THE ERRONEOUS ASSUMPTIONS ARE SAID TO WARRANT REJECTION OF THESE RESEARCHERS' CONCLUSION THAT CAPITAL PUNISHMENT REDUCES HOMICIDE MORE EFFECTIVELY THAN THE ALTERNATIVE PENALTY OF IMPRISONMENT. THE ASSUMPTIONS CRITIZED RELATE TO TIME SAMPLING AND DATA SELECTION, POPULATION SAMPLING, SINGLE-VALUED RATIONALITY (THE CONCEPT OF A RATIONAL PERSON ACTING TO MAXIMIZE PROFITS), ONE-WAY CAUSATION (THE ASSERTION THAT PUNISHMENT IS AN INDEPENDENT VARIABLE AND CRIME IS ITS INVERSELY RELATED DEPENDENT VARIABLE), AND THE FEASIBILITY OF HIGHLY FAIR AND CERTAIN DEATH PENALTIES. TO OBTAIN FINDINGS SUPPORTIVE OF CAPITAL PUNISHMENT, EHRLICH AND YUNKER ARE SAID TO HAVE CHOSEN DATA, TIME SAMPLING, AND METHODOLOGY THAT COULD REASONABLY HAVE BEEN DIFFERENT. ALTERNATIVE CHOICES WOULD HAVE YIELDED CONTRASTING RESULTS. MARKED REGIONAL AND URBAN-RURAL VARIATIONS IN HOMICIDE THAT WERE IGNORED IN THE ANALYSES ARE CITED, AS ARE UNIQUE HISTORICAL DEVELOPMENTS IN 1965-70 THAT MITIGATE AGAINST GENERALIZING ABOUT CAPITAL PUNISHMENT FROM NATIONAL HOMICIDE TRENDS DURING THAT PERIOD. FACTORS IN CRIMES AGAINST PERSONS THAT DO NOT FIT AN ECONOMIC MODEL ARE POINTED OUT. EVIDENCE THAT ABOLITION OF CAPITAL PUNISHMENT ACTUALLY EXPRESSES GREATER OUTRAGE AT KILLING THAN DOES RETENTION OF THE DEATH PENALTY IS OFFERED. THE IMPACT OF UNOFFICIAL PROCEDURES (E.G., PLEA BARGAINING) AND POLICIES ON THE IMPLEMENTATION OF DEATH PENALTY STATUTES IS ALSO CONSIDERED. IT IS CONCLUDED THAT DEATH PENALTY STATUTES PROBABLY WILL NOT DECREASE HOMICIDE RATES AND MAY EVEN FOSTER VALUE SYSTEMS THAT JUSTIFY VIOLENCE. REFERENCES AND SUPPORTING DATA ARE INCLUDED. (LKM)