Galaxies are embedded in dominant dark-matter halos, but it has been difficult to find any visible stellar population that correlates in a simple way with DM halo mass. Systems of globular clusters have now been measured in hundreds of galaxies of all types, ranging from dwarf spheroidals up to supergiant Brightest Cluster Galaxies. This empirical evidence reveals an extremely strong and (surprisingly) near-linear relation between the total halo mass of a galaxy and the total mass in all its globular clusters -- a relation that is unlike any other known stellar population. Why? I will characterize what we know observationally about this correlation and describe early stages of model interpretation.