AFTER THE REPLICATION OF EHRILICH'S MURDER RATE FUNCTION (A REGRESSIVE EQUATION) IS DESCRIBED, THE MODEL IS TESTED FOR TEMPORAL HOMOGENEITY AND EXAMINED FOR SENSITIVITY TO CHANGES IN FUNCTIONAL FORM. THIS RE-EXAMINATION REVEALS THAT THE TIME SERIES MODEL AND DATA USED BY EHRLICH PERMIT NO INFERENCE WHATSOEVER ABOUT THE DETERRENT EFFECT OF CAPITAL PUNISHMENT. HIS FINDINGS ARE ALSO FOUND TO BE EXTREMELY SENSITIVE TO ESSENTIALLY ARBITRARY CHOICES OF MODEL SPECIFICATION AND TO THE PERIOD OVER WHICH THE MODEL IS ESTIMATED. MOREOVER, ESTIMATION OF A SINGLE STRUCTURAL EQUATION SHARPLY LIMITED THE POLICY IMPLICATIONS OF THE ESTIMATED COEFFICIENTS. THE APPENDIX LISTS THE DATA SOURCES USED FOR THIS RE-EXAMINATION. A BIBLIOGRAPHY IS ALSO INCLUDED. (AUTHOR ABSTRACT MODIFIED)