BARON AND RANSBERGER (1978) PRESENTED AN ANALYSIS OF THE RELATIONSHIP BETWEEN THE FREQUENCY OF MAJOR RIOTS AND THE AMBIENT TEMPERATURE DURING THE RIOTS. A TOTAL OF 102 MAJOR RIOTS OCCURRING IN THE UNITED STATES BETWEEN 1967 AND 1971 WERE EXAMINED. THE STUDY CONCLUDED THAT THE LIKELIHOOD OF A RIOT INCREASES WITH RISES IN TEMPERATURE UP TO THE RANGE OF 81-85 DEGREES FAHRENHEIT AND THEN DECREASES SHARPLY WITH FURTHER INCREASES IN TEMPERATURE. THE EVIDENCE TO SUPPORT THIS RELATIONSHIP IS A FREQUENCY DISTRIBUTION OF THE NUMBER OF RIOTS PLOTTED AGAINST TEMPERATURE. THE CRITIQUE CONTENDS THAT THE RELATIONSHIP BETWEEN AMBIENT TEMPERATURE AND RIOTS POSITED IN THE BARON AND RANSBERGER STUDY IS AN ARTIFACT OF THE PARTICULAR WAY THE DATA WERE EXAMINED AND THAT AN APPROPRIATE REANALYSIS SUGGESTS A MONOTONICALLY INCREASING FUNCTION RELATING THE PROBABILITY OF RIOTS AND TEMPERATURE. IT IS ARGUED THAT THE STUDY DID NOT TAKE ACCOUNT OF BASE-RATE DIFFERENCES IN TEMPERATURE. FOR EXAMPLE, IF DAYS IN THE 81 TO 85 DEGREE RANGE ARE MORE COMMON THAN DAYS IN THE 91-95 DEGREE RANGE, RIOTS CAN BE EXPECTED TO BE MORE FREQUENT IN THE FORMER RANGE BECAUSE OF THE GREATER NUMBER OF DAYS WITH THAT RANGE OF TEMPERATURE. APPROPRIATE ANALYSIS SHOWS THAT RIOTS ARE RELATIVELY MORE COMMON IN THE HIGHER TEMPERATURE RANGE, HOWEVER, WHEN THE FEWER DAYS IN THAT TEMPERATURE RANGE ARE TAKEN INTO ACCOUNT. GRAPHIC DATA AND REFERENCES ARE PROVIDED. (RCB)