ONE OF THE PROBLEMS IN INFORMATION STORAGE AND RETRIEVAL SYSTEMS OF TECHNICAL DOCUMENTS IS THE INTERPRETATION OF WORDS USED TO INDEX DOCUMENTS. SEMANTIC TOOLS, DEFINED AS CHANNELS FOR THE COMMUNICATION OF WORD MEANINGS BETWEEN TECHNICAL EXPERTS, DOCUMENT INDEXERS, AND SEARCHERS, PROVIDE ONE METHOD OF DEALING WITH THE PROBLEM OF MULTIPLE INTERPRETATIONS. THIS REPORT SHOWS HOW STATISTICAL DATA ON THE DISTRIBUTION OF OCCURRENCES OF SINGLE WORDS OR WORD PAIRS IN THE TEXT OF A SET OF DOCUMENTS CAN BE USED IN GENERATING SEMANTIC TOOLS, IN PARTICULAR, AN INDEXING VOCABULARY AND A DISPLAY OF RELATIONS AMONG THE TERMS IN THIS VOCABULARY. AN EXPERIMENT IN THIS AREA, WHICH INVOLVED THE TESTING OF SEVERAL STATISTICAL MEASURES AND TECHNIQUES, IS DESCRIBED. THE RESULTS OF THE EXPERIMENT GIVE SOME INSIGHT INTO THE PATTERNS OF LANGUAGE USAGE IN TECHNICAL LITERATURE. THIS DOCUMENT IS AVAILABLE AS AD-664-915 FROM THE CLEARINGHOUSE FOR FEDERAL SCIENTIFIC AND TECHNICAL INFORMATION, SPRINGFIELD, VIRGINIA 22151, $3.00 FOR HARD COPY, $0.65 FOR MICROFICHE, 87 PAGES. (AUTHOR/CM)