The spatial spread within fly photoreceptors of 2 forms of desensitization by bright light have been investigated: the natural process of light adaptation in normal Musca photoreceptors and a receptor-potential inactivation in the no-steady-state (nss) mutant of the sheep blowfly Lucilia. The suction-electrode method used for recording from vertebrate rods was applied to fly ommatidia. A single ommatidium in vitro was partially sucked into a recording pipette. Illumination of the portion of the ommatidium within the pipette resulted in a flow of current having a wave form similar to that of the receptor potential and polarity consistent with current flow into the illuminated region of the photoreceptors. Two 5-microns slits of light, positioned at right angles to the ommatidial axis, were employed to determine the spread of light adaptation or inactivation along the ommatidium. The intensity of a flash of light delivered to one (adapting) slit was adjusted until it produced a criterion fractional reduction in the response to the other (test) slit. The reciprocal of this intensity of the adapting slit was taken as a measure of the effectiveness of the slit in causing light adaptation or inactivation. The effectiveness of the slit in causing light adaptation in normal Musca ommatidia fell as the adapting and test slits were moved farther apart along the ommatidial axis, declining to half its maximal value at a distance of 13 +/- 2 microns. Similar measurements of the effectiveness of a slit in causing light-induced inactivation in the nss mutant of Lucilia also demonstrated localization, declining to half its maximal value at a distance between the slits of 9 +/- 1 microns. Neither light adaptation nor inactivation by the nss mutation, therefore, appear to be mediated by voltage or by a highly diffusible agent. The results are consistent with the idea that inactivation by the nss mutation replaces adaptation in the mutant photoreceptors.