Both wood and cataglyphid ants follow fixed routes fromtheir nest to a feeding site and back again guided by visuallandmarks (Santschi, 1913; Rosengren, 1971; Collett et al.,1992; Wehner et al., 1996). Doing so increases the precisionof their navigation and so reduces the time that they need tospend searching for a goal. Wood ants, like honeybees (vonFrisch, 1967; Collett and Baron, 1994; Chittka et al., 1995;Fry and Wehner, 2002), tend to treat discrete landmarksalong the route as beacons at which they aim (Nicholson etal., 1999), making each landmark an intermediate goal, sopartitioning the route into segments. Many insects approachconspicuous stationary objects under a variety ofcircumstances (fruit flies: Wehner, 1972; Götz, 1994; Straussand Pichler, 1998; locusts: Wallace, 1962; ladybirds: Collett,1988; mantids: Poteser and Kral, 1995) and it seems likelythat in ants and bees beacon aiming is an intrinsic visuo-motor response that has been co-opted as a part of routefollowing. The incorporation of beacon aiming into routesbrings several benefits. For ants learning a route, it meansthat the same path will be repeated from trial to trial, givingants consistent views, thereby speeding up the acquisition ofa route. For ants familiar with a route, it makes routefollowing more robust. For instance, if an ant loses its way,it will automatically be attracted to a prominent beacon andso back to the route, even if the beacon’s appearance isunfamiliar when viewed away from the ant’s usual route.Additionally, the use of discrete landmarks to segment theroute, placing segment boundaries at the landmarks, allowsdefined local vectors to be associated with each landmark(Collett et al., 1998), so that errors of path integration inspecifying the goal are restricted to those of the final localvector (Srinivasan et al., 1997). In the present study, we testexplicitly the hypothesis that an ant’s tendency to aim atbeacons is an important factor in determining the shape ofits route. We then go on to show that the shape of a route,as fixed by a landmark, is reinforced by the learning of othermore distant visual cues additional to those provided by thebeacon. These additional cues can then help channel the antalong the route. Materials and methodsThe antsExperiments were performed on foragers from queen-rightwood ant (Formica rufa L.) colonies, housed in large plastictanks within the laboratory. The temperature was a constant21°C and the laboratory was on a 12h:12h light:dark cycle.When the colony was not being used for experiments, it wasallowed constant access to sucrose solution and water, withprotein (frozen crickets) provided every 2–3days. At the startof training, the colony’s only access to sucrose was throughthe individuals performing the experiment. After a cohort of535The Journal of Experimental Biology 206, 535-541© 2003 The Company of Biologists Ltddoi:10.1242/jeb.00115Many insects have an innate propensity to approachconspicuous objects. We explore how such beacon aimingdetermines the shape of a wood ant’s habitual route. Wefind that a single large black cylinder within an arenabiases the route taken by ants as they run from a startposition at one end of the arena to reach a feeder at theother. Ants learn a stable route with the first segment oftheir trajectory aimed at the cylinder, which becomes anintermediate goal on the way to the feeder. When inoccasional tests the cylinder is removed or displaced, antshead for the usual site of the cylinder. They also aim forthe same site when the cylinder is removed and the ant’snormal start position is changed. This behaviour suggeststhat visual features of the arena are learnt from thevantage point of the cylinder and that this stored snapshotguides the ant to that site. Ants thus reinforce their abilityto reach the cylinder by learning other visual features intheir surroundings that can also steer them to its location.The use of beacon aiming in fixing routes has severalbenefits. Because the same path will be traversed on everytrial, beacon aiming facilitates the acquisition of routes.Beacon aiming also increases the robustness of learntroutes: ants straying from the route will be attracted tothe closest beacon and so regain their habitual paths.Key words: wood ant, Formica rufa, beacon aiming, landmark,navigation, route learning, visual cue.SummaryIntroductionThe influence of beacon-aiming on the routes of wood antsPaul Graham*, Karine Fauria and Thomas S. CollettSchool of Biological Sciences, University of Sussex, Brighton, UK, BN1 9QG*Author for correspondence (e-mail: [email protected])Accepted 29 October 2002536foragers had been selected for an experiment, the colony wasgiven a reduced ration of sucrose during the 10 days(approximately) that the experiment lasted.The arena Experiments were performed in a 220cm×300cm arena thatwas lit by four fluorescent strip lights. The arena was surfacedwith roughened white Perspex. Visual cues external to thearena were reduced by a 280cm-high curtain that surroundedthe arena. One landmark, a black cylinder, 25cm×50cm(diameter × height), was placed in the arena. Ants were carriedfrom the nest on a drinking straw to a small raised startingplatform in the arena, from which they descended by a smallramp to reach the floor. A feeder (a drop of sucrose solutionon a microscope slide) was placed approximately 300cm fromthe starting point. Ants were trained with the cylinder either tothe left or to the right of the direct path from start to feeder(Fig. 1). Training At the beginning of training, 20–30 active ants from thecolony were placed on top of the start rampand the feeder placed at the finish. The first15 ants to reach the feeder were caught andindividually marked with enamel paint.These ants were then allowed to feed and toreturn to the start, where they were caughtagain and returned to the nest. Ants typicallyemerged from the nest mound after a fewminutes and were given another training trial.Tests were introduced after 15 training trials.During the testing phase, ants were giventhree rewarded training runs between non-rewarded tests. Between runs, the arenasurface was wiped down with ethanol to remove possiblepheromone cues.TrackingThe ants’ trajectories were tracked with a camera placed3m above the centre of the arena. The camera (Sony EVI-D30) has movable optics, allowing a high-resolution imageof any part of the arena to be captured. The camera iscontrolled by a PC