The data used in this study were collected as part of a national evaluation of Big Brothers/Big Sisters of America. The evaluation encompassed 1,138 youths who ranged in age from 10 through 16. The youths were randomly assigned to either a mentoring relationship (n=571) or a control group (n=567), both of which were monitored for 18 months. Within the experimental group, 109 youths were not matched during the 18-month period. Analyses were thus conducted with data from the 378 experimental-group youth who were matched with a volunteer mentor and the 472 control group youths. Based on the matched youths' accounts of the mentoring relationship, four distinct types of relationships were identified. The "moderate" group participated in moderate levels of structured conversations around goals and were involved in slightly fewer activities with their mentors than were other groups. The "unconditionally supportive" group reported moderate levels of structure and activities and perceived higher levels of support than did the "moderate" group. The "active" group, on the other hand, reported the lowest levels of structure with their mentors but the highest levels of participation in activities. The "low-key" group ranked relatively high on levels of support and reported moderate structure, but participated in relatively few activities. Relative to the control youths, youths who characterized their mentoring relationships as providing moderate levels of activity, structure, and conditional support derived the largest number of benefits from the relationships. These benefits included decreased alienation from parents, decreased conflict and inequality with friends, and an improved sense of self-worth and school competence. This suggests that a balanced approach to mentoring that involves moderate amounts of structure, activities, and support produces the most benefits. 5 tables, 2 figures, and 43 references