The broad objective of this investigation was to explore the potential and applicability of automatic methods for the indexing of drug-related information appearing in English natural language text and to find out what can be learned about automatic indexing in general from the experience. More specific objectives were the development, implementation, and evaluation of an indexing algorithm which will enable the computer to assign automatically index terms to documents. This final project report describes the automatic indexing method that was developed in which index tags for documents are generated by the computer. The computer scans the text of periodical articles and automatically assigns to them index terms with their respective weights on the basis of explicitly defined text characteristics. A machine file of document references with their associated index terms is automatically produced which can be searched on a coordinate basis for the retrieval of specified drug-related information. A statistical evaluation of the output of the indexing algorithm and information concerning the system's ability to respond to specific queries is given. (Author/JB)