Neuromuscular control and kinematics of intermittent flight in the European starling (Sturnus vulgaris)

Abstract

Electromyographic (EMG) and kinematic data were collected from European starlings (Sturnus vulgaris) flying at a range of speeds from 8 to 18 m s-1 in a variable-speed windtunnel. Their flight at all speeds consisted of alternating flapping and non-flapping phases. Wing postures during non-flapping phases included glides, partial-bounds and bounds. Glides were performed proportionally more often within each speed and were longer in duration than either of the other two non-flapping postures, but the percentage of bounds increased markedly with increasing flight speed. The shift from flap-gliding at slow speeds towards flap-bounding at fast speeds was consistent with reducing mean power output relative to continuous flapping. The starlings often combined more than one non-flapping posture within a single non-flapping period. Transitions between non-flapping postures, as well as transitions between bounds and subsequent flapping, were classified as 'pull-outs'. Pull-outs consisted of an increase in wingspan but no change in wingtip elevation. The pectoralis and supracoracoideus exhibited electrical activity during glides but not during bounds. The scapulohumeralis caudalis was not active during glides, but this muscle and the supracoracoideus were typically active during partial-bounds and pull-out phases. The scapulohumeralis caudalis occasionally showed activity during bounds, which may reflect its role as a humeral retractor. The frequency and duration of non-flapping intervals in starlings were less during EMG experiments than during non-implanted flights. During flapping phases, relative intensity and duration of EMG signal and wingbeat frequency increased with flight speed, whereas flapping or non-flapping cycle duration, the percentage of a cycle spent flapping and the number of wingbeats in a cycle were all greatest at 8 m s-1. Wingbeat amplitude was smaller at intermediate speeds, but differences among speeds were not significant. These variables allowed indirect estimates of power output and suggested that minimum power speed for starlings was near 12 m s-1 and that power output increased at both slower and faster speeds. Within windtunnel speeds, muscle activity changed in relation to wingspan at mid-upstroke, wingtip excursion, wingbeat frequency, acceleration, velocity, altitude and horizontal position.