Data on matched triads of 428 biological mothers, their husbands, and a young adult child interviewed in the second wave of the National Survey of Families and Households are used to examine affective relationships from the perspective of both parent and child. The analysis examines the ways each dyadic relationship depends on relationships with the third member of the triad and whether these processes operate differently for mothers and fathers, for fathers and stepfathers, for daughters and sons. Results show that parents' affect is related significantly to marital quality and the partner's relationship with the child. Children's affect for mothers and for fathers is related to their feelings toward the other parent but not to their parents' marital quality. An important exception is that the child's warmth toward the father is related to the mother's marital warmth. Overall, gender effects are minor, and mothers and fathers are embedded similarly in the family affective system. Stepfathers and stepchildren report significantly lower levels of warmth toward one another than biological fatherchild dyads, but these relationships are not more dependent on the mother than in biological families. The mother's relationship with the child is, however, more independent of her partner when that partner is not the child's biological father Key Words: affect, fathers, marriage quality, mothers, parentchild relationships. Any dyadic relationship in the family can be understood fully only if we understand the family system in which it takes place. For the most part, however, we have lacked the data, the conceptual tools, and the statistical sophistication to implement this insight (Emery & Tuer, 1993; Sabatelli & Bartle, 1995). Too much of our research about families is really about dyads, and, despite important improvements, too much of our research about parents is really about mothers or is based on mothers' reports (Marsiglio, 1993). As a result, our understanding of the interdependencies among mothers, fathers, and children is still rudimentary. Yet, such an understanding is central to many of the questions that drive current research and policy concerns. Are fathers peripheral to the family? Are they dependent, even in married biological families, on their children's mothers to mediate between them and their children? How does marital quality affect mothers' and fathers' relationships with children? Are children's relationships with fathers, even in two-parent biological families, more contingent than their relationships with mothers? Answers to these questions would not only help us understand general family dynamics but would establish an essential baseline against which to assess postdivorce and noncustodial relationships between parents and children. I develop a framework for thinking about the interdependence of family affective relationships and assess it using independent reports on relationship quality from 428 matched triads of mothers, fathers, and a randomly selected young adult child. In addition to establishing gender and generational differences in the family system, the analysis explores whether relationships in stepfamilies are more or less independent than in biological families. BACKGROUND Conceptual Framework The general framework for this study of family relationships is provided by systems theory. This perspective conceives of the family as an interacting set of subsystems (dyads, triads, and so forth), each affecting and being affected by the other subsystems, as well as by systemwide properties (e.g., world views). Although relationships among all subsystems are interdependent and reciprocal, a central assumption of the theory is that subsystems are arranged in a hierarchy of power and influence (Whitchurch & Constantine, 1993). In healthy families, the marital subsystem is assumed to have greater impact on parent-child systems than the parent-child system has on the marriage (Emery, 1988). …