Simultaneous and successive colour contrast expressed in “slow” phototactic behaviour of walkingDrosophila melanogaster

Summary

A behavioural paradigm, the “fly balance” (Fig. 1), allows continuous recording of the position of a walking fly in a Plexiglas tube (Fig. 2). This apparatus was used to study elements of “slow” phototactic behaviour ofDrosophila melanogaster wildtype and some of its mutants. Lights of varying intensity and spectral composition were presented to the two ends of a Plexiglas tube (Fig. 3). The influence of the light stimuli on the walking behaviour and on the mean position of the flies relative to the light sources was evaluated.

If at both ends of the test tube strong UV-lights (375 nm) of equal intensity were presented simultaneously, the addition of visible light (λ>415 nm) at one end of the tube led to a preference of wildtype flies for the pure UV-light (Fig. 10). The “attractiveness” of a UV-light at one end of the tube was enhanced by a visible light present at the opposite end (Figs. 8, 9). These experiments are interpreted as demonstrating the perception of simultaneous colour contrast by flies.

In experiments where lights of different wavelengths were presented successively it became evident that the attraction towards UV-light was strongest shortly after switching from visible light (λ≧425 nm) to UV-light (Fig. 11 a, b). A 1 s exposure to visible light was sufficient to enhance subsequent attraction towards UV-light (Fig. 13). This adaptational effect of visible light was only observed at high UV-intensities (Fig. 14). In UV-light it decreased at a rate which was not positively correlated to the UV-intensity (Fig. 15). This phenomenon of successive contrast confirms the conclusion that UV-light and visible light are perceived as different colours byDrosophila melanogaster wildtype. The colour adaptation effects may enable the visual system of the fly to achieve some colour constancy. Selective adaptation at the photopigment level can be excluded as the sole origin of the described phenomena.

The data support the notion of photopic and scotopic vision being present in flies (Fig. 14), however, this distinctioncannot be accounted for by the photoreceptors alone. Mutants deficient in certain receptor types corroborate the idea that centraland peripheral retinula cells are functional under photopic light conditions (Figs. 6, 10, 14).