Specialization is, without doubt, one of the most important factors in the development of modern science, since beyond a certain level of technicality it is possible, even with intensive application, only to master a limited sector of the total of human knowledge. But some modes of specialization are, at the same time, under certain circumstances, an impediment to the adequate treatment of some ranges of problems.The principal reason for this limitation of the fruitfulness of at least some kinds of specialization lies in the fact that the specialized sciences involve a kind of abstraction. They constitute systematically organized bodies of knowledge, and their organization revolves about relatively definite and therefore limited conceptual schemes. They do not treat the concrete phenomena they study “in general” but only so far as they are directly relevant to the conceptual scheme which has become established in the science. In relation to certain limited ranges of problems and phenomena this is often adequate. But it is seldom, after such a conceptual scheme has become well worked out, that its abstractness does not sooner or later become a crucial source of difficulty in relation to some empirical problems. This is apt to be especially true on the peripheries of what has been the central field of interest of the science, in fields to which some of the broader implications of its conceptual scheme and its broader generalizations are applied, or in which the logically necessary premises of certain of these generalizations must be sought.