The primary visual cortex (V1) of higher mammals such as cats and monkeys shows columns for stimulus features like orientation preference and ocular dominance, but such columns are not found in all groups of mammals. Models suggest that the afferent sampling density in V1 could be a key factor determining whether a columnar organization is established or not. Accordingly, in rodent V1, which – in comparison to other animals – has a low sampling density, there appear to be no orientation columns. However, the situation is less clear for ocular dominance. So far, in vivo imaging studies in mice have not reported an obvious functional clustering for ocular dominance, while experiments in rat V1 revealed eye-specific patches by activity mapping with immediate early genes.Here we use a mouse line with widespread expression of a genetically encoded calcium indicator (GCaMP6s.Niell) to test whether mice show a functional organization for ocular dominance. We performed wide field-of-view, cellular-resolution two-photon calcium imaging throughout cortical layers 2/3, 4 and 5 of binocular V1. In most animals, we observed a patchy organization of eye preference within layer 4, with clear clusters of neurons that preferentially responded to the ipsilateral eye, surrounded by a contiguous region of contra-eye dominance. Extending vertically into layer 2/3 and layer 5, these clusters resembled ocular dominance columns. The observation of ocular dominance columns in the minute binocular visual cortex of mice sets a new boundary condition for computational models explaining the emergence of a columnar organization in the brain.