Failure of stationary but not of flying honeybees (Apis mellifera) to respond to magnetic field stimuli.

Abstract

Trained individual honeybees (Apis melliferd) to feed to repletion from a well of sucrose solution so constructed that shock could be delivered when the proboscis was in contact with the solution. If shock was signaled by vibration of the substrate or by an airstream, the animals learned readily to avoid it by breaking contact briefly, but there was no response to change (either constant or time-varying) in the ambient magnetic field. When, however, a magnetic field anomaly in the region of the food well signaled to flying animals that contact would be punished with shock, hesitation to settle was greater in presence than absence of the anomaly. Parallel results were obtained with light, to which flying Ss clearly responded but to which stationary Ss hardly responded at all. We conclude that stationary bees detect magnetic field stimuli but do not process them successfully in the training situation. In previous experiments (Walker & Bitterman, 1985), it has been shown that flying honeybees {Apis melliferd) can be trained to discriminate earth-strength magnetic fields; this evidence has set the stage for a detailed psychophysical analysis of magnetoreception. Out of an interest in achieving better control of stimulation than is possible in work with flying subjects (which must be depended on to expose themselves to the magnetic fields), we turned in the present experiments to a method of conditioning stationary bees that was developed recently in this laboratory (Abramson, 1986). The work is done with individual foragers that have been pretrained to fly back and forth between the hive and the sill of an open laboratory window, where they feed to repletion on a target containing a well of sucrose solution. During the feeding a conditioned stimulus (CS) is turned on and followed 5 s later by a brief shock (the unconditioned stimulus or US) that the animal can learn to avoid by breaking contact with the sucrose solution. (Typically, it flies up from the target for a short time.) With vibration of the substrate or a stream of air directed at the animal as the CS, conditioning is dependably rapid (Abramson, 1986; Abramson & Bitterman, 1986a, 1986b), and we hoped at the outset of this work that it would be rapid also with magnetic stimuli. In fact, however, the animals hardly responded at all, despite the fact that the magnetic stimuli (produced by passing direct currents through coils below the target) were exactly the same as those discrim