IT IS HYPOTHESIZED THAT SUCH A STABLE IMPRISONMENT RATE SUGGESTS PRESSURE IS GENERATED TO MAINTAIN A CERTAIN RATE. IF PRISON POPULATIONS GET TOO LARGE, POLICE CAN CHOOSE NOT TO ARREST, PROSECUTORS NOT TO PRESS CHARGES, AND JUDGES NOT TO IMPRISON. SIMILARLY, IF THE POPULATION OF PRISONS DROPS BELOW THE STABLE RATE, THEN PRESSURE WOULD DEVELOP TO BEAR DOWN ON CERTAIN KINDS OF ACTIVITY THAT PREVIOUSLY HAD BEEN TOLERATED. DATA USED IN THIS ANALYSIS WERE THE AVERAGE DAILY PRISON POPULATION AND THE TOTAL POPULATION FOR EACH STATE FOR EACH YEAR FROM 1926 TO 1974. DATA ON STATE POPULATIONS WERE DEVELOPED BY LINEAR INTERPOLATION BETWEEN THE DECENNIAL CENSUS YEAR; DATA ON PRISON POPULATIONS WERE COLLECTED FROM THE DEPARTMENTS OF CORRECTIONS IN 47 STATES. TIME TRENDS WERE EXAMINED THROUGH USE OF REGRESSION ANALYSIS; DISCERNIBLE PATTERNS THROUGH TIME-SERIES ANALYSIS. IN EXAMINING THE TRENDS IN STATES PER CAPITA IMPRISONMENT RATES, RESEARCHERS NOTED THAT ALMOST HALF WERE STATIONARY, AND THAT THE TRENDS IN THE REMAINING STATES WERE SMALL. THESE FINDINGS ARE CONSISTENT WITH THE HYPOTHESIS OF THE GENERAL HOMEOSTATIC PROCESS PREVIOUSLY OBSERVED IN THE UNITED STATES AS WELL AS IN OTHER COUNTRIES. IT IS SUGGESTED THAT FUTURE ANALYSES FOCUS ON IDENTIFICATION OF THE ADAPTIVE MECHANISMS BY WHICH THE HOMEOSTATIC LEVEL IS MAINTAINED AND OF THE CAUSES OF THE OBSERVED VARIATION IN THE MEAN IMPRISONMENT RATE ACROSS THE STATES. FOOTNOTES, EXTENSIVE STATISTICAL DATA, GRAPHS, AND FORMULAS ARE INCLUDED IN THE ARTICLE. (LWM)