Summary1. Bumble bees play a vital role in the pollination of many crops and wild¯owers, and plans for their conservation require a knowledge of the dynamics and spatial scale of their foraging¯ights, which are, at present, poorly understood. 2. We investigated the foraging range and constancy of two colonies of bumble bees Bombus terrestris L. on a mixed arable farm using harmonic radar, which has a unique capability to record the trajectories of insects¯ying at low altitude in the ®eld. 3. Foraging bees were ®tted with lightweight radar transponders and tracked as they¯ew to and from the nest to forage. The resulting tracks gave information on length, direction and straightness of foraging routes. Superimposition onto a map of the foraging landscape allowed interpretation of the bees' destinations in relation to the spatial distribution of forage. 4. Outward tracks had a mean length of 275´3 P 18´5 m (n = 65) and a range of 70±631 m, and were often to forage destinations beyond the nearest available forage. Most bees were constant to compass bearing and destination over successive trips, although one bee was tracked apparently switching between forage patches. Both outward and return tracks had a mean straightness ratio of 0´93 P 0´01 (n = 99). The bees' ground speeds ranged from 3´0 m s ±1 to 15´7 m s ±1 (n = 100) in a variety of wind conditions. 5. The results support the hypothesis that bumble bees do not necessarily forage close to their nest, and illustrate that studies on a landscape scale are required if we are to evaluate bee foraging ranges fully with respect to resource availability. Such evaluations are required to underpin assessments of gene¯ow in bee-pollinated crops and wild¯owers. They are also required when making decisions about the management of bees as pollinators and the conservation of bee and plant biodiversity.

Many insects undertake long-range seasonal migrations to exploit temporary breeding sites hundreds or thousands of kilometers apart, but the behavioral adaptations that facilitate these movements remain largely unknown. Using entomological radar, we showed that the ability to select seasonally favorable, high-altitude winds is widespread in large day- and night-flying migrants and that insects adopt optimal flight headings that partially correct for crosswind drift, thus maximizing distances traveled. Trajectory analyses show that these behaviors increase migration distances by 40% and decrease the degree of drift from seasonally optimal directions. These flight behaviors match the sophistication of those seen in migrant birds and help explain how high-flying insects migrate successfully between seasonal habitats.

By using harmonic radar, we report the complete flight paths of displaced bees. Test bees forage at a feeder or are recruited by a waggle dance indicating the feeder. The flights are recorded after the bees are captured when leaving the hive or the feeder and are released at an unexpected release site. A sequence of behavioral routines become apparent: (i ) initial straight flights in which they fly the course that they were on when captured (foraging bees) or that they learned during dance communication (recruited bees); (ii ) slow search flights with frequent changes of direction in which they attempt to ''get their bearings''; and (iii ) straight and rapid flights directed either to the hive or first to the feeding station and then to the hive. These straight homing flights start at locations all around the hive and at distances far out of the visual catchment area around the hive or the feeding station. Two essential criteria of a map-like spatial memory are met by these results: bees can set course at any arbitrary location in their familiar area, and they can choose between at least two goals. This finding suggests a rich, map-like organization of spatial memory in navigating honey bees.dance ͉ communication ͉ localization in navigation ͉ vector orientation ͉ vector map

Dispersal is a key life-history trait, especially in species inhabiting fragmented landscapes. The process of dispersal is affected by a suite of morphological, physiological, and behavioral traits, all of which have a more or less complex genetic basis and are affected by the prevailing environmental conditions. To be able to identify genetic and phenotypic effects on dispersal, movements have to be recorded over relevant spatial and temporal scales. We used harmonic radar to track free-flying Glanville fritillary butterflies (Melitaea cinxia) released in the field and reconstructed their flight tracks for several hours. Flight track lengths for individual butterflies ranged from tens of meters to several kilometers. Butterflies were most mobile at midday and in intermediate temperatures. Flight metabolic rate (MR), measured prior to the tracking, explained variation in mobility at all scales studied. One-third of the variation in the distance moved in one hour could be attributed to variation in flight MR. Heterozygous individuals at a single nucleotide polymorphism in the phosphoglucose isomerase (Pgi) gene moved longer distances in the morning and at lower ambient temperatures than homozygous individuals. A similar genotype x temperature interaction was found to affect the metabolic rate. Our results establish connections from molecular variation in a single gene to flight physiology and movement behavior at the landscape level. These results indicate a fitness advantage to the heterozygous genotype in low temperatures and suggest a mechanism by which varying environmental conditions maintain genetic polymorphism in populations.

Cognitive ethology focuses on the study of animals under natural conditions to reveal ecologically adapted modes of learning. But biologists can more easily study what an animal learns than how it learns. For example, honeybees take repeated 'orientation' flights before becoming foragers at about three weeks of age. These flights are a prerequisite for successful homing. Little is known about these flights because orienting bees rapidly fly out of the range of human observation. Using harmonic radar, we show for the first time a striking ontogeny to honeybee orientation flights. With increased experience, bees hold trip duration constant but fly faster, so later trips cover a larger area than earlier trips. In addition, each flight is typically restricted to a narrow sector around the hive. Orientation flights provide honeybees with repeated opportunities to view the hive and landscape features from different viewpoints, suggesting that bees learn the local landscape in a progressive fashion. We also show that these changes in orientation flight are related to the number of previous flights taken instead of chronological age, suggesting a learning process adapted to changes in weather conditions, flower availability and the needs of bee colonies.

In the 'dance language' of honeybees, the dancer generates a specific, coded message that describes the direction and distance from the hive of a new food source, and this message is displaced in both space and time from the dancer's discovery of that source. Karl von Frisch concluded that bees 'recruited' by this dance used the information encoded in it to guide them directly to the remote food source, and this Nobel Prize-winning discovery revealed the most sophisticated example of non-primate communication that we know of. In spite of some initial scepticism, almost all biologists are now convinced that von Frisch was correct, but what has hitherto been lacking is a quantitative description of how effectively recruits translate the code in the dance into flight to their destinations. Using harmonic radar to record the actual flight paths of recruited bees, we now provide that description.

Automated tracking of bumblebees and computer simulations reveal how bees locate a series of flowers and optimize their routes to visit them all.

Honey bees (Apis mellifera) are regularly faced with the task of navigating back to their hives from remote food sources. They have evolved several methods to do this, including compass-directed "vector" flights and the use of landmarks. If these hive-centered mechanisms are disrupted, bees revert to searching for the hive, but the nature and efficiency of their searching strategy have hitherto been unknown. We used harmonic radar to record the flight paths of honey bees that were searching for their hives. Our subsequent analysis of these paths revealed that they can be represented by a series of straight line segments that have a scale-free, Lévy distribution with an inverse-square-law tail. We show that these results, combined with the "no preferred direction" characteristic of the segments, demonstrate that the bees were flying an optimal search pattern. Lévy movements have already been identified in a number of other animals. Our results are the best reported example where the movements are mostly attributable to the adoption of an optimal, scale-free searching strategy.