After an introductory review of the general nature of the theory of magnetic phenomena and the magnitudes of the atomic forces involved, there is a discussion of Ewing's theory, its results and limitations. The later theory of Weiss is then given briefly in order to fix the concept of the molecular field. In order to elucidate the nature of this field a digression is made to discuss the atomic structure of the ferromagnetic elements and elements having similar structures. With this as a basis the physical nature of the molecular field is discussed at some length. Its relation to the structure of domains, particularly the nature of the boundaries between domains, is brought out.