Item

Abstract

In the blowflyCalliphora flying stationarily in a wind tunnel, compensatory head movements were elicited by rolling the fly about its longitudinal axis (Fig. 1). Responses were recorded on video tape, and evaluated by single frame analysis.
Active head movements were observed in response to visual and mechanosensory stimuli (Fig. 2). They are not made or caused by the head's inertial momentum (Fig. 11).
Gravity, used by walking flies to align their head with the vertical, does not seem to be perceived during flight (Figs. 3–6) but has a passive stabilizing effect upon the flight attitude (Fig. 7).
A difference in aerodynamical load of the two wings elicits a transient head roll partly compensating a banked attitude (Figs. 4–6). The majority of campaniform sensilla at the wing base seems suitable to measure wing load.
Steady roll motion elicits a steady compensatory head roll which does not vanish even after 8 min of rotation at constant angular velocity (Fig. 8). Roll motion is most efficient at high roll speeds (100‡/s<w<2000‡/s). Mechanical motion perception fails if both halteres are disabled by arresting their oscillation or by amputation of the haltere knobs (Fig. 11). Flies with only one haltere intact cannot distinguish pitch from roll, but with respect to the sense of rotation they still respond bidirectionally (Fig. 12). Haltere dynamics and the response characteristics of haltere sensilla are discussed on the basis of recent results.
Head/body coordination is demonstrated in the absence of any roll stimulus (Fig. 3 a). The role of resilience of the neck skeleton, and that of different neck sense organs are discussed.
Mechanosensory roll control inCalliphora depends upon the locomotor state: When walking, the fly aligns its head vertically by gravity perception (Horn 1982). When flying, it controls only fast rotations. Passive attitude stabilization and visual means of control are required to maintain an upright flight attitude and head orientation.