This summary of the Federal Justice Center's experimental rules for nonbinding arbitration offers general descriptions of the rules, the evaluation, the results, and the conclusions. The rules provide for mandatory arbitration of certain classes of civil actions. They set a time limit during which arbitration must occur. The arbitrators' judgment serves as the judgment of the court unless it is rejected by one or both parties within a certain period. The study reviews questionnaires and data from three districts that participated. Arbitration settled about 40 percent of the cases; rejection of the award and subsequent trial occurred in the other 60 percent. In this scenario, court-annexed arbitration can serve as an effective deadline for case preparation and also serve as a stimulus for settlement. The reduction in time and the savings in expense to the parties depend on proper administration of the program. Adherence to time limits is crucial, and corrective action must be taken when faced with delays. Evidence indicates satisfaction on the parts of most litigants and their counsel. Court-annexed arbitration substantially reduces the proportion of cases that ultimately go to trial. This study demonstrates that court-annexed arbitration has been a dramatic success in some places and has produced significant results in others. This procedure deserves wider usage because it is so efficient and effective. Tables, figures, and notes accompany the text. Appendixes include the text of the rules, the questionnaires used in evaluation, and an updated analysis.