The army ant Eciton burchellii boasts more animal associates than any other animal species yet described, but the relationship between army ants and other ant species has only been studied in the context of predation. The waste deposits (middens) of army ant colonies are nitrogen-rich, a potentially high-value nutrient source for leaflitter arthropods. We explored this bottom-up role of army ant middens in the context of tropical ant communities. Our three main questions were 1) Which ant species forage on army-ants middens? 2) How does the bi-phasic life cycle of army ant colonies (affecting midden size, persistence, and abundance) affect which and how many ant species a midden boasts? 3) How do the ants that forage on army ant middens differ across elevations? Across 39 bivouacs we found 36 species of ants foraging on army ant middens. These included highly predatory ants, nitrogen-limited arboreal ants, and fungus-farming ants. Per-midden richness was significantly lower for the usually smaller middens deposited during the nomadic phase and was higher for the typically larger middens deposited during the statary phase. Per-midden richness was not significantly different across elevations, but there was far greater species turnover across elevations than across phases within the same elevational site. Our results suggest that army ant middens are an important resource for a wide variety of tropical ants, informing a better understanding of the complex network of associations revolving around this keystone species.

Three sites on the Pacific slope of the continental divide in Northern Costa Rica were surveyed for army ant midden attendance (Figure 2A): lowland tropical dry forest and tropical moist forest in Santa Rosa National Park (10º53’N, 85º46’W, 100-300 meters above sea level, henceforth “masl”), transitional premontane moist forest near Maritza Biological Station (10º58’N, 85º30’W, 400-650 masl), and premontane to lower montane wet forests in Monteverde (10º18’N, 84º49’W, 1100-1700 masl). At each site we used standard walking-encounter methods on small forest trails to locate army ant raids (Kumar & O'Donnell, 2009), following the direction of army-ant-carried food items back to the bivouac. We determined the phase of each bivouac based on the developmental stage of the developing synchronous brood within, with bivouacs housing larvae deemed nomadic and bivouacs housing pupae deemed statary. Once at the bivouac, middens were located by searching for a small column of worker ants (refuse workers) extending from the bivouac to a pile of debris being actively deposited. Using this method, we surveyed a total of 39 bivouacs in the rainy seasons of 2015 and 2016, finding 30 unique middens among them. Subject colonies in this study represent a subset of those used in a previous collective thermoregulation study (Baudier et al., 2019). For further details on field methods, including how colonies were located, tracked, and delineated see Baudier et al. (2019) and Baudier and Pavlic (2022).

We observed all middens twice a day for as long as the bivouacs were present. Middens were checked for ant foragers in the morning between 07:00 and 10:00 and in the afternoon to evening between 14:00 and 20:00. When the bivouac was in one location for multiple days (indicative of statary bivouacs which left substantial midden piles), surveys continued until one observation (under 24 hours) after the colony had left the bivouac site due to colony emigration. An ant on the midden was considered a midden forager if it was observed mandibulating or carrying material of any kind from the midden. Notes were taken on interspecies interactions and on foraged materials when carried items were visually identifiable (e.g., disarticulated insect tergites). Morphospecies at each midden were identified using a hand lens. Voucher specimens of each morphospecies were collected into 95% ethanol before being pointed and identified using morphological characters. Voucher specimens of all midden-foraging ant species identified in this study have been deposited in the Cornell University Insect Collection, Ithaca, NY, USA (CUIC#000061841-000061987).

References:

Baudier, K. M., D'Amelio, C., Sulger, E., O'Connor, M., & O'Donnel, S. (2019). Plastic collective endothermy in a complex animal society (army ant bivouacs: Eciton burchellii parvispinum). Ecography, 42(4), 730-739.

Baudier, K. M., & Pavlic, T. P. (2022). Multi-level instrumentation of bivouac thermoregulation: current methods and future directions. Artificial Life and Robotics, 27(2), 308-315.

Kumar, A., & O'Donnell, S. (2009). Elevation and forest clearing effects on foraging differ between surface – and subterranean – foraging army ants (Formicidae: Ecitoninae). Journal of Animal Ecology, 78, 91-97.

Attached are .csv files containing raw and subsetted data used to make tables within the manuscript, as well as a list of deposited voucher specimens. Excel or any text editor can be used to open these files.