In the summer following graduation a sample of 125 female college graduates (mean age = 28) completed Cohen & Wills' ISEL (1985) which includes scales measuring four social support functions: belonging (social companionship), appraisal (availability of confidants), tangible (instrumental), and self-esteem support. In the summer and fall subject status on two outcome scales was ascertained: the Psychophysiologic Symptom Scale and the Center for Epidemiologic Studies Depression Scale (CES-D). Reliability of the difference scores suggested that the ISEL scales do not measure entirely different constructs and the ISEL Self-esteem Scale is operationally redundant with the Rosenberg Self-esteem scale and the CES-D. Cross-sectional analyses indicated that the ISEL scales were related to symptoms. By contrast, standard longitudinal and prospective MLR analyses indicated that only the Belonging Scale was significantly related to future symptoms. The issues of confounding support with symptoms and the dimensionality of the subscales were discussed. The study suggests that specific functions of support take on greater importance during major life transitions and that any one supportive behaviour often serves multiple functions.