This study evaluated two self-administered relaxation manuals and a money deposit in the treatment of recurrent, nonphobic anxiety in a college population. Subjects were randomly assigned to a self-monitoring-only control group or one of four active treatment conditions. Subjects in active conditions received a progressive relaxation manual or a manual that called for the client to devise his or her own relaxation method and were assigned to deposit or nondeposit conditions. Improvement did not differ for the two relaxation procedures, but relaxation training groups improved significantly more than selfmonitoring-only subjects on both self-report questionnaires and self-monitored measures of anxiety. The money deposit did not produce greater amounts of relaxation practice or adherence to the program, although subjects in the money deposit condition did report being more relaxed in practice sessions and improved more on two pre-post measures of anxiety. Subjects' locus of control scores were significantly related to a number of practice, adherence, and outcome variables, but subjects' ratings of the likelihood that they would practice and benefit from the program proved to be as good predictors. The study suggests the value of self-monitoring and relaxation practice as treatment for recurrent, nonphobic anxiety.Although considerable progress has been made in our ability to treat phobias (cf. Bandura, 1969; K'azdin & Wilcoxon, 1976;Paul, 1969), we have not yet identified an effective strategy for treating nonphobic, anxious clients. Despite the large number of phobic treatment studies, there have been surprisingly few systematic investigations of how we might effectively assist clients who