The gyroscopic mechanism of the halteres of Diptera

Abstract

The paper gives a detailed anatomical, dynamical and physiological analysis of the gyroscopic mechanism of the halteres of the higher Diptera. (1) A re-examination has been made of the structure of the articular region of the halteres of Lucilia sericata, Calliphora erythrocephala and Eristalis tenax. In these higher Diptera the organ as a whole is free to move only in one plane, by oscillation through an angle of approximately 150° about a horizontal hinge. A secondary articulation distal to the main hinge allows a slight, damped movement about an axis at right angles to the hinge in the plane of oscillation. (2) The frequency of oscillation is determined by the mechanical resonance of the system. The single muscle produces, by its contraction, an upwards movement of the haltere and the downstroke results from the elasticity of the hinge. (3) Stroboscopic observation of the haltere in the living fly shows that the cycle of oscillation consists of two phases of nearly constant angular velocity with rapid reversals at the ends of the stroke. (4) Dynamical analysis of such a mechanical system shows that: (a) when the fly as a whole is not rotating, the secondary articulation ensures that the only forces acting on the basal region of the haltere are the ‘primary’ torques about the main hinge, (b) When the fly is rotating in any plane not that of the oscillation, gyroscopic torques are set up at the base of the haltere about an axis at right-angles to the haltere in the plane of oscillation, (c) The magnitude and periodicity of these torques are different for yawing rotations and for pitch or roll. (5) Since pitching and rolling rotations can be distinguished only by a phase comparison between the gyroscopic torques in the halteres of opposite sides, and since it can be shown experimentally that flight is unimpaired under conditions when the two halteres are oscillating at different frequencies, it is evidently only in the yawing plane that unique indications are given to the fly by the haltere sense organs. (6) A detailed re-examination of the structure of the sensilla groups on the haltere base in the light of recent advances in knowledge of the function of campaniform and chordotonal sensilla suggests a functional classification into three classes: (A) those sensitive to the vertical ‘primary’ forces (scapal plates, Hicks papillae and small chordotonal organ), (B) those sensitive to the lateral gyroscopic forces (basal plate, large chordotonal organ) and (C) those without selective sensitivity (undifferentiated papilla). (7) A method is described of obtaining oscillograph records of impulses in the haltere nerve while the haltere is being oscillated by the pull of its own muscle. (8) When there is no rotation of the body of the fly, the volleys of impulses recorded in the haltere nerve are of the type to be expected if the sensilla groups are being excited by the ‘primary’ forces. (9) When the body of the fly is rotated in yaw or roll the pattern of impulses in the haltere nerve changes in the manner which is to be expected if certain sensilla groups are being excited by the gyroscopic forces. (10) Detailed analysis of the waveform of the oscillograph records at the beginning and end of yawing and rolling rotations reveals differences between the two types of impulse pattern which are consistent with the dynamical analysis. (11) Flash photographs of a haltere-less fly in free flight confirm that such an insect is in that state of spiral instability which is to be expected if there is inadequate stabilization in the yawing plane. (12) A brief comparison of the distribution of sensilla on the bases of the halteres and wings, and a review of what is known of the nature and periodicity of the forces acting on the base of the wings during flight, suggest that it may be possible to trace the stages through which the gyroscopic mechanism of the haltere has evolved from the flight mechanism of the wing.