U tNDOUBTEDLY every educational and research institution is faced with the problem of locating reliable, adequate, and up-to-date sources of information. Standard textbooks and important periodicals are the basic sources. Since textbooks present in permanent form generally accepted theories or facts which probably first appeared in periodical publications at a much earlier date, the latter are the most important sources of current information. Every educational institution, particularly those offering graduate work and research, must have available the most important technical periodicals containing the results of research throughout the world. In addition, all institutions and practicing engineers should subscribe to the most important trade journals. Since the latter publications may not include articles of a highly technical nature, it is apparent that methods must be devised to evaluate these two kinds of publications. In every field of science the passage of a few years sees the birth of new periodicals and changes in the makeup of others which may or may not justify a continuation of their subscription. This paper discusses methods of evaluating periodicals and presents some of the results of an investigation made at the University of California. The objects of the investigation were: (i) to determine a list of the most important periodicals for electrical engineers; (2) to determine which periodicals should be discontinued; (3) to determine which important new or unknown periodicals might be purchased with the funds so released; and (4) to determine a practical solution of the general problem of evaluating scientific journals. The task of evaluating periodicals is a difficult one, and a brief description of several methods is given in the paragraphs immediately following.