My understanding of magnetism is that exists solely as a placeholder to reconcile the different ways charge movements look for observers in relativistic frames...<p>In looking at spin ice, I'm struck by the similarity between the "trail" or "string" between monopoles and the way long bar magnets behave. As a bar magnet gets longer it behaves more and more like 2 monopoles separated by the bar.<p>Here's a little gedanken. Imagine a south magnetic monopole did exist as a single particle at (0,0,0). I'm thinking of the force it would exert on a charged particle (a proton for example) at (1,0,0) moving in the positive Y direction. The magnetic field would point at the origin, and a force would tend to push the particle out of the page... in order for angular momentum to be conserved, the monopole must now spin. If we ignore conservation of angular momentum, and just stick to regular momentum it must still move in the opposite direction of the force on the proton so it would create a sort of weird unequal orbit around a common axis.<p>It's fun to think about these things.