THE STUDY SUBJECTS, WHO WERE SELECTED BY LAWYER-ADMINISTERED VOIR DIRE EXAMINATION FROM 1,500 PERSONS CALLED FOR JURY DUTY AT THE QUEENS COUNTY, N.Y., SUPREME COURT, WERE DIVIDED INTO PANELS OF 4 DIFFERENT SIZE/DECISION RULE COMBINATIONS (23 PANELS FOR EACH COMBINATION). THE JURORS VIEWED A 3-HOUR VIDEOTAPED REENACTMENT OF A TRIAL, IN WHICH THE WORDS OF THE LAWYERS, JUDGES, AND WITNESSES WERE RETAINED. FURTHER REALISM WAS INTRODUCED BY HAVING THE JUDGE ON THE VIDEOTAPE (RATHER THAN THE EXPERIMENTER) GIVE THE JURIES THEIR INSTRUCTIONS. THE JURORS WERE ALLOWED AS MUCH TIME AS THEY REQUIRED TO DELIBERATE, AND CUSTOMARY PROCEDURES FOR HANDLING DEADLOCKED JURIES WERE FOLLOWED. THE ENTIRE EXPERIMENT WAS CARRIED OUT IN A COURTROOM SETTING. IN 11 (23.9 PERCENT) OF THE 6-MEMBER JURIES, THERE WAS A CONSENSUS AMONG 5 OR 6 JURORS BEFORE DELIBERATIONS BEGAN. NO 12-MEMBER JURY BEGAN DELIBERATIONS WITH A CONSENSUS AMONG 11 OR 12 MEMBERS. ONLY 2 (4.3 PERCENT) OF THE 12-MEMBER JURIES STARTED DELIBERATING WITH A CONSENSUS AMONG 10 MEMBERS. DELIBERATION TIMES WERE MUCH SHORTER FOR THE SIX-MEMBER JURIES IN WHICH FIVE OR SIX JURORS WERE ALIGNED PRIOR TO DELIBERATION. HUNG JURIES OCCURRED WITH EQUAL FREQUENCY IN 12-MEMBER UNANIMOUS, 12-MEMBER NONUNANIMOUS, AND 6-MEMBER UNANIMOUS JURIES. ONLY IN THE EXTREME TYPE OF JURY--THE SIX MEMBER NONUNANIMOUS--DID JURIES NEVER END IN HUNG DECISIONS. THERE WERE NO SIGNIFICANT DIFFERENCES IN THE NUMBER AND KIND OF VERDICTS AS A FUNCTION OF JURY SIZE AND TYPES OF DECISIONS. THE FINDINGS INDICATE THAT DIFFERENT DELIBERATION PROCESSES AND GREATER CHANCE FACTORS TAKE PLACE IN 6-MEMBER JURIES AS COMPARED TO 12-MEMBER JURIES. VERDICTS MAY NOT BE THE MOST APPROPRIATE MEASURES OF DIFFERENCES AMONG JURY CONFIGURATIONS AND DECISION RULES, AND UNCOVERING REAL BUT SMALL DIFFERENCES WOULD REQUIRE A LARGE NUMBER OF EXPERIMENTS IN ORDER TO ACHIEVE STATISTICAL SIGNIFICANCE. OTHER MEASURES, SUCH AS THOSE EXPLORED IN THIS STUDY, CAN ILLUSTRATE THE IMPACT OF CHANGES IN JURY SIZE AND DECISION RULE, THUS PROVIDING AN EMPIRICAL BASE FOR POLICYMAKING. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. (LKM)