Individual foraging careers of the Jack Jumper ant, Myrmecia croslandi

Abstract

A solitary foraging ant needs to rely exclusively on her navigational skill set to successfully navigate to and from goals such as the nest and food sites. Of interest is how ants are able to acquire this navigational information at a young age, before they become experienced workers and how this eventually shapes them into becoming efficient foragers. Ants of Myrmecia croslandi are highly visual, solitary foragers that exhibit no evidence of chemical trail following or recruitment. Therefore, a forager leaving the nest for the first time, must do so completely on her own, first by deciding where to go and second by utilising the information she has acquired from the environment to journey between sites. By identifying and individually following ants, I demonstrate that ants exhibit highly individual behaviour in most tasks, from early learning, and daily foraging to navigating from unfamiliar locations. First, I document the spatial and temporal variation in individual foraging behaviour at two nests of M. croslandi over a two-year period. Ants can take variable routes to the same food site and travel the longest distance when they forage on trees. Individual ants depart the nest at different times and a few ants perform multiple trips per day. Surprisingly, not a single ant foraged on consecutive days. By examining the behaviour of inexperienced ants at the nest, I provide a detailed analysis of the learning walks of M. croslandi. Most learning walks take place in the morning with a narrow time window separating the first two learning walks. There are no common bearing or gaze directions between ants, however, (a) in subsequent walks ants always explore directions that they have not previously visited and (b) ants engage in a systematic, saccadic scanning behaviour. I also discuss the significant differences between learning walks of M. croslandi and those previously studied in two other ant species, especially in relation the ‘turn back and look’ behaviour. In displacement experiments, I provide supporting evidence of a quick scanning behaviour that occurs as soon as ants are released. I examine the effect of a conflict in navigational information on successful homing by comparing full and zero vector ants. Zero vector ants are significantly better at navigating home, especially when released at unfamiliar sites. With the aid of the extensive individual foraging histories available to me, I show how in most cases, scene

familiarity plays a role in driving ants home from unfamiliar displacement locations and discuss in detail behaviours that are exceptions to this. Finally, I provide the first evidence of the use of artificial landmarks near the nest in this species, which increases the accuracy with which ants pinpoint the nest entrance, even though they do not appear to make use of such landmarks in the wild and discuss my findings in relation to other ants. I also document the occurrence of re-orientation walks in response to an altered visual environment which show that ants are more directed as a result of re-learning.