.8 40% (n = 4) Crematogaster limata 40% (n = 4) Hirtella racemosa 40% (n = 4) Dolichoderus attelaboides 10% (n = 1) Banisteriopsis sp.2 40% (n = 4) Wasmannia auropunctata 10% (n = 1) Inga sp.5 40% (n = 4) Ochetomyrmex semipolitus 10% (n = 1) Palicourea sp.1 40% (n = 4) Senegalia sp.3 30% (n = 3) Fridericia sp.19 20% (n = 2) Memora sp.2 20% (n = 2) Zygia sp.1 20% (n = 2) Fridericia sp.6 20% (n = 2) Senegalia sp.8 20% (n = 2) Inga punctata 20% (n = 2) Aparisthmium cordatum 10% (n = 1) Fridericia sp.2 10% (n = 1) Inga sp.16 10% (n = 1) Senegalia sp.4 10% (n = 1) Bauhinia sp.4 10% (n = 1) Senegalia sp.7 10% (n = 1) Table 1. Frequency of ant and plant species in generalist core of ant-plant networks sampled in 10 rainforest fragments located in the state of Acre, Brazilian Amazon, between June 2016 and February 2017, n= number of fragments where the species was part of the core of the network. Sociobiology 69(3): e8261 (September, 2022) 7 According to the species and interactions accumulation curves estimated, we collected an average (Mean ± SD) of 84.30 ± 0.02% of the ant species (n = 18.60 ± 3.60 species of the 22.04 ± 4.04 species estimated) and 80.60 ± 0.01% of the plant species (n = 31.90 ± 4.53 species of the 39.61 ± 5.81 species estimated). We observed 76.23 ± 0.01% of the interactions between ants and EFN-bearing plants (n = 63.70 ± 11.06 interactions of the 83.57 ± 14.44 interactions estimated), indicating that we had sampled enough ant and plant species and interactions to describe the ant-plant networks of each plot. We describe the values of the network descriptors (network-size, number of interactions, network specialization – H2’, diversity of interactions – H’, and nestedness – NODF) for each plot in Supplementary Material 4. Eighty percent (n = 8 fragments) of the ant-plant networks presented a significant value for NODF, when compared to null models (Supplementary Material 4). We did not detect any effect of the explanatory variables on the network descriptors. For number of interaction, network size, network specialization and diversity of interaction, the null model, which indicates no detectable effects, was the best selected model (Table 2). For nestedness, the edge irregularity index was the best explanatory variable, however the null model was also equally plausible to explain the observed pattern (Table 2).The results of all candidate models for each ant-plant network descriptor are in Supplementary Material 5. Table 2. Summary of plausible models fitted to explain each ant-network descriptor in response to landscape structure metrics and forest structure (canopy cover) at 10 rainforest fragments located in the state of Acre, Brazilian Amazon. ΔAICc, df, and wAICc indicate the difference in corrected Akaike values, degrees of freedom of the model, and Akaike weights, respectively. The values in bold represent the best model selected for each of the network descriptors evaluated. Response variable Model ΔAICc d.f. wAIC Slope symbol Number of interactions Null model 0.00 2 0.39 Network size Null model 0.00 2 0.59 Network specialization Null Model 0.00 2 0.51 Diversity of interaction Null model 0.00 2 0.52 NODF Edge irregularity index 0.00 3 0.44 - Null model 1.20 2 0.247 Discussion In general, we observed, contrary to our hypotheses, that the structure of ant-plant interaction networks mediated by EFNs remained stable in forest fragments with different landscape configurations. None of the five network descriptors we used (i.e., number of interactions, network size, network specialization, diversity of interactions, and nestedness) were affected by the metrics of the configuration of fragments and vegetation structure. We also did not find evidence that canopy cover shapes the structure of ant-plant interaction networks, suggesting that the self-organization of the networks is independent of local characteristics of canopy openness in our area of study. The generalist core tended to be more constant among the sampled plots as regards ants rather than plants. Of the seven ant species observed in the generalist cores, four (57%) were core species in more than 40% of the networks. Between them, three are competitively superior due to their numerical dominance (C. brasiliensis, C. carinata and Crematogaster limata) (Parr, 2008; Baccaro et al., 2012) and one due to aggressive displacement of competitors (E. tuberculatum) (Hossaert-Mckey et al., 2001; Bächtold & Alves-Silva, 2013). This competitive nature is a feature of ant species of the generalist core (Dáttilo et al., 2014b), and is possibly one of the main factors responsible for the relative constancy of ant species in the studied networks. In addition, most of the ant species of our generalist cores belonged to the functional group Generalist Myrmicinae (Silvestre et al., 2003; Silva & Brandão, 2010), and are highly tolerant of habitat modification (Hoffmann & Andersen, 2003), which allows them to a competitive advantage, even in forest fragments with a high disturbance level. Of the 21 plant species observed in the generalist cores, 12 (57%) were classified as in the central core of highly interacting species in less than 20% of the networks. This lowest constancy in the generalist core of plant species is possibly related to changes in plant species composition due to forest fragmentation, since the reduction in size tends to increase the mortality rate of trees, especially near the forest edge (Laurance, 1991; Ferreira & Laurance, 1997), favoring fast-growing pioneer trees and liana growth (Laurance et al., 2001b; Laurance et al., 2006; Santo-Silva et al., 2016). The plant species turnover in South American tropical forests is also relatively high, with forests 60 km distant from each other (i.e., the maximum distance between our forest fragments), sharing only some 35% of species (Condit et al., 2002). When evaluating the effects of forest fragmentation on the structure of ant-plant networks, we found a structural stability in the face of perturbation. No network descriptor was related to the configuration of forest fragments. This Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes8 stability is possibly related to the low fidelity of ant species when foraging on EFN-bearing plants (Rico-Gray et al.; 1998, Schoereder et al., 2010), and to the constancy of ant species in the generalist cores observed in our study, as previously discussed in the literature (Dáttilo et al., 2013a; Lange et al., 2013; Dáttilo et al., 2014b). In fact, the ant-plant networks mediated by EFNs are shaped by a few generalist species, which are essential for the maintenance of network stability (Díaz-Castelazo et al., 2010; Mello et al., 2011). Dáttilo et al. (2013a) observed that the structure of ant-plant networks mediated by EFNs remains stable throughout space, as does the generalist core of species. Other studies have also detected the structural stability of these ant-plant networks mediated by EFNs over time (i.e., 20 years) (Díaz-Castelazo et al., 2013), between day and night periods (Dáttilo et al., 2014a), and after tropical hurricanes (Sánchez-Galván et al., 2012). More recently, Fagundes et al. (2018) also observed the limited effects of fire disturbances on the structure of ant-plant interaction networks mediated by EFNs. Corro et al. (2019) observed landscape effects on network specialization and on the diversity of interaction of ant-plant networks, but in this case, the authors used a conceptual framework based on co- occurrence records of ant-plant associations to build ant-plant co-occurrence networks. No network descriptors were related to vegetation structure (i.e., canopy cover). According to our hypotheses, we expected an increase in the number of interactions and, consequently, in nestedness, due to the canopy cover reduction. Dáttilo et al. (2013b) found no effects of canopy cover on the nestedness in ant-plant networks mediated by EFNs, in a study also carried out in the Brazilian Amazon. Our results corroborate the discussion presented by Dáttilo et al. (2013b), that perhaps the presence of ants in EFN-bearing plant species is not primarily determined by the amount of nectar, but rather by the nectar quality (i.e., concentration and composition) (Blüthgen & Fiedler, 2004; Lach, 2005). The amount of nectar is positively influenced by the light availability (Szabo, 1980; Kersch & Fonseca, 2005), but also by the water availability (Carroll et al., 2001), which is an abundant factor in tropical rainforests, not directly related to canopy cover. On the other hand, nectar quality, directly related to soil nutrients (Burkle & Irwin, 2009), may influence the behavior and preference of animal species when recruiting this food resource (Parachnowitsch et al., 2019), being therefore an interesting factor to be investigated in structuring of these ant-plant networks. The structural stability of these ant-plant networks mediated by EFNs evidences the resistance or resilience of these interactions in environments with different levels of disturbance. Even small, more irregular, and isolated forest fragments were able to maintain the dynamics of these ant- plant interactions, and therefore should be included in the definition of conservation strategies. Nonetheless, special attention should be directed to the low constancy of plant species in the generalist core. These small changes should be understood as a warning, mainly because the functionality of this protective mutualism (i.e., food secretions in exchange for protection against herbivory) in these areas remains unknown. Moreover, habitat loss and fragmentation often reduce gene flow and genetic diversity in plant populations (Browne & Karubian, 2018), and possibly in ant populations, as gene flow in this group is mainly maintained by male dispersion (Jaffé et al., 2009). This possible reduction in the gene flow, in turn, could negatively affect the long term maintenance of some populations and, consequently, the structure of ant- plant networks mediated by EFNs. It is important to mention that these results were found in open tropical forests (Acre, 2006), and possibly greater effects of forest fragmentation and forest structure can be expected for dense tropical forests. Furthermore, we limited our sampling to three meters high, and we recognize the potentially serious implications of this sampling limitation on our results. We thus suggest that further studies evaluate how ant-plant interactions are structured along a vertical stratification gradient within the forest and to test their vulnerability to fragmentation and habitat loss. In summary, we have provided some important insights into the effects of rainforest fragmentation on the structure of ant-plant interactions mediated by EFNs. In an interaction networks context, the ant species of the generalist core seem to be little affected by forest fragmentation compared to the plant species of the generalist core. Our results regarding the network descriptors indicate that the structure of ant-plant networks mediated by EFNs remains relatively stable in the face of the forest fragmentation, possibly due to the constancy of ant species in the generalist cores. In short, this study makes a valuable contribution to biodiversity and conservation, mainly because we show the vulnerability and robustness of tropical species-rich habitats to forest fragmentation. Acknowledgements The authors thank Christian Frenopoulo for the English review and Dr. Milton Cezar Ribeiro for his important suggestions for the statistical analyses used in this work. The first author thanks the Capes Research Grant received during the period in which she stayed at the Instituto de Ecologia, A.C. (Mexico), and the Instituto de Educação, Ciência e Tecnologia do Acre (IFAC), for financial support throughout her entire doctorate course. The fifth author acknowledges his research grant from CNPq (project 304629/2018-9). References Acre (2006). Governo do Estado do Acre. Programa Estadual de Zoneamento Ecológico-Econômico do Estado do Acre. Zoneamento Ecológico-Econômico do Acre Fase II: documento Síntese-escala 1: 250.000. SEMA, Rio Branco, p 356. Acre (2011). Governo do Estado do Acre. Programa Estadual de Zoneamento Ecológico-Econômico do Estado do Acre. Zoneamento Ecológico-Econômico do Acre: o uso da terra acreana com sabedoria. SEMA, Rio Branco, p 149. Sociobiology 69(3): e8261 (September, 2022) 9 Ahuatzin, D.A., Corro, E.J., Aguirre, A., Valenzuela-González, J., Feitosa, R.M, Ribeiro, M.C., Acosta, J.C., Coates, R. & Dáttilo, W. (2019). Forest cover drives leaf litter ant diversity in primary rainforest remnants within human-modified tropical landscapes. Biodiversity and Conservation, 28: 1091-1107. doi: 10.1007/s10531-019-01712-z Akaike, H. (1974). A New look at the statistical model identification. IEEE Transactions on Automatic Control, 19: 716-723. doi: 10.1109/TAC.1974.1100705 Almeida-Neto, M., Guimarães, P., Guimarães, P.R., Loyola, R.D. & Ulrich, W. (2008). A consistent metric for nestedness analysis in ecological systems: reconciling concept and measurement. Oikos, 117: 1227-1239. doi: 10.1111/j.0030-1299.2008.16644.x Bächtold, A. & Alves-Silva, E. (2013) Behavioral strategy of a lycaenid (Lepidoptera) caterpillar against aggressive ants in a Brazilian savanna. Acta Ethologica, 16: 83-90. doi: 10.1007/s10211-012-0140-2 Baccaro, F.B., de Souza, J.L.P., Franklin, E., Landeiro, V.L. & Magnusson, W.E. (2012). Limited effects of dominant ants on assemblage species richness in three Amazon forests. Ecological Entomology, 37: 1-12. doi: 10.1111/j.1365-2311. 2011.01326.x Bartlett, L.J., Newbold, T., Purves, D.W., Tittensor, D.P. & Harfoot, M.B.J. (2016). Synergistic impacts of habitat loss and fragmentation on model ecosystems. Proceedings of the Royal Society B: Biological Sciences, 283: 20161027. doi: 10.1098/rspb.2016.1027 Bartoń, K. (2016). MuMIn: Multi-model inference. R packages version 1.15.6. https://cran.r-project.org/web/packages/MuMIn/ MuMIn.pdf Bascompte, J., Jordano, P., Melián, C.J. & Olesen, J.M. (2003). The nested assembly of plant-animal mutualistic networks. Proceedings of the National Academy of Sciences of the United States of America, 100: 9383-9387. doi: 10.1073/pnas. 1633576100 Bender, D.J., Tischendorf, L. & Fahrig, L. (2003). Using patch isolation metrics to predict animal movement in binary landscapes. Landscape Ecology, 18: 17-39. doi: 10.1023/A: 1022937226820 Bersier, L.F., Banasek-Richter, C. & Cattin, M.F. (2002). Quantitative descriptors of food-web matrices. Ecology, 83: 2394-2407. doi: 10.1890/0012-9658(2002)083[2394:QDOF WM]2.0.CO Bestelmeyer, B.T., Agosti, D., Alonso, L.T., Brandão, 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, L.T. Alonso & T. Schultz (Eds.). Ants: standard methods for measuring and monitoring biodiversity (pp. 122-144). Washington, DC: Smithsonian Institution Press. Blüthgen, N. & Fiedler, K. (2004). Preferences for sugars and amino acids and their conditionality in a diverse nectar– feeding ant community. Journal of Animal Ecology, 73: 155- 166. doi: 10.1111/j.1365-2656.2004.00789.x Blüthgen, N., Menzel, F. & Blüthgen, N. (2006). Measuring specialization in species interaction networks. BMC Ecology, 6: 9. doi: 10.1186/1472-6785-6-9 Blüthgen, N. (2010). Why network analysis is often disconnected from community ecology: a critique and an ecologist’s guide. Basic and Applied Ecology, 11: 185-195. doi: 10.1016/j. baae.2010.01.001 Bolton, B., Alpert, G.D., Ward, P.S. & Naskrecki P (2006). Bolton’s catalogue of ants of the world (CD). Cambridge: Harvard University Press. Browne, L. & Karubian, J. (2018). Habitat loss and fragmentation reduce effective gene flow by disrupting seed dispersal in a neotropical palm. Molecular Ecology, 27: 3055-3069. doi: 10.11 11/mec.14765 Bruhl, C.A., Eltz, T. & Linsenmair, K.E. (2003). Size does matter – effects of tropical rainforest fragmentation on the leaf litter ant community in Sabah, Malaysia. Biodiversity and Conservation, 12: 1371-1389. doi: 10.1023/A:1023621609102 Burkle, L.A. & Irwin, R.E. (2009). The effects of nutrient addition on floral characters and pollination in two subalpine plants, Ipomopsis aggregata and Linum lewisii. Plant Ecology, 203: 83-98. doi: 10.1007/s11258-008-9512-0 Burnham, K.P. & Anderson, D.R. (2002). Model selection and multi-model inference. New York: Springer. Carvalho, K.S. & Vasconcelos, H.L. (1999). Forest fragmentation in central Amazonia and its effects on litter-dwelling ants. Biological Conservation, 91: 151-157. doi: 10.1016/S0006-3207 (99)00079-8 Carroll, A.B., Pallardy, S.G. & Galen, C. (2001). Drought stress, plant water status, and floral trait expression in fireweed, Epilobium angustifolium (Onagraceae). American Journal of Botany, 88: 438-46. doi: 10.2307/2657108 Condit, R., Pitman, N., Leigh, E.G., Chaves, J., Terborgh, J., Foster, R.B., Núñez, P., Aguilar, S., Valencia, R., Villa, G., Muller-Landau, H.C., Losos, E. & Hubbell, S.P. (2002). Beta- diversity in tropical forest trees. Science, 295: 666-669. doi: 10.1126/science.1066854 Corro, E.J., Ahuatzin, D.A., Aguirre, A., Favila, M.E., Ribeiro, M.C., Acosta, J.C.L. & Dáttilo, W. (2019). Forest cover and landscape heterogeneity shape ant-plant co-occurrence networks in human-dominated tropical rainforests. Landscape Ecology, 34: 93-104. doi: 10.1007/s10980-018-0747-4 Corso, G., Cruz, C.P.T., Pinto, M.P., de Almeida, A.M. & Lewinsohn, T.M. (2015). Binary versus weighted interaction networks. Ecological Complexity, 23: 68-72. https://doi.org/10.1007/s10531-019-01712-z https://doi.org/10.1109/TAC.1974.1100705 https://doi.org/10.1111/j.0030-1299.2008.16644.x https://doi.org/10.1111/j.0030-1299.2008.16644.x https://doi.org/10.1007/s10211-012-0140-2 https://docs.wixstatic.com/ugd/e086b2_2239344d69bb464db034b296317ec4b1.pdf https://docs.wixstatic.com/ugd/e086b2_2239344d69bb464db034b296317ec4b1.pdf https://doi.org/10.1111/j.1365-2311.2011.01326.x https://doi.org/10.1111/j.1365-2311.2011.01326.x https://doi.org/10.1111/j.1365-2311.2011.01326.x https://doi.org/10.1098/rspb.2016.1027 https://cran.r-project.org/web/packages/MuMIn/MuMIn.pdf https://cran.r-project.org/web/packages/MuMIn/MuMIn.pdf doi: https://doi.org/10.1073/pnas.1633576100 https://doi.org/10.1073/pnas.1633576100 http://doi.org/10.1023/A:1022937226820 http://doi.org/10.1023/A:1022937226820 https://doi.org/10.1890/0012-9658(2002)083%5b2394:QDOFWM%5d2.0.CO;2 https://doi.org/10.1890/0012-9658(2002)083%5b2394:QDOFWM%5d2.0.CO;2 https://doi.org/10.1111/j.1365-2656.2004.00789.x https://doi.org/10.1186/1472-6785-6-9 https://doi.org/10.1016/j.baae.2010.01.001 https://doi.org/10.1016/j.baae.2010.01.001 https://doi.org/10.1111/mec.14765 https://doi.org/10.1111/mec.14765 http://doi.org/10.1023/A:1023621609102 https://doi.org/10.1007/s11258-008-9512-0 https://doi.org/10.1016/S0006-3207(99)00079-8 https://doi.org/10.1016/S0006-3207(99)00079-8 https://doi.org/10.2307/2657108 https://doi.org/10.1126/science.1066854 https://doi.org/10.1007/s10980-018-0747-4 Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes10 Dáttilo, W., Guimarães Jr, P.R. & Izzo, T.J. (2013a). Spatial structure of ant – plant mutualistic networks. Oikos, 122: 1643-1648. doi: 10.1111/j.1600-0706.2013.00562.x Dáttilo, W., Rico-Gray, V., Rodrigues, D.J. & Izzo, T.J. (2013b). Soil and vegetation features determine the nested pattern of ant-plant networks in a tropical rainforest. Ecological Entomology, 38: 374-380. doi: 10.1111/een.12029 Dáttilo, W. & Dyer, L. (2014). Canopy openness enhances diversity of ant–plant interactions in the Brazilian Amazon rainforest. Biotropica, 46: 712-719. doi: 10.1111/btp.12157 Dáttilo, W., Fagundes, R., Gurka, C.A.Q., Silva, M.A.S., Vieira, M.C.L., Izzo, T.J., Díaz-Castelazo, C., Del-Claro, K. & Rico-Gray, V. (2014a). Individual-based ant-plant networks: diurnal-nocturnal structure and species-area relationship. PLoS One, 9: e99839. doi: 10.1371/journal.pone.0099838 Dáttilo, W., Díaz-Castelazo, C. & Rico-Gray, V. (2014b). Ant dominance hierarchy determines the nested pattern in ant-plant networks. Biological Journal of the Linnean Society, 113: 405-414. doi: 10.1111/bij.12350 Dáttilo, W., Marquitti, F.M.D., Guimarães, P.R. & Izzo, T.J. (2014c). The structure of ant-plant ecological networks: is abundance enough? Ecology, 95: 475-485. doi: 10.1890/12- 1647.1 Del-Claro, K., Rico-Gray, V., Torezan-Silingardi, H.M., Alves-Silva, E., Fagundes, R., Lange, D., Dáttilo, W., Vilela, A.A., Aguirre, A. & Rodríquez-Morales, D. (2016). Loss and gains in ant-plant interactions mediated by extrafloral nectar: fidelity, cheats, and lies. Insectes Sociaux, 63: 207-221. doi: 10.1007/s00040-016-0466-2 Del-Claro, K., Lange, D., Torezan-Silingardi, H.M., Anjos, D.V., Calixto, E.S., Dáttilo, W. & Rico-Gray, V. (2018). The complex relationships between ants and plants into tropical ecological networks. In W. Dáttilo & V. Rico-Gray (Eds.). Ecological Networks in the Tropics: An Integrative Overview of Species Interactions from Some of the Most Species-Rich Habitats on Earth (pp. 59-71), Springer Publisher. DeSouza, O., Schoereder, J.H., Brown, V.K. & Bierregaard, R.O. (2001). A theoretical overview of the processes determining species richness in forest fragments. In R.O. Bierregaard, C. Gascon, T.F. Lovejoy & A.A. Santos (Eds.). Lessons from Amazonia: the ecology and conservation of a fragmented forest (pp. 13-20), Yale University Press, New Haven. Díaz-Castelazo, C., Guimarães Jr, P.R., Jordano, P., Thompson, J.N., Marques, R.J., Rico-Gray, V. (2010) Changes of a mutualistic network over time: reanalysis over a 10-year period. Ecology, 91: 793-801. doi: 10.1890/08-1883.1 Díaz-Castelazo, C., Sánchez-Galva, I.R., Guimarãe Jr, P.R., Raimundo, R.L.G. & Rico-Gray, V. (2013). Long-term temporal variation in the organisation of an ant–plant network. Annals of Botany, 111: 1285-1293. doi: 10.1093/aob/mct071 Díaz-Castelazo, C., Martínez-Adriano, C.A., Dáttilo, W., Rico-Gray, V. (2020). Relative contribution of ecological and biological attributes in the fine-grain structure of ant-plant networks. Peer J., 8: e8314. doi: 10.7717/peerj.8314 Didham, R.K., Hammond, P.M., Lawton, J.H., Eggleton, P. & Stork, N.E. (1998). Beetle species responses to tropical forest fragmentation. Ecological Monograph, 68: 295-303. doi: 10.1890/0012-9615(1998)068[0295:BSRTTF]2.0.CO;2 Dormann, C.F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J.R.G., Gruber, B., Lafourcade, B., Leitão, P.J., Münkemüller, T., McClean, C., Osborne, P.E., Reineking, B., Schröder, B., Skidmore, A.K., Zurell, D. & Lautenbach, S. (2013). Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36:001-020. doi: 10.1111/j.1600-0587.2012.07348.x Dormann, C.F., Fruend, J. & Gruber, B. (2017). Bipartite: visualizing bipartite networks and calculating some (ecological) indices. R package version 2.08. https://cran.r-project.org/ web/packages/bipartite/bipartite.pdf Ewers, R.M. & Banks-Leite, C. (2013). Fragmentation impairs the microclimate buffering effect of tropical forests. PLoS ONE, 8: 1-7. doi: 10.1371/journal.pone.0058093 Fagundes, R., Lange, D., Anjos, D.V., Lima, F.P., Nahas, L., Corro, E., Silva, P.B.G., Del-Claro, K., Ribeiro, S.P. & Dáttilo, W. (2018). Limited effects of fire disturbances on the species diversity and structure of ant-plant interaction networks in Brazilian Cerrado. Acta Oecologica, 93: 65-73. doi: 10.1016/j.actao.2018.11.001 Falcão, J.C.F., Dátillo, W. & Rico-Gray, V. (2016). Sampling effort differences can lead to biased conclusions on the architecture of ant-plant interaction networks. Ecological Complexity, 25: 44-52. doi: 10.1016/j.ecocom.2016.01.001 Fernández, F. (2003). Subfamília Formicinae. In F. Fernández (Ed,) Introduccion a las hormigas de la region Neotropical (pp. 299-306). Bogotá: Instituto de Investigación de Recursos Biológicos Alexander Von Humbolt. Ferreira, L.V. & Laurance, W.F. (1997). Effects of forest fragmentation on mortality and damage of selected trees in Central Amazonia. Conservation Biology, 11: 797-801. doi: 10.1046/j.1523-1739.1997.96167.x Fischer, J. & Lindenmayer, D.B. (2007). Landscape modification and habitat fragmentation: a synthesis. Global Ecology and Biogeography, 16: 265-280. doi: 10.1111/j.1466-8238.2007. 00287.x Gotelli, N.J. & Colwell, R.K. (2001). Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters, 4: 379-391. doi: 10.1046/j. 1461-0248.2001.00230.x Government of Japan (2010). Nomination of the Ogasawara island for inscription on the world heritage list. http:// https://doi.org/10.1111/j.1600-0706.2013.00562.x https://doi.org/10.1111/een.12029 https://doi.org/10.1111/btp.12157 https://doi.org/10.1371/journal.pone.0099838 https://doi.org/10.1111/bij.12350 https://doi.org/10.1890/12-1647.1 http://dx.doi.org/10.1890/12-1647.1 http://dx.doi.org/10.1890/12-1647.1 https://doi.org/10.1007/s00040-016-0466-2 https://doi.org/10.1890/08-1883.1 http://dx.doi.org/10.1093/aob/mct071 https://doi.org/10.7717/peerj.8314 http://esajournals.onlinelibrary.wiley.com/hub/journal/10.1002/(ISSN)1557-7015/ https://doi.org/10.1890/0012-9615(1998)068%5b0295:BSRTTF%5d2.0.CO;2 https://doi.org/10.1111/j.1600-0587.2012.07348.x https://cran.r-project.org/web/packages/bipartite/bipartite.pdf https://cran.r-project.org/web/packages/bipartite/bipartite.pdf https://doi.org/10.1371/journal.pone.0058093 doi: https://doi.org/10.1016/j.actao.2018.11.001 https://doi.org/10.1016/j.ecocom.2016.01.001 https://doi.org/10.1046/j.1523-1739.1997.96167.x https://doi.org/10.1111/j.1466-8238.2007.00287.x https://doi.org/10.1111/j.1466-8238.2007.00287.x https://doi.org/10.1046/j.1461-0248.2001.00230.x https://doi.org/10.1046/j.1461-0248.2001.00230.x http://ogasawara-info.jp/pdf/isan/recommendation_en.pdf Sociobiology 69(3): e8261 (September, 2022) 11 ogasawara-info.jp/pdf/isan/recommendation_en.pdf. Accessed 15 August 2017 Guimarães, P.R. & Guimarães, P. (2006). Improving the analyses of nestedness for large sets of matrices. Environmental Modelling & Software, 21: 1512-1513. doi: 10.1016/j.envsoft. 2006.04.002 Gustafson, E.J. & Parker, G.R. (1992). Relationship between land cover proportion and indices of spatial pattern. Landscape Ecology, 7: 10-1110. doi: 10.1007/BF02418941 Haan, N.L., Zhang, Y. & Landis, D.A. (2020). Predicting Landscape Configuration Effects on Agricultural Pest Suppression. Trends in Ecology and Evolution, 35: 175-186. doi: 10.1016/j.tree.2019.10.003 Haddad, N.M. (1999). Corridor and distance effects on interpatch movements: a landscape experiment with butterflies. Ecological Applications, 9: 612-622. doi: 10.1890/ 1051-0761(1999)009[0612:CADEOI]2.0.CO;2 Hadley, A.S. & Betts, M.G. (2011). The effects of landscape fragmentation on pollination dynamics: absence of evidence not evidence of absence. Biological Reviews of the Cambridge Philosophical Society, 87: 526-544. doi: 10.1111/j.1469-185X. 2011.00205.x Hill, J.L. & Curran, P.J. (2003). Area, shape and isolation of tropical forest fragments: effects on tree species diversity and implications for conservation. Journal of Biogeography, 30: 1391-1403. doi: 10.1046/j.1365-2699.2003.00930.x Hill, J.K., Gray, M.A., Khen, C.V., Benedick, S., Tawatao, N. & Hamer, K.H. (2011). Ecological impacts of tropical forest fragmentation: how consistent are patterns in species richness and nestedness? Philosophical Transactions of the Royal Society B, 366: 3265-3276. doi: 10.1098/rstb.2011.0050 Hinde, J. & Demétrio, C.G.B. (1998). Overdispersion: Models and estimation. Computational Statistics and Data Analysis, 27:151–170. doi: 10.1016/S0167-9473(98)00007-3 Hoffmann, B.D. & Andersen, A.N. (2003). Responses of ants to disturbance in Australia, with particular reference to functional groups. Austral Ecology, 28: 444-464. doi: 10.1046/ j.1442-9993.2003.01301.x Hölldobler, B. & Wilson, E.O. (1990). The ants. Harvard: Harvard University Press. Holwaya, D.A. & Suarez, A.V. (2006). Homogenization of ant communities in mediterranean California: The effects of urbanization and invasion. Biological Conservation, 127: 319-326. doi: 10.1016/j.biocon.2005.05.016 Hossaert-McKey, M., Orivel, J., Labeyrie, E., Pascal, L., Delabie, J.H.C. & Dejean, A. (2001). Differential associations with ants of three co-occurring extrafloral nectary-bearing plants. Écoscience, 8: 325-335. doi: 10.1080/11956860.2001.11682660 INMET (2016). Instituto Nacional de Meteorologia. IOP Publishing Physics Web: http://www.inmet.gov.br/portal/ index.php?r=home2/index Accessed 13 June 2017 Jaffé, R., Moritz, R.F.A. & Kraus, B. (2009). Gene flow is maintained by polyandry and male dispersal in the army ant Eciton burchellii. Population Ecology, 51: 227-236. doi: 10.1007/s10144-008-0133-1 Johnson, J.B. & Omland, K.S. (2004). Model selection in ecology and evolution. Trends in Ecololy and Evolution, 19: 101-08. doi: 10.1016/j.tree.2003.10.013 Kersch, M.F. & Fonseca, C.R. (2005). Abiotic factors and the conditional outcome of an ant-plant mutualism. Ecology, 86: 2117-2126. doi: 10.1890/04-1916 Klingbeil, B.T. & Willig, M.R. (2009). Guild-specific responses of bats to landscape composition and configuration in fragmented Amazonian rainforest. Journal of Applied Ecology, 46: 203-213. doi: 10.1111/j.1365-2664.2008.01594.x Köppen, W. (1936). Das Geographisches System der Klimate. In W. Köppen & W. Geiger (Eds.). Handbuch der Klimatologie (Kapitel 3 V.1). Berlin: Teil C Ebr Bornträger. Koptur, S., Rico–Gray, V. & Palacios–Rios, M. (1998). Ant protection of the nectaried fern Polypodium plebeium in central Mexico. American Journal of Botany, 85: 736-739. doi: 10.2307/2446544 Lach, L. (2005). Interference and exploitation competition of three nectar-thieving invasive ant species. Insectes Sociaux, 52: 257-262. doi: 10.1007/s00040-005-0807-z Lange, D., Dáttilo, W. & Del-Claro, K. (2013). Influence of extrafloral nectary phenology on ant-plant mutualistic networks in a neotropical savana. Ecological Entomology, 38: 463-469. doi: 10.1111/een.12036 Laurance, W.F. (1991). Ecological correlates of extinction proneness in Australian tropical rainforest mammals. Conservation Biology, 5: 79-89. doi: 10.1111/j.1523-1739.1991.tb00390.x Laurance, W.F., Cochrane, M., Bergen, S., Fearnside, P.M., Delamonica, P., Barber, C., D’Angelo, S. & Fernandes, T. (2001a). The future of the Brazilian Amazon. Science, 291: 438-439. doi: 10.1126/science.291.5503.438 Laurance, W.F., Perez-Salicrup, D., Delamonica, P., Fearnside, P.M., D’Angelo, S., Jerozolinski, A., Pohl, L. & Lovejoy, T.E. (2001b). Rain forest fragmentation and the structure of Amazonian liana communities. Ecology, 82: 105-116. doi: 10.1890/0012-9658(2001)082[0105:RFFATS]2.0.CO;2 Laurance, W.F., Lovejoy, T.E., Vasconcelos, H.L., Bruna, E.M., Didham, R.K., Stouffer, P.C., Gascon,C., Bierregaard, R.O., Laurance, S.G. & Sampaio, E. (2002). Ecosystem Decay of Amazonian Forest Fragments: A 22-Year Investigation. Conservation Biology, 16: 605-618, doi: 10.1046/j.1523-1739. 2002.01025.x http://ogasawara-info.jp/pdf/isan/recommendation_en.pdf https://doi.org/10.1016/j.envsoft.2006.04.002 https://doi.org/10.1016/j.envsoft.2006.04.002 https://doi.org/10.1007/BF02418941 https://doi.org/10.1016/j.tree.2019.10.003 https://doi.org/10.1890/1051-0761(1999)009[0612:CADEOI]2.0.CO;2 https://doi.org/10.1890/1051-0761(1999)009[0612:CADEOI]2.0.CO;2 https://www.ncbi.nlm.nih.gov/pubmed/22098594 https://www.ncbi.nlm.nih.gov/pubmed/22098594 https://doi.org/10.1111/j.1469-185X.2011.00205.x https://doi.org/10.1111/j.1469-185X.2011.00205.x https://doi.org/10.1046/j.1365-2699.2003.00930.x https://www.ncbi.nlm.nih.gov/pubmed/22006967 https://www.ncbi.nlm.nih.gov/pubmed/22006967 http://doi.org/10.1098/rstb.2011.0050 https://doi.org/10.1016/S0167-9473(98)00007-3 https://doi.org/10.1046/j.1442-9993.2003.01301.x https://doi.org/10.1046/j.1442-9993.2003.01301.x https://www.google.com.br/search?hl=pt-BR&tbo=p&tbm=bks&q=inauthor:%22Bert+H%C3%B6lldobler%22 https://doi.org/10.1016/j.biocon.2005.05.016 https://doi.org/10.1080/11956860.2001.11682660 http://www.inmet.gov.br/portal/index.php?r=home2/index Accessed 13 June 2017 https://doi.org/10.1007/s10144-008-0133-1 https://doi.org/10.1016/j.tree.2003.10.013 https://doi.org/10.1890/04-1916 https://doi.org/10.1111/j.1365-2664.2008.01594.x https://doi.org/10.2307/2446544 http://dx.doi.org/10.1007/s00040-005-0807-z https://doi.org/10.1111/een.12036 https://doi.org/10.1111/j.1523-1739.1991.tb00390.x http://doi.org/10.1126/science.291.5503.438 https://doi.org/10.1890/0012-9658(2001)082%5b0105:RFFATS%5d2.0.CO;2 https://doi.org/10.1046/j.1523-1739.2002.01025.x https://doi.org/10.1046/j.1523-1739.2002.01025.x Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes12 Laurance, W.F., Nascimento, H., Laurance, S.G., Andrade, A., Ribeiro, J., Giraldo, J.P., Lovejoy, T.E., Condit, R., Chave, J. & D’Angelo, S. (2006). Rapid decay of tree community composition in Amazonian forest fragments. Proceedings of the National Academy of Sciences of the United States of America, 103: 19010-19014. doi: 10.1073/pnas.0609048103 Laurance, W.F., Camargo, J.L.C., Luizão, R.C.R., Laurance, S.G., Pimm, S.L., Bruna, E.M., Stouffer, P.C., Williamson, G.B., Benítez- Malvido, J., Vasconcelos, H., Houtan, K.S., Zartman, C.E., Boyle, S.A., Didhan, R.K., Andrade, A. & Lovejoy, T.E. (2011). The fate of Amazon forest fragments: A 32-years investigation. Biological Conservation, 144: 56- 67. doi: 10.1016/j.biocon.2010.09.021 Leal, I.R., Filgueiras, B.K.C., Gomes, J.P., Iammuazzi, L. & Andersen, A.N. (2012). Effects of habitat fragmentation on ant richness and functional composition in Brazilian Atlantic forest. Biological Conservation, 21: 1687-1701. doi: 10.1007/ s10531-012-0271-9 Macêdo, M.N.C., Dias, H.C.T., Coelho, F.M.G., Araújo, E.A., Souza, M.L.H. & Silva, E. (2013). Precipitação pluviométrica e vazão da bacia hidrográfica do Riozinho do Rôla, Amazônia Ocidental. Ambiente & Água, 8: 206-221. doi: 10.4136/ambi- agua.809 Marques, T.E.D., Castano-Meneses, G., Mariano, C.S.F. & Delabie, J.H.C. (2015). Interações entre Poneromorfas e fontes de açúcar na vegetação. In J.H.C Delabie, R.M. Feitosa, J.E. Serrão, C.S.F. Mariano & J.D Majer (Eds.), As formigas Poneromorfas do Brasil (pp. 361-374). Ilhéus: Editus. doi: 10.7476/9788574554419.0024 Martensen, A.C., Pimentel, R.G. & Metzger, J.P. (2008). Relative effects of fragment size and connectivity on bird community in the Atlantic Rain Forest: Implications for conservation. Biological Conservation, 141: 2184–-192. doi: 10.1016/j.biocon.2008.06.008 Mello, M.A.R., Santos, G.M.M., Mechi, M.R. & Hermes, M.G. (2011) High generalisation in flower-visiting networks of social wasps. Acta Oecologica, 167: 131-140. doi: 10.1016/j. actao.2010.11.004 Miranda, P.N., Ribeiro, J.E.L.S., Luna, P., Brasil, I., Delabie, J.H.C. & Dáttilo, W. (2019). The dilemma of binary or weighted data in interaction networks. Ecological Complexity, 38: 1-10. doi: 10.1016/j.ecocom.2018.12.006 Newbold, T., Hudson, L.N., Hill, S.L.L, Contu, S., Lysenko, I., Senior, R.A., Börger, L., Bennett, D.J., Choimes, A., Collen, B., Day, J., De Palma, A., Díaz, S., Echeverria-Londoño, S., Edgar, M.J., Feldman, A., Garon, M., Harrison, M.L.K., Alhusseini, T., Ingram, D.J., Itescu Y., Kattge, J., Kemp, V., Kirkpatrick, L., Kleyer, M., Correia, D.L.P, Martin, C.D., Meiri S., Novosolov, M., Pan, Y., Phillips, H.R.P., Purves, D.W., Robinson, A., Simpson, J., Tuck, S.L., Weiher, E., White, H.J., Ewers, R.M., Mace, G.M., Scharlemann, J.P.W. & Purvis, A. (2015). Global effects of land use on local terrestrial biodiversity. Nature, 520: 45-50. doi: 10.1038/nature14324 Parachnowitsch, A.L. Manson, J.S. & Slevtold, N. (2019). Evolutionary ecology of nectar. Annals of Botany, 123: 247- 261. doi: 10.1093/aob/mcy132 Parr, C.L. (2008). Dominant ants can control assemblage species richness in a South African savanna. Journal of Animal Ecology, 77: 1191-1198. doi: 10.1111/j.1365-2656. 2008.01450.x Passmore, H.A., Bruna, E.M., Heredia, S.M., Vasconcelos, H. L. (2012). Resilient Networks of Ant-Plant Mutualists in Amazonian Forest Fragments. Plos One, 7: e40803. doi: 10.1371/journal.pone.0040803 Patton, D.R. (1975). A diversity index for quantifying habitat “edge”. Wildlife Society Bulletin, 3: 171-173. https://www. jstor.org/stable/3781151 Pennington, T.D., Reynel, C. & Daza, A. (2004) Illustrated guide to the trees of Peru. Sherborne, England, D. Hunt. Prevedello, J.A. & Vieira, M.V. (2010). Does the type of matrix matter? A quantitative review of the evidence. Biodiversity and Conservation, 19: 1205-1223. doi: 10.1007/ s10531-009-9750-z Prugh, L.R., Hodges, K.E., Sinclair, A.R.E. & Brashares, J.S. (2008). Effect of habitat area and isolation on fragmented animal populations. Proceedings of the National Academy of Sciences of the United States of America, 105: 20770-20775. doi: 10.1073/pnas.0806080105 QGIS Development Team (2016). QGIS Geographic information system. Open source geospatial foundation project. https://qgis. org/en/site/ Accessed 15 May 2017 R Core Team (2016). R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. https://www.r-project.org/ Accessed 30 July 2017 Radhika, C., Kost, C., Mithöfer, A. & Boland, A.W. (2010). Regulation of extrafloral nectar secretion by jasmonates in lima bean is light dependent. Proceedings of the National Academy of Sciences of the United States of Americ, 107: 17228-17233. doi: 10.1073/pnas.1009007107 Reis Jr., R., Oliveira, M.L. & Borges, G.R.A. (2013). RT4Bio: R Tools for Biologists (RT4Bio). R packages version 1.0. Ribeiro, J.E.L. da S., Hopkins, M.J.G., Vicentini, A., Sothers, C.A., Costa, M.A. da S., de Brito, J.M., de Souza, M.A.D., Martins, L.H.P., Lohmann, L.G., Assunção, P.A.C.L., Pereira, E. da C., da Silva, C.F., Mesquita, M.R. & Procopio, L.C. (1999). Flora da reserva Ducke: guia de identificação das plantas vasculares de uma floresta de terra-firme na Amazonia Central. Manaus: INPA. Rico-Gray, V., García-Franco, J.G., Palacios-Rios, M., Díaz-Castelazo, C., Parra-Tabla, V. & Navarro, J.A. (1998). https://doi.org/10.1073/pnas.0609048103 https://doi.org/10.1016/j.biocon.2010.09.021 http://doi.org/10.1007/s10531-012-0271-9 http://doi.org/10.1007/s10531-012-0271-9 https://doi.org/10.4136/ambi-agua.809 https://doi.org/10.4136/ambi-agua.809 http://dx.doi.org/10.7476/9788574554419.0024 https://doi.org/10.1016/j.biocon.2008.06.008 https://doi.org/10.1016/j.actao.2010.11.004 https://doi.org/10.1016/j.actao.2010.11.004 https://doi.org/10.1016/j.ecocom.2018.12.006 http://dx.doi.org/10.1038/nature14324 https://doi.org/10.1093/aob/mcy132 https://doi.org/10.1111/j.1365-2656.2008.01450.x https://doi.org/10.1111/j.1365-2656.2008.01450.x https://doi.org/10.1371/journal.pone.0040803 https://www.jstor.org/stable/3781151 https://www.jstor.org/stable/3781151 https://doi.org/10.1007/s10531-009-9750-z https://doi.org/10.1007/s10531-009-9750-z https://doi.org/10.1073/pnas.0806080105 https://qgis.org/en/site/ https://qgis.org/en/site/ https://www.r-project.org/ https://doi.org/10.1073/pnas.1009007107 Sociobiology 69(3): e8261 (September, 2022) 13 Geographical and seasonal variation in the richness of ant- plant interactions in Mexico. Biotropica, 30: 190-200. doi: 10.1111/j.1744-7429.1998.tb00054.x Rico-Gray, V. & Oliveira, P.S. (2007). The ecology and evolution of ant-plant interactions. Chicago: University of Chicago Press. Rico-Gray, V., Díaz-Castelazo, C., Ramírez-Hernández, A., Guimarães Jr., P.R. & Holland, J.N. (2012). Abiotic factors shape temporal variation in the structure of an ant-plant network. Arthropod-Plant Interactions, 6: 289-295. doi: 10.1007/ s11829-011-9170-3 Sánchez-Galván, I.R., Díaz-Castelazo, C. & Rico-Gray, V. (2012). Effect of hurricane Karl on a plant–ant network occurring in coastal Veracruz, Mexico. Journal of Tropical Ecology, 28: 603-609. doi: 10.1017/S0266467412000582 Santo-Silva, E.E., Almeida, W.R., Tabarelli, M. & Peres, C.A. (2016). Habitat fragmentation and the future structure of tree assemblages in a fragmented Atlantic forest landscape. Plant Ecology, 217: 1129-1140. doi: 10.1007/s11258-016-0638-1 Schoereder, J.H., Sobrinho, T.G., Madureira, M.S., Ribas, C.R. & Oliveira, P.S. (2010). The arboreal ant community visiting extrafloral nectaries in the Neotropical cerrado savanna. Terrestrial Arthropod Reviews, 3: 3-27. doi: 10.1163/187498 310X487785 Silva, E.R. & Brandão, C.R.F. (2010). Morphological patterns and community organization in leaf-litter ant assemblages. Ecological Monographs, 80: 107-124. doi: 10.1890/08-1298.1 Silvestre, R., Brandão, C.R.F. & Silva, R.R. (2003). Grupos funcionales de hormigas: El caso de los 693 gremios del Cerrado, Brasil. In F. Fernàndez (Ed.). Introducción a las Hormigas de la Región Neotropical (pp. 113-143). Bogotá: Instituto Humboldt. Skole, D. & Tucker, C. (1993). Tropical deforestation and habitat fragmentation in the Amazon: Satellite data from 1978 to 1988. Science, 260: 1905-1910. doi: 10.1126/science.260. 5116.1905 Sobrinho, T.G. & Schoereder, J.H. (2007). Edge and shape effects on ant (Hymenoptera: Formicidae) species richness and composition in forest fragments. Biodiversity and Conservation, 16: 1459-1470. doi: 10.1046/j.1365-2699.2003.00930.x Solar, R.R., Barlow, J., Ferreira, .J, Berenguer, E., Lees, A.C., Thomson, J.R., Louzada, J., Maués, M., Moura, N.G., Oliveira, V.H.F., Chaul, J.C.M., Schoereder, J.H., Vieira, I.C.G., Nally, R.M. & Gardner, T.A. (2015). How pervasive is biotic homogenization in human-modified tropical forest landscapes? Ecology Letters, 18: 1108-1118. doi: 10.1111/ele.12494 Spiesman, B.J. & Cumming, G.S. (2008). Communities in context: the influences of multiscale environmental variation on local ant community structure. Landscape Ecology, 23: 313-325. doi: 10.1007/s10980-007-9186-3 Stouffer, P.C., Strong, C. & Naka, L.N. (2008). Twenty years of understory bird extinctions from Amazonian rain forest fragments: consistent trends and landscape-mediated dynamics. Diversity and Distributions, 15: 88-97. doi: 10.1111/ j.1472-4642.2008.00497.x Sugiura, S. (2010). Species interactions-area relationships: biological invasions and network structure in relation to island area. Proceedings of the Royal Society B: Biological Sciences, 277: 1807-1815. doi: 10.1098/rspb.2009.2086 Szabo, T.I. (1980). Effect of weather factors on honeybee a flight activity and colony weight gain. Journal of Apicultural Search, 19: 164-171. doi: 10.1080/00218839.1980.11100017 Vasconcelos, H.L., Vilhena, M.S., Magnusson, W.E. & Albernaz, A.L.K.M. (2006). Long-term effects of forest fragmentation on Amazonian ant communities. Journal of Biogeography, 33: 1348-1356. doi: 10.1111/j.1365-2699.2006.01516.x Vedovato, L.B., Fonseca, M.G., Arai, E., Anderson, L.O. & Aragão, L.E.O.C. (2016). The extent of 2014 forest fragmentation in the Brazilian Amazon. Regional Environmental Change, 16: 2485-2490. doi: 10.1007/s10113-016-1067-3 Wilson, E.O. (2003). Pheidole in the new world: a dominant, hyperdiverse ant genus. Harvard Cambridge: University Press. https://doi.org/10.1111/j.1744-7429.1998.tb00054.x https://doi.org/10.1007/s11829-011-9170-3 https://doi.org/10.1007/s11829-011-9170-3 https://doi.org/10.1017/S0266467412000582 https://doi.org/10.1007/s11258-016-0638-1 https://doi.org/10.1163/187498310X487785 https://doi.org/10.1163/187498310X487785 https://doi.org/10.1890/08-1298.1 https://doi.org/10.1126/science.260.5116.1905 https://doi.org/10.1126/science.260.5116.1905 http://dx.doi.org/10.1046/j.1365-2699.2003.00930.x https://doi.org/10.1111/ele.12494 https://doi.org/10.1007/s10980-007-9186-3 https://doi.org/10.1111/j.1472-4642.2008.00497.x https://doi.org/10.1111/j.1472-4642.2008.00497.x https://doi.org/10.1098/rspb.2009.2086 https://doi.org/10.1080/00218839.1980.11100017 https://doi.org/10.1111/j.1365-2699.2006.01516.x https://doi.org/10.1007/s10113-016-1067-3 Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes14 Fragment Site Geographic coordinates Area (ha) 1 Senador Guiomard private fragment 10° 3’59.16”S and 67°59’20.12”W 5.26 2 Senador Guiomard private fragment 10° 4’52.34”S and 67°36’2.64”W 27.82 3 Senador Guiomard private fragment 10°6’58.02”S and 67°41’6.90”W 123.16 4 Forestry School 9°59’48.12”S and 67°59’20.12”W 332.15 5 Projeto de Assentamento Walter Arce 9°48’0.46”S and 67°51’26.95”W 681.05 6 Porto Acre private fragment 9°36’28.60”S and 67°34’6.15”W 1072.34 7 Catuaba Experimental Farm 10°04’48.9”S and 67°37’08.6”W 1282.42 8 Embrapa Acre 10°2’17.64”S and 67°40’54.24”W 1871.17 9 Senador Guiomard private fragment 10° 1’24.66”S and 67°35’48.66”W 2894.77 10 Humaitá Reserve 09°45’15.2”S and 67°39’44.9”W 3042.02 Subfamily Ant species Frequency Total F7 F8 F4 F3 F10 F6 F1 F2 F9 F5 Dolichoderinae Azteca chartiflex 0 0 0 0 0 0 0 1 0 0 1 Azteca sp.1 5 2 0 0 0 6 0 6 1 3 23 Dolichoderus attelaboides (Fabricius, 1775) 2 1 0 1 4 5 0 5 0 8 26 Dolichoderus bispinosus (Olivier, 1792) 0 0 0 0 0 8 0 1 0 2 11 Dolichoderus debilis Emery, 1890 0 0 0 0 0 6 10 0 0 0 16 Dolichoderus quadridenticulatus (Roger, 1862) 0 0 0 0 1 0 0 0 0 0 1 Dolichoderus septemspinosus Emery, 1894 1 0 0 0 0 1 0 2 0 0 4 Ecitoninae Eciton mexicanum Roger, 1863 0 0 0 0 0 0 0 1 0 0 1 Ectatomminae Ectatomma tuberculatum (Oliver, 1792) 14 21 1 28 2 0 13 9 24 11 123 Gnamptogenys moelleri (Forel, 1912) 0 0 0 0 0 0 0 0 1 1 2 Gnamptogenys sulcata (Smith, 1858) 0 0 0 0 0 0 0 0 0 1 1 Brachymyrmex heeri Forel, 1874 2 0 8 0 0 0 1 1 2 1 15 Camponotus bidens Mayr, 1870 0 0 0 0 0 1 0 0 0 0 1 Camponotus cingulatus Mayr, 1862 0 0 0 1 0 0 2 0 0 0 3 Camponotus depressus Mayr, 1866 0 0 0 0 1 1 0 0 0 1 3 Camponotus femoratus (Fabricius, 1804) 0 0 0 0 0 1 0 0 0 0 1 Formicinae Camponotus godmani Forel, 1899 0 0 0 0 0 0 0 1 0 0 1 Camponotus latangulus Roger, 1863 4 3 0 1 0 10 2 11 2 0 33 Camponotus nidulans (Fr. Smith, 1860) 0 1 0 0 2 5 0 0 2 2 12 Camponotus prox. novogranadensis 3 0 0 0 0 0 0 0 0 0 3 Camponotus punctulatus andigenus Emery, 1903 1 0 0 0 0 1 0 0 1 1 4 Supplementary Material Appendix S1. Location of the forest fragments sampled in State of Acre, Brazilian Amazon. Appendix S2. Ant species present in the ant-plant networks sampled in 10 rainforest fragments located in the state of Acre, Brazilian Amazon, between June 2016 and February 2017. Sociobiology 69(3): e8261 (September, 2022) 15 Appendix S2. Ant species present in the ant-plant networks sampled in 10 rainforest fragments located in the state of Acre, Brazilian Amazon, between June 2016 and February 2017. (Continuation) Camponotus sexguttatus (Fabricius, 1793) 0 1 0 0 0 0 0 0 0 0 1 Formicinae Camponotus sp.1 0 0 0 0 0 0 0 0 2 1 3 Gigantiops destructor (Fabricius, 1804) 0 0 0 1 0 0 0 0 0 1 2 Nylanderia guatemalensis (Forel, 1885) 0 0 0 0 0 1 0 0 0 0 1 Myrmicinae Cephalotes atratus (Linnaeus, 1758) 0 1 0 0 0 0 0 0 0 0 1 Cephalotes marginatus (Fabricius, 1804) 0 0 0 0 0 0 1 0 0 0 1 Cephalotes pinelii (Guérin-Méneville, 1844) 1 0 0 0 0 0 0 0 0 0 1 Cephalotus opacus Santschi, 1920 0 0 0 0 2 0 0 0 0 0 2 Crematogaster brasiliensis Mayr, 1878 10 2 0 24 27 23 0 69 19 8 182 Crematogaster carinata Mayr, 1862 0 5 26 40 16 12 69 1 0 9 178 Crematogaster curvispinosa Mayr, 1862 0 0 0 0 0 2 0 0 0 0 2 Crematogaster erecta Mayr, 1866 0 0 0 0 0 1 0 0 0 0 1 Crematogaster flavosensitiva Longino, 2003 0 3 0 0 5 2 0 0 0 0 10 Crematogaster limata Fr. Smith, 1858 14 8 1 0 2 16 1 8 12 16 78 Crematogaster longispina Emery, 1890 2 0 0 1 2 0 9 0 0 0 14 Crematogaster nigropilosa Mayr, 1870 1 0 4 1 1 0 0 1 0 0 8 Crematogaster sp.1 0 0 0 0 2 0 0 0 0 0 2 Megalomyrmex balzani Emery, 1894 0 0 0 0 2 0 0 0 0 0 2 Ochetomyrmex semipolitus Mayr, 1878 2 7 2 2 9 1 7 1 2 5 38 Pheidole (gr. Fallax) sp.1 1 0 0 1 0 0 0 0 1 0 3 Pheidole (gr. Fallax) sp.2 2 0 0 0 0 0 0 0 0 0 2 Pheidole (gr. Fallax) sp.3 0 0 1 0 2 0 1 1 0 0 5 Pheidole (gr. Fallax) sp.4 0 1 0 0 1 0 1 2 0 1 6 Pheidole (gr. Flavens) sp.1 0 1 0 0 0 0 0 0 0 0 1 Pheidole radoszkowskii Mayr, 1884 0 0 2 1 1 0 1 0 1 3 9 Solenopsis globularia (Smith, 1858) 0 0 1 0 0 0 1 0 0 0 2 Solenopsis sp.1 0 6 2 0 0 0 0 1 0 0 9 Solenopsis sp.3 1 3 0 0 0 0 1 0 4 0 9 Wasmannia auropunctata (Roger, 1863) 2 5 3 2 8 3 3 0 3 5 34 Ponerinae Neoponera carinulata (Roger, 1861) 0 0 0 0 2 0 0 0 0 0 2 Neoponera unidentada Mayr, 1862 2 0 1 0 2 2 0 0 0 0 7 Odontomachus haematodus (Linnaeus, 1758) 1 0 0 0 0 7 2 0 0 0 10 Odontomachus hastatus (Fabricius, 1804) 0 0 0 0 0 0 0 0 0 1 1 Pseudomyrmecinae Pseudomyrmex oculatus (Smith, 1855) 0 0 0 0 0 2 0 1 1 0 4 Pseudomyrmex tenuis (Fabricius, 1804) 3 1 0 2 0 1 3 2 1 1 14 Subfamily Ant species Frequency Total F7 F8 F4 F3 F10 F6 F1 F2 F9 F5 Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes16 Appendix S3. Plant species present in the ant-plant networks sampled in 10 rainforest fragments located in the state of Acre, Brazilian Amazon, between June 2016 and February 2017. Habitat: T = tree; L = liana; H = herb; ? = undefined. Family Plant species Frequency Total Habitat F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 Anonaceae Anonaceae sp.1 0 0 0 0 1 0 0 0 0 0 1 T Araceae Philodendron sp.1 0 0 0 0 0 0 0 0 0 1 1 L Fridericia sp.1 0 0 0 0 0 0 1 0 0 0 1 L Fridericia sp.2 0 8 7 0 0 0 3 0 2 0 20 L Fridericia sp.3 3 0 0 0 0 0 0 4 0 0 7 L Fridericia sp.4 0 0 0 0 1 0 0 0 0 1 2 L Fridericia sp.5 0 0 1 0 2 1 0 4 0 2 10 L Fridericia sp.6 0 0 0 0 0 0 0 5 0 0 5 L Fridericia sp.7 0 0 0 0 0 0 1 0 3 0 4 L Fridericia sp.8 10 6 0 1 1 11 0 1 0 0 30 L Fridericia sp.9 1 0 0 0 1 0 0 0 2 0 4 L Fridericia sp.10 1 0 0 0 0 0 0 0 0 0 1 L Fridericia sp.11 2 0 0 0 0 1 0 0 0 0 3 L Fridericia sp.12 1 0 0 0 0 0 0 0 2 0 3 L Fridericia sp.13 1 1 2 1 0 0 0 0 1 0 6 L Fridericia sp.14 2 0 0 0 0 0 0 0 0 0 2 L Fridericia sp.15 1 0 0 0 0 1 0 0 0 0 2 L Fridericia sp.16 0 3 0 0 0 0 0 0 1 0 4 L Fridericia sp.17 0 0 0 0 0 1 0 0 1 0 2 L Fridericia sp.18 0 0 0 0 0 1 0 0 0 0 1 L Bignoniaceae Fridericia sp.19 0 0 1 0 7 0 0 0 0 0 8 L Fridericia sp.20 0 0 2 0 0 0 0 0 0 0 2 L Fridericia sp.21 0 1 0 0 0 0 0 2 0 0 3 L Fridericia sp.22 0 0 0 1 0 0 0 0 0 0 1 L Fridericia sp.23 0 0 0 0 1 0 0 0 0 0 1 L Fridericia sp.24 0 0 0 0 2 0 0 0 0 0 2 L Fridericia sp.25 0 0 0 0 0 0 0 1 0 0 1 L Fridericia sp.26 0 0 1 0 0 0 0 3 0 0 4 L Fridericia sp.27 0 0 0 0 0 0 0 1 0 0 1 L Fridericia sp.28 0 0 1 0 0 0 0 0 0 0 1 L Fridericia sp.29 0 0 0 0 1 0 0 0 0 0 1 L Fridericia sp.30 0 0 0 0 0 0 0 0 0 1 1 L Fridericia sp.31 0 0 0 0 1 0 0 0 2 0 3 L Fridericia sp.32 0 0 0 0 0 0 0 2 1 0 3 L Fridericia sp.33 0 1 0 0 0 0 0 0 0 0 1 L Fridericia sp.34 0 0 0 0 1 0 0 0 0 0 1 L Fridericia sp.35 0 1 0 0 3 0 0 1 1 0 6 L Memora sp.1 1 0 0 0 0 0 0 0 0 0 1 L Memora sp.2 0 1 0 0 5 0 0 0 0 0 6 L Memora sp.3 0 0 0 0 0 1 0 0 0 0 1 L Sociobiology 69(3): e8261 (September, 2022) 17 Chrysobalanaceae Hirtella racemosa Lam. 0 0 0 0 9 1 0 6 0 0 16 T Hirtella sp.1 0 0 0 0 0 0 0 1 0 1 2 T Hirtella sp.2 0 0 0 0 0 0 0 0 1 0 1 T Combretaceae Buchenavia sp.1 1 0 0 0 0 0 0 0 0 0 1 T Convolvulaceae Ipomoea philomega (Vell.) House 0 0 0 0 0 0 0 0 1 0 1 L Ipomoea regnellii Meisn. 0 0 0 0 1 0 0 0 0 0 1 L Costaceae Costus scaber Ruiz & Pav. 0 1 0 1 0 0 1 1 0 2 6 H Cucurbitaceae Cucurbitaceae sp.1 0 0 1 0 0 0 0 0 0 0 1 L Gurania sp.1 0 0 1 0 0 1 0 0 0 0 2 L Euphorbiaceae Acalypha sp.1 0 0 0 0 0 0 0 0 4 0 4 ? Aparisthmium cordatum (A. Juss.) Baill. 0 10 0 0 0 0 0 0 0 0 10 T Dalechampia sp.1 1 0 0 0 0 0 0 0 0 0 1 L Omphalea diandra L. 0 0 0 0 0 2 2 0 0 2 6 L Senegalia sp.1 3 0 1 1 3 0 3 1 1 3 16 L Senegalia sp.2 7 4 2 1 0 6 9 3 8 1 41 L Senegalia sp.3 3 7 0 1 2 1 1 0 6 0 21 L Senegalia sp.4 1 0 1 11 0 0 3 0 0 0 16 L Senegalia sp.5 0 0 0 0 0 0 0 0 1 0 1 L Senegalia sp.6 0 0 0 0 1 0 0 0 0 0 1 L Senegalia sp.7 2 0 0 4 0 0 1 0 0 0 7 L Senegaliasp.8 1 0 1 3 1 0 3 1 5 0 15 L Senegalia sp.9 0 1 0 0 0 0 0 0 0 0 1 L Bauhinia sp.1 33 15 16 0 0 9 17 2 7 0 99 L Bauhinia sp.2 0 0 0 1 0 0 0 0 0 0 1 L Bauhinia sp.3 0 0 0 0 0 1 0 0 0 0 1 L Bauhinia sp.4 0 0 0 3 0 0 0 0 0 0 3 L Fabaceae Bauhinia sp.5 0 0 0 0 0 0 0 1 0 0 1 L Bauhinia sp.6 0 0 0 0 0 2 0 0 0 0 2 L Centrosema sp.1 0 0 0 0 0 0 3 0 0 0 3 L Centrosema sp.2 0 0 0 0 0 0 2 0 0 0 2 L Centrosema sp.3 0 0 2 0 0 0 0 0 0 0 2 L Erythrina sp.1 0 0 1 2 0 0 0 0 0 0 3 T Fabaceae sp.1 0 1 3 0 0 0 0 0 0 0 4 ? Inga acreana Harms 0 0 0 0 0 1 0 0 0 1 2 T Inga alba (Sw.) Willd. 0 1 3 1 1 0 0 0 0 0 6 T Inga calantha Ducke 0 0 0 0 1 0 0 0 0 1 2 T Inga capitata Desv. 1 1 0 0 0 2 0 0 0 0 4 T Inga chartacea Poepp. 0 0 1 0 0 0 0 0 0 2 3 T Inga densiflora Benth. 0 1 2 0 0 2 0 1 0 0 6 T Appendix S3. Plant species present in the ant-plant networks sampled in 10 rainforest fragments located in the state of Acre, Brazilian Amazon, between June 2016 and February 2017. Habitat: T = tree; L = liana; H = herb; ? = undefined. (Continuation) Family Plant species Frequency Total Habitat F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes18 Inga edulis Mart. 2 0 1 1 0 0 1 0 1 0 6 T Inga heterophylla Willd. 1 0 0 0 0 0 0 0 0 0 1 T Inga lateriflora Miq. 1 0 2 1 2 0 3 0 1 0 10 T Inga laurina (Sw.) Willd. 6 2 2 4 0 0 0 0 0 4 18 T Inga microcoma Harms 0 0 0 0 1 0 0 0 0 0 1 T Inga punctata Willd. 0 0 0 1 0 0 0 0 0 5 6 T Inga sertulifera DC. 0 0 2 0 3 3 0 0 0 0 8 T Inga suaveolens Ducke 1 0 0 0 0 0 0 0 0 0 1 T Inga tenuistipula Ducke 0 0 0 0 2 0 1 1 0 3 7 T Inga sp.1 0 0 0 0 0 1 0 0 0 0 1 T Inga sp.2 0 0 0 0 0 0 0 0 2 0 2 T Inga sp.3 0 0 0 0 0 2 1 0 0 0 3 T Inga sp.4 1 0 0 1 0 0 0 1 0 2 5 T Inga sp.5 0 0 0 0 1 1 1 6 0 6 15 T Fabaceae Inga sp.6 0 0 0 0 0 0 0 0 0 1 1 T Inga sp.7 0 0 2 0 0 1 0 0 0 0 3 T Inga sp.9 1 0 0 0 0 0 0 0 2 0 3 T Inga sp.10 0 0 0 1 0 0 0 1 0 0 2 T Inga sp.12 0 0 0 2 0 0 0 0 0 0 2 T Inga sp.13 0 0 0 2 0 0 0 0 0 0 2 T Inga sp.14 0 0 1 0 0 0 1 0 1 0 3 T Inga sp.15 0 1 0 0 0 0 0 0 0 0 1 T Inga sp.16 0 0 3 0 0 0 0 0 0 0 3 T Inga sp.18 0 0 0 0 0 1 0 0 0 0 1 T Inga sp.20 0 0 0 0 0 1 0 0 0 0 1 T Senna sp.1 0 0 0 0 1 0 0 0 0 0 1 T Zygia sp.1 0 0 1 0 4 0 0 0 1 0 6 T Zygia sp.2 0 3 0 0 0 0 0 0 0 0 3 T Zygia sp.3 0 1 0 0 0 0 0 0 0 0 1 T Zygia sp.4 0 1 0 0 0 0 0 0 0 0 1 T Lecythidaceae Gustavia augusta L. 0 0 0 0 2 0 0 0 2 0 4 T Lecythidaceae sp.1 0 0 0 0 3 0 0 0 0 0 3 T Loganiaceae Strychnos panurensis Sprague & Sandwith 0 0 0 1 1 0 0 0 0 0 2 L Malpighiaceae Banisteriopsis sp.1 0 0 0 0 0 4 0 0 0 0 4 L Banisteriopsis sp.2 2 0 0 3 0 44 3 2 0 41 95 L Banisteriopsis sp.3 1 1 0 0 0 1 1 0 0 0 4 L Banisteriopsis sp.4 0 1 1 0 0 0 0 0 0 0 2 L Banisteriopsis sp.5 0 9 1 0 0 0 0 0 0 0 10 L Banisteriopsis sp.6 0 0 0 0 3 0 0 0 0 0 3 L Banisteriopsis sp.7 0 0 0 0 1 0 0 0 0 0 1 L Banisteriopsis sp.8 0 0 0 0 0 0 0 0 0 1 1 L Heteropterys sp1 0 1 0 0 2 0 0 0 0 0 3 L Tetrapterys sp.1 1 0 0 0 0 0 2 0 0 4 7 ? Appendix S3. Plant species... (Continuation) Family Plant species Frequency Total Habitat F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 Sociobiology 69(3): e8261 (September, 2022) 19 Malvaceae Byttneria benensis Britton 0 0 0 0 0 5 0 0 1 0 6 L Menispermaceae Abuta sp.1 0 0 0 0 0 0 0 0 1 0 1 L Abuta sp.2 0 0 0 0 0 0 0 1 0 0 1 L Ochinaceae Ouratea sp.2 0 0 0 0 3 1 0 1 0 0 5 T Olacaceae Heisteria sp.1 0 0 0 2 3 0 0 1 0 1 7 ? Passifloraceae Dilkea sp.1 0 0 1 0 0 0 0 0 0 0 1 L Passiflora coccinea Aublet. 1 0 2 0 0 0 0 0 0 0 3 L Passiflora sp.1 1 0 0 0 0 0 0 0 0 0 1 L Passiflora sp.2 0 0 0 0 0 0 1 0 0 0 1 L Passiflora sp.3 0 0 0 0 0 1 0 0 0 0 1 L Passiflora sp.4 0 0 0 0 0 0 0 1 0 0 1 L Polygonaceae Polygonaceae sp.1 26 38 31 0 2 1 2 8 11 0 119 ? Polygonaceae sp.2 0 0 0 0 0 0 1 0 0 0 1 T Rhamnaceae Gouania frangulifolia Radlk. 0 0 0 0 0 0 0 0 1 0 1 L Rubiaceae Tocoyena sp.1 0 0 1 0 0 0 0 0 0 0 1 T Palicourea sp.1 0 3 2 0 1 5 2 6 2 5 26 T Sapindaceae Paullinia sp.1 0 0 0 0 0 0 2 1 0 0 3 L Paullinia sp.2 3 0 0 1 1 0 0 0 0 1 6 L Paullinia sp.3 3 0 0 0 0 0 0 0 0 0 3 L Paullinia sp.5 1 0 0 0 0 0 0 0 0 0 1 L Paullinia sp.6 0 0 0 0 0 0 0 0 1 0 1 L Serjania clematidea Triana & Plach. 0 0 0 0 0 0 0 0 2 2 4 L Solanaceae Solanum sp.1 0 0 1 0 0 0 0 0 0 0 1 ? Vitaceae Cissus sp.1 0 0 0 0 0 0 0 1 0 0 1 L Volchysiaceae Qualea grandiflora Mart. 0 0 2 0 0 0 2 0 0 0 4 T Volchysia sp.1 0 0 0 0 0 1 0 0 0 0 1 T Appendix S3. Plant species... (Continuation) Family Plant species Frequency Total Habitat F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 Appendix S4. Network size, number of interaction, network specialization, diversity of interaction and nestedness (NODF) of the ant-plant networks sampled in 10 rainforest fragments located in the state of Acre, Brazilian Amazon, between June 2016 and February 2017. Fragment Network size Number of interactions Network specialization Diversity of interaction NODF 1 648 128 0.15 3.81 36.70* 2 580 125 0.19 3.70 37.32* 3 518 106 0.17 3.75 32.29* 4 312 52 0.26 3.54 22.78 5 798 81 0.15 4.24 16.40 6 816 118 0.15 4.21 27.57* 7 609 74 0.21 3.96 19.05* 8 576 72 0.20 4.04 20.15* 9 561 79 0.27 3.97 26.52* 10 525 94 0.21 3.78 29.60* Mean ± SD 594.2 ± 136.28 82.90 ± 33.08 0.20 ± 0.04 3.90 ± 0.23 26.84 ± 7.27 *Significant descriptor value (95% CI). Patricia N. Miranda et al. – Ant-plant networks in fragmented landscapes20 Appendix S5. Model selection analysis to explain the five ant-network descriptors in response to landscape structure metrics and forest structure at 10 rainforest fragments located in the state of Acre, Brazilian Amazon. The ΔAICc, df, and wAICc indicate the difference in corrected Akaike values, degrees of freedom of the model, and Akaike weights, respectively. Ed = Edge irregularity index; Co = Connectivity index; Ar = Fragment area; Cc = Canopy cover. The values in bold represent the models selected for each of the network descriptors evaluated. Response variable Model ΔAICc d.f. wAIC Slope symbol Number of interaction null model 0 2 0.396 Ed 0.75 3 0.273 - Co 2.91 3 0.093 - Cc 3.02 3 0.087 - Ar 3.04 3 0.087 - Ed+Co 6.45 4 0.016 -Ed+Co Ed+Cc 6.55 4 0.015 -Ed-Cc Ar+Ed 6.72 4 0.014 -Ar-Ed Cc+Co 8.12 4 0.007 -Cc-Co Ar+Cc 8.33 4 0.006 -Ar-Cc Ar+Co 8.51 4 0.006 -Ar-Co Ed+Cc+Co 15.31 5 0.000 -Ed-Cc+Co Ar+Ed+Co 15.4 5 0.000 -Ar-Ed+Co Ar+Ed+Cc 15.54 5 0.000 -Ar-Ed-Cc Ar+Cc+Co 16.91 5 0.000 -Ar-Cc-Co Ar+Ed+Cc+Co 30.28 6 0.000 -Ar-Ed-Cc+Co Network size null model 0 2 0.597 Cc 2.75 3 0.151 - Co 4.2 3 0.073 + Ar 4.27 3 0.070 - Ed 4.28 3 0.070 + Cc+Co 8.33 4 0.009 -Cc+Co Ed+Cc 8.42 4 0.009 Ed-Cc Ar+Cc 8.66 4 0.008 Ar-Cc Ed+Co 10.07 4 0.004 -Ed+Co Ar+Co 10.12 4 0.004 -Ar+Co Ar+Ed 10.26 4 0.004 -Ar+Ed Ed+Cc+Co 17.32 5 0.000 Ed-Cc+Co Ar+Cc+Co 17.33 5 0.000 Ar-Cc+Co Ar+Ed+Cc 17.42 5 0.000 Ar+Ed-Cc Ar+Ed+Co 19.03 5 0.000 -Ar-Ed+Co Ar+Ed+Cc+Co 32.32 6 0.000 -Ar+Ed-Cc+Co null model 0 2 0.514 Ar 1.63 3 0.228 + Ed 3.88 3 0.074 + Co 3.91 3 0.073 + Network specialization Cc 4.16 3 0.064 + Ar+Ed 7.59 4 0.012 Ar-Ed Ar+Co 7.59 4 0.012 Ar-Co Ar+Cc 7.63 4 0.011 Ar-Cc Sociobiology 69(3): e8261 (September, 2022) 21 Appendix S5. Model selection analysis to explain the five ant-network descriptors in response to landscape structure metrics and forest structure at 10 rainforest fragments located in the state of Acre, Brazilian Amazon. (Continuation) Ed+Co 9.85 4 0.004 Ed+Co Cc+Co 9.87 4 0.004 Cc+Co Ed+Cc 9.87 4 0.004 Ed+Cc Network specialization Ar+Ed+Co 16.59 5 0.000 Ar-Ed-Co Ar+Ed+Cc 16.59 5 0.000 Ar-Ed+Cc Ar+Cc+Co 16.59 5 0.000 Ar+Cc-Co Ed+Cc+Co 18.83 5 0.000 Ed+Cc+Co Ar+Ed+Cc+Co 31.58 6 0.000 Ar-Ed+Cc-Co Diversity of interaction null model 0.00 2 0.521 Ed 2.81 3 0.127 + Co 2.9 3 0.122 + Ar 3.59 3 0.086 + Cc 3.73 3 0.081 - Ed+Cc 6.32 4 0.022 Ed-Cc Cc+Co 7.42 4 0.013 -Cc+Co Ar+Cc 8.54 4 0.007 Ar-Cc Ed+Co 8.69 4 0.007 Ed+Co Ar+Ed 8.74 4 0.007 Ar+Ed Ar+Co 8.79 4 0.006 Ar+Co Ar+Ed+Cc 15.19 5 0.000 Ar+Ed-Cc Ed+Cc+Co 15.29 5 0.000 Ed-Cc+Co Ar+Cc+Co 16.15 5 0.000 Ar-Cc+Co Ar+Ed+Co 17.63 5 0.000 Ar+Ed+Co Ar+Ed+Cc+Co 30.17 6 0.000 Ar+Ed-Cc+Co NODF Ed 0.00 3 0.449 - null model 1.20 2 0.247 Co 2.38 3 0.137 - Ar 4.60 3 0.045 - Cc 5.09 3 0.035 - Ed+Cc 5.84 4 0.024 -Ed+Cc Ar+Ed 5.96 4 0.023 Ar-Ed Ed+Co 6.00 4 0.022 -Ed+Co Cc+Co 8.34 4 0.007 -Cc-Co Ar+Co 8.35 4 0.007 -Ar-Co Ar+Cc 10.42 4 0.002 -Ar-Cc Ar+Ed+Co 14.80 5 0.000 Ar-Ed+Co Ed+Cc+Co 14.83 5 0.000 -Ed+Cc+Co Ar+Ed+Co 14.96 5 0.000 Ar-Ed+Co Ar+Cc+Co 17.32 5 0.000 -Ar-Cc-Co Ar+Ed+Cc+Co 29.80 6 0.000 Ar-Ed+Cc+Co Response variable Model ΔAICc d.f. wAIC Slope symbol _GoBack