Taylor dispersion is a microfluidic analytical technique with a high dynamic range and therefore is suited well to measuring the hydrodynamic radius of small molecules, proteins, supramolecular complexes, macromolecules, nanoparticles and their self-assembly. Here we calculate an unaddressed yet fundamental property: the limit of resolution, which is defined as the smallest change in the hydrodynamic radius that Taylor dispersion can resolve accurately and precisely. Using concepts of probability theory and inferential statistics, we present a comprehensive theoretical approach, addressing uniform and polydisperise particle systems, which involve either model-based or numerical analyses. We find a straightforward scaling relationship in which the resolution limit is linearly proportional to the optical-extinction-weighted average hydrodynamic radius of the particle systems.