Van der Wildt and Vrolijk (1981) Vision Res. 21, 1765-1771, detected propagation of inhibition with small and short flashes, separated in time and space. In this paper, the results measured with flashes of longer duration and/or larger diameter are described and discussed. A model, describing inhibition arising on the flanks of the stimulus; luminance distribution and acting at the top luminance of the probe flash, gives a good description of the results. When non-identical flashes are used, the inhibition proves to be backwards. The propagating inhibition described also seems to give a possible explanation of metacontrast.