Introduction Optical blur is potentially important for visual perception, but how the retina processes blur remains unclear. We know from studies of emmetropization (where a postnatal eye normally adjusts its growth to achieve good focus) that the retina detects both the presence and sign of blur (hyperopic defocus: eye too short, myopic defocus: eye too long), and that it can do this independently of the rest of the central nervous system. What optical cues does it use? The retina could use longitudinal chromatic aberration (LCA), but the sparsity of short wavelength-sensitive ("blue") cones should make it impossible for the retina to use spatial resolution to determine the degree to which blue images are in focus. We propose that the retina uses the temporal flicker in blue as an animal moves itself and its eyes as a proxy for image sharpness. Methods Four groups (n=3 per group) of tree shrews (a diurnal cone dominated mammal closely related to primates) were exposed binocularly to either steady or flickering red (628nm) or blue (464nm) light for 14 hours a day from 11 to 25 days after eye opening. The pattern of flicker approximated the natural temporal pattern on the back of the retina as an animal moves both its eyes and itself. Results Compared with normal eyes, both steady (+3.8D) and flickering red light (+6.4D) powerfully arrested eye growth. Flickering blue light produced significant myopia (-4.5D), but steady blue light was much less myopiagenic (-0.9D). Conclusion These results are consistent with the neural retina using LCA to regulate eye growth during development, and with temporal flicker being used as a proxy for spatial image focus for blue light. Does the retina shares this information with the brain via retinal ganglion cells? Meeting abstract presented at VSS 2015.