To assess the role of quantal limitations and target visibility on vernier acuity, we measured line detection thresholds and vernier thresholds for abutting dark line targets on a uniform background, for a range of retinal illuminance levels and target contrasts. Measurements were made for stimuli presented at the fovea, and at a retinal eccentricity of 2.5 deg. Although the relationship is truly curvilinear, line detection thresholds and vernier thresholds using targets of a fixed contrast, follow a square-root dependence on retinal illuminance over a significant portion of the illuminance range. Once the effect of retinal illuminance on line detection thresholds has been accounted for, there is little further effect of retinal illuminance on vernier thresholds, at least for visibility levels of up to five times the contrast detection threshold. This finding suggests that the spatial mechanisms which mediate changes in the detection threshold and vernier thresholds for abutting targets are similar, and are limited by the same sources of noise. Vernier thresholds for these thin line targets are approximately inversely proportional to target contrast for both retinal loci, and at all retinal illuminance levels tested. However, vernier thresholds for a constant visibility target are always lower for foveal than for eccentric stimuli, even after the effects of increased spatial pooling with increasing eccentricity are accounted for.