they contributed thereto, and vernacular literature being omitted as either derived from Latin works or unlearned and unscientific. [Sidenote: How to study the history of thought.] Very probably I have tried to cover too much ground and have made serious omissions. It is probably true that for the history of thought as for the history of art the evidence and source material is more abundant than for political or economic history. But fortunately it is more reliable, since the pursuit of truth or beauty does not encourage deception and prejudice as does the pursuit of wealth or power. Also the history of thought is more unified and consistent, steadier and more regular, than the fluctuations and diversities of political history; and for this reason its general outlines can be discerned with reasonable sureness by the examination of even a limited number of examples, provided they are properly selected from a period of sufficient duration. Moreover, it seems to me that in the present stage of research into and knowledge of our subject sounder conclusions and even more novel ones can be drawn by a wide comparative survey than by a minutely intensive and exhaustive study of one man or of a few years. The danger is of writing from too narrow a viewpoint, magnifying unduly the importance of some one man or theory, and failing to evaluate the facts in their full historical setting. No medieval writer whether on science or magic can be understood by himself, but must be measured in respect to his surroundings and antecedents. [Sidenote: Definition of magic.]