posted on 2017-03-02, 23:50authored byBruce, Andrew Iain
Ants build structures that facilitate the life of the colony. The structures are created in a highly variable environment by workers that lack centralised communication or a fixed plan. Instead, they use a system of self-organisation where each worker responds to local cues in such a way that a coherent structure emerges. How this process occurs can be both complex and specific to a species and therefore presents significant opportunity for study. One group of ants that creates particularly large, complicated, and sophisticated nests are the leaf-cutting ants. Leaf-cutting ants cut and carry leaf fragments, sometimes over great distances, and use them to grow a fungus garden that serves as a source of protein and nutrients for the colony's brood and adults. In this thesis I examine the regulation of nest construction in leaf-cutting ants: specifically construction of nest space and transport infrastructure, with a focus on nest tunnels and overhead clearance on foraging trails. A literature review places the nest construction behaviour of leaf-cutting ants into the context of the larger group to which they belong, fungus growing ants. It examines what is known and unknown about the construction mechanisms underlying the creation of tunnels, chambers and trails. Subsequently, I asked whether ants could regulate their digging rate in response to the tunnel space they initially encounter. Upon finding that they display higher excavation rates when presented with shorter tunnels than longer tunnels, I tested a prominent hypothesis that a 'digging pheromone' was a key component of the regulatory mechanism. The evidence suggested that it was not and that therefore a new understanding of digging regulation was required. In order to contribute to this new understanding I used the automated tracking system idTracker to follow the movement of ants as they extracted soil from a tunnel. This revealed the importance of interaction rate and arousal to the regulation of tunnel length and I integrated these factors into a conceptual model that could explain digging regulation. As a next step, I examined trail construction in leaf-cutting ants. The overhead method that leaf-cutting ants adopt in transporting leaves causes them to have an elevated clearance. If trails have overhanging obstacles, then the collision of the laden ants with these obstacles can slow their progress, requiring the removal of the obstructions for the sake of efficiency. I showed that the presence of laden ants on the trail is necessary for unladen ants to recognize the need to clear trails to a height suitable for laden ants. Furthermore, I showed that this mechanism is not triggered by the visual cue of an obstruction. These results show the importance of the dynamics of self-organised behaviour in creating large, sophisticated structures and point the way to understandings that could apply to industries as diverse as communication and micro-robotics.