THE FIRST PART OF THE STUDY INVOLVED A 12-MONTH INVESTIGATION (JUNE 1, 1971 TO MAY 31, 1972) OF CRIME RATES FOR HOMICIDE, ATROCIOUS ASSAULT AND BATTERY, AND TOTAL CRIME REPORTED ON A DAILY BASIS. POSSIBLE ENVIRONMENTAL FACTORS PRECIPITATING CRIME WERE ASCERTAINED FROM INTERVIEWS WITH POLICE AND BY PERSONAL OBSERVATION. VARIABLES INCLUDED DAY OF THE WEEK, BAROMETRIC PRESSURE, PHASE OF THE MOON, DATES OF WELFARE AND SOCIAL SECURITY CHECKS, WEATHER CONDITIONS, AND CERTAIN AIR POLLUTION INDEXES. THE SECOND PART OF THE STUDY WAS CONCERNED WITH ONLY HOMICIDE RATES OVER THE 15-YEAR PERIOD FROM JANUARY 1, 1957 TO DECEMBER 31, 1971. BECAUSE OF THE NATURE OF HOMICIDE RECORDKEEPING BY POLICE AND FOR THE SAKE OF SIMPLICITY, NO ATTEMPT WAS MADE TO SORT OUT MULTIPLE HOMICIDES, ARSON VICTIMS, VICTIMS OF POLICE SHOOTINGS, OR OTHER CATEGORIES. THE FINDINGS SUGGESTED THAT SOME AIR POLLUTANT FACTORS MAY AFFECT CRIMINAL ACTIVITY. PART OF THIS APPARENT CORRELATION IS PROBABLY RELATED TO DAY OF THE WEEK, WITH MORE POLLUTION ON WEEKDAYS DUE TO INDUSTRIAL AND TRANSPORTATION ACTIVITIES. THERE WAS A SLIGHT INCREASE IN HOMICIDES ON THE FIRST DAY AFTER THE FULL MOON. TOTAL CRIMES WERE UP ON WELFARE CHECK DAYS BUT NOT ON SOCIAL SECURITY CHECK DAYS. TOTAL CRIMES WERE LOWER ON SATURDAYS. ALL CRIMES AND ASSAULTS DECREASED MARKEDLY BETWEEN DECEMBER AND JANUARY. ASSAULTS DROPPED SHARPLY AFTER THE FIRST OF THE YEAR, BELIEVED TO BE THE RESULT OF THE END OF HOLIDAY SHOPPING AND THE ONSET OF INCLEMENT WEATHER. TEMPERATURE VARIABLES WERE IMPORTANT PROGNOSTICATORS OF TOTAL CRIME AND ASSAULT RATES, AND THERE WAS A HIGH DEGREE OF PARALLELISM BETWEEN OXIDANT LEVELS IN THE AIR AND ASSAULT RATES. SUPPORTING DATA AND REFERENCES ARE PROVIDED. (DEP)