A promising strategy for providing a hospital information system is the gradual development and subsequent integration of well-defined information processing modules. A computer system which is suitable and economical to serve one or more of these modules has been developed. Its unique feature is that it uses an interpretive language for all the application programs both in the developmental and service modes of operation. The interpretive language, which has been designed specifically for data management applications, facilitates the creation and debugging of programs; in addition, it allows the time-sharing of a relatively small computer with a negligible overhead for switching control from one program to the next. The multiterminal, real-time remote-access computer system has been in service operation at the Clinical Chemistry Laboratory of the Massachusetts General Hospital since the early Fall of 1968. Presently, the system accepts typed-in requisition information, creates patient files, and generates schedules for centrifuging blood samples and for the performing of all tests; the system also accepts test results and provides reports on individual patients. In the near future all test results will be manually entered into the machine, which will then generate cumulative test reports each day. The on-line connection of automated laboratory equipment (AutoAnalyzers) will soon follow. Analysis of the system performance shows that the use of an interpretive language is suitable for this application, and it also indicates that the system has sufficient capacity for serving additional information processing needs.