By means of field experiments this study empirically tested two central rationales for pretrial drug testing: that information about drug use among incoming defendants derived from pre-bail drug testing would provide a powerful predictor of likely misconduct during pretrial release; and that drug testing would serve as an effective supervisory tool for minimizing drug use, failure-to-appear, and crime among drug-involved defendants granted conditional pretrial release. Findings show no statistically significant difference in the rates of rearrest between experimental (drug monitoring) and control group defendants in Prince George's County during 120 days of pretrial release. Although proportionately fewer Milwaukee defendants in the experimental group were rearrested during the 90-day pretrial follow-up period compared to the control group, the difference was not statistically significant. In sum, the drug monitoring programs in Milwaukee and Prince George's County did not produce lower rates of pretrial misconduct among program participants when compared to control groups. The assessment process highlighted a number of implementation difficulties in the drug- testing programs studied. These were in the areas of "normal" operational/logistical problems, voluntariness and the rate of testing of incoming defendants, the role of the "structure" of pretrial release decisionmaking and the lower priority of drug involvement, the level of noncompliance among defendants entering the program as a condition of pretrial release, and the difficulties involved in implementing a system of sanctions for noncompliance to be enforced by the courts. 5 tables