At the same time we find the explanation of how there can be groups of religious phenomena which do not belong to any special religion; it is because they have not been, or are no longer, a part of any religious system. If, for some special reason, one of the cults of which we just spoke happens to be maintained while the group of which it was a part disappears, it survives only in a disintegrated condition. That is what has happened to many agrarian cults which have survived themselves as folk-lore. In certain cases, it is not even a cult, but a simple ceremony or particular rite which persists in this way.[73] Although this definition is only preliminary, it permits us to see in what terms the problem which necessarily dominates the science of religions should be stated. When we believed that sacred beings could be distinguished from others merely by the greater intensity of the powers attributed to them, the question of how men came to imagine them was sufficiently simple: it was enough to demand which forces had, because of their exceptional energy, been able to strike the human imagination forcefully enough to inspire religious sentiments. But if, as we have sought to establish, sacred things differ in nature from profane things, if they have a wholly different essence, then the problem is more complex. For we must first of all ask what has been able to lead men to see in the world two heterogeneous and incompatible worlds, though nothing sensible experience seems able to suggest the idea of so radical a duality to them. IV However, this definition is not yet complete, for it is equally applicable to two sorts of facts which, while being related to each other, must be distinguished nevertheless: these are magic and religion.