Spatial structure of ant‐gardens: vertical distribution on host trees and succession/segregation of associated vascular epiphytes

Abstract

Questions: Ant-gardens (AGs) involve a close association between epiphytes and ants with concurrent mutualistic interactions including protection, dispersal and nutrition; however, little is known about the spatial structure patterns that determine their establishment and formation. Our main questions were: (1) do AGs have a particular pattern of vertical distribution on their host trees; (2) does a process of succession of epiphytes occur during development of AGs; and (3) are epiphytes segregated in AGs?Location: Anthropic landscape in southeast Mexico. Methods:We studied AGs built by Azteca gnava ants. We examined their vertical distribution on host trees as well as the diversity, composition and reproductive status of vascular epiphytes associated with the lower, middle and upper zones of small, medium and large AGs.Results: A total of 859 AGs and 10 871 epiphytes, belonging to 26 different species, were recorded. We found that AGs are primarily (75%) located within tree crowns, and that the diversity and composition of epiphytes vary among AG sizes and among AG zones. We infer that the epiphytes that first become established in AGs are the bromeliad Aechmea tillandsioides and the gesneriad Codonanthe uleana, followed by the orchid Epidendrum flexuosum and the cactus Epiphyllum phyllanthus; these species were recorded, either individually or in cooccurrence, in 74% of the AGs examined. The species Ae. tillandsioides, Coryanthes picturata and Epid. flexuosum were most frequent in the upper AG zone, while Epid. pachyrachis was most frequent in the middle AG zone. Conclusions:Our results show that AGs have distinct vertical distribution patterns on their tree hosts and that the establishment of epiphytes in AGs is successive and segregated, suggesting spatial and temporal optimization in the establishment and development of these complex mutualistic systems.

Diversity and interactions of the epiphyte community associated with ant‐gardens are not influenced by elevational and environmental gradients

“…True AG ants determine the establishment of true AG epiphytes (Orivel & Leroy, 2011;Morales-Linares et al, 2018), which have probably been selected for their attractive functional traits (e.g., extrafloral nectar and roots as support; Davidson, 1988;Catling, 1995). As the ant-gardens develop and become larger, the community of associated epiphytes presents a pattern of temporal succession and spatial segregation (Catling, 1995;Morales-Linares et al, 2017). However, non-true AG epiphytes/ants (e.g., opportunistic or parasitic species) can occupy the ant-gardens in a given time period, either fortuitously or through abandonment and death of true AG epiphytes/ants respectively (Davidson, 1988;Kaufmann & Maschwitz, 2006;Morales-Linares et al, 2017).…”

“…Niche ant-gardens (Morales-Linares et al, 2021b). However, epiphyteepiphyte interactions (e.g., co-occurrence) have been little explored (Zotz, 2016), including those species associated with ants (e.g., antgardens; Morales-Linares et al, 2017).…”

“…As the ant-gardens develop and become larger, the community of associated epiphytes presents a pattern of temporal succession and spatial segregation (Catling, 1995;Morales-Linares et al, 2017). However, non-true AG epiphytes/ants (e.g., opportunistic or parasitic species) can occupy the ant-gardens in a given time period, either fortuitously or through abandonment and death of true AG epiphytes/ants respectively (Davidson, 1988;Kaufmann & Maschwitz, 2006;Morales-Linares et al, 2017). For this reason, regardless of the management of the ant-gardens by the AG ants, the epiphyte frequency, degree of dependence and structural role (e.g., long vs short roots) in the ant-gardens suggest the existence of key epiphyte species for the development and maintenance of this mutualistic system.…”

“…We considered these gradients separately because of their relevance in studies of the distribution patterns of vascular epiphytes and arboreal ants (Krömer et al, 2005;Cardelús et al, 2006;Plowman et al, 2020). We then hypothesized that: (a) the richness of associated species will be higher in vegetation types with larger trees, and it will also be positively related to environmental gradients, but not elevation, since the Neotropical ant-gardens usually occur in large trees (Marini-Filho, 1999;Dejean et al, 2000;Morales-Linares et al, 2016) (Catling, 1995;Morales-Linares et al, 2017).…”

Effect of forest disturbance on ant (Hymenoptera: Formicidae) diversity in a Mexican tropical dry forest canopy

“…Ants represent 20% to 60% of the total biomass of canopy arthropods (Davidson et al ., 2003). The canopy offers the ants a great variety of plant resources (e.g., seeds, extrafloral nectar, and nesting sites; Armbrecht et al ., 2004; Yanoviak et al ., 2011; Morales‐Linares et al ., 2017, 2018; Vergara‐Torres et al ., 2018) and the structure of the canopy ant community can be regulated by the availability of these resources, as well as by tree size and canopy connectivity (Blüthgen et al ., 2004; Philpott & Foster, 2005; Klimes et al ., 2012; Yusah & Foster, 2016; Vergara‐Torres et al ., 2018; Adams et al ., 2019a). However, anthropogenic disturbance could reduce the physical structure and diversity of the plant community, negatively affecting both habitat heterogeneity and resource availability for the canopy ants (Andersen, 2018) and for the ground ant species that forage in the canopy (Marques et al ., 2017; Da Silva et al ., 2019).…”