Collective construction of topologically complex structures is one of the triumphs of social behavior. For example, many ant species construct underground nests composed of networks of tunnels and chambers. Excavation by these ‘superorganisms’ depends on the biomechanics of substrate manipulation, the interaction of individuals, and media stability and cohesiveness. To discover principles of robust social excavation, we used X-ray computed tomography to monitor the growth in three dimensions of nests built by groups of fire ants (Solenopsis invicta) in laboratory substrates composed of silica particles, manipulating two substrate properties: particle size and gravimetric moisture content. Ants were capable of nest construction in all substrates tested other than completely dry or fully saturated; for a given particle size, nest volume was relatively insensitive to moisture content. Tunnels were deepest at intermediate moisture content and the maximum tunnel depth correlated with measured yield force on small rod-shaped intruders (a proxy for cohesive strength). This implies that increased cohesive strength allowed creation of tunnels that were resistant to perturbation but did not decrease individual excavation ability. Ants used two distinct behaviors to create pellets composed of wetted particles, depending on substrate composition. However, despite the ability to create larger stable pellets in more cohesive substrates, pellet sizes were similar across all conditions. We posit that this pellet size balances the individual's load-carrying ability with the need to carry this pellet through confined crowded tunnels. We conclude that effective excavation of similarly shaped nests can occur in a diversity of substrates through sophisticated digging behaviors by individuals which accommodate both differing substrate properties and the need to work within the collective.
The authors declare no competing or financial interests.
D.M. participated in the development, design and execution of the experiments, in data collection and analysis, and in manuscript writing. N.G. participated in the development and design of the experiments and in manuscript writing. G.R. and R.K. participated in the execution of experiments, data collection and animal care. M.A.D.G. and D.I.G. participated in the development and design of the experiments, and supervised data analysis and manuscript writing.
This work was supported by NSF PoLS grants 0957659 and 1205878.
Supplementary material available online at http://jeb.biologists.org/lookup/suppl/doi:10.1242/jeb.113795/-/DC1
- © 2015. Published by The Company of Biologists Ltd