A FORECAST IS A PROBABILISTIC STATEMENT ABOUT THE FUTURE STATE OF A NATURAL, TECHNICAL, OR SOCIAL SYSTEM. CRIMINOLOGICAL FORECASTING IS A BRANCH OF SOCIAL FORECASTING THAT DEPENDS HEAVILY ON POPULATION TRENDS. FIVE TYPES OF CRIMINOLOGICAL FORECASTING METHODS ARE COMMONLY USED: (1) 'GUIDELINES FOR THE IMAGINATION' BASED ON PREVIOUS GENERAL EXPERIENCE; (2) QUALITATIVE FORECASTING USING DEVICES SUCH AS THE DELPHI TECHNIQUE, CROSS-IMPACT GAMES, AND MORPHOLOGICAL ANALYSIS; (3) ANALYSIS AND EXTRAPOLATION OF STATISTICAL TIME SERIES; (4) PROJECTION OF ECOLOGICAL RELATIONS INTO THE FUTURE; AND (5) SIMULATION AND SCENARIO WRITING. THE FIRST ATTEMPT AT CRIMINOLOGICAL FORECASTING IN CANADA COMPUTED THE MONTHLY CONVICTION RATES FROM 1949 TO 1967 THEN EXTRAPOLATED LINEAR TRENDS TO PREDICT RATES FOR 1968-1971. A LATER FORECAST PREDICTED THE NUMBER OF ADULTS CHARGED WITH MAJOR OFFENSES BY CALCULATING THE AVERAGE PERCENTAGE CHANGE IN THIS NUMBER FROM 1968-1972 AND ESTIMATING THE NUMBER OF ADULTS CHARGED IN 1973 AND 1974. FOR THE PRESENT STUDY, THE NUMBER OF VARIOUS OFFENSES REPORTED TO THE POLICE IN CANADA FROM 1963-1972 WAS RELATED THROUGH MULTIPLE REGRESSION ANALYSIS TO A PROXY-VARIABLE (THE NATURAL LOGARITHM OF THE CURRENT YEAR MINUS 1960) AND TO THREE DEMOGRAPHIC PREDICTOR VARIABLES (TOTAL POPULATION, PROPORTION OF YOUNG ADULT MALES, AND URBAN POPULATION). THE RESULTING REGRESSION EQUATIONS WERE USED TO ESTIMATE THE CRIME STATISTICS FROM 1973-1976 IN ORDER TO EXAMINE THE PROCEDURE'S ACCURACY. FOR THOSE OFFENSES WHICH HAD BEEN ESTIMATED WITH LESS THAN A 20 PERCENT ERROR, PROJECTIONS WERE MADE BY APPLYING NEW REGRESSION EQUATIONS BASED ON THE 1963-1976 DATA TO DEMOGRAPHIC PROJECTIONS FOR 1977 TO 1982. THE RESULTS SHOW THAT THE OVERALL NUMBER OF OFFENSES KNOWN TO THE POLICE, AS WELL AS THE NUMBER OF HOMICIDES AND RAPES, WILL CONTINUE TO INCREASE, WHILE THE NUMBER OF PROPERTY OFFENSES WILL DECREASE. REFERENCES AND TABULAR DATA ARE PRESENTED. (DAG)