Evidence concerning the contributions of social networks to the subjective well-being of older persons is inconsistent, reflecting the conceptual complexity of social networks and supports. In order to investigate the relative importance of different types of social ties and supports, the distinction between objective and subjective dimensions of social support, and sub-group variations in the implications of social support, older adults (N=1,185) participated in interviews which elicited information about social ties and supports as well as information regarding physical capacity. In addition, demographic data were gathered for all subjects. Data analyses indicated that respondents generally had access to extensive social ties and supports, preferring children more for instrumental support than for expressive support. Neighbors substituted as instrumental helpers while other kin served as confidants. Subjective network assessments had more substantial associations with well-being. Objective network characteristics showed weak associations with measures of well-being, with friends making the greatest contribution. However, there was considerable sub-group variation in the contributions of network characteristics to well-being. Proximate social ties were particularly valuable for the vulnerable elderly. These results suggest that the accessibility of social supports is an important dimension of "person/environment congruence." (Author/AG)