Showing papers by "Horst Mittelstaedt published in 1979"

Interaction of gravity and idiothetic course control in millipedes

Abstract: 1. If their walking plane is tilted, millipedes, in the dark, adopt a compromise course between the (“idiothetic”) tendency to continue going straight ahead and the (“allothetic”) tendency to walk uphill. This compromise direction is attained at about 30 cm of path length after the point of tilt and remains stable within the entire measured extent of about 130 cm, that is 20 to 30 times the animals' length. If the arena is set back into the horizontal after a 25 to 30 cm walk on the inclined plane the millipedes turn into a direction which is approximately equal to the initial one before the tilt. 2. If the animals had run through a bent passage before the tilt the result, at equilibrium, is as expected from a superposition of the two experiments. In the transient phase, the effect of gravity is delayed relative to that of the bent passage. 3. If the arena is tilted, when a millipede walks through a passage that prevents it from turning, and then set back horizontally, the animal turns toward the direction which had been uphill during the tilt. The amount of the turn depends on the direction and the time course of the preceding gravity influence, which acts as if processed by a “leaky” integrator. 4. In the inclined plane at equilibrium, the sine of the animal's deviation (α) from the projection of the vertical into the walking plane is, at first approximation, proportinal to the sine of the deviation (σ) from the animal's (self-adopted) idiothetic direction. More precisely, the pattern of equilibrium directions can be described by the additive superposition of two Fourier sine expansions with declining amplitudes $$\sum\limits_n {A_n \sin n\alpha + } \sum\limits_n {S_n \sin n\sigma = 0.}$$ 5. The courses observed at different angles of arena tilt (ϑ) can be best explained by assuming that the gravity influence is largely proportional to that component of the physical gravity vector which falls into the y-axis of the animal (g·sin α sinϑ, i.e. the “shear-component”), whereas the pressure-component (g·cosϑ) has no effect. 6. The results are discussed with respect to hypothetical underlying mechanisms, in comparison with the general properties of orientation systems, and in view of their methodological consequences.