Recently Mateeff and Hohnsbein (1996 Vision Research36 2873 – 2882) showed that the critical frequency for detection of modulated velocity of motion was invariant with respect to the viewing distance. The critical frequency was a function of the ratio between the mean velocity and the size of the aperture through which the motion was observed. In the present study we examined whether the velocity/aperture ratio affects velocity discrimination and detection of single velocity changes. Six subjects observed a random-dot pattern that could move within an invisible square aperture. In the discrimination task, two motions of 250 ms duration with slightly different velocities were presented with a 1 s interval between them. The subject had to report which of the motions was faster. In the change-detection task the same two motions were presented without an interval between them and the subject had to report whether the change was from a low to a high velocity or vice versa. Mean velocities of 8 and 64 deg s−1 and aperture sizes of 10 and 40 deg were employed in both tasks. Weber fractions were determined by the method of constant stimuli. The discrimination accuracy was not affected by aperture size at either mean velocity. The detection task was also unaffected by aperture size at 8 deg s−1. However, at 64 deg s−1 decreasing aperture size impaired the Weber fractions by a factor of about three. We suggest that the decrease of the lifetime of the dots of the pattern at high velocities and small apertures may be the critical factor for the impairment of the change detection. This factor is of less importance for the velocity discrimination task.