Interaction Intimacy Affects Structure and Coevolutionary Dynamics in Mutualistic Networks

Abstract

The structure of mutualistic networks provides clues to processes shaping biodiversity [1-10]. Among them, interaction intimacy, the degree of biological association between partners, leads to differences in specialization patterns [4, 11] and might affect network organization [12]. Here, we investigated potential consequences of interaction intimacy for the structure and coevolution of mutualistic networks. From observed processes of selection on mutualistic interactions, it is expected that symbiotic interactions (high-interaction intimacy) will form species-poor networks characterized by compartmentalization [12, 13], whereas nonsymbiotic interactions (low intimacy) will lead to species-rich, nested networks in which there is a core of generalists and specialists often interact with generalists [3, 5, 7, 12, 14]. We demonstrated an association between interaction intimacy and structure in 19 ant-plant mutualistic networks. Through numerical simulations, we found that network structure of different forms of mutualism affects evolutionary change in distinct ways. Change in one species affects primarily one mutualistic partner in symbiotic interactions but might affect multiple partners in nonsymbiotic interactions. We hypothesize that coevolution in symbiotic interactions is characterized by frequent reciprocal changes between few partners, but coevolution in nonsymbiotic networks might show rare bursts of changes in which many species respond to evolutionary changes in a single species.

Emergence of structural and dynamical properties of ecological mutualistic networks

“…In addition to simulations of mutualistic communities starting with random interactions networks and then "reorganizing" toward a more optimal state, we also implement a more realistic scenario in which mutualistic communities progressively assemble and are optimized over the course of evolutionary timescales 23,24 . Indeed, we find that the final result is the same, i.e., the final optimized networks display a nested architecture (Supplementary Information).…”

Dedication. Acknowledgments

“…Modular food-webs are defined by dietary resource segregation among consumers [67]. Resource segregation can occur over ecological, but also evolutionary timescales, where coevolutionary relationships may begin to constrain the plasticity of trophic interactions, promoting compartmentalization [65]. In isolated environments, where neighboring systems are similar and invasions are rare, di↵erentiation of resources and the subsequent development of modular interactions may be more likely to occur and be reinforced over time.…”

New record of a very specialized interaction: myrcidris epicharis Ward 1990 (Pseudomyrmecinae) and its myrmecophyte host Myrcia madida McVaugh (Myrtaceae) in Brazilian Meridional Amazon

“…(leguminosae -caesalpinioideae) were found in continuous forest, hosting different ant species (Petini-Benelli and izzo 2011) but never M. epicharis. indeed, Bruna et al (2005) reported that M. madida is extremely rare in areas of continuous forest in central Amazon and several authors have found M. madida in areas near to rivers (Mcvaugh 1956; The symbiotic relationship between ants and myrmecophytes is suggested be specialized and compartmentalized (Guimarães et al 2007). in this type of relationship both parts have different adaptations to select their association (mutualism filters) (Davidson et al 1989;Ferdele 1997;Dáttilo et al 2009b).…”

“…We did not observe ant alates and coccoid symbionts living in the domatias. The symbiotic relationship between ants and myrmecophytes is suggested be specialized and compartmentalized (Guimarães et al 2007). in this type of relationship both parts have different adaptations to select their association (mutualism filters) (Davidson et al 1989;Ferdele 1997;Dáttilo et al 2009b).…”

“…in exchange for shelter and food, in some cases (Beattie 1985;Vasconcelos 1991;leroy et al 2008), the ants defend their host plants against potential herbivores (Fonseca and Ganade 1996;Stapley 1998;Heil 2003). Myrmecophytes and ants have specializations that make the interaction essential to the life cycle of both (Bronstein 2006), so that they have only one or few possible associations (Fonseca and Ganade 1996;Guimarães et al 2007). to investigate whether the specificity of a interaction varies along the geographical range of both mutualistic species may provide insights on ecology and evolution of myrmecophytes (Murase et al 2003;Bronstein 2009;Moraes and Vasconcelos 2009).…”