In a population of hydrogen nuclei, the associated magnetic moments align with an external magnetic field. Each hydrogen nucleus can be in one of two states: a high energy state (antiparallel) and a low energy state (parallel). What bothers me is the orientation of the magnetic moment vectors. I would assume that the north poles (tips of magnetic moment vectors) tend to move towards the south pole of the external magnetic field. But the opposite, i.e., parallel orientation, is of lower energy (see the picture - taken from http://mriquestions.com/magnetic-dipole-moment.html).
Can anyone explain why this is the case?