Ecological Entomology (2013), 38, 463–469 DOI: 10.1111/een.12036 Inﬂuence of extraﬂoral nectary phenology on ant–plant mutualistic networks in a neotropical savanna DENISE LANGE, 1 WESLEY D ´ ATTILO 2 and K L E B E R DEL-CLARO 1 1 Laborat´ orio de Ecologia Comportamental e Interac ¸˜ oes, Instituto de Biologia, Universidade Federal de Uberlˆ andia, Uberlˆ andia, Brazil and 2 Instituto de Neuroetolog´ ıa, Universidad Veracruzana, Xalapa, Mexico Abstract. 1. Temporal variation has been one remarkable feature of ecological interactions. In ant–plant mutualism, this variation is widely known, although little is understood about the mechanisms that shape these variations. 2. This study tested whether or not the phenology of extraﬂoral nectaries (EFNs) inﬂuences the temporal variation of two properties of an ant – plant interaction network. 3. The network under investigation exhibited a nested pattern and low specialisation over most months. Monthly nestedness and specialisation in the network were negatively correlated, both being inﬂuenced by temporal variations in extraﬂoral nectar production of the plant community. The months of highest activity in the nectaries (August – November) were those when the level of generalisation in the network was at its highest. Although there were temporal variations in the properties of the network, the generalist core of the species remained the same over time. 4. The stable core enhances the coevolutionary importance of ant–plant interactions for the community. Thus, it can be concluded that the phenology of EFNs is one effective mechanism shaping the temporal variation in ant–plant interaction. Key words. Ant–plant interactions, coevolutionary process, ecological networks, generalist core, temporal variation. Introduction Spatiotemporal variation has been one of the remarkable features of ecological interactions (e.g. Del-Claro & Oliveira, 2000). However, the mechanisms that shape these variations are not fully understood (Thompson, 2005; Bronstein et al., 2006). Most studies have explored biotic factors to explain the pattern of ecological interaction networks, such as interaction type, relative abundances of species, and body size (see Bascompte, 2009, 2010; Chamberlain et al., 2010; Fontaine et al., 2011), and others have recently begun to evaluate the effects of abiotic factors (e.g. latitude, temperature, ﬁre and precipitation; see Olesen & Jordano, 2002; Rico-Gray et al., 2012; Alves-Silva & Del-Claro, 2013). Nevertheless, owing to the need to consider multiple determinants in the structure and dynamics of natural communities, the factors inﬂuencing the outcome of interactions is far from fully understood. In addition to the processes that inﬂuence interactions in communities, the understanding of who interacts with whom Correspondence: Kleber Del-Claro, Universidade Federal de Uberlˆ andia, Instituto de Biologia, CP 592, CEP38400-902, Uberlˆ andia, MG, Brasil. E-mail: delclaro@ufu.br in these systems is key to drawing valid conclusions about evolutionary processes that occur in natural environments (Thompson, 2005; Ings et al., 2009; Guimar˜ aes et al., 2011). Some species can undergo coevolutionary processes that result in the establishment of strong links with each other. These species have an increased chance of interacting with each other and may give rise to groups of specialists in this interaction (the principal or generalist core of the interaction network) (Thompson, 2005), which are responsible for maintaining the pattern of the interactive network (Bastolla et al., 2009; Mello et al., 2011). Due to the symmetric strength of the interaction, the generalist core can drive the evolution of the whole community, because the core species interacts with virtually all species of the matrix (Bascompte et al., 2006; Jordano, 2010; Guimar˜ aes et al., 2011). Extraﬂoral nectar-mediated ant–plant mutualisms are among the most common and widespread interactions in terrestrial ecosystems (Rico-Gray & Oliveira, 2007). In this type of interaction, plants produce extraﬂoral nectar resources to attract and reward ants that act as biotic defences against herbivores (Oliveira et al., 1987; Nascimento & Del-Claro, 2010; Nahas et al., 2012). Although there are many stud- ies exploring this mutualism, the majority only verify the  2013 The Royal Entomological Society 463