The pupillary response to the onset of bright light (10 000 lx) and darkness (4 lx) is compared in four species of Australian marsupials with different activity patterns. Maximal and minimal Pupil width in relation to eye width, as measured and the rate of change assessed. In the arrhythmic fat-tailed dunnart Sminthopsis crassicaudata, there were marked changes in pupil size with the range of pupil width to eye width of 0.73-0.10 constriction was rapid (54% s(-1)) and dilatation slower at 10% s(-1). The crepuscular honey possum Tarsipes rostratus had a small range of pupil size at 0.56-0.46, with constriction and dilatation being slow at 4% s(-1) and 3% s(-1). respectively. The strictly nocturnal burrowing bettong Bettongia lesueur had a small pupil with a limited size range of 0.21-0.12: constriction at 25% s(-1) outstripped dilatation at 7% s (1). The greater bilby Macrotis lagotis, also strictly nocturnal, had a small and immobile pupil with a fixed pupil width to eye width ratio of 0.08. In conclusion, contrary to most previous assumptions. pupil size is not fixed in marsupials, and not all nocturnal animals possess a wide pupil. Moreover, as a regulatory process. pupillary mobility reflects more than a differentiation between diurnal or nocturnal condition, and seems to correlate with specific visual requirements. The results of this study are discussed in relation to the lifestyle of the species and the associated illumination levels, and as a method of predicting the marsupials' periods of activity in the field.