The Adult Attachment Scale (AAS) (N. Collins and S. Read, 1996) and the Adolescent Relationship Questionnaire (ARQ) (E. Scharfe and K. Bartholomew, 1995) widely used self-assessment measures of attachment behavior. This study investigated the validity of these two measures by administering them concurrently to 117 introductory psychology college students participating in a larger study of the attachment consequences for children of employed and unemployed mothers. The AAS includes three subscales: Depend, Close, and Anxiety. The ARQ contains subscales of: Secure, Fearful, Preoccupied, and Dismissing. Intercorrelations among the seven subscales were calculated. Overall, inspection of the correlation matrix revealed high concurrent validity across several of the subscales. Scores on several were significantly associated, generally in a direction with consistent with predictions. Simultaneously there was evidence of suitable discriminant validity, as subscales within both the AAS and ARQ showed only modest association. There was a significant positive correlation between the ARQ Secure subscale and the AAS Close subscale, and there was a significant positive correlation between the Secure and the AAS Depend subscale. There was a significant negative correlation between Secure and the AAS Anxiety subscale. There were significant strong negative correlations between the ARQ Fearful subscale and the AAS Close and Depend subscales. These and other correlations show that, as predicted, the AAS and the ARQ parallel each other. The Fearful subscale had the strongest intercorrelations among all of the subscales, possibly because it taps a very negative attachment pattern with pervasive effects. (Contains 1 table and 10 references.) (SLD)